Research On Energy Dispatch Strategy Of Distributed PV-Storage System For Near Space Vehicle

The stratospheric space has many natural advantages,such as stable weather condition and high radiation intensity,which are attractive for special application.Development of renewable untilization and power electronics technology provides vast possibility for developing solar-powered unmanned aerial vehicles(SUAV),hovering in the stratospheric space.These longterm flying vehicles are powered by surface adhering photovoltaic arrays during daytime and batteries at night.Because of the large surface curvature and the changeable flight attitude of UAV,the solar radiation and output power of photovoltaic modules are different,which can be divided into small groups to improve the power generation.In addition,in order to make full use of the capacity of airborne battery and improve the power density of energy system,it is necessary to study special structure of PV-battery system to achieve continuous and reliable power supply.Under this background,this thesis studies the energy system of UAV based on PV-battery modules,and focuses on the energy dispatch strategy which is one of its core technologies,to maximize the utilization of photovoltaic power and keep the state of charge of batteries as consistent as possible.Firstly,the load model,photovoltaic model and battery model considering flight altitude and attitude of UAV are established based on the analysis of topology and working principle for energy system.These models lay a foundation for the study of energy dispatch strategy.In order to make full use of the capacity of photovoltaic and battery,the method is studied which determines the output power of each photovoltaic array and battery substring according to the current state of charge of batteries,flight plan and path for next day.This method aims at energy balance of batteries at the end of dispatching cycle,battery life loss and charging energy among groups as constraints.The modified particle swarm optimization algorithm is used to solve the problem.The simulation results show that dispatch strategy can improve the utilization of photovoltaic and optimize the operation of batteries.A real-time energy optimal dispatch which aims to track the day-ahead scheduling of charge state is established to overcome the influence of the deviation between actual power and predicted power of photovoltaic and load,.The energy deviation during each period is allocated to each module according to the day-ahead power plan of the next period,and the output power of each module is calculated in real time according to the adjusted energy Simulation results show that the strategy is effective and real-time.In order to overcome the influence of track changes and module failure,an optimization algorithm for battery energy balance is established in this paper.The algorithm includes two stages.The first stage is a bi-level optimization model without charging among groups.The upper level aims at the energy balance of groups at the end of the scheduling cycle,and the variable is the output power of each group.The lower level aims at the energy balance of modules,and the variable is the output power of each module.The two-level model is solved by particle swarm optimization and interior point method.In the second stage,the group with low energy is allowed to be charged,the model aims at energy balance of these groups at the end of the dispatching cycle to adjust the results of the first stage.The interior point method is used to solve the problem.Correctness and effectiveness of this model are verified by simulation of different scenarios.The operation modes are tested such as the output power change among groups and modules on the experimental platform of PV-battery system,The experimental results prove the schedulability of the system and the effectiveness of the dispatch strategy.