基于鲁棒优化的商业楼宇型微网应急能量管理策略

发布时间：2018-12-16 13:16

【摘要】：在正常情况下,商业楼宇型微网采用并网运行方式,当配网出现故障时,商业楼宇型微网切换至孤岛运行。在微网应急运行的过程中,能量管理系统需兼顾减小停电影响范围与维持微网的可靠持续供电两方面。由于微网孤岛运行后,光伏出力的不确定性依然存在,计及不确定因素的能量管理将更有利于系统的可靠运行。本文针对含有光储系统的商业楼宇型微网,提出了一种基于鲁棒优化的应急能量管理策略。应急能量管理的优化目标为减小停电损失、维持供电时间,且引入权重因子以权衡二者的重要程度,约束条件则考虑到功率平衡、可控负荷与储能系统的运行限制。此外,在优化模型中设置了不确定度以表征鲁棒优化的保守程度,且不确定度的大小决定了不确定因素对问题的影响程度。优化模型中权重因子和不确定度的数值可根据实际情况灵活设定。此外,在鲁棒优化模型的基础上引入了滚动优化的思想,以定时更新光伏的实际出力数据,从而获得更为准确的光伏预测数据。在由上述应急能量管理策略确定可控负荷投入量后,给出了以“先调节、后切除”为原则的可控负荷调控策略,其中,电动汽车采用“柔性调节-刚性调节”两阶段调节方案,空调采用“奇数组-偶数组”轮换式调节方案,由此可得到停电后电动汽车的实时充电倍率与空调的实时功耗。通过算例对比分析可知,该策略可通过合理调整可控负荷投入量,以实现预期目标;通过算例灵敏度分析与不确定度分析可知,该策略可通过灵活设定权重因子与不确定度的数值,以调节两个子目标的重要程度及不确定因素对问题的影响程度。
[Abstract]:Under normal conditions, commercial building microgrids operate in a grid-connected mode. When the distribution network fails, the commercial building-type microgrids switch to islanding operation. In the process of emergency operation of microgrid, the energy management system should take into account two aspects: reducing the influence of power outage and maintaining the reliable and continuous power supply of microgrid. Since the uncertainty of photovoltaic force still exists after the operation of microgrid isolated islands, energy management with uncertain factors will be more favorable to the reliable operation of the system. In this paper, an emergency energy management strategy based on robust optimization is proposed for commercial building microgrids with optical storage systems. The optimization goal of emergency energy management is to reduce the power loss, maintain the power supply time, and introduce the weight factor to weigh the importance of the two factors, and the constraints take into account the power balance, controllable load and energy storage system operation constraints. In addition, uncertainty is set in the optimization model to represent the conservative degree of robust optimization, and the degree of uncertainty determines the influence of uncertainty on the problem. The value of weight factor and uncertainty in the optimization model can be flexibly set according to the actual situation. In addition, the idea of rolling optimization is introduced on the basis of the robust optimization model to update the actual output data of photovoltaic periodically, so as to obtain more accurate photovoltaic prediction data. After determining the amount of controllable load input from the above emergency energy management strategy, a controllable load control strategy based on the principle of "first adjusting, then removing" is presented, in which the "flexible regulation-rigid regulation" two-stage regulation scheme is adopted for electric vehicles. The air conditioning system adopts "odd group-even array" rotation regulation scheme, which can get the real time charging rate of electric vehicle after power failure and the real time power consumption of air conditioning system. Through the comparison and analysis of examples, it can be concluded that the strategy can achieve the expected goal by reasonably adjusting the amount of controllable load input. The sensitivity analysis and uncertainty analysis show that the strategy can adjust the importance of the two sub-targets and the influence of uncertainty on the problem by flexibly setting the value of weight factor and uncertainty.
【学位授予单位】：华北电力大学(北京)
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM73

【参考文献】