Analysis Of Power Load Characteristic And Study On Short-term Load Forecasting In Regional Power System

With the development of the social economy, the power system load characteristics have changed. Load characteristics of the regional power grid analysis are important guarantee for the region peaking, marketing and load forecasting. Collect load date for several years of a city grid in Guangdong Province, repair the missing data, complete the original load data preprocessing.When analyzing power system load characteristics, we need the help of the index system. Establishe an indicator system, use the reliable load data preprocessed before by the index, curve and Comparative method, law in days, months, years to carry out analysis of the regional power grid load characteristics. Load characteristics during the holidays were focuses on individual. Power system load characteristics change is the result of many interacting factors, qualitative analysis of the impact from the economic, social, policy and meteorological aspects, combine with a local real statistical data, use improved gray correlation analysis to conducte regional characteristics factors quantitative analysis, to provide a reference for the regional power grid scheduling, planning and marketing. Due to the holidays load characteristics showed strong specificity, load characteristics influencing factors for the holidays to start specializing.Area grid load characteristics help to improve regional grid load forecasting accuracy. Short term load forecasting research is normally in accordance with the normal day and holidays separately. In this paper, use support vector machine prediction model forecasting normal daily load, and construct training set according to the regional load characteristics; set a prediction method for holidays load forecasting, based on the associated exponential smoothing correction method. Case studies are served to demonstrate the model’s accuracy.