Design Of Speed Sensorless Control System For Asynchronous Motor Based On Passivity

The asynchronous motor has been widely used in industrial production and daily life due to its simple construction, high reliability and low cost, etc. Directing at the properties of asynchronous motor such as nonlinear and strong coupling, there are many control methods and strategies. Passivity-based Control is a nonlinear method which researches system problems based on the viewpoint of energy, and it has advantages of simple control laws and strong robustness when applied in motor control. In recent years it's often to be used in the fields of power electronics and electric drive combined with other methods. Furthermore,speed information requires to be fed back in real time in the closed-loop speed systems,however using speed sensor directly will not only increases the cost of system, but also be restricted to installation technologies at the same time. It has become a goal and direction of AC drive field that how to estimate speed with high precision and low cost, and the creation of speed sensorless technology solves this problem effectively. This dissertation aims at these two hot research topics, combines passivity-based control system with speed sensorless technology, then analyzes and researches thoroughly to this. The specific contents are as follows:(1) In order to facilitate analysis, this dissertation studies mathematical models of asynchronous motor in different coordinate systems and coordinate transformations between them at first. Then it introduces the principles and concepts of passivity detailedly, points out the connections between passivity and stability of the system, derives the conclusion that the motor system satisfies to passivity theorem. And system controllers are designed using passive principles of energy-shaping and damp-injection, the control laws and block diagrams of system are provided in the meantime.(2) As speed estimation and flux observation are keys of speed sensorless method, and the flux observation is accurate or not will influence to the performance of speed estimation system directly, therefore on the basis of compared to the basic voltage and current model flux observers, this dissertation focuses research on the full-order flux model observer,detailedly designs its structure, derives the adaptation laws of speed estimation system based on this method. Then it studies on the stability of designed system, and parameters sensitivity of the full-order flux model observer, offers important reference value to design systems and set parameters.(3) The PI parameters in speed adaptation laws are closely related to the system's performance, and it needs long time to debugging and simulation that select value artificially.This dissertation uses Genetic Algorithm to optimize the PI value for its advantages in parameters self-optimization, this method could reduces blindness of selecting valueartificially in some degree, and it offers optimal value which will make the system perform better.(4) Simulation models of designed systems are built in MATLAB, the correctness and validity have been verified by experiment simulations, the results are analyzed.