MRAS Based Control Method For Induction Motor Drivers

High performance AC drivers have taken the place of DC drivers in most of the electrical drive applications. The induction motor drivers have acquired a great improvement with the great progress of Power Electric, IC design and IT technology. The development trend of AC drivers was to be more intelligent, easy modularization, full digital and higher switch frequency. Applying modern control theory in induction motor drivers has came to be the trait of current developed systems. So the MRAS technology in induction motor drivers has been the focus of this paper.The state space equations of induction motor was systemically summarized in this paper, and then a simulation model was designed based on the state space equations under the Matlab/Simulink environment using S function block.A new simplified feedback matrix design method has been proposed in this paper for the adaptive flux observer of induction motor, although the method can reposition only two poles of the system, but the decrease demand on calculation ability make it very easy to be used.The frequently used speed estimation scheme such as speed-calculate and PID controller methods was also been briefly introduced by this paper. The MRAS based speed estimation was widely considered as the brightest outlook among high performance speed estimation schemes, because of its medium calculation ability requirement and very high precision. A new theory to design speed estimation arithmetic base on the absolutely stability theory of nonlinear system has been proposed by this paper. The MRAS based speed estimation can be explained as a optimize implement of the mentioned theory.The parameter estimation was comprised with two sides: self-recognizing and on-line parameter compensation. As to the former, this paper present a parameter estimation based on the optimize method of the standard quadratic criterion of predicted output. By extending the MRAS theory to parameter estimation, a new online estimation method for the rotor time-constant based on MRAS was presented in this paper. Theory analysis demonstrates the method has a stability of LIYAPUNOV, and its efficiency has been proved through the simulation results. Similar to the speed estimation,absolutely stability theory of nonlinear system can also be used in the stability analysis of parameter estimation.A new simultaneous estimation method of induction machine parameters and velocity has been designed by compounding the speed and parameter estimation based on MARS.Finally an experimental platform of AC motor driver has been established by this research with DSP56F805-a chip intending to be used in motor drivers as its CPU. The ideas of different AC motor control methods programming was discussed and experimented on this platform.