Research On The Evaluation Of New Energy Consumption Capacity Of Energy Internet Based On Edge Intelligence

As a cutting edge concept in the field of energy research,the Internet of Energy(Io E)closely combines decentralized energy and Information Communication Technology(ICT)to improve the intelligence of the power grid.In the Io E,a new model of interconnection and intercommunication has been realized.The Io E improves the intelligence and the automation of the power grid.The purpose of renewable energy sources(RES)accommodation is to increase the utilization rate of RES,and the RES includes wind power and photovoltaic power.Combining edge intelligence technology with the Io E can improve the computational efficiency of renewable energy accommodation.RES has the characteristics of distributed power generation,in order to reduce the cost of communication and improve computing efficiency,a hierarchical edge intelligence framework is designed here.In this framework,we deploy deep learning in the first layer of edge computing to discover the shrinking laws of renewable energy output;based on these laws,combined with the load requirements of different departments,the second layer of edge computing sequence uses deep learning to look for strategies to optimize the accommodation of renewable energy.The continuous deployment of large-scale intelligent measurement units has continuously increased the data generated by the Energy Internet,but unsatisfactory network communication conditions will lead to missing data.In order to solve the missing data and prediction,the first layer of deep learning edge interruption uses a gated recurrent unit algorithm to perceive the characteristics of renewable energy data for data processing.Subsequently,the framework proposed in this paper uses spinning reserve capacity to improve renewable energy ability to absorb energy.The experimental results verify effectiveness of the proposed deep reinforcement learning method for maximizing the accommodation of RES,and deep learning can also compensate for the lack of data and reduce the impact on intelligent algorithms.