An evolutionary triclustering approach to discover electricity consumption patterns in France

An evolutionary triclustering approach to discover electricity consumption patterns in France D. Gutiérrez-Avilés Department of Computer Science, University of Seville Seville, Spain [email protected] J. F. Torres Data Science & Big Data Lab, Pablo de Olavide University Seville, Spain [email protected] F. Martínez-Álvarez Data Science & Big Data Lab, Pablo de Olavide University Seville, Spain [email protected] J. Cugliari ERIC Lab EA 3883, Université Lumière Lyon 2 Lyon, France [email protected] ABSTRACT Electricity consumption patterns are critical in shaping energy policies and optimizing resource allocation. In pursuing a more sustainable and efficient energy future, uncovering hidden consumption patterns is paramount. This paper introduces an innovative approach, leveraging evolutionary triclustering techniques, to unveil previously undisclosed electricity consumption patterns in France. By harnessing the power of triclustering algorithms, this research provides a comprehensive analysis of electricity usage across various dimensions, shedding light on intricate relationships among variables. Using this novel method, the study reveals concealed patterns and offers insights that can inform decision-makers and stakeholders in the energy sector. The findings contribute to a better understanding of electricity consumption dynamics, aiding in developing more targeted and effective energy management strategies. This research represents a significant step forward in the quest for sustainable energy solutions and underscores the potential of evolutionary triclustering as a valuable tool in uncovering complex consumption patterns. CCS CONCEPTS •Computing methodologies → Machine learning;Machine learning approaches;Bio-inspired approaches;Genetic algorithms;Genetic algorithms; KEYWORDS Time series, triclustering, evolutionary algorithm, electricity consumption, patterns ACM Reference Format: D. Gutiérrez-Avilés, J. F. Torres, F. Martínez-Álvarez, and J. Cugliari. 2024. An evolutionary triclustering approach to discover electricity consumption ACM ISBN 978-1-4503-9517-5/23/03...$15.00 https://doi.org/10.1145/3605098.3636034 patterns in France. In The 39th ACM/SIGAPP Symposium on Applied Computing (SAC ’24), April 8 - April 12, 2024, Ávila, Spain. ACM, New York, NY, USA, Article 4, 9 pages. https://doi.org/10.1145/3605098.3636034 1 INTRODUCTION Electricity consumption data analysis is crucial for understanding and optimizing energy usage, a fundamental component of sustainable development and resource management. Extracting meaningful patterns from vast and complex electricity consumption datasets is a challenging task, but it can provide valuable insights for various applications, including energy forecasting, demand-side management, and anomaly detection. This paper presents a novel approach to extracting behavior patterns from France’s electricity consumption data. Our methodology is built upon triclustering, an extension of clustering and biclustering tailored for three-dimensional (3D) datasets. In the context of electricity consumption data analysis, our 3D dataset consists of layers representing years, instances representing time slots, and attributes representing months. A tricluster is a subset of instances, attributes, and layers that exhibit similar behavior patterns under a given criterion. Unlike clustering, which operates on 2D datasets, and biclustering, which considers instances and attributes, triclustering considers the temporal dimension, allowing us to discover behavior patterns that span time, months, and years. In our case, the layer dimension corresponds to the temporal development of electricity consumption instances across attributes, and the criterion for forming triclusters is based on the similarity of their behavior patterns. To implement triclustering, we employ the 𝑇𝑟𝑖𝐺𝑒𝑛 algorithm [ 8 ], an evolutionary metaheuristic inspired by genetic algorithms. 𝑇𝑟𝑖𝐺𝑒𝑛 takes a 3D dataset as input and aims to find a set of triclusters by optimizing specific fitness functions. The algorithm operates by evolving an initial population of potential tricluster solutions over generations using genetic operators such as selection, crossover, and mutation. The fitness function is a crucial aspect of 𝑇𝑟𝑖𝐺𝑒𝑛 , which defines how well a tricluster captures a behavior pattern. We employ three fitness functions: 𝑀𝑆𝑅3𝐷 [ 6 ], 𝐿𝑆𝐿 [ 5 ], and 𝑀𝑆𝐿 [ 7 ], each designed to measure the similarity of behavior patterns across instances, attributes, and layers. These fitness functions enable 𝑇𝑟𝑖𝐺𝑒𝑛 to adapt to different behavior patterns and capture meaningful insights from the electricity consumption data. 386 SAC ’24, April 8–12, 2024, Avila, Spain ©2024 Copyright held by the owner/author(s). This work is licensed under a Creative Commons Attribution International 4.0 License. SAC ’24, April 8 - April 12, 2024, Ávila, Spain D. Gutiérrez-Avilés et al. The results of our analysis provide valuable insights into behavior patterns within the dataset covering the years from 2012 to 2020. We executed the 𝑇𝑟𝑖𝐺𝑒𝑛 algorithm with various control configurations to discover triclusters. Each run aimed to identify ten triclusters, each representing subsets of continuous time slots, months, and years that exhibit similar electricity consumption behavior patterns. Our analysis of the triclustering results focused on two key aspects: coverage and quality. We assessed the coverage of the dataset dimensions by the generated triclusters, analyzing their impact on time slots, months, and years. Additionally, we evaluated the quality of individual triclusters using the 𝑇𝑅𝐼𝑄 [ 4 ] index, which combines graphical quality, Pearson’s correlation quality, and Spearman’s correlation quality. In the following sections, we present detailed descriptions of our methodology, experimental setup, results, and discussions, providing a comprehensive overview of our approach and its implications for electricity consumption data analysis. 2 RELATED WORKS One of the pioneering works in this field introduced an algorithm based on the symmetry property of triclusters [ 31 ]. A more extended and generalized version of this proposal was presented in another study [ 12 ]. Later, Liu et al. [ 11 ] took an evolutionary computation approach, defining the fitness function as a multi-objective measure. The work in [ 30 ] redefined triclusters to uncover gene regulatory relationships, tested on synthetic and real data. Another method, LagMiner [ 28 ], found time-lagged 3D clusters, while [ 14 ] used evolutionary computation for multi-objective triclusters on real data. The work in [ 23 ] introduced a strategy for mining 3D clusters in real-valued data, addressing item count limitations, while [ 10 ] focused on low variance triclusters in quantitative data. Liu et al. [ 15 ] discovered temporal dependency association rules in microarrays, representing gene relationships. In a different application, [ 27 ] developed a triclustering algorithm using statistical methods for 3D short gene expression data time series datasets. In 2014, 𝑇𝑟𝑖𝐺𝑒𝑛 was introduced in [ 8 ]. The authors used an evolutionary approach to discover triclusters in the biological context. However, 𝑇𝑟𝑖𝐺𝑒𝑛 has been successfully applied to different domains. Thus, it was applied to earthquake zonification in [ 17 ] and [ 1 ], to precision agriculture in [ 21 ] and [ 20 ], to the streaming environment in [18] and [19]. Various triclustering algorithms have been developed, each with its own approach. These approaches include iterative searches, distribution parameter identification, pattern mining, and evolutionary multi-objective optimization [ 9 ]. Also in 2018, Kakati et al. [ 13 ] introduced a robust triclustering technique focused on conditionspecific change analysis in HIV-1 progression data, identifying the 38 most responsible genes for this deadly disease. Cuckoo search and particle swarm optimization in triclustering temporal gene expression data were compared in [ 25 ], showing superiority for the latter in three real-world datasets. Consequently, the same authors proposed a hybrid cuckoo search with clonal selection to discover meaningful triclusters in breast cancer gene expression data [ 26 ]. They discovered certain genes with a clear influence on such cancer. Another interesting triclustering-based approach can be found in [ 3 ]. The authors analyzed gene expression microarray data using a restricted neighborhood search during the triclustering process. Later, they used a grained and dynamic deme-based parallel genetic approach to mine, again, gene expression microarray data [2]. In 2022, the authors in [ 24 ] proposed a triclustering method for finding biomarkers in human immunodeficiency virus-1 gene expression data, discovering relevant patterns for different types of patients. Finally, Mondal et al. [ 22 ] introduced a novel data structure suffix forest to design a triclustering algorithm. The approach was applied to the Indian Forest Dataset. The main achievement is discovering triclusters to monitor forest and mangrove cover changes over time. 3 METHODOLOGY This section presents triclustering as a methodology to extract behavior patterns from France’s electricity consumption data and the 𝑇𝑟𝑖𝐺𝑒𝑛 algorithm [ 8 ], an evolutionary metaheuristic to apply triclustering in three-dimensional datasets. 