Font Size: a A A

Power Load Forecasting Method Research And System Development

Posted on:2011-07-19Degree:MasterType:Thesis
Country:ChinaCandidate:Y C ZhangFull Text:PDF
GTID:2192330338488609Subject:Water Resources and Hydropower Engineering
Abstract/Summary:PDF Full Text Request
Power load forecasting has important function in diapatching, programming, planning and designing department of power system, which can provide the effective policy-making support for medium and long-term plan and short-term dispatch of power system. With the development of market economy in China these years, the electric power industry has gradually transitted to the marketability advancement, which set a higher request to the accuracy, timeliness and reliability of load forecasting. In this paper, on the basis of the characteristic analysis and rules identification of the long-term load and short-term load, the year, monthly and daily load forecasting models and probabilistic intelligence forecast method have been established, and power load synthesis forecasting system has been developed. The main research results are as follows:(1) On the basis of the characteristic analysis of the medium and long-term power load, the load sequence change rule has been revealed, the year and monthly load forecasting models has been established by using grey theory and regression analysis method, which has got good forecasting results.(2) In the research of short-term load forecasting, the radial basis function neural network model and comprehensive model based on virtual forecasting principle has been adopted, and through analyzing the chaos characteristic and relatedness with climatic factors of daily peak load, the suitable model has been set up to forecast the daily peak load which can be used to revise daily load forecasting results.(3) By means of the statistics of error characteristics of historical load forecasting, a probability density function model for forecasting errors in load areas is established, and combining with the results of deterministic load forecasting the results of probabilistic load forecasting are obtained, which has carried on the development to the traditional definite load forecast research and enabled the power system decision-makers to realize the possible uncertainties and risk factors faced more completely and make reasonable decisions promptly.(4) Based on the above theoretical research achievement, according to the project application demand of central China power grid load forecasting and quotation decision making support system, the networked systematic load forecasting and management system has been developped. The test results indicated that each functional module of this system has met the constituting requirements and can put into operation.
Keywords/Search Tags:load forecasting, comprehensive model, the radial basis function neural network, chaos characteristic, probabilistic forecasting
PDF Full Text Request
Related items