| As the most common and important electric-driving mechanical equipment in modern industry,motor is the foundation of human society into the electric age.The problems of its operation life,work efficiency and maintenance cost have a great influence on the economic development of the whole society and country.The bearingless permanent magnet synchronous motor(BPMSM)is a new type of motor which integrates magnetic bearing technology with traditional motor.It has the advantages of no friction,no lubrication,long working life,high speed and high precision,which is suitable for a variety of high-speed and high-precision industrial areas.However,the bearingless permanent magnet synchronous motor is a multi-variable and nonlinear system.The decoupling control method to realize the stable suspension and rotation at the same time,as well as the accurate detection of rotational speed and radial displacement have attracted attention of scholars.In this dissertation,to solve the above problems,a bearingless permanent magnet synchronous motor control system based on fuzzy neural network and right inverse system is proposed to realize the decoupling control and the research on the soft sensing of the BPMSM based on the neural network and left inverse system is carried out.The project is supported by Key Research and Development Program of Jiangsu Province(BE2016150)and Jiangsu Province“Qinglan Project”(2014).In this dissertation,the structural characteristics,principle of the internal electromagnetic force,mathematical model,theory of the decoupling control and soft sensing of the BPMSM have been researched.The main works and achievements are as follows:1.By introducing the basic structure,interaction with the internal electromagnetic force and suspension forces operation principle of the bearingless permanent magnet synchronous motor,the mathematical model of suspension force and torque model of the bearingless permanent magnet synchronous motor are deduced and established in this dissertation.2.Aiming at the characteristics of nonlinear and strong coupling of the BPMSM,a novel decoupling control method based on the fuzzy neural network with right inverse system isproposed.Based on the reversibility analysis,by combining the fuzzy neural network,the right inverse system is designed and series connecting it with the original system,the BPMSM control system is pseudo-linearized into two low order linear control systems,and the closed-loop controllers are designed to realize the decoupling control of the suspension and rotation of motor rotor.The simulation model of the decoupling control system is constructed by MATLAB/Simulink software,and the simulation results show that the proposed method is effective.3.In order to improve the detection accuracy and range of the BPMSM rotational speed and radial displacement,the research on the soft sensing based on the left inverse system and neural network is carried out in this dissertation,which avoids the defects of traditional sensors.Through the principle of the left inverse system,the accurate observation with the signal of BPMSM speed and radial displacement can be realized and the neural network can effectively solve the coupling and nonlinear relationship between variables of the left inverse system.The simulation model of the detection system is constructed by MATLAB/Simulink software,and the reliability and accuracy of the method are verified by the simulation results.4.According to the working principle and control requirement of the BPMSM,the software and hardware of the digital control system of motor are designed with TMS320F2812 as the CPU of digital control.Taking the program algorithm in DSP as the core,the feedback signal is sampled,analyzed and processed through the interface circuit,and the current phase and size of the winding are modulated by PWM wave to realize the stable suspension and rotation of the motor rotor,and the variable speed and load experiments are processed to verify the feasibility of the control system. |
PDF Full Text Request