Design And Implementation Of Short-term Prediction Of Power Transmission And Distribution Equipment Active Power Based On SVM

Electric power is an important production goods of the normal operation of society, which modern civilization can not separate from. In terms of national security, economic development, social stability, and all aspects of daily life and production, the electric power industry, as a core industry,plays an important role.Power transmission and distribution equipment plays an important part of power grid, the status of power transmission and distribution equipment has important value in power grid system deployment and control,the normal operation of the grid facility, and accident avoidance. Active power is an important index of power transmission and distribution equipment status.Active power reflects the current situation of power consumption provided by the power transmission and distribution equipment, active power is closely related to social production, economic development and people's daily life,which can reflect the social demand for electricity production materials. Active power index has important practical significance. The prediction of active power is one of the hot spots in the development of power grid system.This paper designs and implements a system which give short-term prediction of power transmission and distribution equipment active power based on SVM. The system can construct the relevant data of energy management system into the feature vector data set of SVM model, and use the data set to train SVM model.To improve the regression prediction accuracy of the model,the system conduct the selection of model's kernel function and the optimization of parameters of model and kernel function.The system use Spark to parallel computing the optimization of parameters in order to accelerate the process.The system predict the active power value of the time node in the future.The system can help the grid staff better on real-time management of the power transmission and distribution Equipment as well as the power grid,to carry out the operation and development plan,it has important influence on the power system security and economic operation.Firstly, this paper introduces the background theory and related technologies, and then describes the functional requirements and non functional requirements of the system. For several key problems of the system, the design of the input feature vector of the SVM model, the kernel function selection of the S VM model, the parameter optimization of the SVM model, and the parallel acceleration of the parameter optimization. This paper conducts researches on them and gives solutions. Then this paper gives the overall design of the system,including the overall architecture module of the system, the overall work flow,module interfaces, and the module interaction sequence diagram. Meanwhile the design and implementation of the core modules of the system is introduced in detail. Finally, the system of short-term prediction of power transmission and distribution equipment active power based on SVM is deployed and tested, test cases are designed for the main function modules of the system and the results of the test are generated, so that the correctness and reliability of the system are tested. At the end of the paper, the design and implementation process of the whole system are summarized, and some conclutions for future improvement are given.