成长鸡群优化RBF神经网络的亚健康诊断模型

Sub-health Diagnosis Model of RBF Neural Network Optimized by Growth Chicken Swarm Algorithm

针对RBF神经网络不能准确计算中心向量和节点宽度,导致RBF神经网络识别亚健康准确率达不到最优的问题,提出一种成长鸡群优化RBF神经网络的亚健康诊断模型.首先,对设备数据进行小波变换去除噪声,提取特征,将数据归一化处理;其次,用混沌搜索策略求得成长鸡群的初始种群;最后,对鸡群算法进行改进,将得到的优化参数输入到RBF神经网络模型进行训练,输出结果.解决RBF神经网络参数择优困难以及识别效果不佳问题.实验结果表明,提出算法收敛速度快、亚健康识别准确率较高.

Since RBF neural network cannot calculate the center vector and node width accurately,the accuracy of RBF neural network in identifying sub-health can't reach the optimal level.A sub-health diagnosis model of RBF neural network for Growth Chicken Swarm algorithm is proposed.Firstly,the equipment data is wavelet transformed to remove noise,extract features,and normalize the data;then,the chaos search strategy is used to obtain the initial population of the Growth Chicken Swarm Optimization;Finally,the optimal results obtained by Growth Chicken Swarm Algorithm are input into the RBF neural network model for training and output the results.It solves the problem that the RBF neural network parameters are difficult to select and poor identification.The experimental results show that the proposed algorithm has fast convergence speed and high accuracy of sub-health recognition.