Centralized Operation Control Method for Power Systems Based on Neural Net-work PSD Algorithm
In order to meet the control requirements of power system centralized control operation under different op-erating conditions and load changes,a power system centralized control operation control method based on neural network PSD algorithm is studied.Based on the probability of loss of load in the power system,the maximum allow-able hierarchical control deviation for centralized control operation is determined as the allowable control deviation for the neural network PSD algorithm.The deviation between the set value and the measured value of the power system's centralized control operation control quantity is used as input to the PSD controller.By utilizing the hidden layer of neural networks,the input of the controller is subjected to proportional element,summation element,and differential element operations.The gain coefficient of the neural network is adjusted using self-tuning rules.Through forward and reverse transfer operations,centralized control operation and control of the power system are achieved.The exper-imental results show that using this method to control the centralized operation of the power system,the control con-version ratio of the power system is maintained in the range of 1.25～1.4,and the fault occurrence rate is below 6%.
PSD algorithmpower systemcentralized control operationcontrol methodsforward operation
万方数据
维普
