Research On Nonlinear Backstepping Controller Design And Optimization Of Permanent Magnet Synchronous Motor For Ship Electric Propulsion

The ship electric propulsion system is becoming the future trend of the ship propulsion system, due to the advantages of sving space, reducing fuel consumption and improving ship performance. The propulsion motor is a key component of the ship electric propulsion. The performance of the propulsion motor determines performance of the ship electric propulsion. The permanent magnet synchronous motor (PMSM) can meet the demands of the electric propulsion technology, due to the high energy density of the permanent magnet synchronous motor. The permanent magnet synchronous motor is a nonlinear and strongly coupled system with the load torque variation for the complex ocean environments. So the nonlinear control algorithm for permanent magnet synchronous motor is the key to improving the ship electric propulsion system performance.The nonlinear controller design for permanent magnet synchronous motor and the nonlinear controller optimization algorithm are studied, based on adaptive backstepping control, fuzzy control theory and particle swarm optimization (PSO). The main contributions are as follows:1. The model of ship electric propulsion system is built by analyzing the structure and principle of the various parts. The simulation results of the permanent magnet synchronous motor with the load torque disturbance and parameters uncertainty are the foundation of nonlinear adaptive backstepping controller design and optimization for permanent magnet synchronous motor.2. The adaptive integral backstepping controller with differential term for permanent magnet synchronous motor is proposed in order to solve the problem of load torque disturbance and d-q axis inductance uncertainty, and the stability of the designed sytem is analyzed. The adaptive law with the differential term and the control law with integral term of d-q axis current error are deduced by constructing the suitable Lyapunov function, and the anti-windup integrator is added to improve the dynamic performance. The controller is verified by simulation and experiment.3. Combining backstepping control with the fuzzy theory, the fuzzy self-tuning backstepping controller is designed for permanent magnet synchronous motor in order to improve the response of speed and the robustness. The parameters of backstepping controller are tuned by fuzzy inference module. The design method for fuzzy inference module is proposed by the condition of the motor running status analyzed, including the scale factors, quantization factors, membership functions and fuzzy rules. The controller is verified by simulation and experiment.4. It is difficult to tune the parameters of the adaptive integral backstepping controller with differential term, because of many parameters of controller. The adaptive weighted particle swarm optimization (AWPSO) algorithm for tuning the parameters of the controller is proposed. The ranges of the adaptive weight and the acceleration factors are determined by the convergence analysis of adaptive weighted particle swarm optimization. Subsequently the single-objective optimization algorithm for solving multi-objective optimization problem is proposed, and the fitness function is obtained by this method. The results of simulation and experiment show the effectiveness of this algorithm.5. A novel hybrid particle swarm optimization (HPSO) is proposed for the fuzzy self-tuning backstepping controller optimization, and the algorithm convergence is proved. The section of local best is added to standard particle swarm velocity update formula, using the adaptive acceleration factors and dynamic neighborhood topology, in order to prevent premature and guaranteed convergence. Meanwhile it has ability of local search and escaping from the local extremum, due to mutation operation and multiple restart mechanism. Combined with feature of fuzzy self-tuning backstepping controller, the search space, the inertia weight and ranges of acceleration factors are determined based on the hybrid particle swarm optimization and the convergence condition. The simulation and experiment results show the method can efficiently search the optimal parameters of controller.6. The permanent magnet synchronous motor experiment platform is built using the dSPACE sytem (DS1103 singl-board controller). The software and hardware are desgined for this platform. The torque disturbance is simulated by the torque variation of the load motor. The adaptive integral backstepping controller with differential term, the fuzzy self-tuning backstepping control and optimization algorithms for the controllers are verified by experiments based on this platform. Finally, research work presented in this dissertation has been concluded comprehensively and the further research aspect is pointed out.