Power Quality Management of Interconnected Microgrids using Model Predictive Control

IFAC PapersOnLine 53-2 (2020) 12918–12923 ScienceDirect Available online at www.sciencedirect.com 2405-8963 Copyright © 2020 The Authors. This is an open access article under the CC BY-NC-ND license . Peer review under responsibility of International Federation of Automatic Control. 10.1016/j.ifacol.2020.12.2121 10.1016/j.ifacol.2020.12.2121 2405-8963 Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗ Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: felix.gar[email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: ser[email protected]) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: bor[email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: pedro.roncer[email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗ S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Copyright © 2020 The Authors. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0 ) F. Garcia-Torres et al. / IFAC PapersOnLine 53-2 (2020) 12918–12923 12919 Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Power Quality Management of Interconnected Microgrids using Model Predictive Control  F. Garcia-Torres ∗S. Vazquez ∗∗ C. Bordons ∗∗∗ I. Moreno-Garcia ∗∗∗∗ A. Gil ∗∗∗∗ P. Roncero-Sanchez † ∗Centro Nacional del Hidrogeno, Puertollano (Ciudad Real), Spain, (e-mail: [email protected]). ∗∗ Electronic Engineering Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: serg[email protected]s) ∗∗∗ Systems Engineering and Automatic Control Department, Universidad de Sevilla, Sevilla, Spain, (e-mail: [email protected]) ∗∗∗∗ Department of Electronic and Computer Engineering, Universidad de Cordoba, Cordoba, Spain, (e-mail: [email protected]) †Systems Engineering and Automatic Control Department, University of Castilla-La Mancha, Ciudad Real, Spain, (e-mail: [email protected]) Abstract: In this paper, the power quality of interconnected microgrids is managed using a Model Predictive Control (MPC) methodology which manipulates the power converters of the microgrids in order to achieve the requirements. The control algorithm is developed for the microgrids working modes: grid-connected, islanded and interconnected. The results and simulations are also applied to the transition between the different working modes. In order to show the potential of the control algorithm a comparison study is carried out with classical Proportional-Integral Pulse Width Modulation (PI-PWM) based controllers. The proposed control algorithm not only improves the transient response in comparison with classical methods but also shows an optimal behavior in all the working modes, minimizing the harmonics content in current and voltage even with the presence of non-balanced and non-harmonic-free threephase voltage and current systems. Keywords: Interconnected systems, Harmonics, Predictive Control, Power System Control. 1. INTRODUCTION Microgrids can be seen as a key technology to improve power quality and reliability (PQR). Their ability to work in grid-connected or islanded mode is specially adequate to supply electricity to sensitive loads. An introduction to the main problems and solutions of power quality in microgrids can be found in Guerrero et al. (2012). As can be seen in the aforementioned paper, most of the solutions for power electronics devices in microgrids are based on PI-PWM controllers which present slow transient response. Microgrids are not only electrical power systems essentially based on renewable energy systems but also they usually content sensitive loads to PQR losses. In environments with sensitive loads, fast response against problems related with PQR is required. MPC controllers are based on future behavior of the system, achieving fast dynamic response improving in this way the transient response of PI-PWM controllers, see Vazquez et al. (2016). This work has been partially supported by the Ministry of Economy and