摘要
为了对刮板输送机负载进行更有效的监测与控制,建立了基于BP神经网络的电流预测模型,结合自适应学习速率和附加动量法改进标准BP算法中连接权值和阈值的梯度下降更新算法,仿真实验表明平均相对预测误差为1.280 3%,效果良好。在刮板输送机WinCC远程监控平台中实现所述预测功能,结果显示,该系统能够实时预测工作面刮板输送机的运行电流。
In order to monitor and control the load of the scraper conveyer effectively, the paper established the current prediction model based on BP neural network, and improved the gradient descending updating algorithm of weights values and thresholds in standard BP algorithm by applying adaptive learning rate and additional momentum method. Simulation results showed the average relative prediction error was 1.280 3% with excellent effects. The prediction function was realized in Win CC remote monitoring platform of the scraper conveyor, the results showed the system online predicted the operation current of the scraper conveyor in working face.
出处
《矿山机械》
2015年第10期17-21,共5页
Mining & Processing Equipment