3.1 Triclustering This methodology consists of finding triclusters from a threedimensional (or 3 𝐷 ) dataset, which means a set of layers composed of instances, each composed of several attributes (or features). Considering the previous 3 𝐷 dataset statement, a tricluster is defined as a subset of instances, attributes, and layers from a whole collection of a 3 𝐷 dataset dimension whose values are similar under a given criterion. Triclustering is typically interpreted as an extension of clustering and biclustering. In a clustering problem, given a common two-dimensional (2 𝐷 ) dataset (a collection of instances composed of features), the goal is to find clusters or subsets of instances whose values along all the features show a similar behavior under a given criterion. Extending the clustering technique, biclustering emerges intending to discover biclusters, which are subsets of instances whose values along a subset of features depict a similar behavior under a given criterion. Finally, in the last extension of the former methodology, the environment is changed by adding the third dimension (layers) to the dataset, and the objective is to find triclusters, meaning subsets of instances whose values along a subset of features depict a similar behavior along a subset of layers under a given criterion. This technique, as demonstrated in Section 2, is highly versatile and, therefore, can be applied to various problems. The definition of instances, attributes, and layers within the 3 𝐷 -dataset, as well as the criteria for grouping triclusters, are the key factors in solving the problem. Typically, in most triclustering problems, the layer dimension of the 3 𝐷 -dataset corresponds to the time points representing the temporal development of its instances across attributes, and the criterion for forming triclusters is that their values exhibit a behavior pattern. As triclustering is considered an np-complete problem, metaheuristic approximations offer a proper approach to performing the technique; in that sense, the TriGen algorithm, an evolutionary 387 Evolutionary triclustering approach for electricity consumption in France SAC ’24, April 8 - April 12, 2024, Ávila, Spain implementation of the triclustering technique, has been used for this research. 3.2 The TriGen algorithm This algorithm, whose initial version was presented in [ 8 ], implements a triclustering technique. Given a 3 𝐷 -dataset as input, 𝑇𝑟𝑖𝐺𝑒𝑛 finds a set of triclusters by optimizing a specific criterion, as is shown in Eq. 1. 𝑇𝑟𝑖𝐺𝑒𝑛(𝐷3𝐷)={𝑇1,𝑇2, ...,𝑇𝑖, ...,𝑇𝑁}(1) 𝑇𝑟𝑖𝐺𝑒𝑛 was implemented following the genetic algorithm paradigm; its general workflow can be found in Figure 1. For each requested solution (tricluster), an evolutionary process is applied based on the evaluation performed by a fitness function, previously named criterion. In other words, the algorithm evolves them over specific generations using genetic operators starting from an initial population of potential solutions (individuals). These operators include: Figure 1: TriGen algorithm. • Initial population: In this phase, the initial individuals of the populations are built. For each new individual in the initial population, a subset of instances, attributes, and layers are randomly selected from the input 3𝐷-dataset. • Selection: This operator implements elitism within the population by advancing the best individuals to the next generation. A portion of the remaining individuals is also promoted to ensure genetic diversity. The algorithm employed for executing this operator is the tournament method; population members are initially divided into three groups and subsequently sorted by their fitness levels. A percentage of the population is chosen from these three ordered groups. These selected individuals are directly advanced to the next generation and will serve as suitable parents for reproduction by applying the crossover operator. • Crossover: This process merges the genetic material of two individuals to create two new ones, potentially generating improved solutions for the next generation. Two individuals are randomly selected for reproduction from those chosen by the selection operator. From these two parent individuals, denoted as 𝑃1 and 𝑃2 , two offspring, 𝐶𝐻1 , and 𝐶𝐻2 , are derived as illustrated in Eq. 2 and Eq. 3. The first new individual, 𝐶𝐻1 , is formed by combining the instance, 𝐼 , coordinates from 𝑃1 , the attribute coordinates from 𝑃2 , and the layer coordinates from 𝑃1 . Conversely, the second individual, 𝐶𝐻2 , comprises the instance coordinates from 𝑃2 , the attribute coordinates from 𝑃1, and the layer coordinates from 𝑃2. 𝑃1={𝑃1𝐼, 𝑃1𝐴, 𝑃1𝐿}and 𝑃2={𝑃2𝐼, 𝑃2𝐴, 𝑃2𝐿}(2) 𝐶𝐻1={𝑃1𝐼, 𝑃2𝐴, 𝑃1𝐿}and 𝐶𝐻2={𝑃2𝐼, 𝑃1𝐴, 𝑃2𝐿}(3) • Mutation: This process involves making minor modifications to the genetic makeup of a randomly selected group of individuals. These alterations introduce variations in the solutions, potentially resulting in improved outcomes for future generations. The individuals produced through the crossover operation are candidates for mutation. An individual can change with a certain probability, such as removing or adding a random instance, attribute, or layer coordinate. In the context of the 𝑇𝑟𝑖𝐺𝑒𝑛 genetic process, an individual represents a tricluster, with its genetic material consisting of a subset of instances, attributes, and layers from the input 3 𝐷 -dataset. Consequently, the described evolutionary process is applied to a population of triclusters and their collections of instances, attributes, and layers (genetic material). The fitness function is a key aspect of the 𝑇𝑟𝑖𝐺𝑒𝑛 algorithm (and genetic algorithms in general). This function, the objective optimized by the algorithm, defines how good a potential solution (individual) is. It corresponds to the tricluster formation criterion mentioned in Section 3.1. 