Competitiveness of Spain with the financial support under grant DPI2016-78338-R (Project CONFIGURA) and partially supported by Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) Yazdi and Hosseinian (2019) introduce a novel Smart Branch to compensate power quality disturbances using a finite control set-model predictive controller (FCS-MPC). Jayachandran and Ravi (2019) present a decentralized model predictive hierarchical control strategy for islanded AC microgrids. An MPC controller is applied to the voltage control of an islanded microgrid in Garcia-Torres et al. (2015) for the case of non-linear loads. This method is expanded in Bordons et al. (2020) developing an MPC methodology to the cases of microgrids with non-linear and unbalanced loads in both grid-connected and islanded mode. The interconnection of microgrids can be considered as a way to increase the robustness in power supply overall when these microgrids have to work islanded from the main grid. This mode of operation of microgrids has hardly been studied to date. In this paper, the work presented in Bordons et al. (2020) is expanded to be used in the case of interconnected microgrids working under a blackout of the main grid. The topology of the interconnected microgrids object of this paper is shown in Fig. 1. As can be seen in Fig. 1, there are three intelligent power switches (IPS) installed to isolate or connect the working mode of each microgrid with the main grid and/or with the neighbor microgrid. Copyright © 2020 The Authors. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-nc-nd/4.0 ) S2,a N N S1,a S2,b S1,b S2,c S1,c S2,n S1,n  N            MICROGRID A GRID IPS-A       N         POWER INVERTER S2,a N N S1,a S2,b S1,b S2,c S1,c S2,n S1,n             MICROGRID B IPS-B       N         POWER INVERTER IPS GRID PCC PCC Fig. 1. Interconnected microgrids object of study Fourier’s Transform Zth (k) ,  ,  ,  ,  ),,,(min 1111 ncba SSSSfJ = 15 15 Park’s Transform PLANT MODEL 1/20 MPC Applied to the Different Control Levels of Renewable Energy Microgrids with Multiple Energy Storage System      , 1  , 1 , 1 U,  Islanded/ Grid-connected Fig. 2. Block Diagram 2. CONTROLLER DESIGN In AC microgrids, the final power quality obtained in the microgrid depends on the exchanged power flow between its local devices connected and the grid. An appropriate energy storage system (ESS) connected to a voltage source inverter (VSI) can be used to enhance the power quality of the microgrid. The final result not only depends on the topology of the VSI but also on its control system. With the aim to obtain an optimal power flow in the microgrid, a four wire VSI with active neutral control is selected in order to integrate unbalanced and non-linear loads. The VSI response is improved using an innovative MPC controller to manage the power quality of the microgrids and their power exchange with the main grid or with the neighbor microgrids. The block diagram of the controller is exposed in Fig. 2. For the Park’s transformation, it is considered that the d-axe is aligned with the voltage reference. 2.1 Predictive Model The first step of the controller design is to calculate the equivalent output Thevenin’s impedance at each sample instant Tsfor every phase Zth out,α(k)|α=a,b,c for the fundamental frequency, using the phasors for output voltages Uout,α and currents Iout,α: Zth out,α(k)= Uout,α(k) Iout,α(k)−Igrid,α(k) =Rth out,α(k)+jXth out,α(k) (1) An equivalent inductance or capacitance can be obtained. When sign(Xth out,α(k)) =sign(Rth out,α(k)) using relationship (2) and using the expression given in (3) when sign(Xth out,α(k)) =sign(Rth out,α(k)): Lth out,α(k)=Xth out,α(k) 2πf ;Cth out,α(k)=0 (2) Lth out,α(k)=0; Cth out,α(k)=−Xth out,α(k)·2πf (3) The predictive model of the inverter as function of the switching vector s(k)=[S1a(k)S1b(k)S1c(k)S1n(k)]T can be obtained with the expressions (4)-(6). vout,α(k+ 1) = Vdc ·(S1α(k+ 1) −S1n(k+ 