摘要
Abstract
Firstly, based on seven aspects, namely the image, the expectation, the perception on power quality, the perception on quality of service (QoS), the perceived value, grumble and allegiance, an index system to comprehensively evaluate the satisfaction of electric power customers is established. Then the feasibility of optimizing BP neural network by fish swarm algorithm is analyzed and the procedures for the optimization of BP neural network by fish swarm algorithm are reseached. Finally, according to evaluation data of electric power customer satisfaction in five regions in 2009 and based on the assessment of expert scoring, the customer satisfactions are carried out by neural network and fish swarm-optimized algorithm neural network respectively. During the convergence there are 130 times for the former to close to the error figure about 0.1 and only 10 times for the latter to stay at local optima; when error figure is 0.001, after 168 times of training the former reaches the target and only after 88 times of training the latter reaches the target. The results show that the proposed method is effective for the evaluation on electric power customer satisfaction, and the method of fish swarm algorithm-optimized BP neural network is accurate, fast, simple and easy.关键词
鱼群算法/BP神经网络/电力客户满意度/综合评价Key words
fish swarm algorithm/ BP neural network/ electric power customer satisfaction/ comprehensive evaluation分类
信息技术与安全科学
