主动型配电网日前调度策略研究

Study on day-ahead dispatch strategy of active distribution network

相对于传统配电网,主动型配电网可以合理利用其双向调度功能,充分发挥分布式发电容量,在保护环境的同时还可以提高电力公司的效益。以电力公司效益最大为调度目标,针对主动型配电网调度问题,该文计及输电和高压配电网网损的影响,提出了一种配电网日前优化调度模型,通过该模型对微电网电源交互功率和分布式电源进行优化调度,同时确定最佳渗透率,不仅可以有效地削峰填谷,还可以根据最佳渗透率调整分布式电源的功率,使电力公司效益最大化。算例研究表明,电网中接在距离电源点较近的节点下的主动型配电网,通过该文提出的优化调度模型,得出渗透率为21.14%时,电力公司效益最高。该研究可为主动型配电网日前调度策略以及分布式电源最佳渗透率的制定提供参考。

Comparing with the traditional distribution network, optimization scheduling of an active distribution network no longer passively conducts power distribution. Because controllable and adjustable distributed power can be scheduled as a resource to provide electricity power for the medium voltage distribution network, and micro-grid power can be used as supply-side resources or load-side resources to participate in scheduling, thus medium voltage distribution network became an active distribution network that can be a bidirectional dispatch electricity distribution network type. Thus, because the distributed generation and micro-grid are connected to the medium voltage distribution network, it constitutes an active distribution network. If power companies rationally use a bidirectional dispatch function of an active distribution network, not only the profit of the power company is increased but also the environment is protected. The major issue of this paper is how to maximize the profit of a power company by rationally scheduling the micro-grid and distributed power in an active distribution network. Research of foreign scholars has mainly focused on the purchase power plan of power companies in the electricity market. That is maximizing profit of a power company by making an optimization purchase power plan under the conditions that account for the distributed generation and load shedding. China’s power is one of public utilities. It is impossible for us to take measures of load shedding for the benefit of a power company. Moreover, the market mechanism of the power distribution side is also uncertain. However, a medium/low voltage distribution network connected by distributed power and micro-grid of different investors has become a trend under the policy support. Therefore, for the power company, it is different from other states how to schedule distributed power and micro-grid. In fact, after the micro-grid and distributed power are connected to a medium/low voltage distribution network, electricity sales of the power company are bound to decrease and operating income will accordingly be reduced, which is an avoided problem in previous research. All scheduling models proposed in the studies are based on the goal of the lowest supply power cost, and are not concerned with who gets the operating income. However, this is the most significant problem of power companies. In this paper, a novel day-ahead scheduling model of active distribution network is supposed. The goal of this model is maximizing the profit of the power company. And it is more objectively described as purchase power costs and sale power revenues of Chinese power companies with an active distribution network. Moreover, this model takes into account the various power prices of different distributed power, the incentive policies of government and government subsidies of the power company. More importantly, because the network losses of distribution lines in high voltage network before the inlet of medium voltage distribution network is correlated strongly with its electrical distance. This model accounts for the profit of the power company which network losses are reduced owing to use of distribute power and micro-grid. Using this model, the active medium voltage distribution networks are scheduled in different locations, which not only can achieve greater efficiency of power companies through accessing the distribute power and micro-grid at the location away from the main network, but also don’t face losses in close supply power with the new energy accesses through reasonable scheduling. In this paper, examples of IEEE33 node demonstrate the effectiveness and feasibility of the model. The results can provide a reference for optimization scheduling of the active distribution network.