1)) +TsRLN+LN Ts·iLN(k+ 1) −LN iLN(k) Ts −TsRLfα +Lfα Ts·iLfα(k+1)+Lfα iLfα(k) Tsα=a,b,c (4) iLf,α(k+ 1) = Cf Ts+CfRf vout,α(k+ 1) Ts TsRth,µgrid α(k)+Lth,µgrid α(k)vout,α(k+ 1) +Lth,µgrid α(k) TsRth,µgrid α(k)+Lth,µgrid α(k)·iout,α(k) −Cf Ts+CfRf vout,α(k)+ CfRf Ts+CfRf iCfα(k)α=a,b,c (5) 12920 F. Garcia-Torres et al. / IFAC PapersOnLine 53-2 (2020) 12918–12923 iLN(k+ 1) = 1−2C+ TsTsRLN+LN Ts−1 ·   α=a,b,c Cf,α Ts vCf,α(k+1)+C+ Ts Vdc + α=a,b,c Ts TsRth,µgrid α(k)+Lth,µgrid α(k)vout,α(k+ 1) −C+ Ts (vC+(k)+vC−(k)) − α=a,b,c Cf,α Ts vCf,α(k) Lth,µgrid α(k) TsRth,µgrid α(k)+Lth,µgrid α(k)·iout,α(k) −2C+ TsLN iLN(k) Ts +Vdc ·S1n(k+ 1) (6) In grid-connected mode the following equation has to be added: igrid,α(k+ 1) = Lgridigrid,α(k)+ Ts RgridTs+Lgrid (vgrid,α(k+ 1) −vout,α(k+ 1))α=a,b,c (7) 2.2 Cost Function for the Islanded Mode In this working mode the inverter object of this study has to manage the voltage waveform with respect to magnitude, frequency, harmonics content and phase equilibrium. In order to achieve these criteria, the cost function expressed in (8) is divided into three main parts: Jwave isl which manages the waveform of the output voltage, Jharm isl which minimizes the harmonics content and Jbal isl which controls the balance between phases. In order to use the predictive model of the inverter, the assumption that between two sample instants Zth out,α(k+ 1) = Zth out,α(k) has to be used in the predictive model of the inverter. min s(k)Jisl(k)=min s(k)Jwave isl (k)+Jharm isl (k)+Jbal isl (k)(8) Jwave isl (k)=  α=a,b,c winst isl vout,α(k+ 1) −vref out,α(k+ 1)2 +wcycle isl,α e(Uout,α(k+ 1)) −e(Uref out,α(k+ 1))2 +wcycle isl,α m(Uout,α(k+ 1)) −m(Uref out,α(k+ 1))2 (9) At each sample instant, the voltage reference is calculated and imposed in the first term of (9) minimizing the difference between the predicted voltage and the calculated reference. In order to minimize the steady-state error the second term of (9) is added, correcting this error with the complete fundamental cycle computation. Jharm isl (k)=  α=a,b,c wv isl,α (∆vout,α(k+ 1))2 +wi isl,α (∆iout,α(k+ 1))2 +wcap isl (vC+(k+ 1) −vC−(k+ 1))2 (10) The first and second term of (10) minimize the voltage and current abrupt variations between two sample instants avoiding the harmonics content in both voltage and current. The third term manages the balance of voltage for the neutral point. Jbal isl (k)= β=b,c,a  α=a,b,c wbal isl (|Uout,α(k+ 1))|−|Uout,β (k+ 1)|)2 (11) When unbalanced loads are connected to the inverter the obtained voltage magnitude can take different values for each phase . In order to control these situations the term expressed in (11) is included in the cost function. 2.3 Cost Function for the Grid-Connected Mode In grid connected mode, it is assumed that due to the fact that the voltage reference is imposed by the main grid. Under the assumption of robustness in the voltage waveform provided by the main grid and considering that Park’s transformation is a rotational reference frame, it is considered that between two sample instants the dqovoltage is constant. Under this assumption and using the predictive model can be obtained the output currents iout,γ(k+ 1) igrid,γ (k+ 1) of each phase. The controller receives the set-point for the exchange of active and reactive powers with the main grid. Due to the fact that Uref grid,α and ϕref grid,α are imposed by the main grid and supposed constant between two sample instants , it can easily obtained the reference current Iref grid,α with the following equations: Pref grid,α(k)=|Uref grid,α(k)||Iref grid,α(k)| 2cos(ϕref grid,α(k)) (12) Qref grid,α(k)=|Uref grid,α(k)||Iref grid,α(k)| 2sin(ϕref grid,α(k)) (13) The current references are calculated as follows: iref grid,α(k+ 1) = |Iref grid,α(k+ 1)|sin(2πf(k+1+Dα)+ϕref grid,α(k+ 1)) (14) A digital delay Dαhas to be included which is adaptive with Zth out,α(k). As done for the case of islanded mode, the cost function