A fourth-order power system chaos control method based on bidirectional optimized bp neural network sliding mode variable structure

Abstract

The invention is a bidirectional optimized BP neural network sliding mode variable structure fourth-order power system chaos control method. The invention is based on the chaotic mathematical model of the fourth-order power system, and uses a control method combining bidirectional optimization BP neural network and sliding mode variable structure control to control the chaotic oscillation of the power system. In the neural network, the present invention optimizes the excitation function used in the forward direction of the algorithm and the learning rate used in the reverse direction, thereby effectively suppressing the chattering phenomenon of the sliding mode variable structure control and making the chaotic oscillation control more ideal; Compared with the traditional sliding mode variable structure control, the present invention not only retains the excellent characteristics of BP neural network and sliding mode variable structure control, but also is more effective under the premise of overcoming the false saturation phenomenon of the neural network, improving the generalization ability and reasonably speeding up the learning process. The chattering phenomenon of sliding mode variable structure control is greatly weakened, so that the power system chaos control can achieve better results.

Description

technical field [0001] The invention relates to the technical field of fourth-order power system chaos control, and relates to a fourth-order power system chaos control method with bidirectional optimization of BP neural network sliding mode variable structure. Background technique [0002] With the interconnection of large power grids and long-distance power transmission becoming the development trend of contemporary power grids, the structure of power systems is becoming increasingly complex. As a typical nonlinear system, the nonlinearity of the power system itself will cause chaotic oscillations under certain circumstances, which is a non-periodic and irregular low-frequency oscillation. At the same time, emergencies and uncertain factors may also cause During the operation of the system, chaotic oscillation occurs. At present, low-frequency oscillations have been observed many times at home and abroad, and the harm caused by chaotic oscillations cannot be ignored. Whe...

AI Technical Summary

Problems solved by technology

As one of the most widely used neural networks, the multi-layer feed-forward neural network (BP neural network) based on BP algorithm also has some problems, such as false saturation of the neural network, insufficient generalization ability, and unsatisfactory learning speed. It will directly affect the application effect of the neural network, and even the chattering problem of the latter cannot be solved after combining the sliding mode variable structure control, so it will not be able to achieve the ideal control of chaotic oscillation.

Images