PDF(1756 KB)
PDF(1756 KB)
PDF(1756 KB)
考虑需量防守和调峰调频的园区混合储能双层迭代鲁棒规划
Two-Layer Iterative Robust Planning Framework for Hybrid Energy Storage in Industrial Parks Considering Balancing Demand Power Defense，Peak Shaving， and Frequency Regulation
【目的】工业园区配置储能不仅有助于降低电度电费和需量功率电费，还可通过参与电网调峰调频服务增加额外收益。然而，光伏发电的随机波动特性对储能系统在需量功率管理、调峰调频等多场景协同应用方面提出了严峻挑战。尽管电化学储能在平抑光伏发电随机波动方面具有毫秒级响应特性，但其高昂的投资成本与有限的循环寿命制约了其在用户侧规模化应用；相较而言，重力储能凭借其低成本与长寿命优势，在日均投资成本方面展现出显著竞争力。针对上述问题，提出了一种面向园区需量功率管理与调峰调频的混合储能系统双层迭代鲁棒规划方法。【方法】上层模型基于信息间隙决策理论（information gap decision theory，IGDT），以系统年运行成本最小化为目标，优化重力-电化学储能容量配置，并将结果反馈至下层模型；下层模型采用模型预测控制（model predictive control，MPC）方法，通过滚动优化实现混合储能对需量功率的自适应动态控制，并以需量功率防守是否成功为判据，判断是否需反馈至上层进行容量重新配置。【结果】所提方案使园区需量管控的最大购电功率降低比例提升49.4%，调峰调频综合性能指标提高42%，系统年运行成本降低21%。【结论】所提方案为工业园区储能系统规划提供了重要的理论依据。
[Objective] The deployment of energy storage in industrial parks not only reduces electricity costs and demand power charges but also generates additional revenue through participation in grid peak shaving and frequency regulation services. However， the stochastic volatility of photovoltaic （PV） generation presents significant challenges for the collaborative application of energy storage systems in demand power management and peak shaving. While electrochemical energy storage demonstrates millisecond response capabilities in mitigating the random fluctuations of PV generation， its high investment costs and limited cycle lifespan restrict its large-scale application on the demand side. In contrast， gravitational energy storage is highly competitive in terms of daily investment costs due to its low cost and long operational lifespan. To address these issues， this study proposes an innovative two-layer iterative robust planning method for a hybrid energy storage system aimed at managing demand power charges and facilitating peak shaving and frequency regulation in industrial parks. [Methods] The upper-layer model， based on information gap decision theory （IGDT）， seeks to minimize the annual operating costs of the system by optimizing the capacity allocation of gravitational and electrochemical energy storage， with the results transmitted to the lower-layer model. The lower-layer model employs model predictive control to achieve adaptive dynamic control of demand charges through rolling optimization. The success of demand power defense serves as the criterion for determining whether feedback is needed for capacity reallocation in the upper layer. [Results] Simulation results indicate that the proposed approach enhances the accuracy of demand management by 49.4%， improves the comprehensive performance index for peak shaving and frequency regulation by 42%， and reduces the annual operating costs of the system by 21%. [Conclusions] The proposed method provides significant theoretical support for the planning of energy storage systems in industrial parks.
需量防守 / 重力-电化学混合储能 / 调峰调频 / 信息间隙决策理论（IGDT） / 模型预测控制（MPC）
demand defense / gravity-electrochemical hybrid energy storage / peak shaving and frequency regulation / information gap decision theory（IGDT） / model predictive control（MPC）
| [1] |
吴限, 李卫东, 李正文, 等. 大扰动下考虑电化学储能的主动频率响应优化控制策略[J]. 电力系统自动化, 2023, 47(17): 118-127.
|
| [2] |
李军徽, 张靖祥, 穆钢, 等. 辅助服务市场下独立储能调峰调频协同优化调度[J]. 中国电机工程学报, 2025, 45(2): 650-664.
|
| [3] |
戴汉扬, 肖雄, 宋新立, 等. 适用于大电网动态仿真的电化学储能电站多场景仿真建模及应用[J]. 电网技术, 2023, 47(5): 1828-1836.
|
| [4] |
|
| [5] |
|
| [6] |
李嘉媚, 艾芊, 殷爽睿. 虚拟电厂参与调峰调频服务的市场机制与国外经验借鉴[J]. 中国电机工程学报, 2022, 42(1): 37-55.
