Cooperative Planning Models And Methods Of Active Distribution Systems

Due to the accelerated deployment of renewable energy sources(RESs)in active distribution system(ADS)and rapid development of flexible loads such as electrical vehicles(EVs),ADS has become the key technology for promoting the development of smart grid and future energy system for sustainable and emission free energy supply.However,the traditional deterministic planning methods based on the worst and rare scenarios lead to economic inefficiency and are unable to deal with the uncertainties and variabilities associated with high penetration variable RESs and flexible loads.It is no longer valid for ADSs to accommodate these widespread RESs and flexible loads.Therefore,it urgently needs to develop the appropriate ADS planning methodologies and models to tackle a number of key problems brought by the high proportion RESs and flexible loads.In order to take the key theoretical and application issues of ADS planning into adequate consideration,the emphasis of this dissertation is given on the mathematical models and optimization methodologies of ADS planning issues.In this dissertation,based on the multi-level programming theory,several novel cooperative planning models and methodologies are introduced to satisfy different ADS planning issues from different perspectives.Compared to the previous research works,the main research contents and primary contributions of this study are summarized as follows.The multi-level programming problems and decentralized decision-making problem are analyzed including principles,models and solution algorithms.Then,the optimization performance of particle swarm optimization(PSO)algorithm is investigated and analyzed by the optimization problems of several standard multi-dimension and multi-modal functions.On this basis,the optimization process of multi-level programming problems based on PSO is introduced.Then the feasibility and availability of it are verified by several different types of bi-level programming problems,including continuous linear bi-level programming model,integer linear bi-level programming model,and nonlinear bi-level programming model.In order to tackle the high-level uncertainties and variabilities and take full consideration of the time-variable feature simultaneously,the effective multi-scenario tool is provided to establish high-quality typical scenarios based on the k-mean clustering algorithm and the cluster validity index of Davies Bouldin.Thus,the different probabilistic sequence operation scenarios are simulated,which can deal with high-level uncertainties and capture the time-complementarity between different RESs,as well as the time-dependence correlation between power outputs of RESs and power consumptions of load demands.In order to optimal allocate energy storage system(ESS)in ADS,which is integrated with wind generation,solar photovoltaic,and multi-microgrid,a fuzzy multi-objective bi-level optimization problem(single-objective/multi-objective)is proposed.Firstly,the power characteristic model of micro-grid is introduced to represent the time-variable feature of micro-grids.For purpose of considering the different roles of ESS in planning issues,the operation strategy of ESS is modeled as a fuzzy multi-objective model and integrated into the proposed planning model by means of bi-level programming model.The proposed bi-level planning model achieves the collaborative optimization between long-term planning issues and short-term operation issues.Then,a hybrid algorithm is introduced based on differential evolution and PSO to solve the proposed bi-level mixed integer problem.Finally,the IEEE-33 bus benchmark system is utilized to investigate the availability and effectiveness of the proposed model and the hybrid algorithm.In order to allocate variable distributed RESs and ESSs in ADSs collaboratively,a multi-objective,bi-level cooperative planning model(multi-objective/multi-objective)is proposed.Firstly,the mathematical model charging loads of EVs is provided to represent the temporal characteristics of EVs.Then,considering the key roles to improve economic efficiency and reliability,a fuzzy multi-objective optimal operation strategy model of ESSs is developed and integrated into the proposed cooperative planning model by means of bi-level programming.In order to solve the proposed bi-level mixed integer problem effectively,several modified measures are introduced to improve the PSO performance and obtain the high-quality Pareto front,including the tent chaotic mapping,the fast non-dominated sorting approach,and the crowding-distance assignment.Results based on the IEEE-33 bus benchmark system and the PG&E 69 benchmark system are presented and used to verify the availability and effectiveness of the proposed planning model.Furthermore,several key issues are explored and analyzed,including the impacts of EVs charging load demands and ESSs operation strategy on the planning model,as well as the time-dependence correlation between power outputs of RESs and power consumptions of load demands.In order to optimize the planning of RESs,ESSs and distribution networks in ADS cooperatively,a multi-objective,tri-level cooperative planning model(multi-objective/single-objective/single-objective)is proposed under the modern electricity market environment of ADS.Firstly,the different planning objectives and benefits of multi-stakeholder are analyzed,and then the feasibility of collaborative decision making between different stakeholders are discussed.On this basis,the proposed multi-objective,tri-level cooperative planning model are developed,where the upper level and the middle level serve to model the planning problems from different perspectives of multi-stakeholders.The lower level serves to model the operation aspect of ESSs.In order to balance the multiple objectives of the cost reduction,the reliability improvement,and the green penetration promotion,a modified Pareto-based PSO is employed to solve the proposed optimization problem.Finally,results based on the IEEE-33 bus benchmark system are presented and discussed,where the availability and the effectiveness of the proposed planning model are verified.