Quantitative Light Pollution Analysis Based On K-Means++ Cluster Analysis and Neural Networks

Abstract

The purpose of this study is to investigate the problem of light pollution, to quantitatively analyze a wide range of light pollution and to simulate the effect of regional control strategies. First, TANGY model was introduced to transform the light pollution evaluation from simple luminous flux calculation to comprehensive evaluation. TOPSIS evaluation method optimized by entropy weight method and BP neural network optimized by genetic algorithm were used to calculate the light pollution index. Secondly, three strategies to solve light pollution are proposed, and the implementation model is constructed by cluster analysis and regression modeling. The optimal strategy is determined by gradient enhanced decision tree algorithm, and the prediction model of light pollution intervention strategy is constructed. This study provides an effective reference for the formulation of light pollution prevention strategies in different regions, and is expected to have practical significance in the field of environmental protection.