in grid-connected mode is divided into three parts: min s(k)Jconn(k)=min s(k)Jwave conn (k)+Jharm conn (k)+Jbal conn(k) (15) Jwave conn = α=a,b,c winst conn igrid,α(k+ 1) −iref grid,α(k+ 1)2 +wcycle conn,α e(Igrid,α(k+ 1)) −e(Iref grid,α(k+ 1))2 +wcycle conn,α m(Igrid,α(k+ 1)) −m(Iref grid,α(k+ 1))2 (16) The procedure to formulate (16) is similar to the one carried out for (9). At each sample instant, the current reference is calculated and imposed in the first term of (16), minimizing the difference between the predicted current exchange with the main grid and the reference F. Garcia-Torres et al. / IFAC PapersOnLine 53-2 (2020) 12918–12923 12921 iLN(k+ 1) = 1−2C+ TsTsRLN+LN Ts−1 ·   α=a,b,c Cf,α Ts vCf,α(k+1)+C+ Ts Vdc + α=a,b,c Ts TsRth,µgrid α(k)+Lth,µgrid α(k)vout,α(k+ 1) −C+ Ts (vC+(k)+vC−(k)) − α=a,b,c Cf,α Ts vCf,α(k) Lth,µgrid α(k) TsRth,µgrid α(k)+Lth,µgrid α(k)·iout,α(k) −2C+ TsLN iLN(k) Ts +Vdc ·S1n(k+ 1) (6) In grid-connected mode the following equation has to be added: igrid,α(k+ 1) = Lgridigrid,α(k)+ Ts RgridTs+Lgrid (vgrid,α(k+ 1) −vout,α(k+ 1))α=a,b,c (7) 2.2 Cost Function for the Islanded Mode In this working mode the inverter object of this study has to manage the voltage waveform with respect to magnitude, frequency, harmonics content and phase equilibrium. In order to achieve these criteria, the cost function expressed in (8) is divided into three main parts: Jwave isl which manages the waveform of the output voltage, Jharm isl which minimizes the harmonics content and Jbal isl which controls the balance between phases. In order to use the predictive model of the inverter, the assumption that between two sample instants Zth out,α(k+ 1) = Zth out,α(k) has to be used in the predictive model of the inverter. min s(k)Jisl(k)=min s(k)Jwave isl (k)+Jharm isl (k)+Jbal isl (k)(8) Jwave isl (k)=  α=a,b,c winst isl vout,α(k+ 1) −vref out,α(k+ 1)2 +wcycle isl,α e(Uout,α(k+ 1)) −e(Uref out,α(k+ 1))2 +wcycle isl,α m(Uout,α(k+ 1)) −m(Uref out,α(k+ 1))2 (9) At each sample instant, the voltage reference is calculated and imposed in the first term of (9) minimizing the difference between the predicted voltage and the calculated reference. In order to minimize the steady-state error the second term of (9) is added, correcting this error with the complete fundamental cycle computation. Jharm isl (k)=  α=a,b,c wv isl,α (∆vout,α(k+ 1))2 +wi isl,α (∆iout,α(k+ 1))2 +wcap isl (vC+(k+ 1) −vC−(k+ 1))2 (10) The first and second term of (10) minimize the voltage and current abrupt variations between two sample instants avoiding the harmonics content in both voltage and current. The third term manages the balance of voltage for the neutral point. Jbal isl (k)= β=b,c,a  α=a,b,c wbal isl (|Uout,α(k+ 1))|−|Uout,β (k+ 1)|)2 (11) When unbalanced loads are connected to the inverter the obtained voltage magnitude can take different values for each phase . In order to control these situations the term expressed in (11) is included in the cost function. 2.3 Cost Function for the Grid-Connected Mode In grid connected mode, it is assumed that due to the fact that the voltage reference is imposed by the main grid. Under the assumption of robustness in the voltage waveform provided by the main grid and considering that Park’s transformation is a rotational reference frame, it is considered that between two sample instants the dqovoltage is constant. Under this assumption and using the predictive model can be obtained the output currents iout,γ(k+ 1) igrid,γ (k+ 1) of each phase. The controller receives the set-point for the exchange of active and reactive powers with the main grid. Due to the fact that Uref grid,α and ϕref grid,α are imposed by the main grid and supposed constant between two sample instants , it can easily obtained the reference current Iref grid,α with the following equations: Pref grid,α(k)=|Uref grid,α(k)||Iref grid,α(k)| 2cos(ϕref grid,α(k)) (12) Qref grid,α(k)=|Uref grid,α(k)||Iref