The Control Strategy Of Controlled Loads With Smoothing Response To Wind Power In Active Power Distribution Systems

With the increasing depletion of conventional energy sources and the deteriorating environment,more and more people pay attention to renewable energy,distributed power access to the proportion of distribution network is growing.Due to its intermittent and volatility,the active distribution network is different from the traditional distribution network,which contains distributed energy,controllable loads and energy storage system.The active distribution network can change the distribution network topology and trend,reduce the impact of intermittent energy,power distribution fluctuations in the distribution network to achieve optimal system operation.First,the paper introduces the theoretical basis,including the demand response and active power distribution network,and introduces the development of the demand response at home and abroad,which appears the intelligence,flexibility and coordination of the active distribution network.Then,taking the active distribution network as the research environment,this paper constructs the system model of the active distribution network which contains the wind power and the typical load,and static and dynamic models which causes the electricity peak.Secondly,wavelet-BP neural network optimized by genetic algorithm is used to predict the output of wind power,and the prediction and remodeling of the signals after wavelet decomposition are carried out.The prediction results are better than what we get by using the traditional BP neural network.Loading curve characteristics improve fuzzy c-means clustering algorithm,which makes the clustering effect more practical,then we will analysis the characteristics of the clustering results index.Finally,loading controllability analysis is carried out on the characteristics of the subordinate industry in load classification.A multi-agent wind power control model is established by using controllability of the wind load.A multi-objective function under multi-condition constraint is established.The particle swarm optimization algorithm verifies the feasibility and economy of the control method.The control strategy of the controllable load for smooth response wind power in the paper,considering the power consumption reliability of the demand side,embodies the interaction of distributed energy,demand side response and power distribution system.In order to build a flexible,economical and integrated and coordination of the smart grid this paper provides a meaningful reference.