摘要
针对常规遗传算法的缺陷,提出了一种基于改进遗传算法和RBF神经网络相结合的控制方法。该方法对RBF神经网络的隐层中心值和宽度进行了优化,用递推最小二乘法训练隐层和输出层之间的权值。最后在双电机驱动伺服系统中进行了仿真试验,结果表明所提出的控制策略是有效的。
According to the deficiency of conventional genetic algorithm（GA）,a new control method based on improved GA which combining with RBF neural network is proposed.The method was used to optimize the centers and widths of RBF hidden layer.Recursion least square method was used to train the weights between hidden layer and output layer.Finally,the simulation experiment in dual-motor driving servo system shows the effectiveness of the proposed control strategy.
出处
《井冈山大学学报（自然科学版）》
2011年第4期76-80,共5页
Journal of Jinggangshan University (Natural Science)
基金
安徽省高校自然科学研究项目(KJ2011B186)
