Research On Optimal Scheduling Of Unit Commitment With Large-scale Flexible Access To Plug-in Electric Vehicle Load

In recent years,with the improvement of availability of charging infrastructure,the continuous maturity of V2 G technology and the incentives of favorable policies and measures,the number of plug-in electric vehicle(PEV)is increasing.The charging and discharging of electric vehicles have the characteristics of randomness and intermittentness.Its large-scale access to the power grid will bring many adverse effects to the safety planning,operation and market operation of the power system,which may cause load overload and exacerbate power imbalances.However,electric vehicles can also be used as a distributed energy storage,and the stored electrical energy can be fed back to the grid through V2 G technology.Therefore,based on the different charging strategies of electric vehicles,they are connected to the power system to the optimal objectives of economy and environment has become an urgent problem.And the high-dimensional,non-linear,multi-constrained and mixed integer features of the system become necessary considerations in the optimization process.Therefore,in this thesis,the unit status and the PEV charge and discharge load are taken as the dispatching objects,improves the impact of PEV load on the power system,and achieve the optimal objective functions of economy,environment and the satisfaction of electric vehicle users.Firstly,according to the charging strategy and system characteristics in different scenarios,the traditional unit combination model is improved and a new mathematical model is established;Then,the binary competitive particle swarm optimization(BCSO)and parallel optimization algorithm framework are proposed to reduce the generation cost;The charging weight factor is designed to explore the impact of the proportional relationship between demand-side load and unschedulable load on the system,and achieve the effect of "peak cutting and valley filling".An appropriate algorithm for solving a multi-objective optimization problem is selected through experimental analysis,and a suitable scheduling reference scheme is selected for different scenarios by designing a decision scheme.The details are as follows:(1)In this thesis,a single objective optimization model based on PEV charging strategy and a multi-objective optimization model based on PEV charging and discharging strategy are proposed,which all based on the traditional unit combination model,and combining with different charging strategies and load distribution scenarios of electric vehicles.At the same time,in order to achieve the binary optimization in the single-objective model,based on the CSO algorithm and combining the binary strategy,the BCSO algorithm is proposed and applied,and also verify the algorithm performance.(2)In this thesis,when considering the charging strategy of PEV,according to the different load scheduling requirements in different places,the charging weight factor is designed to divide the charging load into demand-side load and non-schedulable load.And the possible impact of the different proportion relationship between the two on the system is discussed.Considering the mixed coding problem in this case,a parallel optimization algorithm framework is proposed to optimize the unit combination problem of accessing charging loads of different scales.It effectively realizes the function of "peak cutting and valley filling",and saves economic cost and reduces the influence of electric vehicle load on the system.(3)In this thesis,considering whether the users are satisfied with the charging freedom of the scheduled electric vehicle load,the charging deviation function is proposed,and a new super multi-objective solution problem is formed by combining the traditional unit commitment multi-objective optimization problem.Due to the strong mutual exclusion of multiple objectives under the model,the common MOEA / D,NSGA-II and NSGA-III algorithms are used to optimize the multi-objective problem,and the algorithm suitable for solving the problem is selected through experimental comparison.(4)Finally,according to the economic and environmental evaluation indexes of power system in practical engineering problems,and the different needs of users,the decision-making scheme is designed according to the method of normalization and weighted function sum,and the Pareto front obtained from the above algorithm is selected to provide the most appropriate reference scheme for decision makers.