DESIGN SUPPORT TOOL FOR FLEXIBLE DRINKING WATER TREATMENT PLANTS

CCWI 2025 - 21st Computing & Control for the Water Industry Conference, Sheffield, UK DESIGN SUPPORT TOOL FOR FLEXIBLE DRINKING WATER TREATMENT PLANTS Grigorios Kyritsakas1, Dirk Vries2, Tavishi Guleria2, Luuk Rietveld1, Lydia Vamvakeridou-Lyroudia2,3 1 Delft University of Technology, Stevinweg 1, 2628 CN, Delft, the Netherlands 2 KWR Water Research Institute, Groningenhaven 7, 3433 PE,Nieuwegein, the Netherlands 3 University of Exeter, Harrison Building, North Park Road, EX4 4QF, Exeter, UK [email protected] ABSTRACT The increasing challenges posed by climate change and emerging contaminants necessitate a more adaptable and resilient approach to drinking water treatment. This paper presents a novel Design Support Tool aimed at guiding, assisting and simplifying the preliminary design of small-scale and flexible Drinking Water Treatment Plants, with the selection of the “best” combination of treatment trains adapted to their specific needs (water quantity and quality). The tool integrates a comprehensive database of treatment technologies, considers various water sources. In its final version, the tool will incorporate a multi-objective optimization of treatment processes based on cost and spatial footprint. The current Python-based prototype and development of the web application are discussed, highlighting its potential to aid process technologists and decision-makers in achieving high water quality standards with enhanced adaptability and robustness. Keywords: Drinking water treatment plant, design support tool, water treatment trains INTRODUCTION The provision of safe and high-quality drinking water faces growing complexities due to factors such as climate change and the proliferation of emerging contaminants, including pesticides, pharmaceuticals, disinfection by-products, heavy metals, and pathogenic microorganisms. Therefore, the design of a drinking water treatment plant (DWTP) is a complex process that requires careful selection of treatment technologies to meet regulatory standards for drinking water quality. Additionally, water utilities must consider various factors, including construction and operational costs, land availability, environmental impact, and system complexity, while flexibility in the design parameters is also desirable [1]. To address these challenges, this paper presents a design support tool (DST) for the preliminary design of a small and flexible DWTP, aiming to guide, assist and simplify the design process. The DST leverages a structured, automated methodology to assist water utilities and process engineers in identifying viable treatment trains tailored to raw water characteristics and site-specific constraints. Key performance indicators—including capital and operational costs, energy use, CO₂ emissions, and spatial footprint—are incorporated to support decision-making. This paper outlines the DST's conceptual development, computational workflow, and early validation results. CCWI 2025 - 21st Computing & Control for the Water Industry Conference, Sheffield, UK METHODS The Design Support Tool (DST) is developed to facilitate the preliminary design of small and flexible Drinking Water Treatment Plants (DWTPs). Its methodology is structured to guide process technologists and decision-makers through a systematic selection and evaluation process. Core Functionality and Data Integration The DST operates by allowing users to specify raw water quality characteristics, desired treatment goals, and relevant process and scale parameters. Based on these inputs, the tool employs an automated procedure to identify suitable treatment stages and technologies. A comprehensive database underpins this functionality, containing key characteristics of state-of-the-art treatment technologies sourced from literature and regulatory specifications (e.g., EPA's Drinking Water Treatability Database [1],[2],[3]). This database includes information on: • Treatment Stages and Technologies: Grouped by treatment purpose (e.g., screening, ammonia removal, coagulation/flocculation, floc removal, rapid sand filtration, OMP removal, NOM removal, softening, aeration, conditioning, disinfection, ultrafiltration/microfiltration). • Deciding factors: a pre-specified decision tree is translated into database tables where each row specifies a certain question or answer (here referred to as deciding factor) and is dependent on a parent deciding factor. Questions and answers are queried from the