基于神经网络的异步电动机温升预测分析  

Prediction and Analysis of Temperature Rise of Asynchronous Motor Based on Neural Network

作　　者：王水发[1] 陈德为[2] Wang Shuifa;Chen Dewei(Fujian Chuanzheng Communications;Fuzhou University)

机构地区：[1]福建船政交通职业学院 [2]福州大学

出　　处：《哈尔滨师范大学自然科学学报》2020年第1期27-31,共5页Natural Science Journal of Harbin Normal University

基　　金：福建省中青年教师教育科研项目(JZ180348)。

摘　　要：异步电动机绕组温升受电机功率、负载、结构、材料和外部环境等众多因素的影响,导致其温升难以用传统的公式进行计算和预测.针对这一问题,基于大量的实验数据,建立并优化了异步电动机温升的神经网络数学模型.采用Matlab软件中的神经网络函数,编写计算程序训练学习网络模型.最终验证了神经网络模型在异步电动机温升的预测是可行的,精度基本满足实际应用需求.这不仅为异步电动机温升的预测预报和优化设计奠定基础,而且对于其他电机电器的研究具有重要的现实意义.The winding temperature rise of asynchronous motor is affected by many factors such as power,load,structure,material and external environment,which makes it difficult to calculate and predict with traditional formula.To solve this problem,in this paper,the neural network mathematical model of asynchronous motor temperature rise is established and optimized,which is based on a large number of experimental data.Using the neural network function in the software of matlab,the calculating program is compiled to train and learn the network model.Finally,it is verified that the neural network model is feasible in predicting the temperature rise of asynchronous motor,and the accuracy can basically meet the practical application requirements.It not only lays a foundation for predicting and optimizing the temperature rise of asynchronous motor,but also has important practical significance for the research of other motor and electrical appliances.

关 键 词：异步电动机 神经网络 温升 预测 

分 类 号：TM3[电气工程—电机]