Title: Evaluating electric environmental issues using BP neural network with optimised hidden layer nodes
Authors: Dongxiao Niu, Jialiang Lv, Hongyan Liu
Addresses: School of Business Administration, North China Electric Power University, Baoding City, Hebei Province 071003, China. ' Department of Business Administration, North China Electric Power University, Baoding, Hebei, China. ' Department of Business Administration, North China Electric Power University, Baoding, Hebei, China
Abstract: BP neural network with optimum hidden layer nodes were used to evaluate the power environmental issues in China. Furthermore, Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) method is used to optimise the number of hidden layer nodes of BP neural network. The evaluation indices were determined through the training of the BP neural network with the samples. This new method is validated through a case study where power environmental issues were evaluated, and high accuracy rate and simple computation complexity were achieved. Therefore, the number of hidden layer nodes can be determined more scientifically and evaluation results derived by our method tend to be more accurate.
Keywords: TOPSIS; back propagation; neural networks; hidden layer nodes; combination weight; environmental issues; electricity markets; power supply; China.
International Journal of Global Environmental Issues, 2009 Vol.9 No.3, pp.227 - 238
Published online: 09 Jul 2009 *Full-text access for editors Access for subscribers Purchase this article Comment on this article