Parameters Identification On Speed-less Sensor DTC System Based On Chaos-genetic Algorithms

In the middle of 1980's, Depenbrock of Germany putted forward the DTC of asynchronism motor at first. It got more attention at once. The vector of voltage is selected according to the error of torque and flux. The switch of converter is controlled directness, so the system's good control characteristic is obtained. It is applied widely. However, there are problems of bad low speed performance, low reliability and adding cost because of sensor's use. In this paper, the author made a study of these problems. The technique of sensorless is study hotspot all along in order to reduce sensor's influence to system. It is studied in this paper.To fulfill high performance direct torque control of an induction motor, it is necessary to know precisely parameters of induction motor, such as rotor speed and stator resistance. So wavelets and neural network are combined and wavelet neural network (WNN) is proposed in this paper. Through WNN identifier, stator resistance and rotor speed are identified. The new thought of WNN optimized by chaos-genetic algorithms is proposed. Chaos-genetic algorithms provide an effect way for very complex problem optimization in virtue of chaos mechanism, especially for the problem which can't be solved or abate by traditional math. There is important meaning importing chaos-genetic algorithms to WNN optimization.Simulation model of direct torque control system with WNN identifier is built by matlab/simulink. We can see from simulation results that WNN can identify the change of rotor speed and stator resistance precision and DTC system with WNN identifier shows excellent static and dynamic performance. Simulation shows that the scheme of replacing speed sensor with WNN speed identifier in the DTC system, with no speed sensor is effective and feasible, and direct torque control system with WNN stator resistance detector resolves the problem of low speed performance.Finally, we design hardware and soft system for DTC basis on simulation, adopting the control special chip of TMS320LF2407 made in company TI.