|
| [7] |
洪烽, 梁璐, 逄亚蕾, 等. 基于机组实时出力增量预测的火电-飞轮储能系统协同调频控制研究[J]. 中国电机工程学报, 2023, 43(21): 8366-8377.
|
| [8] |
胡梦凯, 张高航, 李凤婷, 等. 考虑频率安全及调峰需求的风电高渗透系统两阶段优化调度方法[J]. 电网技术, 2025, 49(10): 4365-4375.
|
| [9] |
李相俊, 刘晓宇, 韩雪冰, 等. 电化学储能电站数字化智能化技术及其应用展望[J]. 供用电, 2023, 40(8): 3-12.
|
| [10] |
徐赏, 刘继春. 基于中长期电量滚动均衡化的售电公司全周期多时间尺度购售电策略[J]. 现代电力, 2023, 40(4): 505-513.
|
| [11] |
郑晓东, 陈皓勇, 段声志, 等. 基于场景概率驱动的输电网和储能分布鲁棒规划[J]. 电力自动化设备, 2022, 42(6): 170-178.
|
| [12] |
钱敏慧, 陈宁, 牟佳男, 等. 基于分频及模型预测控制的高比例风电电力系统风储协同频率控制策略[J/OL]. 电网技术. (2026-02-13)[2026-03-02]. https://doi.org/10.13335/j.1000-3673.pst.2025.0045.
|
| [13] |
吴卓睿, 张萌, 管晓宏. 面向电力系统快速频率响应的数据与模型驱动预测控制[J]. 自动化学报, 2025, 51(10): 2337-2346.
|
| [14] |
文军, 徐若晨, 裴杰, 等. 一种重力压缩空气储能系统及其工作方法: CN202010038208.8[P]. 2020-06-05.
|
| [15] |
王团结, 何信林, 牛利涛, 等. 一种基于重力储能技术的海上风力发电系统: CN111486057A[P]. 2020-08-04.
|
| [16] |
杨天鑫, 黄云辉, 何珍玉, 等. 基于多时间尺度调节的构网型储能电站定容选址优化配置[J]. 电力系统自动化, 2024, 48(23): 54-64.
|
| [17] |
赵浩辰. 电力市场环境下电化学储能商业模式研究[D]. 北京: 华北电力大学, 2024.
|
| [18] |
廖东颖, 朱丹蕾, 谢保胜, 等. 参与辅助服务的电化学储能系统调峰调频技术及机制探究[J]. 电工技术, 2022(6): 127-131.
|
| [19] |
袁铁江, 郭建华, 杨紫娟, 等. 平抑风电波动的电-氢混合储能容量优化配置[J]. 中国电机工程学报, 2024, 44(4): 1397-1405.
|
| [20] |
梅书凡, 檀勤良, 代美. 考虑风光出力季节性波动的储能容量配置[J]. 电力工程技术, 2022, 41(4): 51-57.
|
| [21] |
黄际元, 李欣然, 常敏, 等. 考虑储能电池参与一次调频技术经济模型的容量配置方法[J]. 电工技术学报, 2017, 32(21): 112-121.
|
| [22] |
谢德清, 徐雨红, 李海强, 等. 辅助火电机组AGC调频的混合储能容量配置优化[J]. 电工技术, 2023(19): 15-18.
|
| [23] |
丁涛, 谢刘双飞, 黄雨涵, 等. 西北地区新型储能发展面临的关键问题及建议[J]. 电力系统自动化, 2026, 50(5): 16-32.
|
| [24] |
|
| [25] |
|
| [26] |
刘峰伟, 陈佳佳, 魏延康, 等. 基于连续时间尺度的储能运行提升光储型园区需量防守能力[J]. 高电压技术, 2024, 50(12): 5621-5629.
|
| [27] |
梁宸华. 混合储能参与调峰和频率响应的电力系统优化调度方法研究[D]. 南宁: 广西民族大学, 2024.
|
| [28] |
王辉, 董宇成, 夏玉琦, 等. 考虑阶梯碳-绿证互认与重力储能的矿区综合能源系统优化调度[J]. 中国电力, 2025, 58(7): 54-67.
|
| [29] |
靳雯皓, 刘继春. 平滑风电功率波动的混合储能系统容量优化配置[J]. 分布式能源, 2017, 2(2): 32-38.
