The Research Of Power System Voltage Stability Evaluation Based On F Decision Tree

Now the recognition of the importance in that power system voltage stability study is growing, whether the voltage stability is directly related to the safe operation of the power system can be, so how to rapidly and accurately for evaluation and analysis of voltage stability state is great valueble.This paper aiming at this problem using the theory of neural network and fuzzy neural network theory combined with decision tree algorithm to evaluate voltage steady state, and using the application of decision tree, neural network decision tree and fuzzy neural network decision tree three kinds of methord to analyze the problem of voltage stability.This article selects relatively important node in the system of the phase Angle as the decision tree of candidate attributes, the system of running data as the training sample, establishing the decision tree model of the unknown operation condition of the system voltage stability evaluation, and decision tree algorithm is verified by simulation example in the application of voltage stability analysis is effective. But the decision tree of the processing results show that there is a judgment of nodes with large error rate. Aiming at this problem, in this paper, combining the neural network, fuzzy neural network technology and the decision tree algorithm, forming a hybrid decision tree model, improved the original decision tree model, this paper constructing the neural network decision tree model suitable for the evaluation and analysis the power system voltage stability, this method is verified by the example simulation in evaluating the effectiveness of the system voltage stability, and improve the decision tree classification accuracy of power system voltage stability state.This paper also puts forward a kind of fuzzy neural network decision tree model is applied for simulation verify that this method can effectively reduce the voltage steady state error. tree method can rapidly and accurately assess whether the power system voltage is stable, but the fault rate of neural network decision tree and fuzzy neural network decision tree method is lower.