The Load Prediction Of Urban Power Grid With Wind Power Generation Integrated

With the development of urban economy,more research work is demanded on load forecasting analysis of urban power network.Accurate and reliable urban power network load forecasting can make convenient for the dispatch of power system and be essential for the economic operation of the power system.With the wide application of renewable energy,it is necessary to conduct research on issues about absorbing wind power generation units into urban power network.This thesis has incorporated large amount of research works on load forecasting topic and the influence brought to the urban power network by the adoption of wind power generators.This thesis first detailed describes the method for short-term and ultra-short-term load forecasting of urban power network.At first,monthly load characteristic,daily load characteristic,ensemble daily load characteristic,industrial and residential commercial load characteristic are under studied.Then influencing factors including both micro-meteorological factors such as temperature,sunlight radiation and some government-policies are taken into consideration.Meanwhile,to overcome some shortcomings in ultra-short-term load forecasting method such as edge effect and dynamical property,a new artificial neuro-network for urban power network load forecasting model was proposed by improving the training strategy.Finally,both short-term and ultra-short-term urban power network load forecasting experiments were conducted.Results show that the proposed method is both accurate and can be very applicable.Research work on medium-long-term load forecasting of urban power network are also included in this thesis.Economy factor,income level and consumption attitudes,industrial restructuring,power supply ability and other government policy factors' influence on urban power network load condition are first analyzed.After that,a multi-variable grey MGM(1,N)load forecasting model based on multiple linear regression was proposed for medium-long-term load forecast.The test results proved this newly proposed method has high accuracy and good applicability.Lastly,the load developing trend for different areas of urban power system was analyzed by this method.Some strategies and suggestions which regard to the serious situation faced by the urban power network are also proposed.This thesis analyzed the influence of urban power network brought by the adoption of wind turbine generators from the perspective of time scales.For the short-term and ultra-short-term load forecasting of urban power network,an Elman artificial network method based on prime component analysis was used to conduct the forecast of wind farm and an equivalent load model concept was proposed.An equivalent load forecasting model of urban power network was constructed based on the wind farm output-load forecasting characteristic.Evidence was provided to prove the advantages of this equivalent load forecasting model by a simulation of an ultra-short-term load forecasting after the adoption of a wind farm in one area.For the medium-long-term load forecasting of urban power network,more emphasis was put on macro perspectives and some concerning policies and laws.