基于改进最小二乘支持向量机的电力系统短期负荷预测-电力系统及其自动化专业论文.docxVIP

万方数据 万方数据 摘要 Title: Short-term Load Forecasting Based on Improved Least Squares Support Vector Machine Major: Electrical Power System and Automation Name: Di Wang Signature: 1)J 协1j.- Supervisor: Prof. Lixiao Ya。 Signature:先沁。丁ω Abstract Short-term load forecasting is an important basis for the scheduling of power system security， and economic operation. Load forecasting accuracy will have a direct impact on its operation of reliability，economy and supply quality of power system. Therefore to improve the accuracy of precision ，finding an appropriate load forecasting method has important value. 1n the paper first describes the current applications situation of the current load forecasting， summarized the characteristics and inf1uencing factors of load forecasting ，and Summed up the common method of short-term load forecasting，as well as analysis of the advantages and disadvantages of different methods. ，then introduced statisticallearning the。可and the principle of SVM as the basis of support vector machine (SVM ) theory，SVM regression model is derived; this paper adopted least squ缸es support vector machine (LSSVM) model ，according to the historicalload data and meteorological data of a certain area of somewhere，Analyzed the various factors affecting the forecast，surnmarized the regul町ity of load change ， Corrected the 飞bnormal data in the historicalload data，normalized the relevant factors to be considered by load forecasting. The two 、p町缸neters of LS-SVM have a great impact on the model， however，the current solution is still based on experience. 1n this regard ，this paper adopted particle sw田m optimization algorithm to optimized the model p缸ameters，made the test set eηor as the basis for the decision，to achieve optimal choice of model p缸ameters，enabled prediction accuracy improved. Actual examples show that predicted convergence of this method is good，it with high precision and fast training speed. Key words: Short-term Load