𝑇𝑟𝑖𝐺𝑒𝑛 implements the fitness function of a tricluster 𝑇 defined by a weighted average shown in Eq. 4. 𝐹𝑖𝑡 (𝑇)= 𝑤𝑓∗𝐹(𝑇) + 𝑤𝑠∗𝑆(𝑇𝐼,𝑇𝐴,𝑇 𝐿) + 𝑤𝑜∗𝑂(𝑇, 𝑆𝑜𝑙) 𝑤𝑓∗𝑤𝑠∗𝑤𝑜 (4) Where 𝑆(𝑇𝐼,𝑇𝐴,𝑇𝐿) is the size of the tricluster 𝑇 regarding its instances, attributes, and layers. 𝑂(𝑇, 𝑆𝑜𝑙) measures the overlapping of the evaluated tricluster concerning the discovered solutions, 𝑆𝑜𝑙 . 388 SAC ’24, April 8 - April 12, 2024, Ávila, Spain D. Gutiérrez-Avilés et al. The main component of the 𝐹𝑖𝑡 evaluation procedure is the 𝐹(𝑇) , which measures how the tricluster 𝑇 accomplishes the formation criterion. Finally, 𝑤𝑓 , 𝑤𝑠 , and 𝑤𝑜 are the weights that establish the influence of each member in the evaluation process; 𝑇𝑟𝑖𝐺𝑒𝑛 set these weights, as default, with the values shown in Eq. 5. 𝑤𝑓=0.8,𝑤𝑠=0.1,𝑤𝑜=0.1(5) Focusing on the 𝐹(𝑇) member, 𝑇𝑟𝑖𝐺𝑒𝑛 implements three tricluster criteria, all designed to measure the similarity of the behavior patterns depicted by the values of a tricluster across its subset of instances, features, and layers. These fitness functions are: •𝑀𝑆𝑅3𝐷 : This function measures the homogeneity of triclusters across all three dimensions (instances, attributes, and layers) and was introduced in [ 6 ], initially for solving biological problems. •𝐿𝑆𝐿 : Introduced in [ 5 ], this function uses the angle of the least squared line of each pattern line within the tricluster to assess the similarity of all lines, representing the overall pattern. •𝑀𝑆𝐿 : For each line in the pattern depicted by the tricluster, this function computes the angles between each pair of points and compares them to gauge the similarity of the pattern. This function and its computation process can be detailed in [7]. In summary, 𝑇𝑟𝑖𝐺𝑒𝑛 takes an input 3D-dataset and aims to find a specific number of triclusters, as defined by the input parameters. It does so by optimizing a particular fitness function (described above) through an evolutionary process for each requested tricluster. 4 RESULTS Several experiments were conducted to discover behavior patterns (triclusters) in France’s electricity consumption data using the 𝑇𝑟𝑖𝐺𝑒𝑛 algorithm. The following sections present the materials used for running the experiments and the analysis of the discovered patterns. 4.1 Experimental Setup The input for the experiments consists of France’s electricity consumption data, which was obtained from the website of France’s transmission system operator (www.rte-france.com/). These data were downloaded in separate files, with one file per year, covering the years from 2012 to 2020 (inclusive). Each file contains the consumption data (in TWh) distributed in a semi-hourly (30-minute frequency) time series spanning from the first to the last day of the respective year. In other words, each file (corresponding to a year) contains a time series ranging from January 1st at 00:00h to December 31st at 23:30h. 4.1.1 3 𝐷 -Dataset Definition. The data has a 3 𝐷 -array structure, resembling a cube, where the rows (instances) represent the time slots in a semi-hourly time series for a month. Specifically, it covers the time slots from the 1st day at 00:00h, progressing through each half-hour interval, until the 31st day at 23:30h. Figure 2 shows how the columns (attributes) represent the months of the year, and the layers correspond to the years, spanning from 2012 to 2020. Therefore, a cell located in the 3 𝐷 -dataset at coordinates [𝑖, 𝑎, 𝑙] contains the electricity consumption value for France for the year 𝑙 , in the month 𝑎 , at the time slot 𝑖 (a specific day of the month 𝑎 , at a specific time). Figure 2: France electricity consumption 3𝐷-dataset. 4.1.2 𝑇𝑟𝑖𝐺𝑒𝑛 Execution. The 𝑇𝑟𝑖𝐺𝑒𝑛 algorithm was executed multiple times to identify behavior patterns within the 3 𝐷 -dataset. The algorithm was executed with various control configurations for the evolutionary process, and each run requested ten triclusters. In this research, the obtained triclusters represent subsets of continuous time slots (rows/instances), subsets of months (columns/attributes), and subsets of years (layers) in such a way that their electricity consumption values reveal behavior patterns. 