MULTI-CRITERION OPTIMIZATION OF ADAPTIVE CONTROL SYSTEMS FOR WATER TREATMENT PROCESSES USING FUZZY LOGIC

Abstract

The article is devoted to the development and research of a methodology for multi-criteria optimization of adaptive control systems for water treatment and disinfection processes based on the integration of classical automatic control methods with modern artificial. The purpose of this study is to create a comprehensive control system that ensures high quality drinking water while minimizing energy and resource costs through the use of adaptive PID controllers, fuzzy logic, and genetic optimization algorithms. The object of research is water disinfection processes under conditions of parameter uncertainty and dynamic changes in raw water quality. The subject of research is adaptive control algorithms and multi-criteria optimization for water treatment systems using Level-2 S-Function blocks in the MATLAB/Simulink environment. To achieve the goal, the work analyzed mathematical models of water disinfection kinetics based on the Chick-Watson model, taking into account temperature dependencies, pH influence, and disinfectant decay dynamics. Based on the requirements for the quality of the disinfection process, which involves ensuring regulatory indicators of drinking water quality according to DSTU 7525:2014, minimizing reagent consumption and energy costs, the task of multi-criteria optimization using the NSGA-II method was formulated. Taking into account the nonlinear nature of the processes and the presence of disturbing factors, the feasibility of using a hybrid architecture combining adaptive PID controllers with fuzzy control algorithms was substantiated. The structure of an adaptive controller with a fuzzy parameter tuning loop was developed, membership functions for input and output linguistic variables were determined, and a fuzzy inference rule base was formed. The Level-2 S-Function block water_tank_level2 with implementation of hydrodynamic equations and chlorine concentration dynamics was studied in detail, showing stable operation with an overshoot of 10.55% for water level and 2.06% for chlorine concentration. Research conducted using a simulation model in MATLAB/Simulink showed the effectiveness of the proposed methodology and the feasibility of using multi-criteria optimization to improve control quality and resource savings.