风电输出功率具有波动性，为减小其对电力系统运行的影响，提高电力系统的电能质量，提出了平抑功率波动的混合储能系统容量优化配置方法。采用滑动平均法滤除波动功率，平滑风电输出功率，滑动平均法中窗口长度的选择将影响功率平滑效果及混合储能系统容量配置，依据并网功率波动限制为约束确定窗口长度值，实现其最优选择。采用频谱分析的方法分解波动功率，以年均综合成本最小为目标函数确定超级电容器和蓄电池各自补偿频段，进而确定其实时充放电功率。最后，建立了考虑蓄电池寿命损耗的混合储能系统成本模型，通过案例分析验证本方案的可行性以及经济优越性。
Wind power have the characteristic of generation output variability. To reduce the impact of power fluctuations on grids and improve the power quality of power system, this paper puts forward the capacity optimization configuration of hybrid energy storage system for smoothing wind power fluctuation. The paper adopts moving-average method to filter the fluctuation of power and smooth wind power. The window length of moving-average method will affect the smoothing effect of power and the capacity configuration of hybrid energy storage system, so we consider the fluctuation limit of grid power as constraints to determine the window length, then achieve its optimal choice. We use the method of spectrum analysis to decompose the fluctuation of wind power, and take the minimum annual comprehensive cost as the objective function to determine the compensation frequency bands of the super capacitor and the battery respectively, and then determine the charge and discharge power. Finally, we establish a cost model of hybrid energy storage system considering battery life loss, and verify the feasibility and economic superiority of the program through case analysis. |
| [30] |
史沛然, 李彦宾, 江长明, 等. 第三方独立主体参与华北电力调峰市场规则设计与实践[J]. 电力系统自动化, 2021, 45(5): 168-174.
|
| [31] |
李军徽, 侯涛, 严干贵, 等. 计及调频成本和荷电状态恢复的多储能系统调频功率双层优化[J]. 中国电机工程学报, 2021, 41(23): 8020-8032.
|
| [32] |
谢小荣, 马宁嘉, 刘威, 等. 新型电力系统中储能应用功能的综述与展望[J]. 中国电机工程学报, 2023, 43(1): 158-168.
|
| [33] |
付勇. 含风储的外送型电网一次调频控制策略研究[D]. 沈阳: 沈阳工业大学, 2024.
|
| [34] |
桑丙玉, 杨波, 李克成, 等. 模型差异性混合储能系统建模及功率分配[J]. 太阳能学报, 2024, 45(1): 433-441.
该文从锂电池和燃料电池的混合储能特性出发, 针对已有文献的拓扑结构, 对比选取适用于功率分配的混合储能拓扑模型和建模方法, 并在组合过程中通过分析选取最合适的变换方式组成混合储能模型, 最后通过仿真验证所提模型可在进行功率分配的前提下提高储能系统荷电状态偏移, 延长储能系统寿命。
Starting from the hybrid energy storage characteristics of lithium batteries and fuel cells, this paper compares and selects hybrid energy storage topology models and modeling methods suitable for power distribution according to the topology structures of existing literature, and selects the most suitable hybrid energy storage topology model through analysis and selection during the combination process. A hybrid energy storage model is formed by appropriate transformation methods. Finally, it is verified by simulation that the proposed model can improve the state of charge offset of the energy storage system and prolong the life of the energy storage system under the premise of power distribution.
|
| [35] |
刘又榕, 林顺富, 沈运帷, 等. 计及电动汽车参与多元需求响应的微电网多时间尺度优化调度模型[J]. 电力建设, 2023, 44(10): 51-62.