4.1.3 Tricluster Evaluation. The results of each 𝑇𝑟𝑖𝐺𝑒𝑛 experiment, which consists of a collection of triclusters, are evaluated from two perspectives: overall coverage of the input 3 𝐷 -dataset dimensions by the generated triclusters and the individual quality of the triclusters. Concerning the first aspect, coverage graphs for the dimensions of the 3 𝐷 -dataset (time slot, month, and year) are generated. These graphs illustrate which coordinates are selected for each obtained tricluster, providing insight into the influence of each tricluster on each dimension. 𝑇𝑅𝐼𝑄 (𝑇)= 𝑤𝑔𝑟 ∗𝐺𝑅𝑄(𝑇) + 𝑤𝑝𝑒 ∗𝑃𝐸𝑄 (𝑇) + 𝑤𝑠𝑝 ∗𝑆𝑃𝑄(𝑇) 𝑤𝑔𝑟 ∗𝑤𝑝𝑒 ∗𝑤𝑠𝑝 (6) The second aspect involves assessing the individual quality of the triclusters. This uses the 𝑇𝑅𝐼𝑄 index, a tricluster quality metric introduced in [ 4 ]. 𝑇𝑅𝐼𝑄 is defined in Eq 6 as a weighted average of three indexes, these are: •𝐺𝑅𝑄 : It evaluates the visual quality of the tricluster, which quantitatively represents its qualitative aspect—namely, how consistent and similar its members are. This approach is commonly employed in research to visually validate results 389 Evolutionary triclustering approach for electricity consumption in France SAC ’24, April 8 - April 12, 2024, Ávila, Spain by graphically illustrating the triclusters across their three components: instances, attributes, and layers [16, 29, 31]. •𝑃𝐸𝑄 and 𝑆𝑃𝑄 : These metrics assess the typical correlation among the tricluster’s values and are commonly referenced in the literature [ 12 ]. Correlation quantifies the interdependence among the instances, attributes, and layers within the triclusters. Specifically, 𝑃𝐸𝑄 relies on Pearson’s correlation coefficient, while 𝑆𝑃𝑄 is grounded in Spearman’s correlation coefficient. 𝑤𝑔𝑟 =0.8,𝑤𝑝𝑒 =0.1, 𝑤𝑠𝑝 =0.1(7) These three indexes are combined into a final value, 𝑇𝑅𝐼𝑄 , by applying the weights described in Eq. 7. 𝑇𝑅𝐼𝑄 ranges from lower (0.0) to higher (1.0) to indicate the quality of each tricluster. Additionally, the behavior patterns of the triclusters are visually represented in two views: the month view, where the consumption curves for specific periods (time slots on the x-axis) of each selected month are depicted, separated by the selected years, and the year view, where the consumption curves for specific periods (time slots on the x-axis) of each selected year are depicted, separated by the selected month. 4.2 Discussion In this section, we discuss the effectiveness of applying the triclustering methodology through the 𝑇𝑟𝑖𝐺𝑒𝑛 algorithm to extract consumption behavior patterns by examining the results of two experiments: experiment A and experiment B. 4.2.1 Coverage Analysis. In terms of the time slot dimension (rows), Figure 3a illustrates how the triclusters in experiment A cover short periods while leaving long periods without coverage. This behavior occurs because 𝑇𝑟𝑖𝐺𝑒𝑛 prioritizes searching in short periods, as guided by its control configuration. This pattern contrasts with the month coverage, as shown in Figure 4a, where the triclusters cover all the months except January. On the other hand, Experiment B exhibits complete coverage of time slots, as demonstrated in Figure 3b. In this case, the triclusters are larger in terms of time slots, indicating that 𝑇𝑟𝑖𝐺𝑒𝑛 diversifies its search along this dimension to identify longer periods. However, Figure 4b illustrates that the month coverage is lower than Experiment A. Both experiments complement each other; Experiment A offers good coverage of months, while Experiment B excels in covering time slots. Consequently, their respective triclusters provide valuable insights into the dimensions being analyzed. Regarding the year dimension, Figure 5 reveals that both experiments achieve complete coverage across all layers. 4.2.2 Quality analysis. The 𝑇𝑅𝐼𝑄 values for Experiment A’s triclusters are presented in Table 1. The average 𝑇𝑅𝐼𝑄 is 0 . 90 with a standard deviation of 0 . 016, indicating a satisfactory experiment in terms of quality. The best solution, tricluster #4, achieves a 𝑇𝑅𝐼𝑄 of 0 . 93, demonstrating excellent graphical quality ( 𝐺𝑅𝑄 at 0 . 96) and a high level of correlation among its values, as confirmed by its 𝑃𝐸𝑄 and 𝑆𝑃𝑄 values. Regarding Experiment B, its 𝑇𝑅𝐼𝑄 values, as shown in Table 2, have a mean of 0 . 866 with a standard deviation of 0 . 014, making it Table 1: TRIQ values of Experiment A Tricluster TRIQ GRQ PEQ SPQ #1 0.91 0.96 0.71 0.69 #2 0.90 0.96 0.61 0.69 #3 0.88 0.91 0.73 0.72 #4 0.93 0.96 0.86 0.75 #5 0.92 0.95 0.75 0.80 #6 0.92 0.96 0.78 0.75 #7 0.89 0.93 0.75 0.73 #8 0.91 0.97 0.62 0.74 #9 0.90 0.93 0.76 0.79 #10 0.91 0.96 0.64 0.75 Table 2: TRIQ values of Experiment B Tricluster TRIQ GRQ PEQ SPQ #1 0.88 0.91 0.73 0.74 #2 0.86 0.89 0.74 0.71 #3 0.87 0.92 0.65 0.67 #4 0.87 0.92 0.66 0.68 #5 0.87 0.92 0.67 0.67 #6 0.89 0.93 0.71 0.70 #7 0.88 0.91 0.79 0.74 #8 0.86 0.91 0.64 0.67 #9 0.86 0.90 0.70 0.68 #10 0.83 0.91 0.52 0.51 another noteworthy experiment. The best solution is tricluster #6, with a 𝑇𝑅𝐼𝑄 of 0 . 89. While it shares a graphical quality of 0 . 93 with tricluster #4 in Experiment A, its correlation indexes are slightly different, with 0.71 for Pearson and 0.70 for Spearman. 