Research On Current Loop Control Strategy With Excellent Performance Of AC Servo System

Nowadays,science and technology are gradually advancing,and the requirements of industrial production automation for AC servo systems are becoming more and more diversified and more stringent.AC servo systems need to work stably with superior performance in complex and constantly changing harsh environments.Automation companies urgently need AC servo systems that are highly efficient,low-cost,and can adapt to a variety of complex working environments.In practical applications,there will be interference from non-linear problems such as temperature floating,mutual friction of equipment,dead zone and saturation of equipment components,and the variability of the load.These few reasons make it impossible for the PID controller with fixed variables of the current loop to make the control system always run in the best condition.In the classic control strategy,researchers generally use the method of manually trying to compile PID control parameters.This strategy is difficult to operate and requires the relevant researchers to operate.It will not only consume a lot of time but also consume a lot of manpower,and is more efficient Low,difficult to operate in actual industrial applications.To this end,this subject explores and studies the parameter self-adjustment of the current loop PI controller of the AC servo drive system to obtain the optimal proportional coefficient K_pand integral coefficient K_i,so that the AC servo system can adapt to more complex working conditions.,In order to improve industrial production efficiency,reduce industrial production labor costs,and meet user requirements for high-performance indicators.The existing basic particle swarm optimization algorithm,with poor global convergence,may end the particle search too early in the actual industrial application practice,which is the more obvious fatal drawback of the existing basic particle swarm optimization algorithm.In this paper,we try to optimize the existing basic particle swarm optimization algorithm by combining randomly generated weight values with the principle of superiority and inferiority.However,the combination of randomly generated weight values and particle swarm optimization algorithm will make the speed of particle search efficiency worse,and the convergence efficiency of the algorithm decreases,this paper applies the principle of superiority and inferiority to solve this problem.This paper analyzes and explores the mathematical essence of the time-absolute error product integral ITAE equation,which in fact solves the magnitude of the absolute value of the area enclosed by the command signal straight line and the output signal curve.In this paper,the time absolute error product integral ITAE criterion is used as the fitness function of the particle swarm algorithm to find and filter the parameters of the current loop PI controller of AC servo system based on this deduction.In order to verify the degree of integration,implementability and adaptiveness of neural network control methods with classical control methods,this paper discusses and investigates the specific application and functional implementation of single neuron adaptive control and radial basis function neural network adaptive control in real servo current loop.In this paper,the self-tuning function of the parameters of the current loop PI controller is implemented in the AC servo control drive that has been commercially applied in industrial production.The parameters obtained by the self-tuning method in this paper enable the AC PMSM to operate stably at speeds far above the specified speed.