Integrated Energy Conservation And Loss Reduction Technology Of Distribution Network Based On Neural Network

In China, energy conservation and environmental protection is a long-term strategy of economic and social development. Effectively reducing the line loss of distribution networks has far-reaching significance in achieving economic operation of modern power system. At present, the line loss rate of our distribution networks is at a medium level in the world.Compared with developed countries; the line loss rate is still high.There is a large room for energy loss reduction. Deep study of distribution network energy loss reduction measures, using new algorithms and techniques, Application of more scientific means of line loss management, all these are able to minimize unreasonable power loss, reduce the line loss rate and improve the level of power system economic operation. Reducing distribution system losses can reduce electricity costs, improve economic efficiency, and tap the power supply capability of distribution equipment. At the same time, it is very favorable to the energy use, environmental protection and the optimal allocation of resources. Therefore, reducing the energy loss, saving electricity, improving power transmission efficiency and effectiveness of the relevant departments and enterprises, will result in very substantial economic benefits. These are also required by the energy crisis, construction of "saving" society, and consistent with state policies.This paper starts from the actual situation of the distribution network. Based on the study of energy-saving measures to reduce distribution network loss and summary of previous experience, this paper proposes integrated energy conservation and loss reduction technology of distribution networks. Artificial neural network algorithm is applied to load forecasting, distribution network reconfiguration, reactive power compensation. Based on the results from load forecasting, analyze current trends of distribution network, give a more reasonable distribution network topology and reactive power compensation plan, ultimately achieving the purpose of reducing network losses. The main contents are as follows:This paper researches on energy reduction measures of distribution networks from energy saving transformer, reactive power compensation, network optimization operation, energy saving alteration, distribution network reconfiguration five aspects. We study on energy loss reduction principles and effects of various measures, attention problems in the practical application. Reactive power compensation and distribution network reconfiguration need small investment, have good loss reduction effect, great benefits, more complex issues to be considered, and high research value.Integrated energy conservation and loss reduction technology includes three major aspects:load forecasting, distribution network reconfiguration, reactive power compensation. After analyzing and comparing various algorithms of the three areas, BP neural network is adopted in the whole process of load forecasting, distribution network reconfiguration and reactive power compensation, achieving the purpose of reducing network losses. The learning and training rules of BP network, learning rate and the number of hidden neurons design problems are focused on. Compared with other methods, neural network has powerful learning algorithm, multi-input and output parallel processing, nonlinear mapping, distributed storage, fault-tolerant and other advantages. BP neural network has mature theory and is widely used.Artificial neural network algorithm is applied to load forecasting, distribution network reconfiguration, reactive power compensation, forming distribution networks integrated energy conservation and loss reduction technology. According to the results of load forecasting, with the goal of reducing the network loss, distribution network optimized topology and reactive power compensation plan are given. The model building, implementing, and programming works of load forecasting, distribution network reconfiguration, reactive power compensation based on neural network are completed.Taking a 1O0kV hand in hand distribution line in Heze region, Shandong province as the example, the calculations related to energy loss reduction are completed, technical details and effects of loss reduction are also analyzed, verifying the feasibility and effectiveness.