摘要
针对传统预测冲击矿压方法存在网络训练时间较长、容易陷入局部极小值、容易早熟等问题,提出了采用克隆选择遗传算法预测冲击矿压的方法,详细介绍了克隆选择遗传算法优化BP神经网络权值和阈值的基本步骤,并利用Matlab7.1在PC机上建立网络模型进行仿真实验,仿真结果表明,该方法有效提高了冲击矿压预测的准确性。
To solve problems of long time of network training, easy to fall into local minimum and easy to early maturity existed in traditional method of impulsion pressure prediction, the paper proposed a method using CLGA to predict impulsion pressure. It introduced basic steps using CLGA to optimize weight and threshold of BP neural network in details and made simulation experiment using MatlabT. 1 to build network model on PC. The simulation result showed that the method can improve accuracy of impulsion pressure prediction effectively.
出处
《工矿自动化》
2010年第3期39-41,共3页
Journal Of Mine Automation
基金
辽宁省教育厅基金资助项目(20060392)
