Research On DG Hosting Capacity Evaluation And Improvement Of Distribution Network

With the proposal of China’s "dual-carbon" goals,renewable energy represented by wind and solar power is continuously penetrating into the distribution network in the form of Distributed Generation(DG).However,the high penetration rate of DG brings many uncertainties to the operation of the distribution network.Meanwhile,the rapid development of energy storage systems(ESS),the use of a large number of grid control equipment,and the emergence of intelligent soft open points(SOP)have a huge impact on the acceptance of DG.Against this background,it is of great significance to accurately evaluate and efficiently improve the hosting capacity of DG in the distribution network.In order to ensure the safe and stable integration of DG into the distribution network and fully utilize its potential,this thesis focuses on the evaluation and improvement of the hosting capacity of DG in the distribution network.Firstly,in order to characterize the correlation between DG outputs and between DG outputs and load demand,a time-series scenario set was generated using the correlation matrix method and Latin Hypercube Sampling(LHS)method.Meanwhile,in order to improve the computational efficiency of subsequent research work and solve the problems of difficulty in determining the starting clustering center of common clustering algorithms and low clustering quality in processing massive data,a comprehensive distance-based improvement of Affinity Propagation(AP)clustering algorithm was proposed,which uses Euclidean distance,DTW distance,and differential cosine distance,significantly improving the clustering quality.Secondly,in order to balance the wind-solar-load correlation and the impact of active management measures on the evaluation of DG accommodation capacity,a DG accommodation capacity evaluation model was established based on the typical scenarios with correlation,aiming to maximize the DG access capacity and taking into account four active management measures: On-Load Tap Changer(OLTC),Capacitor Banks(CB),Static VAR Compensation(SVC),and Demand Response(DR)based on incentives.After linearizing and second-order cone relaxation of the model,the commercial solver CPLEX was used for solution,which can quickly obtain accurate evaluation results.Finally,considering the synergistic control effects of ESS and SOP on DG output absorption and the load growth during the planning period,a multi-stage planning model for improving the hosting capacity of DG based on the collaborative planning of ESS and SOP is established on the basis of the above.The upper layer of the model is the planning layer,which configures DG,ESS,and SOP with the minimum sum of investment and operation costs and the maximum hosting capacity of DG as the objective function.The lower layer is the operation layer,which controls various active control devices with the minimum voltage deviation as the objective function.The hybrid algorithm of NSGA-Ⅱ and second-order cone programming is used for solving,and the improved IEEE 33-node test system is used for verification.The results show that the proposed model has good economy,can effectively improve the hosting capacity of DG,and can also improve the system operation level.