| With rapid development of the economy, the shortage of electric energy has become the bottleneck restricting the economic development. How to effectively improve the power load forecasting precision is the key to solve the problem. Mid-long term load forecasting is an important part of the power system planning, and it is the foundation of economic operation of the power system. Mid-long term load forecasting provide reliable decision-making basis for electricity plan management, power network operation mode, installation maintenance plans, electric power construction planning, electric power production, power network operation, grid fund balance and balance of human resource demand and supply. Precise load forecast is beneficial to improve the safety and stability of the power grid operation, effectively reduce the power generating costs, ensure the electricity demand, so as to improve the economic benefit and social benefit of power system. How to effectively improve the power load forecasting accuracy is an critical question we have to discuss. The research of this problem is of great theoretical and practical significance.Based on the former research, this paper studied the related theory on medium and long term power load forecasting and load forecasting technology in the following aspects.(1) In this paper, the basic concept of power load forecast, the classification of electric power load forecasting, the principles of load forecasting and characters of load forecasting was introduced. On the basis of this, basic program of power load forecasting on power load forecasting is researched. Besides, the classic load forecast technology, the load forecast technology based on experience and the power load forecasting based on new technologies is also studied in this paper.(2) The gray prediction technology was analyzed in this paper, then the defects of traditional gray forecasting model was put forward. The formation and development of the gray system theory was introduced in this paper, and the basic concepts, the basic principles and basic contents of gray system were studied in the paper. On the basis of the above study, the paper analyzed the gray prediction technology and the steps of traditional GM (1,1) predicting model, and then the defects existing in the traditional GM (1,1) prediction model was proposed, namely respectively analyzed the reasons of lower precision from the aspects of the model application scope and the preprocess of the historical data. So the methods of improve the forecasting model itself and preprocess the historical data sequence using the sequence operator were put forward.(3)The load forecasting model Nonlinear Gray Bernoulli Model based on genetic algorithm (GA-NGBM) was put forward in this paper. First, the modeling steps of Nonlinear Gray Bernoulli Model (NGBM) were stated in detail. Second, how to improve the prediction of NGBM model was researched, and the paper proposed that the historical data preprocess, the time response function parameter and the power parameter of NGBM model are main factors that may affect the prediction accuracy. According to the problems proposed above, the paper put forward that preprocess the original data using the weakening buffer operator and at the same time the genetic algorithm was employed to optimize the two parameters in the model with the purpose of searching for the optimal parameters of the model, which could make up for the insufficient caused by given the parameters on experience. Based on the above research, the Nonlinear Gray Bernoulli Model based on genetic algorithm load forecasting model was put forward in this paper, and the model prediction process was stated in detail.(4) The GA-NGBM power load forecast model simulation effect was test in this paper. The history load data of Hebei south part grid was employed to inspect the model forecasting effect. The numerical results and error analysis illustrated that the model has a favorable prediction effect in mid-long term power load forecasting. |
PDF Full Text Request