Uniform Demagnetization Fault Diagnosis Research Of Permanent Magnet Synchronous Motor Based On ALN-MRAS Flux Observer

Permanent Magnet Synchronous Linear Motor(PMSLM)has the advantages of direct displacement method without intermediate conversion mechanism,high thrust density,high positioning accuracy,which satisfy high speed large stroke motion and micro level dynamic accuracy.It has important application value in the various processing field of laser cutting machine and high-end CNC machine tools.The magnetic field size of PMSLM permanent magnets is often the key factor that determines the operating performance of the motor,and the performance of the motor directly affects the quality of product production and processing.The permanent magnet of PMSLM is made of neodymium iron boron material.Under the severe working conditions such as high temperature,overload,poor heat dissipation conditions,etc.In addition,the natural aging of the permanent magnet and the corrosion of chemicals,etc.,can easily cause the permanent magnet of the motor to occur.Irreversible demagnetization failure,which directly affects the output performance of the motor,which in turn affects the stability and high-precision output of the entire servo control system.Therefore,it is necessary to perform online demagnetization fault diagnosis on the motor and accurately identify the degree of demagnetization,so as to adopt fault-tolerant control compensation measures for the entire servo system in time to achieve the maintenance of motor positioning accuracy and high-quality thrust output.This subject takes the dual secondary ironless core PMSLM applied in the laser cutting machine as the research object,to solve the problem of uniform demagnetization fault diagnosis and demagnetization degree estimation under the condition of multiple working conditions of the motor,and carried out the following research:(1)Aiming at the problems of PMSLM uniform demagnetization fault diagnosis and demagnetization degree identification,design PMSLM uniform demagnetization fault diagnosis system.According to the characteristics of PMSLM's topological structure and the description of the control system,the problems to be solved for the uniform demagnetization fault are analyzed.By analyzing the problem,the function of designing a uniform demagnetization fault diagnosis system and the main components of the design of a uniform demagnetization fault diagnosis system are designed.(2)A magnetic flux observer based on neural network algorithm is studied to realize real-time online monitoring of magnetic flux under multiple operating conditions.Based on three common coordinate transformation relationships,the mathematical model of PMSLM in a synchronous rotating coordinate system is derived.Introduce the principle of sliding mode control,design the speed controller of the PMSLM control system,and prove its stability using the Lyapunov stability principle.Based on the principle and structure of the model reference adaptive system,the MRAS flux linkage observer is established,and the adaptive law satisfying the stability of the entire system is solved according to the Popov superstability principle.An adaptive linear neuron algorithm based on neural network,which can adjust the weight online,is used to design the adaptive law of MRAS,and establish the ALN-MRAS flux linkage observer.By analyzing the influence of resistance and inductance errors on flux linkage observation,a feedback gain matrix is designed and the parameters are tuned.(3)A method of uniform demagnetization fault presetting is studied,and the accurate value of flux linkage under different demagnetization degrees is simulated to provide a reference value for the fault diagnosis system to realize demagnetization fault diagnosis and demagnetization degree identification.According to the modeling process of the finite element method,the PMSLM finite element model under normal conditions is established,and the correctness of the built model is verified by simulation.Through calculation,the Br parameter of all permanent magnets is changed to achieve uniform demagnetization fault presetting.The PMSLM finite element failure models under different degrees of demagnetization were established,and the simulation results obtained the accurate value of flux linkage under different demagnetization failures.(4)The PMSLM-based uniform demagnetization fault diagnosis algorithm is designed to realize uniform demagnetization fault diagnosis and demagnetization degree identification.According to the degree of uniform demagnetization,four types of demagnetization faults are defined.The PMSLM uniform demagnetization fault diagnosis algorithm is designed by combining the flux linkage values observed by the ALN-MRAS flux linkage observer in real time,and the precise flux linkage values at different demagnetization degrees obtained by the finite element preset fault.By comparing simulation experiments,the accuracy and performance of the ALN-MRAS flux linkage observer under multiple operating conditions are verified.(5)A PMSLM demagnetization experiment platform based on the AD5435 hardware-in-the-middle was built,and a uniform demagnetization failure prototype was made.Two types of faults,weak demagnetization and severe demagnetization,are set.Thin blocks(1,2,3mm)of different thicknesses are printed by a 3D printer,and the thin blocks are placed under the back iron,by increasing the air gap between the upper and lower permanent magnets.Under single-condition and multi-condition conditions,the unmodified flux linkage observer is introduced for comparison,and the demagnetization prototype is tested respectively.The experimental results are analyzed to prove the high efficiency of the ALN-MRAS flux linkage observer studied in this topic.In summary,the ALN-MRAS flux linkage observer studied in this topic can not only observe the flux linkage value online in real time,and the observation accuracy is as high as 98% or more,and under multiple operating conditions,the fluctuation of the flux linkage observation error is small,Load adaptability and speed adaptability are better.Therefore,based on this method,a PMSLM uniform demagnetization fault diagnosis system is designed.The demagnetization fault is preset by finite elements to obtain the accurate value of the flux linkage at different demagnetization degrees.Using the uniform demagnetization fault diagnosis algorithm,not only the demagnetization fault diagnosis,but also the realization the degree of demagnetization is estimated.