Study On Electromagnetic Loss And Thermal Model Of Permanent Magnet Synchronous Motor Used In Drive System

Permanent magnet synchronous motor has been widely used in electric drive systems such as electric vehicles and ships due to its advantages of high power density,high efficiency,and wide speed range.At present,high-speed motor is an important development trend of permanent magnet synchronous motor.However,as the speed increases,motor iron loss and copper loss increase rapidly,which leads to excessive temperature rise of the motor.As a result,short-circuit of windings and irreversible demagnetization of permanent magnets may occur.Therefore,it is of great significance to calculate or predict the motor loss and temperature field and analyze its influence factors.However,most of the existing studies only focus on the influence factors of loss,and few literatures analyze the factors on the motor temperature field.Finite element analysis is an important method for the study of motor electromagnetic field.Based on the commercial software Motor-CAD,a finite element model of the motor is established.The variation of motor iron loss and output performance from 11 factors of control parameters,structural parameters,and material parameters are obtained and analyzed by this model.A finite element model of windings is also established to analyze the influence factor on winding AC lossBased on the basic theory of heat transfer,the main heat transfer paths in the motor is analyzed and a lumped parameter thermal network model of the motor is established to calculates the temperature distribution in the motor based on the loss calculation results.Different influence factors on motor temperature distribution are also analyzed by calculating this model.The research results have reference value for the motor design and thermal management.The neural network model can solve complex non-linear problems and does not need to establish a specific model.There have been few applications in the prediction of motor temperature by neural network model,but no research has been done on the prediction of electromagnetic loss.In this paper,4 and 3 layers of BP neural networks are built respectively to train,predict and verify the electromagnetic loss and the temperature of each part.The results show that this model can achieve a good prediction accuracy which provides an efficient and simplified way to calculate the electromagnetic loss and temperature distribution of the motor.