grid,α(k)| 2sin(ϕref grid,α(k)) (13) The current references are calculated as follows: iref grid,α(k+ 1) = |Iref grid,α(k+ 1)|sin(2πf(k+1+Dα)+ϕref grid,α(k+ 1)) (14) A digital delay Dαhas to be included which is adaptive with Zth out,α(k). As done for the case of islanded mode, the cost function in grid-connected mode is divided into three parts: min s(k)Jconn(k)=min s(k)Jwave conn (k)+Jharm conn (k)+Jbal conn(k) (15) Jwave conn = α=a,b,c winst conn igrid,α(k+ 1) −iref grid,α(k+ 1)2 +wcycle conn,α e(Igrid,α(k+ 1)) −e(Iref grid,α(k+ 1))2 +wcycle conn,α m(Igrid,α(k+ 1)) −m(Iref grid,α(k+ 1))2 (16) The procedure to formulate (16) is similar to the one carried out for (9). At each sample instant, the current reference is calculated and imposed in the first term of (16), minimizing the difference between the predicted current exchange with the main grid and the reference calculated. In order to minimize the steady state error the second term of (16) is added correcting this error with the complete fundamental cycle calculus done for the current exchange with the main grid expressed in Fourier’s domain. Jharm conn (k)=  α=a,b,c wv conn,α (∆vout,α(k+ 1))2 +wi conn,α (∆igrid,α(k+ 1))2 +wcap isl (vC+(tk+1)−vC−(tk+1))2 (17) The second part of the cost function in grid-connected mode (17) minimizes the harmonic injection in current to the grid, as well as the voltage variations in the microgrid. It also balances the neutral point of the inverter. Finally, when unbalanced loads are connected to the microgrid they can affect to balance in the active and reactive power injected to main grid. For this purpose the term of the cost function expressed in (18) is included. Jbal conn(k)= β=b,c,a  α=a,b,c wbal conn(Pgrid,α(k+ 1)) −Pgrid,β (k+ 1)))2 + β=b,c,a  α=a,b,c wbal conn(Qgrid,α(k+ 1)) −Qgrid,β (k+ 1)))2 (18) 2.4 Cost Function for the Interconnected Mode The interconnected mode can be considered as a hybrid mode between the connected and the islanded mode since there does not exist a main grid who imposes the references in voltage and frequency but there can be energy exchange between the interconnected microgrids. Due to the fact that there is not a main grid both microgrids have to work controlling the voltage and the frequency, the socalled multi-master mode. min s(k)J(X) inter(k)=min s(k)J(X),wave isl (k)+J(X),harm isl (k) +J(X),bal isl (k)+  γ=a,b,c (i(X)→(Y) γ(k))2  (19) The notation (X) refers to the microgrids (A) and (B) and the terminology (X)→(Y) makes reference to the exchange between the microgrid (X) and the microgrid (Y), being iexch (X)→(Y)the exchanged current between the microgrid (X) and the microgrid (Y). Notice that this term achieves to syncronize in frequency both microgrids and also to equilibrate the voltage magnitude between both microgrids without being necessary any kind of communication between the interconnected microgrids. 3. SIMULATION RESULTS The simulations are carried out using Simpower using T=1µs as sample period while the controller acts each T= 20µs. The different values for the simulation and power inverter components are exposed in Table 1. Table 1. Components value Parameter Value Filter inductance Lf1 mH Filter inductance resistance RLf0.1 Ω Filter capacitor Cf0.5 mF Filter capacitor resistance RCf0.1 Ω DC link voltage Udc 950 V Neutral inductance LN2.5 µF Neutral inductance resistance RLN0.1 Ω Neutral balancing capacitors C+,C −6600 µF Grid connection line inductance Lgrid 0.1 mH Grid connection line resistance Rgrid 0.1 Ω Slave inverter line inductance Linv 0.1 mH Slave inverter line resistance Rinv 0.1 Ω Non-linear load line inductance Lnon 0.1 mH Non-linear load line resistance RLnon 0.1 Ω Non-linear load dc resistance Rnon 60 Ω Non-linear load dc capacitor Cnon 6.6 mF Unbalanced load phase a resistance Ra1 MΩ Unbalanced load phase b resistance Rb10 Ω Unbalanced load phase c resistance Rc10 Ω Unbalanced load phase b inductance Lb1 mH Unbalanced