为深入挖掘电动汽车的可调度潜力,缓解含高比例新能源微电网的供能压力,结合多元需求响应技术,提出一种考虑电动汽车资源参与的微电网多时间尺度优化调度模型。在日前调度阶段,一部分电动汽车资源与价格型需求响应技术相结合,以用户的综合满意度为目标进行优化。微电网基于价格型电动汽车资源的调度计划,以经济低碳成本最低与灵活性满足度最大为目标进行优化,确定各侧可调资源的调度安排。在日内调度阶段,另一部分电动汽车资源与激励型需求响应技术相结合。微电网能量管理中心作为领导者,以运营成本最小为目标,激励型电动汽车群作为跟随者,以用电成本最小为目标,构建微电网日内主从博弈模型进行滚动优化,双方基于补贴价格与用能策略进行博弈。最后,基于某微电网场景进行仿真验证,结果表明所提模型能够降低用户用电成本,减少负荷曲线的峰谷差,实现新能源的全消纳。
To exploit the schedulable potential of electric vehicles (EVs) efficiently and relieve the energy supply pressure of microgrids with a high proportion of new energy sources, a multi-time-scale optimization scheduling model of the microgrid is proposed, considering the participation of EV resources combined with the multi-demand response technology. In the day-ahead scheduling stage, some EV resources are combined with the price-based demand response technology to optimize comprehensive user satisfaction. Based on the scheduling plan of the EV resources based on price, the microgrid is optimized focusing on minimizing economic cost, ensuring low-carbon expenditure, and maximizing flexibility satisfaction, and the scheduling arrangement of the adjustable resources on each side is determined. In the intra-day scheduling phase, another portion of the EV resources is combined with the incentive demand response technology. The microgrid energy management center, as the leader, aims to minimize the operating costs, and the incentive EV group, as the follower, aims to minimize the electricity costs. The intra-day master-slave game model of the microgrid was constructed for rolling optimization, and the two sides played the game based on a subsidized price and energy use strategy. Finally, a simulation verification was conducted based on a microgrid scenario, and the results show that the proposed model can reduce the electricity cost of users, reduce the peak-valley difference of the load curve, and realize the full absorption of new energy. |
| [36] |
高天, 王祖凡, 方舒扬, 等. 含齿轮变速与链式传动机构的斜坡重力储能系统能效分析模型与实验验证[J]. 储能科学与技术, 2025, 14(2): 688-698.
为了研究传送链式斜坡重力储能系统(transmission chain slope gravity energy storage system，TCS-GESS)充放电过程的能量转换效率及各环节损耗占比，针对系统质量块移动、机械传动以及电气驱动环节，推导了各环节损耗数学表达式及相应的能效计算方法，在此基础上，建立了基于MATLAB/Simulink的TCS-GESS能效分析模型。以一套2.2 kW重力储能样机为例，设计了充放电工况下的实验方案并实测了不同负载条件下系统各环节能效，从速度、机械功率、充放电功率、传动损耗以及电机损耗五个维度与能效计算结果进行对比，验证了能效分析模型的准确性与实用性。结果显示，随负载增加系统效率逐渐提高，其中链条损耗占比较大，齿轮箱和齿轮盘损耗变化不大且占比较小，电机损耗占比中等且充/放电工况下随着加载均有所增加。额定负载工况下充放电效率分别为59.5%和37.4%，系统效率为23.2%；进一步对具有相同传动机构、不同功率等级下重力储能系统充放电效率进行预测，结果表明系统容量低于1 MW时充放电效率将低于68%，容量高于10 MW时系统能效提升潜力有限，即采用文中传动机构的重力储能单机系统最佳功率范围宜选取在1～10 MW。
To study the energy efficiency and the loss proportion of each link in the charging and discharging process of the transmission chain slope gravity energy storage system (TCS-GESS), the mathematical expression of each link loss and the corresponding energy efficiency calculation method were derived for the mass block movement, mechanical transmission, and electrical driving links of the system. Furthermore, an energy efficiency analysis model for TCS-GESS using MATLAB/Simulink was established. An experimental scheme was designed under charging and discharging conditions and measured the energy efficiency of each link of the system under different load conditions using a 2.2 kW gravity energy storage prototype. The accuracy and practicability of the energy efficiency analysis model were verified by comparing the results of the energy efficiency calculated with the five dimensions of speed, mechanical power, charging and discharging power, transmission loss, and motor loss. The results show that the system efficiency gradually increases with an increase in load. Among them, the chain loss accounts for a large proportion of each loss link. The loss of gearbox and gear plate was not significantly altered and accounts for a small proportion. The motor loss accounts for a medium proportion and increases as the load increases under charging and discharging conditions. Under rated load conditions, the charge and discharge efficiencies were 59.5% and 37.4%, respectively, and the system efficiency was 23.2%. Furthermore, the charging and discharging efficiency of the GESS with the same transmission mechanism and different rated powers was predicted. The results indicate that the charging and discharging efficiency was <68% when the system capacity was <1 MW, and the energy efficiency improvement potential of the system was limited when the capacity was >10 MW. Based on the findings from this study, this paper recommends selecting the optimal power range for a GESS utilizing the same transmission mechanism within the range of 1—10 MW. |
| [37] |
王海伟, 陈晨, 陈朔, 等. 基于多站融合的“光储充换”协同优化目标研究[J]. 电力系统装备, 2021(17): 166-167.
