A Bearingless Permanent Magnet Synchronous Motor And Its Nonlinear Control

Bearingless motors are a new type of motors combining the characteristics of conventional motors and magnetic bearings. In recent years, with the development of power electronic technology, digital signal processing technology, and modern control theory, the theory and related technology of bearingless motors are being continuously developed and improved. Bearingless motors have demonstrated the potential application prospects in canned pumps, high speed precision mechanical processing, aeronautics and astronautics, flywheel energy storage, life science, vacuum technique and so on. Compared with other types of bearingless motors, the bearingless permanent magnet synchronous motor （BPMSM）, due to its advantages of simple structure, small volume, light weight, reliable operation, high efficiency and high power density, has become one of the hot spots in the research field of bearingless motors presently. Under the susports of National High Tech Research and Development Plan of China （2007AA04Z213）, and Foundation Project of Technology Innovation for Graduate in Jiangsu Province （CX10B₂70Z）, a comparative systematical and deep study of the BPMSM is made, including mathematical model, electromagnetic analysis, nonlinear modelling of flux-linkage, nonlinear decoupling control, digital control system implementation and so on. The main research works of this dissertation are as follows:1. The Maxwell force and Lorentz force in bearingless motors are elaborated, and then the, producing principle of radial suspension forces in the BPMSM is introduced. According to the magnetic circuit principle, the inductance coefficients and the inductance matrix of the BPMSM is analyzed and deduced in detail using the analysis method. On the basis of the inductance matrix and the imaginary displacement principle, the analytical models of the radial suspension forces and electromagnetic torque are established.2. The method of finite element parameterized modeling for the BPMSM is proposed. Based on introducing the ANSYS finite element parameterized modeling method, the finite element analysis model of the BPMSM is built. By using the parameterized model, the size and parameters of the motor structure, the amplitude and phase of the torque winding currents and suspension force winding ones can be easily modified, which is beneficial to the analysis and design requirements of the BPMSM. Thus, the corresponding static electromagnetic characteristics, including magnetic field analysis, permanent magnet （PM） flux-linkage, back electromotive force （back-EMF）, winding inductances, radial suspension force, cogging torque, and static torque are deduced. Specially, the winding inductance characteristics are discussed in detail, including the static inductance and incremental one.3. A new nonlinear modeling method of the flux-linkage for the BPMSM based on least squares support vector machine regression is proposed. According to the strongly nonlinear relationships among the flux-linkage, rotor position angle, rotor eccentricity, torque winding current and suspension winding current, a nonlinear model of the BPMSM’s flux-linkage is established using the the least squares support vector machine regression method. The nonlinear modeling of the flux-linkage can character the nonlinear characteristics of the BPMSM, and has the advantages of high precision, quick convergence, and the powerful predictive ability. Moreover, the nonlinear model can be used for the online control of the BPMSM in the future.4. A novel decoupling control strategy based on neural network inverse system method is proposed for the BPMSM, which is a multi-variable, nonlinear and stongly coupled system. The reversibility of the BPMSM system is proved, and then a neural network is used to approach the inverse model of the BPMSM system. Combining the neural network inverse system with the BPMSM, the system is decoupled into two independent linear displacement subsystems, and a linear speed subsystem. The design of outer loop controller is easier, so the whole system control performance is further improved. The simulation results show that good decoupling effect and dynamic performance can be achieved by using the neural network inverse system control strategy.5. According to the deficiency of the traditional rotor-field-oriented control method for the BPMSM with the current regulated pulse width modulated （CRPWM） inverter, a control strategy based on the space vector pulse width modulation （SVPWM） method is proposed. In order to improve the control accuracy and response speed of the BPMSM system, the current closed-loops are adopted for both torque winding control and suspension force winding control. Based on digital signal processor TMS320F2812, a digital drive and control system is built, which supply a lot of information for the experimental platform development of high efficiency and high reliability motor drive. And then, the experimental studies are carried out, and the experimental results verify that the stable suspension and operation of the BPMSM can be successfully achieved by using the SVPWM control strategy. Finally, based on the digital drive and control system, detailed experiment scheme for nonlinear decoupling control by using the neural network inverse system method is proposed, which establishes a solid foundation for developing the high performance control of the BPMSM further.