4.2.3 Behavior pattern analysis. This section discusses the behavior patterns depicted by the best tricluster for Experiments A and B. Tricluster #4 from Experiment A exhibits a behavior pattern spanning from the 15th at 5:00h to the 16th at 5:30h. In the month view, as depicted in Figure 6a, the consumption curves for the tricluster months, specifically July and October, are examined across each tricluster year (2015, 2017, and 2018). July and October both exhibit a consistent pattern throughout all years. From the 15th at 5:00h to the 15th at 12:30h, consumption displays an upward trend, followed by a gradual descent with smooth peaks until the 15th at 19:30h. Subsequently, another peak emerges, with a subsequent peak at 22:30h on the same day. Finally, consumption declines until the 16th at 5:30h. This behavior maintains consistency between July and October, albeit with varying scales, as October’s values are higher than those of July. In the year view, shown in Figure 6b, the consumption curves for the tricluster years (2015, 2017, and 2018) are depicted for each tricluster month (July and October). The behavior patterns in the year view align with those observed in the month view. This emphasizes the similarity in behavior between the yearly series for each month and highlights the differences in scale while maintaining the same behavior across the years. 390 SAC ’24, April 8 - April 12, 2024, Ávila, Spain D. Gutiérrez-Avilés et al. Tricluster #1 Tricluster #2 Tricluster #3 Tricluster #4 Tricluster #5 Tricluster #6 Tricluster #7 Tricluster #8 Tricluster #9 Tricluster #10 5th, 12:30h 6th, 12:30h 8th, 5:00h 8th, 9:30h 9th, 5:00h 9th, 9:30h 9th, 13:00h 10th, 13:00h 11th, 23:30h 12th, 10:30h 12th, 23:30h 13th, 10:30h 15th, 5:30h 16th, 5:30h 18th, 17:00h 19th, 19:00h 25th, 21:30h 26th, 4:30h 26th, 21:30h 27th, 4:30h (a) Experiment A. Tricluster #1 Tricluster #2 Tricluster #3 Tricluster #4 Tricluster #5 Tricluster #6 Tricluster #7 Tricluster #8 Tricluster #9 Tricluster #10 8th, 14:30h 10th, 0:30h 10th, 15:00h 10th, 21:00h 11th, 2:00h 11th, 9:00h 11th, 16:00h 12th, 16:30h 18th, 14:00h 19th, 0:00h 20th, 10:00h 21st, 0:30h 21st, 6:30h 21st, 11:30h 23rd, 2:00h (b) Experiment B. Figure 3: Triclustering coverage in the time slot’s dimension. Regarding Experiment B, the month view of Tricluster #6’s behavior patterns can be found in Figure 7a. Here, the consumption curves for the tricluster months, specifically January and November, are depicted across each tricluster year (2013, 2018, and 2020). The consumption curves span from the 11th at 1:30h to the 21st at 11:30h and depict a pattern characterized by prominent peaks and descents. Notably, this tricluster covers 501 time slots, significantly larger than the 50 grouped time slots of the previous tricluster. The observed peaks and descents repeat approximately every 48 time slots in both series (January and November) and maintain a similar profile for each year. As in the previous tricluster, there is a notable change in scale, with January exhibiting higher values. Moreover, similar consumption values are evident in the central period of 2020 and at the beginning and end of 2013. The year view, represented in Figure 7b, confirms the recurring peaks and descents across all three-year lines for both months. This view highlights a significant scale change from the 17th at 8:30h to the 19th at 13:00h between 2018 and the other two years for January. A similar behavior is also observed for November, from the 13th at 21:00h to the 16th at 17:00h between 2018 and the other two years. 5 CONCLUSIONS AND FUTURE WORKS In this study, we have introduced an innovative approach using evolutionary triclustering techniques to analyze electricity consumption patterns in France, and we have shed light on how and why consumption patterns change over time. Our findings offer valuable insights into the underlying dynamics of electricity consumption and have significant implications for energy policy formulation and efficient resource allocation. However, it is important to acknowledge certain limitations in our study. Firstly, the availability and quality of electricity consumption data can vary, which may affect the precision of our results. Additionally, while we have identified significant patterns, we have not delved deeply into the underlying causes of these patterns. For this reason, in future work, we will investigate the underlying causes of the identified electricity consumption patterns (economic and/or climatic) by incorporating exogenous variables. Moreover, the algorithm efficiency will be improved to tackle even larger and more complex datasets. ACKNOWLEDGMENTS This work was supported by Grant PID2021-124571OB-I00 funded by MCIN/AEI/ 10.13039/501100011033, by “ERDF A way of making 391 Evolutionary triclustering approach for electricity consumption in France SAC ’24, April 8 - April 12, 2024, Ávila, Spain Tricluster #1 Tricluster #2 Tricluster #3 Tricluster #4 Tricluster #5 Tricluster #6 Tricluster #7 Tricluster #8 Tricluster #9 Tricluster #10 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec (a) Experiment A. Tricluster #1 Tricluster #2 Tricluster #3 Tricluster #4 Tricluster #5 Tricluster #6 Tricluster #7 Tricluster #8 Tricluster #9 Tricluster #10 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec (b) Experiment B. Figure 4: Triclustering coverage in the months’ dimension. Europe” and the project PID2020-117954RB of the Spanish Ministry of Science and Innovation. REFERENCES [1] J. L. Amaro-Mellado, L. Melgar-García, C. Rubio-Escudero, and D. Gutiérrez- Avilés. 