Study Of Speed Sensorless Vector-Controlled Servo Drives Of IM

Several issues in the speed sensorless flux-oriented vector-controlled induction motor (IM) servo drive system are studied in this paper, which include the asynchronous motor parameter identification at standstill, the rotor flux observation, rotor speed estimation, the selection of the reference flux, the control of the flux, speed and the current, and the implementation of the speed sensorless vector drives based on DSP.First, a review of the development of power electronics and the control of AC drives is presented. Some difficulties are pointed out in the sensorless vector controlled induction motor servo drives.On the basis of the induction motor (IM) models in the field orientation, a new approach to identify the parameters of induction motors at standstill which is based on the discrete model reference adaptive system (MRAS) is proposed. From the IM state equations, the identification model is obtained and the recursive identification algorithm is derived based on discrete MRAS principles. The design methodology such as the choice of test signal, sampling frequency and initiative value of the gain matrix is given. The simulation and based on DSP is implemented to validate the good results of the IM parameter identification.A closed-loop rotor flux observer with nonlinear speed dependent gain is presented after the discussion of the parameter sensitivity of open-loop rotor flux observer based on voltage and current models of IM. The nonlinear gain which is characterized by two special frequency values, comer frequency and break frequency, determines the bandwidth of the closed-loop observer and allows for a exponential transition to take place from current model to voltage model which makes a best use of the rotor flux estimating attributes of two models. In (?)e speed estimation, two methods frequently used in speed-sensorless vector drives which are based on speed dependent phenomenon and spatial saliency phenomenon are discussed. In the back EMF based speed estimation, a closed-loop flux observer MRAS (MRAS-CLFO) method is used. Its equivalent PID controller parameters are derived by using small signal analysis. The relation of speed estimation performance with the PID controller is analyzed. In the space based speed estimation, an adaptive notch filter is designed to extract the fundamental frequency from the rotor slot harmonics under steady state. This additional speed information is used to tune the mechanical system parameters and rotor time constant and the parameter correction algorithm is given.The optimal selection of the rotor flux reference is discussed based on standard differential equation model of the induction motor which produce the maximum torque at constant speed subject to voltage and current limits. Both the motor and generator mode are considered. Upon the optimal flux reference, the Fuzzy control of the rotor flux is used to regulate the flux with a good dynamic response. The detail design methodology is given. In order to obtain a robust speed regulation, a fuzzy sliding mode speed control with torque feedforward compensation is proposed. The sliding mode controller (SMC) is designed and its optimizing by fuzzy logic which reduces the chattering of SMC is given. The robust character is maintained by introducing the torque feedforward control.A robust current regulation in vector-controlled induction motor drive based on adaptive internal mode control (IMC) is proposed. The IMC performance and its design procedure is given after discussion of the robust performance and H2 as well as H∞ control. The current regulator is obtained using the IMC design procedure based on the model of IM in synchronous rotation frame. The performance of the IMC current regulator is analyzed and the on-line parameters estimation such as stator resistance and leakage inductance are presented. The implementation of IMC which prevent it from integration wind-up is discussed. The simulation results compared with PI regulation verify the good performance of the current controller.Finally, the implementation of the speed sensorless direct rotor flux-oriented vector-controlled IM drive based on DSP (TMS320C50) is presented. The hardware and software which accomplish the rotor flux observation and speed estimation, as well as the IGBT gate drive circuits using IR2233 are designed in detail. The Simulation experiments of the with the speed sensorless vector-controlled IM drive system which are based on Matlab Simulink and DSP system validate our design and analysis above.The author hope that the work above would be helpful to the design and application of the sensorless vector-controlled IM drives.