DST web application. • Physical and Operational Parameters: Spatial footprint (m2), energy consumption (kWh per year), and head losses (m) per treatment stage. Footprint calculations are derived from established engineering practices. • Cost Data: Information for calculating Total Costs of Ownership (TCO) per m3 of treated water, often integrated with external tools. • Environmental impact: CO2 emissions are calculated per treatment stage for both the construction phase and the operation phase. In the latter, these are calculated based on the selected energy source (e.g. renewable energy, fossil fuels etc.) Design Process and Optimization The DST workflow follows a structured approach: 1. Nominal Flow Identification: Calculation of the DWTP's nominal flow (m3/day) based on population numbers, consumption (default 125 L per person per day), industrial/agricultural needs, leakage percentage (default 20%), and peak factors (default 1.5). The tool also allows for user-defined population expansion and consumption pattern differences. 2. Water Source Selection: Users select the available drinking water source (groundwater, surface water, or riverbank water). 3. Treatment Stage Identification and Selection: Based on the water source and its characteristics (e.g., oxygen levels, ammonia, hardness, algae, NOM, turbidity, OMPs), the tool guides the user in defining and identifying required treatment processes from a list of feasible options. The DST directs the user to specific solutions based on their selection following a decision tree approach as presented in figure 1. CCWI 2025 - 21st Computing & Control for the Water Industry Conference, Sheffield, UK 4. Treatment Technology Selection and Ordering: Users select specific technologies within each identified stage. 5. Multi-Objective Optimization: The final version of the DST (currently under development) will include an optimization module, primarily focusing on minimizing cost, use of chemicals and spatial footprint. 6. Performance Evaluation: The tool calculates and presents overall performance scores for the selected treatment train across all weighted criteria, along with detailed output performance indicators like area footprint. Figure 1. Decision Tree that DST uses for the identification of treatment stages and treatment trains System Architecture The first version of the DST is developed as a Python notebook. The current version is developed as a Django web application. The web application framework consists of: • Front-end: A web interface built with HTML, CSS (using Bootstrap for styling), and JavaScript, providing interactive forms for user input and displaying results. • Back-end, with a: CCWI 2025 - 21st Computing & Control for the Water Industry Conference, Sheffield, UK 1. Communication Layer (ORM): Utilizes Django's Object-Relational Mapper (ORM) to manage data interactions with the database. REST APIs are developed using the Django REST framework to generalize and abstract communication. 2. Database: An SQLite database stores treatment technologies, stages, chemical compounds, removal efficiencies, scenario settings, and user management data. 3. Calculation Modules: modules perform the calculation of treatment capacity (head loss and nominal flow) and uncertainty propagation. RESULTS AND DISCUSSION An early development version of the DST demonstrates the capability to generate preliminary designs for Drinking Water Treatment Plants (DWTPs) based on user-defined inputs and a comprehensive database of treatment technologies. User Interaction and Preliminary Design Workflow The current Python-based prototype (figure 2) allows users to interact through a command-line interface, demonstrating the core workflow: 1. Input of population numbers, water consumption, leakage estimation, and industrial/irrigation needs to determine nominal flow. 2. Selection of water source type. 3. Identification of required treatment stages based on water quality parameters (e.g., algae concentration, turbidity, NOM, hardness). 4. Selection of specific treatment technologies within each stage (e.g., Cascade Mixing for Coagulation, Dissolved Air Flotation for Floc Removal, Cationic IEX for Softening, Dual Media Filtration for Sand Filtration, GAC for OMP Removal, Chlorination for Disinfection). 5. Display of the proposed treatment train, listing the sequence of treatment steps and their selected technologies. 