Optimization Algorithms For Demand-side Energy Management And Economic Dispatch In Smart Grid

With the development of society,people’s requirement for power quality is constantly improving and the number of distributed power resources is also increasing,which make the traditional power grid difficult to meet the needs of the new era,the smart grid has become the reform trend of the current power industry.As the key technologies of smart grid,demand-side energy management and economic dispatch are the hot areas of research in recent years.Demand response encourages consumers to actively participate in the regulation of the power grid system and change the original mode of electricity usage,which is an important mechanism to ensure the stable operation of the power grid.Economic dispatch is one of the most fundamental problems in the power grid system.Through advanced control technologies and optimization methods,economic dispatch rationally allocates various distributed power resources,improves the economy and reliability of power grid operation,it is of great significance to the development of smart grid.The demand-side energy management and economic dispatch schemes of smart grid are designed in this paper based on various optimization methods,and the main research contents are summarized as follows:Firstly,taking into account the operating states of the loads,a demand-side energy management problem based on intelligent optimization algorithm is studied.An energy management optimization model is formulated considering the relationship between the operating states and the energy consumption of the loads under forecast errors.The sum of the consumers’ discomfort costs and the energy cost of the provider is minimized with the operating states limits and the supply constraints satisfied.The optimization model is applied to distributed heating,ventilation,and air conditioning(HVAC)systems,and the discomfort costs of the consumers are formulated based on the Fanger thermal comfort model.The bacterial colony chemotaxis(BCC)algorithm is applied to solve the optimization problem,so that the optimal temperature settings of the consumers and the optimal energy supply of the provider are obtained.Thesimulation results demonstrate that the BCC algorithm shows good convergence performance,and the total costs are reduced with the balance between supply and demand achieved by the proposed scheduling strategy.Secondly,considering the price oscillations in electricity markets,a demand-side energy management problem based on dual method is studied.An energy management scheme which can schedule the supply and the demand simultaneously for distributed HVAC systems of residential buildings is proposed with the purpose of reducing the total costs of the cosnsumers and the energy provider.The Fanger thermal comfort model is simplified by the quadratic fitting method to formulate the discomfort costs of the consumers.An energy management algorithm based on dual-gradient algorithm is proposed to obtain the optimal temperature settings,the optimal energy supply and the optimal electricity price in real time.In particular,due to the ubiquity of price oscillations in electricity markets,the impact of price oscillations on the performance of the proposed energy management algorithm is analyzed and examined,and the convergence condition of the algorithm with price oscillations is established.The simulation results verify the validity of the algorithm.Finally,a distributed algorithm based on gradient descent and consensus protocol is studied for the economic dispatch problem in power grids.The economic dispatch optimization model is established to minminize the total generation costs with the energy supply and demand reaching the balance.A discrete-time distributed economic scheduling strategy is designed with the gradient descent in combination with the consensus algorithm.Through the theoretical analysis,it is revealed that the proposed algorithm can converge to the optimal solution of the primal problem by choosing the appropriate initial values and step size.The simulation results show the validity and scalability of the algorithm.