Fuzzy logic and multi-objective optimization-based water level regulation model for the Great Lakes of North America

Abstract

In this study, a series of models was developed to ascertain and maintain the optimal water levels of the Great Lakes, aiming to reconcile the diverse requirements of stakeholders. Initially, a fuzzy logic algorithm was applied to calculate the weighting coefficients of five stakeholder categories, facilitating the creation of a multi-objective optimization framework to harmonize all demands and thereby determine the optimal water level for each lake. Subsequently, an analysis of the geographical and meteorological data of the Great Lakes informed the construction of a dynamic equilibrium model for water levels, grounded in the calculation of influx and efflux rates to sustain the optimal water level. Following this, simulated tests were conducted using historical data, with Principal Component Analysis (PCA) utilized to assess the model’s pre-optimization and post-optimization, providing a clear comparison of outcomes. This article aims to develop and apply optimization models to balance the diverse needs of stakeholders in the Great Lakes, manage water levels scientifically, enhance water resource management efficiency, and promote regional sustainable development.