Research On The Load Frequency Control Of Power System Based On Neural Network Adaptive Inverse System

The main goal of power system load frequency control is to keep the change of system frequency in the specified error range,with the lowest possible cost to provide sufficient matching of the load.The power system is composed of interconnected large generating units,which are usually defined as the control area of the power system.Interconnected power systems rely on automatic generation control to ensure that the output of the generator follows the change of the power load demand.Uninterrupted power supply and low running cost are the important indexes of power system.Therefore,it is very important to study the influence of large interconnected centralized power generation and the renewable energy generation on the reliability and safety operation of the transmission system in the high voltage transmission system.In this paper,taking the load frequency of power system as the research object,designed several neural network control schemes as follow:According to the problem that the load frequency will be instable after multi-area interconnected power systems are suffered from wind power and load disturbance,a kind of load frequency control(LFC)of multi-area interconnected power system is proposed,based on cloud neural network adaptive inverse system.On the basis of analyzing the active power output characteristics of a single area power system,the load frequency control model of interconnected power system with multi-area active power output is established.Adaptive inverse control is used to solve the contradiction between system response and disturbance inhibition effectively.The cloud model is introduced into the adaptive inverse system to construct the cloud neural network identifier.The advantages of cloud model are utilized to deal with the uncertainty such as fuzziness and randomness,the identification ability of neural network is further improved.The simulation results show that by designed cloud neural network adaptive inverse system,not only a good dynamic response can be got,but also the wind power and load disturbance can be reduced to minimum.According to the problem that the large fluctuations of load frequency will happenafter multi-area interconnected power systems are suffered from wind power and load disturbance,a kind of load frequency control(LFC)of multi-area interconnected power system is proposed,based on terminal sliding mode fuzzy neural network.The terminal sliding mode is introduced into the adaptive inverse system to construct the fuzzy neural network identifier.Taking advantages of terminal-sliding-mode can be tracked with zero steady-state error within limited time,and the identification ability of neural network is further improved.The simulation results show that by designed terminal-sliding-mode fuzzy neural network adaptive inverse system,not only a good dynamic response can be got,but also the wind power and load disturbance can be reduced to minimum.