6. Output of performance indicators, such as footprint per unit and total footprint, providing immediate feedback on the design's physical implications. Figure 2. The current interface of the DST tool CCWI 2025 - 21st Computing & Control for the Water Industry Conference, Sheffield, UK Proposed Treatment Trains The tool successfully identifies and proposes treatment trains tailored to specific raw water sources (groundwater, surface water, riverbank filtrate) and their characteristics. For instance, for groundwater, the tool can suggest aeration and rapid filtration for iron, manganese, and ammonium removal, with optional remineralization or softening. For surface water, it can propose coagulation, flocculation, floc removal, rapid sand filtration, and OMP removal steps, with disinfection. The DST's automated procedure ensures that the sequence of treatment stages adheres to established engineering principles, adapting the order based on selected technologies (e.g., placing PAC dosing before sedimentation). Spatial footprint calculation, head loss and Energy consumption estimation The tool estimates a footprint area required following examples from the literature [2],[3]. These calculations include the required areas for the storage chemicals and equipment per treatment stage. The total footprint of the DWTP is calculated as a sum of the footprint required per area. Finally, additional space estimations are made for secondary areas such as offices, roads for commuting and other facilities. The same approach is followed for the calculation of the head losses and energy consumption per treatment step. The head losses are provided in meters of water column (mH2O) and the energy consumption in kWh in a year (365 days). Costs calculation and environmental impact estimation The cost of the final treatment train is calculated in terms of total costs of ownership per m3 of treated water and operational costs per m3 of treated. The costs are calculated following 3 existing tools for the cost estimation ‘Kostenstandaard’ tool (https://kostenstandaard.nl/) developed by RHDHV [4], the Drinking Water Treatment Technology Unit Cost Models developed by EPA [5], and expert knowledge from KWR. Regarding the environmental impact estimation, CO2 emissions are calculated following the Greenhouse Gas (GHG) standard protocols [6]. The prototype DST effectively showcases the decision-making logic and calculation capabilities of the DST, providing a proof-of-concept for its utility in supporting preliminary DWTP design. A demonstration of the current version of the DST can be viewed in the following YouTube link: https://www.youtube.com/watch?v=PtWiVbBbQWo&t=10s&ab_channel=gorekyr itsakas CONCLUSIONS This paper presents a novel DST that enables rapid, informed, and transparent preliminary design of DWTPs. By integrating automated process selection, performance estimation, and optimization functionalities, the DST empowers engineers and utilities to evaluate multiple design alternatives based on quantitative metrics. Future development will focus on: CCWI 2025 - 21st Computing & Control for the Water Industry Conference, Sheffield, UK • Transitioning the tool to a Django-based web platform to improve accessibility and user interface • Finalising the CO2 emission calculation module • Developing an optimization module with the aim to minimize the spatial footprint and chemicals usage while keeping the water quality constrains. Ultimately, the DST aspires to support a paradigm shift toward adaptive, decentralized, and sustainable drinking water treatment solutions. REFERENCES [1] J. C. Crittenden, R. R. Trussell, D. W. Hand, K. J. Howe, and G. Tchobanoglous, MWH's Water Treatment: Principles and Design, John Wiley & Sons, 3rd edition 2012. [2] U.S. Environmental Protection Agency, "Drinking Water Treatability Database (TDB)," [Online]. Available: https://tdb.epa.gov. [Accessed: 02-06-2025]. [3] M. L. Davis, Water and Wastewater Engineering: Design Principles and Practice, 2nd ed., New York: McGraw-Hill Education, 2020. [4] Kostenstandaard.nl, "[Online]. Available: https://kostenstandaard.nl/." [Accessed: 02-06-2025]. [5] U.S. Environmental Protection Agency, " Drinking Water Treatment Technology Unit Cost Models" [Online]. Available: https://www.epa.gov/sdwa/drinking-water-treatment-technology-unit-costmodels [Accessed: 02-06-2025]. [6] International Water Association, "Greenhouse Gas Emissions and Water Resource Recovery Facilities," IWA, 2023. Available: https://iwanetwork.org/wp-content/uploads/2023/09/GHG-Emissions-and-Water-ResourceRecovery-Facilities-digital.pdf [Accessed: 24-06-2025].