Research On Stability And Dynamic Performance Improvement For Sensorless Induction Motor Ultra Low Speed Operation

Induction motors have been widely used in the industrial applications for the advantages of simple structure,high reliability,low cost,etc.Speed sensors are commonly used in the industry to estimate the motor speed for speed closed-loop control.However,the speed sensors not only increase system cost and size,and reduce system reliability,but also limit system applications where encoders cannot be installed.The speed sensorless control system removes expensive and fragile speed sensors and enhances the system reliability.Thus,it is a new research field for the industrial development.However,the speed sensorless induction motor vector control system requires high accuracy of speed information.While the traditional methods have an unstable region at ultra-low speed operation,which cannot guarantee the long-time stable operation of induction motors.Besides,the dynamic performance is usually ignored.Thus,it is still very challenging for the control of sensorless induction motor at ultra-low operation.For this challenging,the main contents of this paper are as follows:First,based on the induction motor model in the three-phase static frame,the dynamic mathematical model of induction motor in the two-phase synchronous frame is derived according to the coordinate transformation theory.On this basis,the induction motor vector control system is designed based on the rotor field oriented control.The adaptive full-order observer is constructed using the induction motor state equation,and the speed adaptive law is derived according to the Lyapunov stability theory.Four different feedback gain design methods are analyzed for adaptive full-order observer.The analysis shows that the first three feedback gain design methods cannot guarantee the stability of the full-order observer in the full speed range,while the fourth method can guarantee the stability based on the principle of speed estimation stability configuration.But it ignores the dynamic performance of the full-order observer.Second,the existing feedback gain design methods of adaptive full-order observer are purely based on stability theory and ignore the dynamic performance.Considering this problem,this paper proposed a novel feedback gain design method with variable coefficient.So all zero poles of the observer have negative real parts.On this basis,the range of the variable coefficient is further limited by the analysis of the zero diagram and the bode diagram.Then,the stability and dynamics of the full-order observer can be guaranteed simultaneously.Furthermore,the robustness of the proposed method is analyzed.Finally,the experimental results show that this method can improve the instability and dynamic performance of the sensorless induction motor in the low-speed power generation regionFinally,an auxiliary variables-based method is proposed,which can guarantee the stability of the system in the full speed range without complex feedback gain design.The mathematical model of induction motor and adaptive full-order observer are constructed with the defined auxiliary variable.The speed adaptive law of the extend speed adaptive observer is deduced by Lyapunov stability theory.The stability of the observer is analyzed by the zero-pole diagram.Then,the speed adaptive law is further optimized to solve the problem of estimating speed error caused by ignoring the error term of auxiliary variables.The experimental results show that compared with the traditional full-order observer design method,the proposed method can improve the system stability effectively without the feedback gain design.