基于模型预测控制的微电网与配电网有功优化调度研究

发布时间：2018-05-04 19:13

本文选题：微电网 + 配电网　；参考：《北京交通大学》2017年硕士论文

【摘要】：随着全球范围内化石能源储备不断下降,环境污染问题日益严峻,利用可再生能源的分布式发电作为一种清洁的产能方式,逐渐成为应对上述问题的重要手段而受到广泛关注。微网是承载分布式发电接入配电网的有效载体,随着分布式电源在区域配电网中的渗透率迅速提升,研究配电网与微网的协调优化调度以实现可再生能源发电的消纳、降低分布式发电对配电网的负面影响、提升电网运行的经济性等成为研究热点。本文首先对分布式电源和微网的发展进行探讨,总结了国内外微网的发展情况,分析了微网和主动配电网在优化调度方面的研究进展,结合实际背景阐述了本文的研究意义。其次,分析了配电网传统电力调度方式在高比例分布式电源并网的场景下存在的缺陷和不足,将微网协调控制策略的分层控制与能量管理相结合,提出由配电网调度层、微网集中控制层和就地控制层组成的配电网与微网三层协调调度架构。然后,对基于模型预测控制的微网多时间尺度优化调度进行研究。通过分析微网中各分布式电源和储能的运行特性,建立其成本模型;在日前调度阶段,以由微网运维成本、环境治理成本和燃料成本的构成综合运行成本最低为优化目标,运用改进的遗传算法求解微网的经济优化运行方案;日内优化阶段,提出了基于线性状态空间模型预测控制算法的微网多时间尺度滚动优化策略,修正了微网内可调度单元的日内短时间尺度的运行计划,实现了微网联络线功率的平稳跟踪。最后,根据本文提出的配电网与微网的三层协调调度架构,对配电网与微网的协调调度展开研究。分析了分布式电源接入对配电网网损和节点电压的影响,以配电网运行网损最小为优化目标,考虑分布式发电就地消纳、电压安全、集中式储能等运行约束,利用改进的支路电流前推回代法和遗传算法对模型求解,完成对配电网层的优化调度,确定微网的日前运行计划。利用仿真算例进行对比分析,验证了方法的有效性。
[Abstract]:With the decrease of fossil energy reserves in the world, the problem of environmental pollution is becoming more and more serious. As a kind of clean production capacity, distributed power generation using renewable energy has gradually become an important means to deal with these problems. Microgrid is an effective carrier for carrying distributed generation to the distribution network. With the permeability of distributed generation in regional distribution network increasing rapidly, the coordinated optimal dispatching between distribution network and microgrid is studied to realize the absorption of renewable energy generation. Reducing the negative impact of distributed generation on distribution network and improving the economic performance of power grid become the research focus. In this paper, the development of distributed power generation and microgrid is discussed, the development of microgrid at home and abroad is summarized, the research progress of microgrid and active distribution network in optimal dispatching is analyzed, and the significance of this paper is expounded in combination with the actual background. Secondly, this paper analyzes the defects and shortcomings of the traditional power dispatching mode of distribution network under the scenario of high proportion of distributed generation connected to the grid, and combines the hierarchical control of micro-grid coordination control strategy with energy management, and proposes the dispatching layer of distribution network. A three-layer coordinated dispatching architecture of distribution network and micro-grid composed of centralized control layer and local control layer. Then, the multi-time scale optimal scheduling of microgrid based on model predictive control is studied. By analyzing the operating characteristics of each distributed power source and energy storage in microgrid, the cost model is established, and in the pre-day scheduling stage, the minimum integrated operating cost is considered as the composition of operation and maintenance cost, environmental treatment cost and fuel cost. The improved genetic algorithm is used to solve the economic optimization scheme of microgrid, and in the in-day optimization stage, a multi-time rolling optimization strategy based on linear state space model predictive control algorithm is proposed. The operation plan of the scheduling unit in the microgrid on a short time scale is corrected, and the steady tracking of the tie-line power of the microgrid is realized. Finally, according to the three-layer coordinated dispatching architecture of distribution network and microgrid, the coordinated dispatching between distribution network and micro-network is studied. The influence of distributed generation access on distribution network loss and node voltage is analyzed. The minimum network loss of distribution network is considered as the optimization objective, and the operational constraints such as local absorption of distributed generation, voltage security and centralized energy storage are considered. The improved branch current forward and back substitution method and genetic algorithm are used to solve the model to complete the optimal dispatching of the distribution network layer and to determine the daily operation plan of the microgrid. The effectiveness of the method is verified by a simulation example.
【学位授予单位】：北京交通大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM73

【参考文献】