Research On Speed Sensorless Vector Control Of Induction Motor

With the development of electronics, microelectronic technology and modem control theory, many kinds of electric drive system are researched and developed by more and more domestic and foreign scholars. Owing to its inherent advantages, induction motor plays an important role in the area of electric drive field. In the drive system of induction motor with high performance, the close loop of speed is absolutely necessary. Remarkable progress has been achieved in the theory and practice of induction motor vector control system with speed sensor. But because of the existence of the speed sensor such as rotating coder, the installation of speed sensor is quite troublesome and the demand of the installation is quite high. The scale and amount of induction motor vector control system with speed sensor is limited by these disadvantages. Speed sensorless vector control system can solve the problems mentioned above and reduce the cost of the system, which is meaningful both to improve the reliability and adaptability of the AC drive system and to expand the AC drive system's application range.The dynamic model of induction motor is firstly discussed in this paper. And then researches are done on the theory of flux oriented vector control system and flux open-loop vector control system without speed sensor. By comparing and analyzing the existed scheme for the speed sensorless vector control, the paper presents a new scheme of estimating speed by torque current differential. With the help of MATLAB, the validity of this method is proved. And the simulation results prove that the system using this method has good static and dynamic performance.The performance of vector control system without speed sensor depends on the speed estimation, and the precision of speed estimation is relied on the precision of motor parameter. In this paper, a new method, which makes use of inverter itself resources, is presented to identify the induction motor parameters by injecting voltage exciting signals and testing the current response. Induction motor parameters are determined by the DC test, the single-phase test, no-load test and dynamic test. A fully automatic measurement for the stator resistance, rotor resistance, leakage stator, rotor leakage inductance, mutual inductance and moment of inertia is developed by the use of software calculation current and voltage which are detected by inverter. In order to get more accurate measured parameters, in the PC using the least squares algorithm or Median-average filter to process sample date to improve the accuracy of the parameters identification. The experiments demonstrate that the accuracy of the induction motor parameters meet the speed sensorless vector control requirements.Stator current decoupling is realized by the method of nonlinear decoupling based on dynamical mathematic model of rotor-field-oriented control,which is applied to design nonlinear decoupling controller for counteracting the influence of stator coupled current. Current and speed PI controllers were designed by optimal tuning methods. Simulation results showed that these controllers are able to achieve a good performance of desired dynamic and steady state responses.Based on XC164CS, the experiment platform designed for induction motor vector control system is introduced and developed in this paper and hardware and software design are introduced in detail. Based on this platform, experiments of parameters identification, parameter self-tuning of PI controller and induction motor vector control system without speed sensor are realized one bye one.The identified results of parameters are accurate. With them, parameters of PI controller of the vector control system are designed. The results show that the system using selected parameters has a good performance. The scheme of estimating speed by torque current differential is accurate. The experiment results show that speed sensorless vector control system using this method has the good dynamic and static performance.