Research on Lake Water Level Control Based on Simulated Annealing Multi-Objective Planning Modeling

This paper focuses on the water level management problem of the Great Lakes, and solves the problem of solving the water level control problem by constructing a network flow model and an intelligent perception model. First, we constructed network maps of the Great Lakes and rivers using the collected water level data, and derived the optimal water level of the Great Lakes at any time of the year through a multi-objective planning model based on simulated annealing. Based on this, we constructed network flow maps and optimization-oriented models for the lakes and rivers. Next, we gradually added dam influences and continued with simulated water level control. Evaluation based on the entropy weight-TOPSIS model yielded an optimization model score of 0.743, while the actual water level score was 0.428, which indicated that our model better controlled the water level. Finally, based on the AHP model in order to determine the weights of each stakeholder, it was found that shipping companies have the greatest influence on the water level of Lake Ontario. Also, through multiple linear regression modeling, we found that the four major lakes and the St. Lawrence River have a greater influence on Lake Ontario.