In order to reduce the production cost of the water plant,achieve the goal of"carbon peak and carbon neutrality",and reduce the impact of the uncertainty of wind and solar output on the power system,this paper establishes the energy management system of the smart water plant,and proposes a long short-term memory neural network based on digital twin and fully adaptive noise integration empirical mode decomposition method and sparrow search algorithm optimization as a prediction model to improve the prediction accuracy of short-term wind and solar output,and provide a data basis for decision-making layers to guide the power dispatch in smart water plants.Using a smart water plant in northwest China as a simulation example,the experimental results show that the proposed method can effectively improve the prediction accuracy of short-term wind and solar output and reduce the operation and maintenance cost of the water plants.
关键词
智慧水厂/数字孪生/能源管理/神经网络
Key words
smart water plant/digital twins/energy management/neural networks
维普
万方数据