PMSM Direct Drive Elevator Door System Based On Fuzzy PI Parallel Control

The application of PMSM in elevator door control system is an inevitable trend. Compared with elevator door control system with traditional induction motor, PMSM door system does not install decelerator, which save space and makes the installation more flexible. But the lack of decelerator causes that the moment of inertia converted to the rotor shaft becomes larger. For door control system with Synchronous door knife, the moment of inertia changes greatly when the motor is loaded or not. Besides, The different weight of the door can also lead to large differences of moment of inertia. The changing of moment of inertia during a wide range will affect dynamic performance of the system. In order to solve this problem, this paper introduces fuzzy logic control to combine with the traditional PI control and designs Fuzzy PI parallel controller to expand the adaptability of the variation range of the door system's moment of inertia. It also improves the robustness and rapidity of the door system.First of all, this paper summarizes our elevator industry structure and the development tendency of door control system. Then it introduces composition of door system and function of each part. Based on the study of PMSM's mathematical model, speed- current double closed loop vector control system is given. We design fuzzy control rules according to the consequence of door system caused by variation of moment of inertia. Based on the rules, fuzzy PI control in parallel is designed. It is used for the speed control in the double closed loop vector control system. Simulation which reflects the real situation in use is done by the simulation model built in MATLAB / Simulink. The simulation results were analyzed, which demonstrates the effectiveness of the parallel fuzzy PI control. At last, based on theoretical analysis and simulation results on above, code a control progress, design three control systems including PI control, fuzzy control and fuzzy PI control in parallel, do engineering verification on door system for experiment, door system weight of 65 kg for use and door system weight of 140 kg for use respectively.Experimental results show that, PI control need to use two sets of parameters when moment of inertia in large changes, it has to at the expense of the rapidity of open process to ensure the stability of the close process. And the debugging of parameters is complex and hard to decide which group parameters are better. On the contrary, fuzzy PI parallel control can adapt to a larger range of variation of moment of inertia, the debug processer is easier, and the control effect is better.