Short Term Electricity Load Forecasting Of Qian'an Power Supply Company

Power system load forecast has the important significance to the society as a whole, it is reasonable arrangements for the premise of a lot of work, such as grid how to plan and set it, how the production operation. So it can save a lot of resources to reduce costs and ensure that the power system and many production activities on the economic and social security of the operation. Power system short-term load forecasting is an important reference for the power system plan, marketing, market trading, scheduling and other departments, rational scientific prediction of stable operation of the power system of major significance. With the reform of the electricity system, as well as energy saving and environmental protection requirements of the community now, which for the safe operation of the power system to bring more stringent requirements, and therefore predict the power system for all electrical equipment of the power consumed by the entire society has a greater significance.Short term electric power load forecasting, its forecasting time range is within one year, and the load capacity of the future is based on monthly, daily, weekly and hourly load forecasting. In this paper, the short-term electric load forecasting is carried out in this paper. Power system short-term load forecasting is affected by the influence of many factors, such as the overall situation of the social, economic development, population size, climate change and other, of diversity, complexity and uncertainty of power load forecasting, the impossible achieve accurate pre-measurement and control, so the historical data recorded in the data were detailed observation and analysis, and uses the best prediction according to the historical data and the specific reality, which became the load is the most effective way to predict the future. The on current status of Qian'an power load forecasting were studied in detail, and a detailed analysis of the theoretical basis of short-term power load forecasting and analysis and comparison of the advantages and disadvantages of every single forecasting method, combination forecasting method, based on the has better properties of combination forecasting method for short-term power load forecasting model.From the characteristics of county-level power load has, by influencing factors to consider, this paper selects from the regression analysis, trend moving average method, RBF neural network method to three single forecasting method are combined, Qian'an short-term power load by the month as a unit to the maximum load, minimum load respectively, were combined to predict. In the absence of data take interpolation method to fill a vacancy, in terms of weight coefficient in the most commonly used the equal weighted average method to carry on the combination forecast.Finally, based on the characteristics of Qian'an electric power load, short-term power load in Qian'an city was predicted by autoregressive moving average method, trend method, RBF neural network method and combination of the three methods. Comparison of single forecasting results and the combination forecasting results show that the combination forecasting accuracy is better than the single forecast combination forecasting method has higher accuracy, thus of Qian'an short-term power load forecasting error is smaller, Qian'an power company power for short-term negative charge as a case study, which achieved satisfactory prediction accuracy, to achieve the desired purpose.