load phase c capacitor Cc0.1 mF 3.1 Comparative between controllers The first simulation is used to compare the results in both grid-connected and islanded mode, as well as the transition between modes using an MPC-controller and a PI-PWM-controller for a single microgrid working in both modes: grid-connected and islanded. In this simulation the non-linear and the unbalanced loads are connected to the microgrid in all the sample instants. Both controllers receive the next references for the power exchange with the main grid: [Pref grid,α,Q ref grid,α]=[−15000 W,−9000 Var]∀t≤0.5s [Pref grid,α,Q ref grid,α] = [15000 W,9000 Var]∀t≥0.5s Between t∈[1s, 1.5s] a fault in the main grid occurs so the transition to islanded mode is required, restoring the connection of the microgrid with the main grid for t>1.5s. The comparison between the results obtained in the reference tracking for the active and reactive power between the MPC and the PI controller can be found in Fig. 3. As can been seen in the figure, the PI controller presents a longer transient response while the MPC controller reaches the given references in just two cycles of the fundamental frequency. In Fig. 4, the comparison between the THD results for the MPC and PI controller are exposed. As can be seen, despite the presence of nonlinear and unbalanced loads the current waveforms present a low content of harmonics in the MPC-controller while the PI controller is not able to minimize the harmonic content in the current waveform. During the instants t= 1 s and t=1.5 s, a grid blackout occurs and the power inverter works in islanded mode. The comparison between the behavior of the power inverter with the MPC and the PI-PWM controllers can be seen in Fig. 5 and Fig. 6 where the voltage magnitude and the phase values are shown. As it occurs for the case of gridconnected mode, a better transient response is obtained in the case of the MPC controller. A better response is also obtained for THD values of the voltage at the Point 12922 F. Garcia-Torres et al. / IFAC PapersOnLine 53-2 (2020) 12918–12923 of Common Coupling (PCC) in the case of the MPC controller as can be seen in Fig. 7. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 −2 −1 0 1 2x 104 Time (s) Active/Reactive Power (W/Var) Pa(MPC) Pa(PI−PWM) Qa(MPC) Qa(PI−PWM) Fig. 3. Comparison of the results for the active and reactive power exchange with the main grid between the MPC and PI-PWM controllers for phase a 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 10 20 30 40 Time(s) THD (%) THD(Ia) MPC THD(Ia) PI−PWM THD(Ib) MPC THD(Ib) PI−PWM THD(Ic) MPC THD(Ic) PI−PWM Fig. 4. Comparison of the THD values for the current exchange with the main grid between the MPC and PI-PWM Controllers 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 215 220 225 230 235 240 245 Time (s) Voltage Magnitude (Vrms) Mag Ua (MPC) Mag Ua (PI−PWM) Fig. 5. Voltage Magnitude for phase A at the PCC 1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 1.4 1.45 1.5 −100 −50 0 50 100 Time (s) Voltage Phase (º) MPC PI−PWM Fig. 6. Absolute voltage phase angle value of the voltages at the PCC during the blackout of the main grid 3.2 Power Quality Management Results for interconnected microgrids working without presence of grid The aim of the second simulation launched is to evaluate the behavior of the presented controller for the case of interconnected microgrids working under a grid blackout. 