基于粒子群优化算法和RBF神经网络的水质评价方法研究  

Reach of Water Quality Evaluation Method Based on Particle Swarm Optimization and RBF Neural Network

作　　者：宫玉琳[1] 文大化[2] 张静鹤 

机构地区：[1]长春理工大学电子信息工程学院,长春130022 [2]中国科学院长春光学精密机械与物理研究所,长春130033 [3]空军航空大学,长春130062

出　　处：《长春理工大学学报（自然科学版）》2014年第5期141-145,共5页Journal of Changchun University of Science and Technology(Natural Science Edition)

摘　　要：将粒子群优化算法和RBF神经网络相结合,建立了基于神经网络的水质评价模型,实现了对水质的合理评价。通过采用粒子群优化算法对RBF神经网络的参数进行优化,提高了神经网络的收敛速度和精度,进一步提高了水质评价方法的精确程度。通过与传统的神经网络水质评价方法的对比,验证了本文方法的可靠性和优越性。In this paper,a water quality evaluation model was established based on neural network combined with parti-cle swarm optimization algorithm and RBF neural network,realized the reasonable evaluation of water quality. By opti-mizing the parameters of the particle swarm optimization algorithm of RBF neural network, the neural network im-proves the convergence speed and accuracy, to further improve the accuracy of water quality evaluation methods. By comparing the neural network water quality evaluation with the traditional method, the presented method is reliable and superiority.

关 键 词：粒子群优化算法 RBF神经网络 水质评价 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]