Theoretical Line Loss Rate Prediction And Analysis Of Power Network

In recent years,with the rapid development of China’s society and economic,we proposes a higher level for grid operation level. Massive loss reduction measures have been applied to the grid,and it has played a positive role in improving power waste. With the intensification of the global energy crisis,The 12th Five Year Plan proposed adhere to building a resource-saving and environment-friendly society to accelerate the transformation of economic development as an important focal point. In order to respond positively to the policy of the Party Central Committee, establish a conservation-oriented society,minimize the loss of power generated in the transmission process,so the Line loss prediction becomes very important.Distribution network connected to the power system, transmission system and user, with the rapid development of the power system,the inevitability, complexity and uncertainty of Line loss has posed a serious challenge to the safe operation of power systems and power quality. Therefore, accurately predicting the rate of line loss,helping relevant staff to develop assessment indicators which accord realistic situation and planning,and playing an effective value on energy have become practical issues which needed to analyze and solve urgently. Depth study of line loss rate forecast has a very important to improve coordination operation capabilities on power system and promote the sustainable and healthy development of power.This paper focuses on the following research mainly centered on the theoretical line loss rate prediction based on the data of history theoretical line loss rate:1. With the characteristics of high precision, neural network model is a common nonlinear model which can better cope with the fluctuations of sequence. Neural network model is used to predict theoretical line loss rate.Theory of Markov is applied to process error data,and then RBF-Markov model is built to predict theoretical line loss rate.2. With the advantages of fast convergence speed, strong learning ability and good generalization ability, SVM can effectively predict the change trend of sequence. Based on the principle of SVM, the regression model of SVM is built to predict the theoretical line loss rater directly.3. Studying on model parameter optimization problems of support vector machine (SVM). It analyzes the effect of penalty factor and nuclear parameters and the impact of the performance of SVM. the genetic optimization algorithm is applied to propose the optimized support vector machine forecasting model (GA-SVM) based on genetic algorithm to deal with the insufficiency of SVM modeling. Finally, we use GA-SVM prediction model for theoretical line loss rate prediction.4.Combined RBF-Markov model and GA-SVM model,the probabilistic model is established to study theoretical line loss rate.For the critical issue which the probability density is rarely to solve in the probabilistic model,non-parametric kernel density estimation is used to estimate probability density function of theoretical line loss rate,and then the probabilistic model is established to obtain confidence intervals.