2021. Generating a seismogenic source zone model for the Pyrenees: A GIS-assisted triclustering approach. Computers & Geosciences 150 (5 2021), 104736. [2] B. S. Biswal, A. Mohapatra, and S. Vipsita. 2020. Triclustering of gene expression microarray data using coarse grained and dynamic deme based parallel genetic approach. Evolutionary Intelligence 13 (2020), 475–495. [3] B. S. Biswal, S. Patra, A. Mohapatra, and S. Vipsita. 2020. TriRNSC: triclustering of gene expression microarray data using restricted neighbourhood search. IET Systems Biology 14 (2020), 323–333. [4] D. Gutiérrez-Avilés, R. Giráldez, F. J. Gil-Cumbreras, and C. Rubio-Escudero. 2018. TRIQ: a new method to evaluate triclusters. BioData Mining 11, 1 (2018), 15. [5] D. Gutiérrez-Avilés and C. Rubio-Escudero. 2014. LSL: A new measure to evaluate triclusters. In Proceedings of the IEEE International Conference on Bioinformatics and Biomedicine. IEEE, 30–37. [6] D. Gutiérrez-Avilés and C. Rubio-Escudero. 2014. Mining 3D Patterns from Gene Expression Temporal Data: A New Tricluster Evaluation Measure. The Scientific World Journal 2014 (2014), 1–16. [7] D. Gutiérrez-Avilés and C. Rubio-Escudero. 2015. MSL: A Measure to Evaluate Three-dimensional Patterns in Gene Expression Data. Evolutionary Bioinformatics 11 (1 2015), EBO.S25822. [8] D. Gutiérrez-Avilés, C. Rubio-Escudero, F. Martínez-Álvarez, and J.C. Riquelme. 2014. TriGen: A genetic algorithm to mine triclusters in temporal gene expression data. Neurocomputing 132 (2014), 42–53. Tricluster #1 Tricluster #2 Tricluster #3 Tricluster #4 Tricluster #5 Tricluster #6 Tricluster #7 Tricluster #8 Tricluster #9 Tricluster #10 2012 2013 2014 2015 2016 2017 2018 2019 2020 (a) Experiment A. Tricluster #1 Tricluster #2 Tricluster #3 Tricluster #4 Tricluster #5 Tricluster #6 Tricluster #7 Tricluster #8 Tricluster #9 Tricluster #10 2012 2013 2014 2015 2016 2017 2018 2019 2020 (b) Experiment B. Figure 5: Triclustering coverage in the year’s dimension. [9] R. Henriques and S.C. Madeira. 2018. Triclustering algorithms for threedimensional data analysis: A comprehensive survey. Comput. Surveys 51, 5 (2018), 1–43. [10] Z. Hu and R. Bhatnagar. 2010. Algorithm for Discovering Low-Variance 3-Clusters from Real-Valued Datasets. In Proceedings of the IEEE International Conference on Data Mining. 236–245. [11] J. Liu, Z. Li, X. Hu, and Y. Chen. 2008. Multi-objective evolutionary algorithm for mining 3D clusters in gene-sample-time microarray data. In Proceedings of the IEEE International Conference on Granular Computing. 442–447. [12] Haoliang Jiang, Shuigeng Zhou, Jihong Guan, and Ying" Zheng. 2006. gTRI- CLUSTER: A More General and Effective 3D Clustering Algorithm for Gene- Sample-Time Microarray Data. In Proceedings of the Data Mining for Biomedical Applications. 48–59. [13] T. Kakati, H. A. Ahmed, D. K. Bhattacharyya, and J. K. Kalita. 2018. THD-Tricluster: A robust triclustering technique and its application in condition specific change analysis in HIV-1 progression data. Computational Biology and Chemistry 75 (2018), 154–167. [14] Z. Li, X. Hu, and Y. Chen. 2008. Multi-objective evolutionary algorithm for mining 3D clusters in gene-sample-time microarray data. In Proceedings of the IEEE International Conference on Granular Computing. 442–447. [15] Y. C. Liu, C. H. Lee, W. C. Chen, J. W. Shin, H. H. Hsu, and V. S. Tseng. 2010. A novel method for mining temporally dependent association rules in three-dimensional microarray datasets. In Proceedings of the IEEE International Computer Symposium. 759–764. [16] Yu-Cheng Liu, Chao-Hui Lee, Wei-Chung Chen, J. W. Shin, Hui-Huang Hsu, and Vincent S. Tseng. 