Study On Second-order Sliding Mode Based MRAS-Type Sensorless Vector Control Of Asynchronous Motor

With the rapid development of social economy, energy shortage and environment pollution problem are becoming increasingly outstanding. In recent years, the electric vehicles, as a clean and efficient means of transportation, are receiving more and more attention. And the performance of the electric vehicles depends mainly on the performance of the drive motor, which leads to the demands for further development and improvement of motor drive control. Due to the development of vector control technology, and the significant merits of asynchronous motor such as simple structure, low manufacture and maintenance cost and wide speed range, asynchronous motor has been widely applied in electric vehicle. For electric vehicles, the vector control system with speed sensor leads to a higher cost, and bad conditions including mechanical vibration and electromagnetic interference also affects the reliability of speed signal detection. Therefore, speed sensorless vector control technology is now becoming an essential technique to the high performance motor drive systems, with the key to the technology being speed and flux observation. Accurate flux observation at the same time has the crucial significance on the performance of the vector control.The speed sensorless vector control based on the rotor field-oriented is discussed and researched in this paper, mainly including the observation of speed and rotor flux. First of all, based on the mathematical model of asynchronous motor, this paper analyzes the two typical vector control schemes, namely the indirect vector control system and direct vector control system. Then,in view of the insufficient and defects existing in the traditional model reference adaptive system (MRAS) speed identification, this paper presents an speed identification scheme which is constructed by combing a second-order sliding mode (SM) observer with a design on MRAS.The method observes an intermediate variable by the second-order Super-Twisting SM observer which does not need to speed information and is used as the reference model of MRAS speed observer. And the adaptive model of MRAS observer comes from the current model of the flux. Based on all above, Popov hyper-stability theory is introduced to design the speed adaptive rate which is used to adjust the adaptive model, and to gain speed information. Moreover, on the basis of realizing speed identification, this paper designs the rotor flux observation scheme based on the second-order Super-Twisting SM observer, which is used to gain the rotor flux accurately to achieve the direct field orientation. Finally, this paper builds MATLAB simulation models and experimental platforms for vector control system of a 5.5kW asynchronous motor, and the results of the simulations and experiments validate the feasibility and effectiveness of speed identification, rotor flux observation schemes.