Multi-objective Optimal Dispatch Of Active Distribution Networks With Distributed Energy Storage

The growing demand for energy,the depletion of traditional energy sources,and the increasing environmental pollution have driven more and more attention to renewable energy.Nowadays,how to improve the ability to absorb renewable energy in distribution networks is an important part of the smart grids development.Traditional distribution networks cannot cope with the problem of large-scale distributed renewable energy integration,and active distribution network(ADN)technology has emerged.ADN coordinated control distributed generation,energy storage,and load within the system through advanced measurement,communication,and intelligent control technologies.ADN has an important practical significance in achieving effective management of distribution network power flow,and promoting large-scale grid-connected renewable energy integration.The ADN optimal dispatch strategy is the core technology for achieving effective control of distributed resources and ensuring the security,stability,and economic operation of the system.However,with the increasing penetration of intermittent renewable energy sources,the increasing uncertainties have made distribution network dispatch more difficult.In addition,energy storage and flexible loads have coupling properties at different time periods,and include various constraints.After participating in ADN dispatch,comprehensive optimization of multiple periods within the scheduling period is required.The control dimension and search space of optimal ADN dispatch are increased.The search process for optimal solution becomes more complicated.Therefore,the control variables,optimization functions,and constraints of ADN optimal dispatch have changed greatly compared with the traditional grid scheduling.The optimal dispatch model and method based on the traditional optimal power flow cannot be directly used in the ADN.Based on the above background,in order to improve economy,save energy and reduce emissions,reduce network loss,and improve voltage quality,this paper studies the multi-objective optimal dispatch model and optimization algorithm of ADN under the scenario of distributed energy storage participation.Firstly,aiming at the difficulty in solving ADN optimal dispatch model,an improved multi-objective HTL-MOPSO algorithm is proposed,which strengthens the search ability and provides an algorithm basis for the efficient solving of the subsequent ADN optimal dispatch model.To solve the problem of reasonable allocation of energy storage in ADN,a distributed energy storage configuration method based on dual-layer multi-objective optimization is proposed to lay the foundation for dispatch strategies research of energy storage in ADN.Then,in order to solve the renewable energy prediction error problem,a robust multi-objective optimal dispatch method for ADN with distributed energy storage is proposed.Finally,based on the above research,a multi-objective optimal dispatch method considering demand response is proposed,and the optimal dispatch of “generation-net-load-storage” coordination in ADN is realized.The main contents research works are summarized as follows:(1)Aiming at solving the complex problems of the ADN optimal dispatch model,the HTL-MOPSO algorithm is proposed to enhance convergence and diversity.First,through the introduction of simulated teaching and learning strategies,the information exchange among individuals in the population is further enhanced and the global search capability of the particle swarm optimization(PSO)is improved.Secondly,an improved Euclidean distance based circular crowding sorting strategy is proposed to maintain the diversity of elite particles and improve algorithm convergence.Based on the above strategies,a performance-enhanced HTL-MOPSO algorithm is proposed.Finally,a large number of typical multi-objective test functions are simulated and compared with other representative multi-objective optimization algorithms.It is verified that the proposed algorithm has the good convergence and diversity.This part provides the algorithm basis for the subsequent chapters to efficiently solve complex ADN multi-objective optimal dispatch problems.(2)Aiming at the allocation issue of distributed energy storage in ADN,a joint multi-objective optimization method for energy storage configuration and operation is proposed.Taking into account the influence of the pre-storage configuration and the post-operation of the energy storage in the ADN,a two-layer coupled multi-objective optimal configuration model for energy storage is constructed.The proposed model takes the minimization of the overall cost and voltage deviation of the energy storage configuration as the objective function and takes energy storage access position and capacity as decision variables.The outer layer of the model optimizes energy storage configuration solutions.The inner layer implements the optimal operation of the system by formulating the distributed generation output,charge and discharge of energy storage,and reactive power output strategies in accordance with the corresponding configuration scheme to match the decision of the outer layer.HTL-MOPSO is respectively used in the outer and inner layers of the model.Finally,a simulation example shows that the energy storage optimal configuration method proposed in this paper can provide decision-makers with multiple choice schemes that takes into account both economical and voltage indicators.The result confirms that HTL-MOPSO has better search ability than other algorithms.This part lays the foundation for subsequent energy storage research in the optimal dispatch strategy in ADN.(3)Aiming at the prediction errors of the renewable energy generation,an optimal dispatching method for ADN robust economic environment benefits with distributed energy storage is proposed.Based on the predictive value of wind power output,the probability distribution of forecasting error of wind power output is considered.Based on worst-based optimization theory,a robust multi-objective optimization method is proposed to deal with wind power forecasting error.Then,with the objective functions of minimizing the operation cost of the distribution network and the polluting gases emission,taking into account the constraints of energy storage and safe,stable operation of the distribution network,a robust multi-objective optimal dispatch model for ADN is constructed to achieve a coordinated and optimized operation of distributed generation and energy storage.The simulation example verifies that the proposed robust dispatch method can realize the self-adaptive adjustment according to the uncertain wind power output.The comparison and analysis of algorithms proves that the HTL-MOPSO algorithm has the good search ability.(4)Aiming at the demand side participation in power grid scheduling problem,a multi-objective coordinated optimal dispatch method of “generation-net-load-storage” coordination in ADN is proposed.First,the author analyzed the relationship between electricity consumption and electricity price elasticity of users.From the perspective of distribution network managers,a demand response model is developed taking time-of-use price as the decision variables,and subjects to users' electricity consumption and electricity bills constraints.Then,based on the demand response,considering the minimization of distribution network operating cost and network loss as the objective function,and considering the constraints of energy storage and safe,stable operation of the distribution network,a multi-objective optimal dispatch model for distribution network is constructed.This model achieves the goal of coordinated optimal control and comprehensive management of “generation-net-load-storage” in ADN.The simulation example verifies that the optimal dispatch model.It verified that demand response can simultaneously achieve peak load reduction and valley filling,reduce the economic cost of the distribution network and reduce the network loss.Meanwhile,with the greater demand response participation,peak shaving,reduction of the economic costs and power loss of distribution networks is more obvious.The example analysis proves that HTL-MOPSO has certain advantages in model solving.