2010. A novel method for mining temporally dependent 392 SAC ’24, April 8 - April 12, 2024, Ávila, Spain D. Gutiérrez-Avilés et al. Time slot Consumption (TWh) 35000 40000 45000 50000 55000 60000 65000 15th, 5:00h 15th, 6:30h 15th, 8:30h 15th, 10:30h 15th, 12:30h 15th, 14:30h 15th, 16:30h 15th, 18:30h 15th, 20:30h 15th, 22:30h 16th, 0:30h 16th, 2:30h 16th, 4:30h 16th, 5:30h 2015 35000 40000 45000 50000 55000 60000 65000 2017 35000 40000 45000 50000 55000 60000 65000 2018 Jul Oct (a) Month view. Time slot Consumption (TWh) 35000 40000 45000 50000 55000 60000 65000 15th, 5:00h 15th, 6:30h 15th, 8:30h 15th, 10:30h 15th, 12:30h 15th, 14:30h 15th, 16:30h 15th, 18:30h 15th, 20:30h 15th, 22:30h 16th, 0:30h 16th, 2:30h 16th, 4:30h 16th, 5:30h Jul 35000 40000 45000 50000 55000 60000 65000 Oct 2015 2017 2018 (b) Year view. Figure 6: Tricluster #4 from Experiment A behavior patterns. Time slot Consumption (TWh) 40000 50000 60000 70000 80000 90000 11th, 1:30h 11th, 18:00h 12th, 11:00h 13th, 4:00h 13th, 21:00h 14th, 14:00h 15th, 7:00h 16th, 0:00h 16th, 17:00h 17th, 10:00h 18th, 3:00h 18th, 20:00h 19th, 13:00h 20th, 6:00h 20th, 23:00h 21st, 11:30h 2013 40000 50000 60000 70000 80000 90000 2018 40000 50000 60000 70000 80000 90000 2020 Jan Nov (a) Month view. Time slot Consumption (TWh) 40000 50000 60000 70000 80000 90000 11th, 1:30h 11th, 18:00h 12th, 11:00h 13th, 4:00h 13th, 21:00h 14th, 14:00h 15th, 7:00h 16th, 0:00h 16th, 17:00h 17th, 10:00h 18th, 3:00h 18th, 20:00h 19th, 13:00h 20th, 6:00h 20th, 23:00h 21st, 11:30h Jan 40000 50000 60000 70000 80000 90000 Nov 2013 2018 2020 (b) Year view. Figure 7: Tricluster #6 from Experiment B behavior patterns. association rules in three-dimensional microarray datasets. In 2010 International Computer Symposium (ICS2010). 759–764. https://doi.org/10.1109/COMPSYM. 2010.5685410 [17] F. Martínez-Álvarez, D. Gutiérrez-Avilés, A. Morales-Esteban, J. Reyes, J.L. Amaro- Mellado, and C. Rubio-Escudero. 2015. A novel method for seismogenic zoning based on triclustering: Application to the Iberian Peninsula. Entropy 17 (2015), 5000–5021. Issue 7. 393 Evolutionary triclustering approach for electricity consumption in France SAC ’24, April 8 - April 12, 2024, Ávila, Spain [18] L. Melgar-García, D. Gutiérrez-Avilés, C. Rubio-Escudero, and A. Troncoso. 2020. High-Content Screening Images Streaming Analysis Using the STriGen Methodology. In Proceedings of the ACM Symposium on Applied Computing (SAC ’20). Association for Computing Machinery, 537–539. [19] L. Melgar-García, D. Gutiérrez-Avilés, C. Rubio-Escudero, and A. Troncoso. 2021. Discovering three-dimensional patterns in real-time from data streams: An online triclustering approach. Information Sciences 558 (5 2021), 174–193. [20] L. Melgar-García, D. Gutiérrez-Avilés, M. T. Godinho, R. Espada, I. S. Brito, F. Martínez-Álvarez, A. Troncoso, and C. Rubio-Escudero. 2021. Discovering spatiotemporal patterns in precision agriculture based on triclustering. In Proceedings of the International Conference on Soft Computing Models in Industrial and Environmental Applications. 226–236. [21] L. Melgar-García, D. Gutiérrez-Avilés, M. T. Godinho, R. Espada, I. S. Brito, F. Martínez-Álvarez, A. Troncoso, and C. Rubio-Escudero. 2022. A new big data triclustering approach for extracting three-dimensional patterns in precision agriculture. Neurocomputing 500 (8 2022), 268–278. [22] K. C. Mondal, M. Ghosh, R. Fajriyah, and A. Roy. in press. Introducing suffix forest for mining tri-clusters from time-series data. Innovations in Systems and Software Engineering (in press). [23] K. Sim and Z. Aung. 2010. Discovering Correlated Subspace Clusters in 3D Continuous-Valued Data. In Proceedings of the IEEE International Conference on Data Mining. 471–480. [24] T. Siswantining, A. Bustamam, D. Sarwinda, S. M. Soemartojo, M. A. Latief, E. A. Octaria, A. T. M. Siregar, O. Septa, H. S. Al-Ash, and N. Saputra. 2022. Triclustering method for finding biomarkers in human immunodeficiency virus- 1 gene expression data. Mathematical Biosciences and Engineering 19 (2022), 6743–6763. [25] P. Swathypriyadharsini and K. Premalatha. 2019. Comparison of cuckoo search and particle swarm optimisation in triclustering temporal gene expression data. International Journal of Swarm Intelligence 4 (2019), 55–72. [26] P. Swathypriyadharsini and K. Premalatha. 2022. Hybrid Cuckoo Search with Clonal Selection for Triclustering Gene Expression Data of Breast Cancer. IETE Journal of Research 69 (2022), 2328–2336. [27] Alain Tchagang, Sieu Phan, Fazel Famili, Heather Shearer, Pierre Fobert, Yi Huang, Jitao Zou, Daiqing Huang, Adrian Cutler, Ziying Liu, and Youlian Pan. 2012. Mining biological information from 3D short time-series gene expression data: The OPTricluster algorithm. BMC bioinformatics 13 (04 2012), 54. [28] X. Xu, Y. Lu, K. L. Tan, and A. Tung. 2009. Finding Time-Lagged 3D Clusters. In Proceedings of the IEEE International Conference on Data Engineering. 445–456. [29] Xin Xu, Ying Lu, Kian-Lee Tan, and Anthony K. H. Tung. 2009. Finding Time- Lagged 3D Clusters. In 2009 IEEE 25th International Conference on Data Engineering. 445–456. https://doi.org/10.1109/ICDE.2009.80 [30] Y. Yin, Y. Zhao, B. Zhang, and G. Wang. 2007. Mining Time-Shifting Co-regulation Patterns from Gene Expression Data. Lecture Notes in Computer Science 4505 (2007), 62–73. [31] Lizhuang Zhao and Mohammed J Zaki. 2005. triCluster: an effective algorithm for mining coherent clusters in 3D microarray data. In Proceedings of the ACM SIGMOD International Conference on Management of data. 694–705. 394