Research On Overload Thermal Protection Of Asynchronous Motor Based On ANSYS

As a typical driving force and driving device,asynchronous motors are widely used in industrial production and life.It is of great theoretical value and practical significance to be able to accurately understand the variation law of temperature rise under asynchronous motor overload operation,to protect the motor from thermal overload,and to improve the safety and stability of its operation.This paper takes squirrel cage three-phase asynchronous motor as the research object,uses ANSYS to analyze the magnetic field and temperature field of the motor,and combines the neural network to predict the temperature rise of the motor,and studies the overload heat of the motor combined with the finite element analysis and the neural network.Protection method.main tasks as follows:(1)The two-dimensional model of the motor was established through ANSYS,and the relevant equivalent thermal conductivity was analyzed and calculated in detail.The meshing and loading of the motor were performed in the Workbench platform,and the transient thermal analysis of the motor was performed.The heat radiation model of the motor is established.Aiming at the problem that the load of the motor is constantly changing with time,the load-time curve is used in the transient thermal analysis to more intuitively express the relationship between the load and the time.Through the operation of ANSYS feeding back useful values after each step of calculation,the accuracy of the time integration algorithm is improved.(2)Based on the established two-dimensional finite element model of the motor,the magnetic field line distribution and temperature field distribution of the three-phase asynchronous motor under different load operation conditions are simulated and studied.The comparison between the actual data and the measured data shows that the current error of the motor under various operating conditions is ±0.05 A,the iron loss error is ±0.08 W,the copper loss error is ±0.4W,and the stator winding temperature error is ± 1.5 ?,In order to verify the validity of the simulation model and the correctness of the analysis method,it also provides data and model support for the intelligent prediction of motor temperature rise based on neural network.(3)In order to make up for the shortcomings of ANSYS analysis of the temperature field of the motor,which is generally only applicable to the design stage of the motor and has poor real-time performance,this paper proposes a neural networkbased intelligent temperature rise prediction method for the motor.The specific implementation steps of this method are: firstly,a neural network model for predicting the temperature rise of the asynchronous motor during overload operation is designed;secondly,by analyzing the influence of Levenberg-Marquardt,Fletcher-Reeves,PolakRibiere and quasi-Newton algorithm on the network prediction effect,The LevenbergMarquardt with the best effect is selected as the training method of this particular network;finally,the neural network after training is used to predict the temperature rise of the motor.The simulation results show that the method proposed in this paper can control the temperature rise error of the squirrel-cage three-phase asynchronous motor within ±0.2°C,so as to realize the overload thermal protection of the motor.The temperature prediction provides a theoretical basis.