Distribution Network Expansion Planning Considering A Distributed Hydrogen-thermal Storage System In High Proportion Pv Scenarios

Under the background of PV power generation whole-county development,the large-scale access of high proportion distributed PV power brings strong source-side stochasticity and volatility to the distribution network.In order to reduce the uncertainty and improve the PV consumption rate,this paper proposes a bi-level expansion planning model for a distributed hydrogen-thermal storage system(DHTSS)with a high proportion of PV scenarios.First,the refined model construction of distributed system of distribution network is carried out.In order to solve the problem of insufficient accuracy of traditional PV output model in high proportion PV power generation scenarios,the physical model of each link of irradiation transformed PV output is studied,and the inner coupling relationship of each physical model is analyzed.The refined physical model chain of distributed PV system is constructed based on angle of incidence(AOI)model,plane of array irradiance(POA)conversion model,PV system PV cell temperature model,PVWatts losses model,and the PV system model.To solve the problem of source-load imbalance during the peak PV output time in high proportion PV power generation scenarios,which makes it difficult to effectively consume large-scale irradiated resources,a distributed hydrogen-thermal storage system operation model is established based on hydrogen-based multi-energy flow multi-energy coupling,which includes electric heat pump(EHP),thermal storage tank(TST),alkaline electrolyzer(AE),hydrogen storage tank(HST),solid oxide fuel cell(SOFC),electric refrigerator(ER),and lithium bromide refrigerator(LBR).Then,a multi-grids joint scenarios generation model is constructed.To effectively portray the high proportion of PV power generation and multiple load energy use uncertainty,a two-layer coupled meteorological clustering model based on the temperature-meteorological and temperature-load correlation analysis results is established.The historical temperature data are clustered in the upper layer to effectively delineate the source-load differential temperature scenario.The historical wind speed,relative humidity and barometric pressure data are clustered in the lower layer to refine the source-load uncertainty scenarios.For the irradiation,temperature and multi-energy load fluctuation characteristics of each grid under each meteorological scenario,a multi-source load multi-meteorological multi-node joint scenarios generation model is constructed based on ACWGAN-GP.The historical power load data are screened and reorganized to obtain 95% peak load scenarios,extreme high temperature scenarios and extreme low temperature scenarios to build small probability high load scenarios generation model.The historical data of high output distributed PV power are screened and reorganized to build high PV output scenarios generation model.Finally,in order to improve the power supply reliability of the distribution network in little probability energy use scenarios,the objective of power supply reliability in the little probability high load scenarios is proposed.To further solve the problem of large-scale PV local consumption in high PV output scenarios in the area to be expanded,the objective of PV consumption in high PV output scenarios is proposed.The Timing Voltage Sensitivity Index(TVSI)is introduced to optimize the siting of distributed hydrogen-thermal storage systems.A multi-objective bi-level expansion planning model for distribution networks is constructed to reasonably plan the capacity of each equipment of distributed hydrogen-thermal storage systems,distribution lines and low-voltage transformers to be expanded in distribution networks,so as to improve the economics and reliability of distribution network expansion planning in the context of high proportion PV development.The proposed distribution network expansion planning model is designed to improve the economics of distribution network expansion planning,operational reliability of little probability high load scenarios and PV consumption rate of high PV output scenarios.The experimental analysis shows that the proposed multi-objective bi-level expansion planning model of distribution network can balance the conflicting issues of planning economy of typical scenarios,reliability of power supply in little probability high load scenarios and PV consumption rate in high PV output scenarios,which can well solve the problem of economy,reliability and PV consumption of distribution network in high proportion PV scenarios.