1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 1.4 1.45 1.5 0 2 4 6 8 10 Time(s) THD (%) MPC Phase A PI−PWM Phase A MPC Phase C PI−PWM Phase C MPC Phase C PI−PWM Phase C Fig. 7. THD values for the voltages at the PCC during the blackout of the main grid In this case, the IPS-A and IPS-B are connected and IPS-grid is disconnected (see Fig. 1). In the case of the microgrid (A) the non-linear loads are connected during all the sample instants of the simulation and the unbalanced loads are connected for these sample instants t>0.1s. In the microgrid (B) the unbalanced loads are connected during all the sample instants and the non-linear loads are connected at t>0.1s. In Fig. 8 and Fig. 9 a comparison between the obtained results for the voltage magnitudes for every phase of each microgrid are shown. The current consumption can be observed in Fig.12. As can be seen, for the sample instants t∈[0.10,0.12] in Fig. 9 a more robust behavior is obtained in the case of working interconnected where the voltage magnitudes of each microgrids are always |U(X) out,γ|>200 for both microgrids. In Fig.13 the obtained results for the current exchange between both microgrids are shown. As can be seen, each microgrid manages its own loads without nearly non-affection to the neighbor microgrid. As can be seen in Fig. 10 and Fig. 11, the presence of non-linear and unbalanced loads and the changes in current demand at each microgrid, as well as the interaction between microgrids do not affect to the THD content in voltage or to the balance between phases guaranteeing the power quality supply to the loads connected to both microgrids. 0 0.05 0.1 0.15 0.2 0.25 0.3 180 200 220 Voltage Magnitude (Vrms) (A)-a (A)-b (A)-c (B)-a (B)-b (B)-c 0 0.05 0.1 0.15 0.2 0.25 0.3 Time (s) Fig. 8. Voltage Magnitude per phase and microgrid in mode non-interconnected and grid-islanded Fig. 9. Voltage Magnitude per phase and microgrid in mode interconnected and grid-islanded F. Garcia-Torres et al. / IFAC PapersOnLine 53-2 (2020) 12918–12923 12923 of Common Coupling (PCC) in the case of the MPC controller as can be seen in Fig. 7. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 −2 −1 0 1 2x 104 Time (s) Active/Reactive Power (W/Var) Pa(MPC) Pa(PI−PWM) Qa(MPC) Qa(PI−PWM) Fig. 3. Comparison of the results for the active and reactive power exchange with the main grid between the MPC and PI-PWM controllers for phase a 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 0 10 20 30 40 Time(s) THD (%) THD(Ia) MPC THD(Ia) PI−PWM THD(Ib) MPC THD(Ib) PI−PWM THD(Ic) MPC THD(Ic) PI−PWM Fig. 4. Comparison of the THD values for the current exchange with the main grid between the MPC and PI-PWM Controllers 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 215 220 225 230 235 240 245 Time (s) Voltage Magnitude (Vrms) Mag Ua (MPC) Mag Ua (PI−PWM) Fig. 5. Voltage Magnitude for phase A at the PCC 1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 1.4 1.45 1.5 −100 −50 0 50 100 Time (s) Voltage Phase (º) MPC PI−PWM Fig. 6. Absolute voltage phase angle value of the voltages at the PCC during the blackout of the main grid 3.2 Power Quality Management Results for interconnected microgrids working without presence of grid The aim of the second simulation launched is to evaluate the behavior of the presented controller for the case of interconnected microgrids working under a grid blackout. 1 1.05 1.1 1.15 1.2 1.25 1.3 1.35 1.4 1.45 1.5 0 2 4 6 8 10 Time(s) THD (%) MPC Phase A PI−PWM Phase A MPC Phase C PI−PWM Phase C MPC Phase C PI−PWM Phase C Fig. 7. THD values for the voltages at the PCC during the blackout of the main grid In this case, the IPS-A and IPS-B are connected and IPS-grid is disconnected (see Fig. 1). In the case of the microgrid (A) the non-linear loads are connected during all the sample instants of the simulation and the unbalanced loads are connected for these sample instants t>0.1s. In the microgrid (B) the unbalanced loads are connected during all the sample instants and the non-linear loads are connected at t>0.1s. In Fig. 8 and Fig. 9 a comparison between the obtained results for the voltage magnitudes for every phase of each microgrid are shown. The current consumption can be observed in Fig.12. As can be seen, for the sample instants t∈[0.10,0.12] in Fig. 9 a more robust behavior is obtained in the case of working interconnected where the voltage magnitudes of each microgrids are always |U(X) out,γ|>200 for both microgrids. In Fig.13 the obtained results for the current exchange between both microgrids are shown. As can be seen, each microgrid manages its own loads without nearly non-affection to the neighbor microgrid. As can be seen in Fig. 10 and Fig. 11, the presence of non-linear and unbalanced loads and the changes in current demand at each microgrid, as well as the interaction between microgrids do not affect to the THD content in voltage or to the balance between phases guaranteeing the power quality supply to the loads connected to both microgrids. 