Font Size: a A A

Research On Key Technologies Of Microgrid Based On Demand Side Response

Posted on:2022-03-24Degree:MasterType:Thesis
Country:ChinaCandidate:Y H CaiFull Text:PDF
GTID:2512306527970289Subject:Control Science and Engineering
Abstract/Summary:
With the increasingly serious problems of energy shortage and environmental pollution,the importance of the research and utilization of renewable energy is increasing day by day,and more and more new energy sources are involved in the power system through the microgrid system.But at the same time,the intermittency and variability of new energy sources bring new challenges to the economic and stable operation of the microgrid system.To address this situation,this article focuses on key technologies based on demand-side response for a microgrid composed of wind power,water electrolysis device,hydrogen-oxygen fuel cells and microgrid loads.The specific steps are as follows:First,analyze the operating characteristics of each system unit in the microgrid,and study its corresponding mathematical model.Secondly,in response to the unstable characteristics of load forecasting with strong randomness and many affected factors,different forecasting tools and suitable forecasting methods are used.The load forecasting model based on an improved back propagation neural network,the gray neural network forecasting model based on Greymodel(GM)(1,1)and the load forecasting model based on an improved long shortterm memory(LSTM)neural network are included.These three forecasting models are simulated and verified through calculation examples.The comparison reveals that the load forecasting model based on the improved LSTM neural network outperforms the other models in terms of prediction accuracy,reliability,and the goodness of fit.Then,the operation model of the microgrid is analyzed based on the demand-side response.The analysis process is mainly divided into three parts: the first part is to analyze the function of each module in the microgrid operation model;the second part is to study the load transfer strategy and solve it using greedy-genetic based optimization algorithm;The third part is to study the microgrid operation control strategy,starting from the energy storage system scheduling strategy,constructing the coordinated optimal operation strategy of the microgrid,and using the genetic algorithm to optimize and solve the microgrid coordinated optimal operation strategy.Afterwards,to correct the errors and disturbances in the operation of the microgrid,a feedback correction system is introduced to form a closed-loop control,and the process of microgrid operation is continuously optimized on a rolling basis.Finally,the simulation analysis of microgrid operation based on demand-side response is carried out to observe the effect of the optimization algorithm by using the measured data from a region in northern China and a region in Europe as examples.After optimizing the microgrid system through the demand-side coordinated operation optimization algorithm of the microgrid,the consumption rate of new energy is improved,and the stability of the microgrid operation is improved while achieving the maximum local consumption of new energy.
Keywords/Search Tags:Demand side response, Multi-source microgrid system, Improved LSTM neural network, Energy management, Wind power consumption
Related items