Font Size: a A A

Distributed Energy Storage Optimal Allocation And Life Cycle Economic Evaluation Method In Distribution Network

Posted on:2020-08-03Degree:MasterType:Thesis
Country:ChinaCandidate:Y F LiuFull Text:PDF
GTID:2392330623460131Subject:Electrical engineering
Abstract/Summary:PDF Full Text Request
Renewable energy utilization has been agreed in principle considering the urgent need of developing low-carbon economy,promoting ecological civilization and realizing sustainable development in the world.Distributed generation(DG)technology based on wind and photovoltaic renewable energy generation has been widely studied and applied,thus the operation mode of distribution network has evolved from passive to active.The increasing access of high penetration renewable distributed generation and multi-type terminal load to distribution network result in the problems of voltage overshoot,reverse power flow and imbalance power.As a regulatable resource with source-load characteristics,distributed energy storage can be compatible with intermittent renewable energy,adjust load peak-valley difference,and break the asynchrony between energy production and consumption.It is an effective method to deal with such problems.Aiming at the application of distributed energy storage access to distribution network,This paper studies the coordinate planning model of distributed energy storage and distribution network,efficient solution method for mixed integer nonlinear optimization problem and energy storage benefit evaluation method based on life cycle theory.The research is based on distribution network structure,equipment characteristics and typical scenarios.Following work has been accomplished.(1)A multi-stage coordinated planning model for distribution network considering distributed energy storage and distributed generation is proposed.In terms of power flow constraints,based on the traditional Distflow power flow model and directional Distflow power flow model for radial power flow,an extended Distflow power flow model for non-radial power flow is established,and the second-order conic relaxation of the quadratic constraints in the model is proposed.In terms of network topology constraints,under the conditions of network connectivity and open-loop,a set of constraints based on virtual power flow and treelike open-loop structure is proposed to ensure that the distribution network operates in radial structure.The objective of optimization is to minimize the present value of total cost including construction cost,generation cost,maintenance cost and penalty cost.Considering the optimal allocation of distributed energy storage and distributed generation,the construction and expansion of stations and lines,a multi-stage coordinated planning model of distribution network is established.(2)An improved Benders decomposition method for multi-stage coordinated planning model of distribution network is proposed.The large-scale mixed integer non-linear optimization problem represented by the original model is decomposed into a main problem and sub-problems according to different time scales.The main problem is the construction problem with the planning stage as the time scale,and the sub-problem is the operation problem with a scenario in the planning stage as the time scale.The main and sub-problems are mixed integer optimization problems.The sub-problems have no dual problems,so the optimal cut sets can not be generated by the conventional Benders decomposition method.This paper studied the generation method of the optimal cut sets in the iteration process.The relaxation method of the sub-problems is proposed.The optimal cut sets of the sub-problems are established to continuously optimize the integer variables of the sub-problems,so that the optimal cut sets of the main problems can be generated.Iterative process between main problem and sub-problems is presented.(3)A distributed energy storage cost-benefit evaluation method based on life cycle theory is proposed.The intra-day wind-light-load characteristic curves are clustered by the peak density clustering method.The multi-dimensional state chain of inter-stage price-wind-light-load is generated by the Markov Chain Monte Carlo method.The multi-dimensional state scene tree is generated and scene reduction is carried out based on the correlation coefficient of eigenvectors.Thus typical operation scenarios are obtained.The simulation method of energy storage lifetime is studied.The energy storage capacity decay rate model is established considering charge-discharge ratio and cycle number,and the equivalent cycle life statistics method considering discharge depth is proposed.A cost-benefit model of energy storage and the calculation method of economic indicators are presented,and a life cycle based energy storage evaluation method is proposed.
Keywords/Search Tags:distributed energy storage, distribution network planning, improved Benders decomposition, cost-benefit assessment, capacity decay analysis
PDF Full Text Request
Related items