Research On Real-time Control Optimization Algorithm For Power Demand Side Under Uncertain Environment

Since the 21 st century,the increasing electricity demand and the rapid growth in the number of new types of loads,the issue of energy management on the demand side of electricity has received critical challenges and extensive attention.Along with the concern of carbon dioxide emissions from excessive reliance on traditional fossil fuels（coal,oil,natural gas）,energy conservation and emission reduction have become the top priority of power demand-side energy management.Currently,the application of distributed energy and distributed energy storage technologies are considered to be the effective way to alleviate the energy consumption problem and reduce carbon emissions,as well as to provide more power revenue for the electricity demand side.However,due to the inherent stochastic and intermittent properties of renewable energy sources（wind and solar）,as well as the uncertainty of power consumption behavior on the power demand side makes it difficult to accurately predict their statistical knowledge information,which undoubtedly brings great challenges to power scheduling and cost optimization on the electricity demand side.Therefore,it is very necessary to study the energy management of electricity demand side under uncertain environment（without knowing the statistical knowledge of renewable energy generation,electricity price and electricity demand）in order to promote the development of energy saving and demand side management.In this paper,we study the energy scheduling and cost optimization problems on the electricity demand side in an uncertain environment,and our main work is as follows.1.Firstly,the research background and significance of this topic are introduced in detail,and the research results of domestic and foreign research on the electricity demand-side energy management scheme are summarized in depth,and the ideas and methods of cost reduction and efficiency enhancement are elaborated.In view of the unknown statistical knowledge of renewable energy generation,electricity price and electricity demand,a queue model is established to characterize their stochastic arrival process,and the relevant theoretical knowledge of the queue is elaborated.Finally,the theoretical methods required for solving the stochastic optimization problem are introduced to provide a theoretical basis for the research of the electricity demandside energy management.2.Investigate dynamic energy trading and load dispatch management schemes for end-users in smart grids.It is assumed that the end-user is equipped with renewable energy generators and storage batteries,and maintains real-time communication with the grid through the smart meter to obtain tariff-related information.The end-user long-term time-averaged profit maximization problem is investigated under the combined consideration of unknown statistical knowledge of renewable energy generation,electric load and tariff and satisfying the battery capacity boundary and the load delay constraint.After a series of problem corrections and transformations,a dynamic energy trading and load scheduling algorithm based on an improved Lyapunov optimization method is designed to solve the profit maximization problem.In addition,in the performance analysis section of the algorithm,it is theoretically demonstrated that the proposed algorithm ensures that the optimization results are close to optimal without the need to know the statistical characteristics of renewable energy generation,electricity load,and electricity price.Finally,numerical simulations show that the proposed algorithm outperforms other algorithms in terms of both profit maximization and average load waiting time delay,which verifies its effectiveness.3.Considering that the energy structure on the demand side of electricity presents the form of multiple energy combinations coexisting,based on the above research,a multi-energy management framework including different energy sources such as traditional electricity,heat and natural gas is proposed to expand the traditional demand-side energy management problem of single energy supply to the problem of joint supply and consumption of multiple energy sources.According to different energy demand relationships,the coupling and complementary relationship between multiple energy sources and the coordination and optimization are studied under the premise of satisfying energy demand,so as to achieve the lowest total operating cost on the demand side while achieving the optimal comprehensive energy efficiency.