Research On Control Strategy Of Linear Induction Motors Based On Model-Free Technology

In order to accelerate the construction of the economic basement and superstructure,and to achieve the “13th Five-Year Plan” faster and better,more and more special motors are produced and manufactured and applied in practice.As one of them,linear induction motor has the advantages of strong climbing ability,low working noise,small turning radius and less emission pollution.It has been splendid in urban rail transit,elevator transportation,household appliances,and national military.However,due to the unique end effects of the LIM,the coupling between the state quantities is more serious,and the mathematical model becomes more complicated than the conventional rotary motor.As a typical method in the field of data-driven,the model-free adaptive control does not rely on the internal model of the controlled system and relies only on the input and output data of the system.It has the advantages of simple controller design,strong adaptability and fast self-learning.It has been practiced in the field of smart parking,multi-agent,aircraft control and other fields.Based on the above social and scientific research background,together with the theory of model-free control method,this paper designs a series of schemes to effectively control the LIM considering the load variation caused by the actual working condition of the motor,the time-varying system parameters caused by the temperature change,the insufficient modeling,and the input saturation phenomenon.The main research contents of this paper are as follows:Firstly,based on the existing research literature,this paper summarizes the basic structure and characteristics of several kinds of LIMs,and studies the basic model of LIM under the consideration of its unique end effects.According to the classical indirect vector control principle,the mathematical model of the d-q axis LIM is established.Two representative simulation examples verify the effectiveness of this method.However,the problems of slow follow-up and low control accuracy need to be solved.Secondly,velocity as the most important parameter of the motor,is the primary goal of this paper.After transferring the mathematical model of the d-q axis LIM to a Nonlinear AutoRegressive with eXogenous input(NARX)model.the NARX model is linearized by the compact form dynamic linearization(CFDL)technique.Considering the load disturbance phenomenon existing in the actual operation process,the load disturbance is considered as the system input.In the controller design process,the error surface is changed into the integral sliding mode surface,and an improved model-free adaptive integral sliding mode control method is adopted to ensure the fast and stable follow-up performance.In addition,considering the input saturation problem of the actuator,an anti-windup compensator is introduced to solve the problem of output non-following due to input saturation.Thirdly,considering the LIM as a high-order system,the control performance of the system will be more significant when the controller of each state is designed and the stability of the entire closed-loop system is guaranteed.Using the discrete-time backstepping method,a model-free adaptive control method for high-order systems is established.In order to solve the computational expansion and non-causal problems inherent in discrete-time backstepping schemes,discrete-time command filters and filter compensators are introduced.At the same time,anti-saturation compensators are designed for the input saturation problem in the design process of each order controller.Simulation examples demonstrate the effectiveness of the scheme.Finally,because LIM is a multi-input multi-output system,it is obviously impractical to consider velocity as the sole target of control objective.Therefore,the flux linkage is the secondary goal of control design.Matching the velocity and flux linkage as the output matrix,applying the pseudo-partial derivative matrix,and introducing the anti-saturation compensator matrix,the model-free adaptive integral sliding mode control is established for multi-input multi-output systems.