|
| [38] |
徐永乐, 许仪勋, 朱明光. 考虑绿证-碳交易的含氢综合能源系统分布鲁棒优化[J]. 电力系统及其自动化学报, 2026, 38(2): 1-11.
|
| [39] |
薛雨润, 吴军, 吴明阳, 等. 计及多形式储能接入的氢能园区综合能源系统优化规划方法研究[J]. 电网与清洁能源, 2025, 41(12): 127-138.
|
| [40] |
邓程程, 沈晓东, 唐西胜. 考虑多种场景下的移动-固定式混合储能优化配置方法[J]. 电网与清洁能源, 2025, 41(10): 56-65.
|
| [41] |
刘正茂, 朱司丞, 黄忠梁, 等. 城市高压配电网分区储能规划方法[J]. 浙江电力, 2025, 44(4): 122-132.
|
| [42] |
于琳琳, 陈姝彧, 张凯南, 等. 考虑最低惯量需求的受端电网储能规划研究[J]. 电网与清洁能源, 2024, 40(10): 149-158.
|
| [43] |
董福贵, 孟子航, 郗来昊, 等. 面向薄弱配电网的农村多能互补系统储能协同优化配置[J]. 电力工程技术, 2024, 43(5): 13-26.
|
| [44] |
冯弋舟, 吴志, 李新煜, 等. 光-蓄-储混合储能系统灵活性提升与容量规划[J]. 电力工程技术, 2024, 43(5): 27-36.
|
| [45] |
卞海红, 李灿, 童宇轩. 共享储能模式下电动汽车充电站双层优化运行策略[J]. 电力工程技术, 2024, 43(5): 170-180.
|
| [46] |
郑圣, 谭书平, 张清周, 等. 基于负荷场景多层聚类的储能精细化规划研究[J]. 浙江电力, 2024, 43(2): 79-87.
|
| [47] |
崔文倩, 魏军强, 赵云灏, 等. “双碳”目标下含重力储能的配电网多目标运行优化[J]. 电力建设, 2023, 44(4): 45-53.
在“双碳”背景下,新型电力系统承载电力系统低碳转型的重要任务。为了更好地促进新能源的消纳,同时有效促进需求响应参与电力系统运行,针对主动配电网的运行优化,提出了基于重力储能和需求响应的主动配电网多目标优化模型。目标函数一考虑主动配电网运行和用户用能的总成本,目标函数二选取主动配电网的等效碳排放成本。由于风光出力的不确定性,利用鲁棒优化理论将不确定优化模型转换为确定性优化模型,同时构建了基于NSGA-Ⅱ算法的运行优化求解方法。最后,在三种场景下对模型进行求解,结果表明系统配置重力储能,同时开展需求响应除了能减少碳排放量之外,还能达到削峰填谷的目的,系统有较好的运行经济性。
Under the background of “double carbon target ”, the new power system carries the important task of low-carbon transformation of power system. In order to better promote the absorption of new energy and effectively promote the demand response to participate in the operation of power system, this paper proposes a multi-objective optimization model of active distribution network considering gravity energy storage and demand response. Objective function 1 considers the total cost of active distribution network operation and user energy consumption, and objective function 2 is selected as the equivalent carbon-emission cost of active distribution network. Due to the uncertainty of wind power output, the uncertain optimization model is converted to the deterministic optimization model by using the robust optimization theory, and the operation optimization solution method based on NSGA-Ⅱ algorithm is constructed. Finally, the model is solved in three scenarios. The results show that the system equipped with gravity energy storage and the demand response at the same time can not only reduce carbon emissions, but also achieve the purpose of peak-cutting and valley-filling, and the system has better operation economy. |
| [48] |
程鑫, 许亮, 周姝灿, 等. 基于新能源出力保证率轨迹灵敏度分析的储能配置方法[J]. 电力系统自动化, 2020, 44(13): 25-31.
|
| [49] |
李刚, 赵治钧, 戴浩男, 等. 考虑天气不一致性与日内出力特性变化的光伏集群功率短期预测方法[J/OL]. 高电压技术. (2026-02-02)[2026-03-02]. https://doi.org/10.13336/j.1003-6520.hve.20251419.
|
| [50] |
王莲, 简子淋, 张晓斌, 等. 基于CNN-BiLSTM和多目标函数优化的省级光伏日内保障出力预测[J]. 高电压技术, 2025, 51(9): 4846-4855.
|
利益冲突声明（Conflict of Interests）： 所有作者声明不存在利益冲突。
AI小编
/
| 〈 |
|
〉 |
AI 导读
