Research on Management of Lake Water Levels Based on Multi-Objective Genetic and PID Control Algorithm

Abstract

A comprehensive optimization approach utilizing a multi-objective genetic algorithm and PID control algorithm is proposed for addressing water level management in lake systems. Taking the Great Lakes of North America as a case study, this research investigates dynamic regulation methods for lake water levels. Through analysis of monthly water level data from the lakes, an optimal water level calculation method specific to each lake is determined. Subsequently, a water level optimization model is developed with primary objectives of ensuring drinking water supply, supporting navigation, and preserving the ecological environment. The non-dominated sorting genetic algorithm (NSGA-II) is employed to solve the model and determine the optimal water levels for the Great Lakes. Recognizing the challenge of simultaneously achieving optimal water levels in the actual Great Lakes system, a PID control algorithm is utilized for water level adjustments. By establishing a dynamic water level change model, a PID controller is designed and its parameters are calibrated through numerous experiments to effectively maintain water levels within the optimal range. This research offers a scientific foundation and practical methodology for enhancing water level management in the Great Lakes, thereby contributing significantly to the sustainable utilization of lake water resources.