考虑出力区间预测和储能优化配置的风电场调频容量估计两阶段模型

Two-Stage Model for Estimation of Wind Farm Frequency Modulation Capacity Considering Power Probabilistic Prediction and Optimal Configuration of Energy Storage

摘要

风电场配置储能参与调频辅助服务具有显著意义,既能大幅增加风电场收益,又能有效提升电力系统的灵活性与稳定性．然而,当前风电场在实际应用中面临风电场出力预测精准度欠佳、储能容量配置缺乏合理性、参与调频容量估计难度较大等诸多问题．为此,针对上述问题,本文提出一种考虑出力区间预测和储能优化配置的风电场调频容量估计两阶段模型．阶段1模型主要针对风电场出力预测精度不高的难题,采用时间卷积网络(TCN)-长短期记忆(LSTM)网络-Transformer(TCN-LSTM-Transformer)分位数回归模型进行风电功率区间预测,其中,TCN能够高效提取出丰富的时序特征,LSTM进行时序模型建模,Transformer则有效捕捉数据的长时间依赖性,三者协同,精准获知风电场功率的波动区间；进一步从风电场的历史弃风数据出发,利用改进的蝴蝶算法实现风电场储能容量优化配置,确保在风电功率出现波动时,有足够的储能容量满足电力系统的调频需求；在阶段1基础上,阶段2模型基于风阻限值估算风电场在不同置信概率下的最佳调频容量,进一步优化风电场的经济效益和系统稳定性．最后,选取我国南方某地区风电场的实际数据进行仿真验证,结果表明,与点预测方法相比,所提出的两阶段模型估计误差下降55.6%,经济效益提升2%．该案例充分证明所提方法可为风电场接入下电力系统的灵活性改造提供有效技术支撑．

Abstract

The configuration of energy storage in wind farms to participate in frequency modulation(FM) auxiliary services is of great significance because this can considerably increase the income of wind farms and effectively improve the resulting power systems’ flexibility and stability. However, in their actual application in wind farms, they currently face several challenges, including poor accuracy of wind farm output prediction, lack of rationality in energy storage capacity allocation, and greater difficulty in FM capacity estimation. To address this issue, a two-stage model of FM capacity estimation for wind farms is proposed, considering output interval prediction and the optimized configuration of energy storage. The Stage 1 model addresses the problem of the low output prediction accuracy of wind farms, and the TCN-LSTM-Transformer quantile regression model is used to predict the wind power interval. Specifically, TCN can efficiently extract rich time-series characteristics, LSTM performs time-series model modeling, and Transformer effectively captures the long-term dependence of data. These three components work together to accurately determine the fluctuation range of the power of the wind farm. Furthermore, starting from the wind farm’s historical wind abandonment data, the improved butterfly algorithm is used to optimize the allocation of the wind farm’s energy storage capacity. This ensures sufficient energy storage capacity to meet the power system’s frequency modulation needs when the power of the wind power fluctuates. Then, building upon Stage 1, the Stage 2 model estimates the wind farm’s optimal FM capacity under different confidence probabilities based on the wind resistance limit, further optimizing the wind farm’s economic benefits and system stability. Finally, the actual data of a wind farm located in southern China are selected for simulation verification. The results indicate that the estimation error of the proposed two-stage model is reduced by 55.6%, and the economic benefits are increased by 2% compared with the point prediction method. This case fully proves that the proposed method can provide effective technical support for the flexible transformation of power systems driven by wind farms.

关键词

风电场调频容量估计 / 改进蝴蝶算法 / 储能优化配置 / TCN-LSTM-Transformer分位数回归 / 风电功率预测

Key words

wind farm frequency modulation capacity estimation / improved butterfly algorithm / optimal configuration of energy storage / TCN-LSTM-Transformer quantile regression / wind power prediction

参考文献

[1] 娄素华, 杨天蒙, 吴耀武, 等. 含高渗透率风电的电力系统复合储能协调优化运行[J]. 电力系统自动化, 2016, 40(7):30-35.

[2] Lou Suhua, Yang Tianmeng, Wu Yaowu, et al. Coordinated optimal operation of hybrid energy storage in power system accommodated high penetration of wind power[J]. Automation of Electric Power Systems, 2016, 40(7):30-35(in Chinese).

[3] He Y Y, Liu R, Li H Y, et al. Short—term power load probability density forecasting method using kernel—based support vector quantile regression and Copula theory[J]. Applied Energy, 2017, 185:254-266.

[4] 杨楠, 周峥, 陈道君, 等. 基于非参数核密度估计的风功率波动性概率密度建模方法[J]. 太阳能学报, 2019, 40(7):2028-2035.

[5] Yang Nan, Zhou Zheng, Chen Daojun, et al. Research of modeling method based on non—parametric kernel density estimation of probability of wind power fluctuations[J]. Acta Energiae Solaris Sinica, 2019, 40(7):2028-2035(in Chinese).

