| The high performance vector controlled motor requires speed closed loop control, but thespeed sensor is hard to fix and the cost increased obviously, besides, it is inevitable to beaffected and misbehave. So, the research on speed sensor-less vector control system forinduction motor becomes a hot topic. Based on the state space equations of induction motor,this paper presents a full order observer to estimate speed and identify parameters, in this way,the speed sensor-less vector control system is built.Based on the theory of rotor flux field-oriented control of induction motor, the traditionalspeed estimation methods using Model Reference Adaptive System is introduced firstly. Thenthe full order observer method used for speed estimation is presented, and the speed adaptivelaw is deduced from the Lyapunov stability theory. Considering that the accurate fluxestimation relies on the reliable parameters, this paper presents a static off-line motorparameters identification method, it is suitable to many occasions where running test is notpossible. According to the fact that parameters are readily to change while the motor isrunning, especially the rotor time constant, so this paper also presents an on-line identificationof rotor time constant based on full order observer. In addition, a method to identify speed androtor time constant simultaneously is also introduced, the adaptation scheme is designedseparately.The system based on full order observer is simulated on Matlab to identify the speed androtor time constant. And the simulation results show that the identification value can performfast and accurate tracking of their real value, and the appropriate adaption law couldeffectively improve the rate of the algorithm convergence.Finally, the vector control system based on DSP chip TMS320F28332is designed. Thestructure designs of software and hardware is presented. An incremental debugging method isadopted to verify the system, and the experimental results and waveform are analyzed. |
PDF Full Text Request