Loss Minimization And Fast Response Control Of Inverter Fed Asynchronous Motor Drive

Variable Frequency Variable Speed (VFVS) technology has becoming the primary method to save energy and improve operation performance with its excellent variable speed behavior, remarkable energy-conservation and wide applicability. However, when operation points deflect from rated point, the efficiency of the motors will not be optimal whether by V/F control or vector control, especially in the case of light load and low speed. In recent years, along with tenser international energy situation and increasing requirement of high operation performance, the loss minimization control of IM attracts more and more attention. Now, the research of electric vehicles is no doubt a hotspot, whose drive-train system strictly demands both the least losses and fast response. In fact, to meet both demands is so hard that there has not been any satisfied solution yet. Some existed problems, especially fast response problem, are still to be discussed.As to the cases such as electric vehicles, which have a high demand not only for energy- conservation but also for good control performance, there are three vital problems to solve to realize the optimal control of their drive-train systems. First is how to obtain optimal rotor flux linkage for loss minimization control; Second is to design fast response control strategy to avoid the descending of dynamic performance caused by the flux linkage optimal control; Third is to design high performance closed loop control strategy to make the system keep excellent variable-speed behavior under the situations of time-varied parameters, variable flux linkage and variable load, etc. Under the background of electric vehicles, this paper studies on the loss minimization control strategy and fast response control strategy in detail, and designs a high performance closed-loop controller by the aid of the self-disturbance rejection theory. Simulations and experiments results prove the validity of these new methods, whilst fruitful innovative results are obtained, which provide effective solutions for the efficiency optimal control of the electric vehicles' drive-train systems.In this paper, the development and research situation of the VFVS and its control technology are discussed at first, so are the drive-train system in electric vehicles and the motor's loss minimization control. After that, the loss minimization control strategies have been analyzed in detail, which spring out recently. The research hotspots in this field are pointed out.On the basis of detailed analysis of the asynchronous motor's loss behavior and its mathematical model including the iron losses, a kind of loss model control strategy is deduced and the variety law of the optimal flux linkage is illustrated. And then, an experimental system is designed based on the TMS320LF2407A DSP and the strategy above, and a large number of corresponding experiments are completed. The experimental results show that, the proposed strategy can well satisfy our demand under stable state and speed changing. To be noted, the energy conservation effect in the light-load, high-speed situation is superior to that in the heavy-load, low-speed situation. To solve the problem that the control precision of the strategy above is influenced by the parameters of asynchronous motor, the gradual memory erase RLS algorithm is introduced into the parameter estimation of asynchronous motor's loss model, followed by the loss minimization control strategy considering the motor parameter's time-variant,which improves the system's precision and robustness.Based on gradient algorithm and golden section method, the search controllers are designed to optimize the efficiency of the asynchronous motor to avoid the parameter change's influence to loss minimization control. The simulation results indicate that the convergence speed of golden section method is obviously superior to that of the gradient algorithm; however in the former search process, the fluctuation of flux linkage is not reduced. So the improved golden section algorithm is proposed, the validity of which is verified through theoretical analysis and experiments. Combined the advantages both of loss model controller and the search controller, a novel hybrid loss minimization control strategy is presented. The steps are as follows: first, get the approximate optimal flux linkage according to the loss model; second, find operation point with the minimal input power by on-line search. Simulation and experiment verify that, this hybrid strategy, which has rapid convergence speed and robustness, is a promising method to efficiency optimization of asynchronous motor.In the traditional loss minimization control system, the improvement of the efficiency is at the cost of deterioration of the dynamic response. Aimed at the problem above, a fast response control strategy is studied based on the current dynamic assignment. The experiment results indicate that this method will decrease the dynamic speed dropping and reduce dynamic adjustment time. Inspired by the ideas of voltage space vector and direct torque control, a new strategy which directly composes the voltage vector is proposed, which realizes the high behavior control in the dynamic process of load variety. Different from both the direct torque control and the classical vector control, this scheme combines vector control's good stable performance and direct torque control's fast dynamical response, whilst it does not increase any hardware in control system. Simulation results show that, this scheme, which has excellent dynamic performance, notably reduces the speed-fall of loss minimization control system when load changes. It is indeed an efficient solution to the contradiction between the loss minimization and fast response control.The asynchronous motor is a complicated, nonlinear, multi-variable, parameter time-variant controlled object. Especially in the drive-train system with loss minimization control strategy to adopt, the change of flux linkage will inevitably bring perturbations into the system and make it unstable. Therefore, a closed loop controller with high performance and its parameter-adjustable scheme are designed by means of nonlinear self-disturbance rejection control theory. Experiment results illustrate that, compared to the classical PI controllers, the proposed controller has faster speed-tracking performance and better anti-disturbance performance, and provides good references for closed loop strategy selection in loss minimization control system.