[6] Jeon J, Taylor J W. Using conditional kernel density estimation for wind power density forecasting[J]. Journal of the American Statistical Association, 2012, 107(497):66-79.

[7] 王森, 孙永辉, 周衍, 等. 计及误差时间相依性的风电功率超短期条件概率预测[J]. 电力自动化设备, 2022, 42(11):40-46.

[8] Wang Sen, Sun Yonghui, Zhou Yan, et al. Ultra—short term conditional probability prediction of wind power considering error time dependence[J]. Electric Power Automation Equipment, 2022, 42(11):40-46(in Chinese).

[9] Zhou B W, Ma X G, Luo Y H, et al. Wind power prediction based on LSTM networks and nonparametric kernel density estimation[J]. IEEE Access, 2019, 7:165279-165292.

[10] 杨锡运, 张艳峰, 叶天泽, 等. 基于朴素贝叶斯的风电功率组合概率区间预测[J]. 高电压技术, 2020, 46(3):1099-1108.

[11] Yang Xiyun, Zhang Yanfeng, Ye Tianze, et al. Prediction of combination probability interval of wind power based on naive Bayes[J]. High Voltage Engineering, 2020, 46(3):1099-1108(in Chinese).

[12] 庞传军, 尚学伟, 张波, 等. 基于改进梯度提升算法的短期风电功率概率预测[J]. 电力系统自动化, 2022, 46(16):198-206.

[13] Pang Chuanjun, Shang Xuewei, Zhang Bo, et al. Short—term wind power probability prediction based on improved gradient boosting machine algorithm[J]. Automation of Electric Power Systems, 2022, 46(16):198-206(in Chinese).

[14] 殷豪, 黄圣权, 孟安波, 等. 基于长短期记忆网络分位数回归的短期风电功率概率密度预测[J]. 太阳能学报, 2021, 42(2):150-156.

[15] Yin Hao, Huang Shengquan, Meng Anbo, et al. Short—term wind power probability density prediction based on long short term memory network quantile regression[J]. Acta Energiae Solaris Sinica, 2021, 42(2):150-156(in Chinese).

[16] 庞昊, 高金峰, 杜耀恒. 基于时间卷积网络分位数回归的短期负荷概率密度预测方法[J]. 电网技术, 2020, 44(4):1343-1350.

[17] Pang Hao, Gao Jinfeng, Du Yaoheng. A short—term load probability density prediction based on quantile regression of time convolution network[J]. Power System Technology, 2020, 44(4):1343-1350(in Chinese).

[18] Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[C]// Proceedings of the 31st International Conference on Neural Information Processing Systems. Red Hook, USA, 2017:6000-6010.

[19] 林铮, 刘可真, 沈赋, 等. 考虑海上风电多机组时空特性的超短期功率预测模型[J]. 电力系统自动化, 2022, 46(23):59-66.

[20] Lin Zheng, Liu Kezhen, Shen Fu, et al. Ultra—short—term power prediction model considering spatial—temporal characteristics of offshore wind turbines[J]. Automation of Electric Power Systems, 2022, 46(23):59-66(in Chinese).

[21] 张越, 臧海祥, 程礼临, 等. 基于自适应时序表征和多级注意力的超短期风电功率预测[J]. 电力自动化设备, 2024, 44(2):117-125.

[22] Zhang Yue, Zang Haixiang, Cheng Lilin, et al. Ultra—short—term wind power forecasting based on adaptive time series representation and multi—level attention[J]. Electric Power Automation Equipment, 2024, 44(2):117-125(in Chinese).

[23] 盛四清, 金航, 刘长荣. 基于VMD—WSGRU的风电场发电功率中短期及短期预测[J]. 电网技术, 2022, 46(3):897-904.

[24] Sheng Siqing, Jin Hang, Liu Changrong. Short—term and mid—short—term wind power forecasting based on VMD—WSGRU[J]. Power System Technology, 2022, 46(3):897-904(in Chinese).

[25] 刘军, 甘乾煜, 张泽秋, 等. 考虑储能电池运行寿命的风电功率波动平抑方法研究[J]. 电网技术, 2023, 47(5):2098-2107.

[26] Liu Jun, Gan Qianyu, Zhang Zeqiu, et al. Research on wind power fluctuation suppression method considering the operating life of energy storage battery[J]. Power System Technology, 2023, 47(5):2098-2107(in Chinese).

[27] 胡泽春, 夏睿, 吴林林, 等. 考虑储能参与调频的风储联合运行优化策略[J]. 电网技术, 2016, 40(8):2251-2257.

[28] Hu Zechun, Xia Rui, Wu Linlin, et al. Joint operation optimization of wind—storage union with energy storage participating frequency regulation[J]. Power System Technology, 2016, 40(8):2251-2257(in Chinese).

[29] 赵爱云, 陈宽. 储能电池参与风电调频控制策略[J]. 通信电源技术, 2018, 35(7):29-35, 214.

