Wavelet Neural Network Method Of Voltage Security Assessment Indicator In Power System

Power system are characterized by the growing size and complexing of enormous number of generating units, loads and transmissing lines that include in the system. With expanding of electric power system scale, demands to reliability are higher and higher. Voltage collapse has become one of the important problems, which threatens the running safety in large modern electric power system. Studying voltage security is still an important tasking that electric workers face.At first, this paper analyzes the actuality of voltage stability research and the measure of voltage security assessment in power system , introduces simply current voltage security indicator and compares their merits and demerits. Then using VCPI (voltage collapse proximate indicator) as voltage security assessment indicator.The solution of voltage security assessment involves prediction, pattern recognition classification and fast solution, which are tasks well suited for neural network technology. This paper establishes firstly a ANN (artificial neural network) model for voltage collapse proximity indicator estimating. According to the forecast results, this model can estimate VCPI based on system state after being rained. But, the precision and study rate of BP algorithm is poor. Hence, this paper proposes a WNN (wavelet neural network) model for voltage collapse proximity indicatorestimating. The nervous cells function is the basis of nonlinear wavelets. A wavelet network composed by the wavelet basis function is computed by an expansion and contraction factor and a translation factor to reach the global best approximation effect. By the wavelet neural network which has been trained, we can calculate on-line voltage collapse proximity indicator. It can be seen from the simulation results, this method is useful for early prediction of the voltage collapse phenomenon in power system, and should be a fast , real-time tool for voltage security assessment. Comparing ANN model and WNN model, it can be seen that WNN model has higher precision and faster computation speed.At last, this paper discusses the measures which can prevent voltage collapse and enhance voltage stability, and summarizes the methods to improve the voltage stability in the aspects of power system planning and design, dispatching and operation, and so on.