0 0.05 0.1 0.15 0.2 0.25 0.3 180 200 220 Voltage Magnitude (Vrms) (A)-a (A)-b (A)-c (B)-a (B)-b (B)-c 0 0.05 0.1 0.15 0.2 0.25 0.3 Time (s) Fig. 8. Voltage Magnitude per phase and microgrid in mode non-interconnected and grid-islanded Fig. 9. Voltage Magnitude per phase and microgrid in mode interconnected and grid-islanded Fig. 10. Absolute voltage phase angle value per phase and microgrid in mode interconnected and grid-islanded 00.05 0.1 0.15 0.2 0.25 0.3 0 10 20 30 Time (s) Voltage THD (%) Phase (A)-a Phase (A)-b Phase (A)-c Phase (B)-a Phase (B)-b Phase (B)-c Time (s) Fig. 11. Voltage THD per phase and microgrid in mode interconnected and grid-islanded Fig. 12. Current per phase and microgrid in mode interconnected and grid-islanded 4. CONCLUSIONS The main aim of the present study has been the implementation and validation under simulation of a methodology to manage the power quality in interconnected microgrids acting when they are grid-connected or under a grid blackout where they have to work interconnected but islanded from the main grid. The control algorithm is based on an MPC-controller applied to a four-wire three-phase VSI with active control of the neutral point which works as master of a microgrid with unbalanced and non-linear loads and generators connected. The simulation results show the potential of the presented MPC-controller in comparison wiht classical PI-PWM controllers solving the transient response problems of traditional methods. As can 0 0.05 0.1 0.15 0.2 0.25 0.3 -100 0 100 Current (A) (B)->(A)-a 0 0.05 0.1 0.15 0.2 0.25 0.3 -100 0 100 Current (A) (B)->(A)-b 0 0.05 0.1 0.15 0.2 0.25 0.3 -100 0 100 Time (s) Current (A) (B)->(A)-c Fig. 13. Current exchange per phase between microgrid (A) and microgrid (B) be seen, the developed methodology is improved with its application to the case of interconnected microgrids acting islanded from the main grid. ACKNOWLEDGEMENTS This work has been carried out with the financial support of the European Regional Development Fund (ERDF) under the program Interreg SUDOE SOE3/P3/E0901 (Project IMPROVEMENT) and the financial support by Spanish Ministry of Science, Innovation and Universities under grant DPI2016-78338-R (Project CONFIGURA). REFERENCES Bordons, C., Garcia-Torres, F., and Ridao, M.A. (2020). Model predictive control of microgrids. Garcia-Torres, F., Bordons, C., and Vazquez, S. (2015). Voltage predictive control for microgrids in islanded mode based on fourier transform. In 2015 IEEE International Conference on Industrial Technology (ICIT), 2358–2363. IEEE. Guerrero, J.M., Loh, P.C., Lee, T.L., and Chandorkar, M. (2012). Advanced control architectures for intelligent microgrids-part ii: Power quality, energy storage, and ac/dc microgrids. IEEE Transactions on Industrial Electronics, 60(4), 1263–1270. Jayachandran, M. and Ravi, G. (2019). Decentralized model predictive hierarchical control strategy for islanded ac microgrids. Electric Power Systems Research, 170, 92–100. Vazquez, S., Rodriguez, J., Rivera, M., Franquelo, L.G., and Norambuena, M. (2016). Model predictive control for power converters and drives: Advances and trends. IEEE Transactions on Industrial Electronics, 64(2), 935–947. Yazdi, F. and Hosseinian, S. (2019). A novel smart branch for power quality improvement in microgrids. International Journal of Electrical Power & Energy Systems, 110, 161–170.