[30] Zhao Aiyun, Chen Kuan. Control strategy of energy storage battery participating in wind power frequency modulation[J]. Telecom Power Technology, 2018, 35(7):29-35, 214(in Chinese).

[31] Savkin A V, Khalid M, Agelidis V G. A constrained monotonic charging/discharging strategy for optimal capacity of battery energy storage supporting wind farms[J]. IEEE Transactions on Sustainable Energy, 2016, 7(3):1224-1231.

[32] 齐晓光, 姚福星, 朱天曈, 等. 考虑大规模风电接入的电力系统混合储能容量优化配置[J]. 电力自动化设备, 2021, 41(10):11-19.

[33] Qi Xiaoguang, Yao Fuxing, Zhu Tiantong, et al. Capacity optimization configuration of hybrid energy storage in power system considering large—scale wind power integration[J]. Electric Power Automation Equipment, 2021, 41(10):11-19(in Chinese).

[34] 杨立滨, 曹阳, 魏韡, 等. 计及风电不确定性和弃风率约束的风电场储能容量配置方法[J]. 电力系统自动化, 2020, 44(16):45-52.

[35] Yang Libin, Cao Yang, Wei Wei, et al. Configuration method of energy storage for wind farms considering wind power uncertainty and wind curtailment constraint[J]. Automation of Electric Power Systems, 2020, 44(16):45-52(in Chinese).

[36] 王欣, 谭永怡, 秦斌. 改进 MOGOA 及其在风储容量优化配置中的应用[J]. 电力科学与技术学报, 2024, 39(2):159-169.

[37] Wang Xin, Tan Yongyi, Qin Bin. Improved multi—objective grasshopper algorithm applied in optimal capacity allocation of energy storage system in wind farms[J]. Journal of Electric Power Science and Technology, 2024, 39(2):159-169(in Chinese).

[38] 刘颖明, 王瑛玮, 王晓东, 等. 基于蚁狮算法的风电集群储能容量配置优化方法[J]. 太阳能学报, 2021, 42(1):431-437.

[39] Liu Yingming, Wang Yingwei, Wang Xiaodong, et al. Optimization of storage capacity allocation in wind farm cluster based on ant lion optimization algorithm[J]. Acta Energiae Solaris Sinica, 2021, 42(1):431-437(in Chinese).

[40] 李滨, 邓有雄, 陈碧云. 含超短期风功率预测增强处理的风储系统超前滚动优化控制策略[J]. 电网技术, 2021, 45(6):2280-2287.

[41] Li Bin, Deng Youxiong, Chen Biyun. Advanced rolling optimization control strategy for wind storage system with enhanced ultra—short—term wind power prediction[J]. Power System Technology, 2021, 45(6):2280-2287(in Chinese).

[42] Kou P, Liang D L, Gao L, et al. Probabilistic electricity price forecasting with variational heteroscedastic Gaussian process and active learning[J]. Energy Conversion and Management, 2015, 89:298-308.

[43] 杨文强, 常彬. 计及多影响因素的发电侧混合储能系统容量配置方法及配置工具[J]. 储能科学与技术, 2022, 11(10):3246-3256.

[44] Yang Wenqiang, Chang Bin. Research on the configuration method & tool for the hybrid energy storage system on the power generation side[J]. Energy Storage Science and Technology, 2022, 11(10):3246-3256(in Chinese).

[45] Ghorbani N, Babaei E. Exchange market algorithm[J]. Applied Soft Computing, 2014, 19:177-187.

[46] 武佳卉, 邵振国, 杨少华, 等. 数据清洗在新能源功率预测中的研究综述和展望[J]. 电气技术, 2020, 21(11):1-6.

[47] Wu Jiahui, Shao Zhenguo, Yang Shaohua, et al. Review and prospect of data cleaning in renewable energy power prediction[J]. Electrical Engineering, 2020, 21(11):1-6(in Chinese).

[48] Taylor J W. A quantile regression neural network approach to estimating the conditional density of multiperiod returns[J]. Journal of Forecasting, 2000, 19(4):299-311.

[49] 梅简, 张杰, 刘双宇, 等. 电池储能技术发展现状[J]. 浙江电力, 2020, 39(3):75-81.

[50] Mei Jian, Zhang Jie, Liu Shuangyu, et al. Development status of battery energy storage technology[J]. Zhejiang Electric Power, 2020, 39(3):75-81(in Chinese).

[51] 杨锡运, 刘雅欣, 邢国通, 等. 基于风电概率预测的风电场调频容量估计方法[J]. 太阳能学报, 2022, 43(7):409-416.

[52] Yang Xiyun, Liu Yaxin, Xing Guotong, et al. Method of estimation frequency regulation capacity of wind farm based on wind power probability prediction[J]. Acta Energiae Solaris Sinica, 2022, 43(7):409-416(in Chinese).

基金资助

国家电网有限公司科技资助项目(5100-202323430A-3-2ZN)