Evaluation of Regional Water-Saving Level Based on Support Vector Machine Optimized by Genetic Algorithm

Abstract

The evaluation of regional water-saving level can provide scientific theoretical support for steadily promoting the implementation of a national water-saving priority strategy. Referring to the water consumption statistics of 31 provinces (except Hong Kong, Macao and Taiwan) in China in 2018, 14 easily accessible and comprehensive indexes were selected to establish an index system of regional water-saving level and a water-saving level evaluation model based on support vector machine optimized by genetic algorithm (GA-SVM) was constructed to analyze the national regional water-saving level from different perspectives. The results showed that the water-saving level in China presented a spatial distribution characteristic with Beijing City, Henan Province and Zhejiang Province as the center and gradually decreased outward. From the perspective of regionalization, the water-saving level in North China, Central China and Southeast China was higher, while the water-saving level in Northwest China, Southwest China and Northeast China need to be improved. Therefore, the national water-saving level is generally at a medium level and effective water-saving work and water-saving schemes should be carried out according to different regions and industries.

Keywords

• genetic algorithm
• support vector machine
• index system
• water-saving level