计及精细化氢能利用的综合能源系统多时间尺度鲁棒优化策略

Multi-Time-Scale Robust Optimization Strategy for Integrated Energy System Considering the Refinement of Hydrogen Energy Use

能源结构改革背景下,构建以氢能驱动、电-热为主体的综合能源系统(HEH-IES)具有重要意义。为提高含电氢耦合单元的IES运行灵活性,同时减小功率波动对系统的影响,提出含多元储能与综合需求响应的氢能精细化利用两阶段调度方法。首先,分析电制氢(P2H)两阶段运行过程与电氢耦合单元运行特性,对氢能的用能过程与设备进行精细化建模;其次,计及多元储能与综合需求响应提出了日前-日内两阶段多时间尺度优化策略,日前阶段充分考虑能源系统的不确定性,构建数据驱动的分布鲁棒优化(DRO)模型,日内阶段遵从日前计划,考虑多能流在灵活性调节时间尺度上的差异,通过多时间尺度的滚动优化降低功率波动的影响;最后,仿真算例证明了所提模型与策略对提升IES的运行灵活性具有积极作用。

The local flexibility of the high percentage of new energy power system is seriously insufficient,and it is urgent to establish a new energy structure system that is clean,efficient and flexible.Hydrogen energy as a secondary energy source with diverse and efficient conversion forms. Refined modeling of hydrogen energy utilization is a key issue to study the flexibility value of electric-hydrogen coupled units. At the same time, the energy system has uncertainty and fluctuating variability on multiple time scales, and the existing studies are too adventurous or conservative in considering the uncertainty at the day-ahead stage. In response to the above problems, a robust optimization strategy is proposed for the multi-timescale distribution of the integrated energy system taking into account the refined hydrogen energy utilization. Firstly, the two-stage operation process of P2H is considered, and the refinement modeling of hydrogen energy use process and equipment is carried out by taking into account the operating characteristics of electrolyzer, hydrogen fuel cell and other equipment. The operating states of PEM are divided into shutdown, cold standby, overload, variable load and low load states, and a mixed integer linear mathematical model of the electrolyzer with variable load start-stop characteristics is established, taking into account the loss of hydrogen output during the cold start of PEM. In order to improve the operational flexibility of the cogeneration unit, the adjustable thermoelectric ratio of CHP and HFC is considered to decouple the thermoelectric linkage, and an adjustable thermoelectric ratio heat model of the cogeneration unit is established. Secondly, to reduce the power fluctuation caused by the deviation of wind power and multi-energy load forecasts in the day-ahead and intra-day, a two-stage optimization model of day-ahead scheduling and intra-day rolling is established. In the day-ahead stage, a data-driven distribution robust optimization model is established, and the probability distribution is constrained by the composite norm to adjust the conservativeness of the model;in the intra-day stage, the differences in the time scales of flexibility regulation of multi-energy flows are considered, and the impact of power fluctuations is reduced by rolling optimization on multiple time scales. In the case simulation, five scenarios are set up for comparative analysis in the day-ahead phase, and the proposed multi-timescale model is compared with day-ahead programming (DA-P) in the intra-day phase, and the conservativeness of the data-driven DRO model is investigated, leading to the following conclusions: (1) The intermediate energy ladder losses are avoided after refining the two-stage operation process of P2H. The efficiency and flexibility of hydrogen energy utilization are fully exploited, and the comprehensive energy utilization of IES is significantly improved. (2) The proposed PEM mixed integer linear model and adjustable cogeneration model can adjust the equipment output in real time according to the load, which promotes the wind power consumption and improves the operating economy. (3) The data-driven DRO model proposed in the day-ahead stage fully takes into account the uncertainty of the energy system based on historical data samples, and its conservativeness is influenced by the number of reduction scenarios and sample size. It has a better ability to resist the fluctuation of uncertainty forecast error in the intra-day correction phase. (4) The intra-day phase takes into account the differences in the forecast characteristics of different energy sources, and smoothes out power fluctuations by regulating different energy coupling devices on a sub-time scale, effectively reducing wind power volatility and operating costs.