长春理工大学学报（自然科学版）
長春理工大學學報（自然科學版）
장춘리공대학학보（자연과학판）
JOURNAL OF CHANGCHUN UNIVERSITY OF SCIENCE AND TECHNOLOGY(NATURAL SCIENCE EDITION)
2014年
5期
141-145
,共5页
粒子群优化算法%RBF神经网络%水质评价
粒子群優化算法%RBF神經網絡%水質評價
입자군우화산법%RBF신경망락%수질평개
particle swarm optimized algorithm%RBF neural network%water quality evaluation
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。
將粒子群優化算法和RBF神經網絡相結閤，建立瞭基于神經網絡的水質評價模型，實現瞭對水質的閤理評價。通過採用粒子群優化算法對RBF神經網絡的參數進行優化，提高瞭神經網絡的收斂速度和精度，進一步提高瞭水質評價方法的精確程度。通過與傳統的神經網絡水質評價方法的對比，驗證瞭本文方法的可靠性和優越性。
장입자군우화산법화RBF신경망락상결합，건립료기우신경망락적수질평개모형，실현료대수질적합리평개。통과채용입자군우화산법대RBF신경망락적삼수진행우화，제고료신경망락적수렴속도화정도，진일보제고료수질평개방법적정학정도。통과여전통적신경망락수질평개방법적대비，험증료본문방법적가고성화우월성。
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.
