基于伊藤过程的电制氢合成氨负荷随机最优控制  

Flexible Load Stochastic Optimal Control of Wind Power-Based Hydrogen Production and Ammonia Synthesis Systems Based on the ItôProcess

作　　者：杨国山 朱杰 宋汶秦 邱一苇 周步祥[2] YANG Guoshan;ZHU Jie;SONG Wenqin;QIU Yiwei;ZHOU Buxiang(Economic and Technological Research Institute of Gansu Electric Power Corp.,Lanzhou 730050,China;College of Electrical Engineering,Sichuan University,Chengdu 610065,China)

机构地区：[1]国网甘肃省电力公司经济技术研究院,甘肃兰州730000 [2]四川大学电气工程学院,四川成都610065

出　　处：《中国电力》2023年第7期66-77,共12页Electric Power

基　　金：国网甘肃省电力公司重点科技项目(基于氢电高效转化的氢能系统参与电网调峰调频辅助服务的研究,SGGSJY00NYJS2200024);国家自然科学基金资助项目(基于伊藤理论的高比例可再生能源电力系统分析与控制技术研究,51907099)。

摘　　要：风电制氢进而合成氨(power to ammonia,P2A)是规模化消纳可再生发电资源,实现电力与化工行业碳减排的潜在技术路线之一。利用电制氢作为媒介,P2A可作为大型工业负荷参与电网能量平衡调节。然而,P2A负荷受化学工艺及过程控制的限制,负载调控惯性较大,当风电出力偏离预测轨迹时P2A负荷难以快速响应。为此,提出计及风电出力时序不确定性的P2A负荷随机最优控制方法。首先建立P2A系统柔性调控的状态空间模型。其次,考虑合成氨工段的调节惯性与风电出力时序相关性的耦合影响,基于伊藤过程建模风电出力的不确定性,构造随机动力学约束的P2A系统优化控制模型。之后,基于动态轨迹灵敏度分解,将随机动力学优化问题变换为确定性二阶锥规划,并采用随机模型预测控制(stochastic model predictive control,SMPC)滚动求解,有效避免了传统基于随机抽样模拟的方法计算复杂度高、求解效率低的问题。算例分析表明,与确定性控制相比,所提方法能够充分发挥合成氨柔性生产的优势,提升P2A负荷消纳波动性风电的能力。Wind power-based hydrogen production and ammonia synthesis(P2A)is one of the potential technical routes for largescale renewable energy consumption and carbon emission reduction in the power and chemical industries.Taking power-to-hydrogen production as a medium,P2A can participate in grid balancing regulation as a large-scale industrial load.However,due to the limitation of the chemical process and control of the ammonia synthesis load,the inertia of load regulation is large.When the wind power output deviates from the predicted trajectory,it is difficult for the P2A load to respond quickly.Thus,a stochastic optimal control method for the P2A load considering temporally correlated uncertainty of wind power is proposed.Firstly,the flexible regulation state-space model of the P2A system is established.Second,considering the coupling of the stochastic process of wind power output and the adjustment inertia of the ammonia synthesis section,the Itôprocess is used to model the stochastic process of wind power output,and a stochastic dynamics-constrained optimal control model of the P2A system is constructed.Then,the stochastic optimization problem is transformed into a deterministic second-order cone programming by the trajectory sensitivity decomposition and solved by the stochastic model predictive control(SMPC)in a rolling-horizon manner,avoiding the disadvantages of high computational complexity and low efficiency of the traditional sampling-based stochastic control methods.The case study shows that,compared with deterministic control,the proposed method fully takes advantage of the flexible production of ammonia synthesis and greatly improves the ability of P2A Load to consume the fluctuating wind power.

关 键 词：风电制氢 合成氨 绿氨 伊藤过程 时序相关性 随机模型预测控制 

分 类 号：TQ113.2[化学工程—无机化工] TQ116.2[电气工程—电力系统及自动化] TM614