摘要
在能耗制动方式下,制动时间是影响异步电机停车定位精度的重要参数。针对制动时间与其影响因素之间的非线性关系,提出基于BP神经网络的制动时间预测方法。通过实验采集样本数据,并对其做归一化处理;然后建立预测制动时间的BP网络模型;最后利用训练样本对BP网络进行训练,并将测试样本输入已经训练好的网络进行仿真。仿真结果表明,该方法有很好的预测效果。
The braking time is an important parameter which affects the positioning accuracy of asynchronous motor under the energy consumption braking mode.According to the nonlinear relationship between the braking time and its influencing factors,a braking time prediction method based on BP neural network was proposed.The samples were collected by experiment and normalized.The BP neural network model was established to predict braking time.The training samples were used to train the BP network,and the test samples were input into the trained network for simulation.The simulation results show that the method has good prediction effect.
作者
韩亮
HAN Liang(Department of Electrical Engineering,Shanxi Institute of Mechanical and Electrical Engineering,Changzhi Shanxi 046011,China)
出处
《机床与液压》
北大核心
2018年第19期56-58,共3页
Machine Tool & Hydraulics
基金
山西省高等学校科技创新项目(20161128)
关键词
BP神经网络
异步电机
制动时间预测
BP neural network
Asynchronous motor
Braking time prediction
