Integrated Structure And Control Design Of Multi-motor Servo System Based On Multi-objective Optimization

With the rapid development of servo technology,single motor servo system is more and more difficult to meet the demand of high power and large inertia servo system.Thus,using multi-motor servo system to drive the load becomes one key research and development direction of the servo system.However,how to design a controller to guarantee the tracking performance of the system,how to achieve fast synchronization of multi-motor,and how to reduce the energy consumption of the system are the technical difficulties of multi-motor servo system.In addition,traditional design method of controller is "throw off the wall",which designs structural parameters firstly and control parameters secondly.This will inevitably lead to the mismatch of the structural part and the control part,so that the system’s tracking performance,synchronization performance and energy consumption cannot achieve the best performances.This paper mainly investigates the control parameters and structural parameters integrated design method for the four-motor servo system.The main research contents and achievements are listed in the following:(1)According to the backlash nonlinearity of the four-motor servo system,mathematical model of four-motor servo system containing backlash is established by adopting the dead zone function as the backlash model.The PID controller and the constant bias torque are designed to realize the load tracking,simulation results have been conducted to show the drawbacks coming from the separately design of structural parameters and control parameters for the servo system.(2)The integrated structure and control design of four-motor servo system is investigated based on multi-objective optimization of weighted sum.Four-motor servo system’s tracking performance,synchronization performance and energy consumption are summed based on the weights determined by delphi method,to reduce the number of those.Then tracking controller’s PI parameters,angle difference bias torque parameters,synchronization controller’s PI parameters and structural parameters are integrated design by using the improved harmony algorithm.Simulation results show that this approach can search an optimal solution among the good system performances.(3)To handle the subjectivity problem and the single optimal solution problem by adopting the delphi method and weighted sum method,an improved NSGA2 algorithm is developed by introducing the Pareto optimal concept to search the optimal solutions of the tracking controller parameters,bias torque parameters,synchronization controller parameters and structural parameters in the integrated design and obtain the Pareto optimal set.The unique optimal solution is selected from the Pareto optimal set by the decision technology.Simulation’s result indicates this algorithm is very effective.(4)Aiming at the measured system states and the unknown nonlinearities,an extended state observer is designed to combine with the sliding mode control or the backstepping control to achieve accurate load tracking of the four-motor system.Moreover,based on the design of the controller,the multi-objective optimization algorithm is used to optimize the structural parameters and control parameters simultaneously.Therefore,the integrated structure-control design of four-motor servo system can be realized.