Parameters Optimization Of Auto Disturbance Rejection Controller For Load Frequency Controller Of Power Systems

As the increasing scale and complexity of power system, a power grid interconnection between regions is also growing. The actual power output control problem of generation units response system changes of frequency and tie line power within a specific range is known as the load frequency control (LFC). It’s a very important factor in the operation of the power system, any sudden load disturbance can cause deviation of tie line power and frequency fluctuations. For the widespread disturbances problem in the power system, all kinds of advanced control methods such as optimal control, variable structure control, adaptive control, robust and intelligent control also got great application. This paper study the linear active disturbance rejection controller (LADRC) of load frequency control in power system and the main contents are as follows:Firstly, because of the complexity and coupling problem of the power system, the thesis adopts LADRC technology for LFC system design. LADRC controller can regard the various uncertainty of the power system as disturbance. And then with estimation and compensation of the linear extended state observer(LESO), the controller can quickly eliminate the disturbance, and obtain good decoupling effect. But the manual parameter tuning still requires a lot of time and the performance of the system is not optimal, meanwhile with the increasing number of setting parameters for multiarea power system, the parameters plays a more important role on dynamic performance of the system, so this paper use a DE(differential evolution) algorithm to optimize the controller parameters.Secondly, for the generation rate constraint (GRC) problem of power system, this thesis adopted a anti-GRC scheme. On the basis of the LADRC,the competation signal is sent to LESO and added to the LADRC controller as a new disturbance. Usually, competation gain is setted by trial. Based on the improved differential evolution algorithm, compensation gain is added to the optimization process as a new parameter. The result shows that through the improved differential evolution algorithm, in terms of rapidity and dynamic performance, power generation rate constraint problem got obvious improvement.Finally, this paper study the generator dead band problem. In order to obtain better dynamic characteristics, the original algorithm was improved. The basic idea is under the premise of maintaining the diversity of the population, strengthening the global search ability of the algorithm at initial stage and the local search ability at later optimization.So that the convergence speed and accuracy of the algorithm was improved. The improved differential evolution (IDE) algorithm is used to tune LADRC parameters. Simulation results show that the optimized LADRC can achieve good control performance.Several examples of interconnected areas power system simulation show that: By using improved differential evolution algorithm, the LFC controller parameters was optimized. For power system nonlinear problem such as generation rate constraint and dead zone characteristics, control effect has been greatly improved.