基于Markov链的园区随机功率多场景预测模型

doi:10.3969/j.issn.2097-0706.2023.01.002

摘要/Abstract

摘要：

为了促进园区分布式能源的消纳，针对园区随机功率的预测问题提出了一种基于马尔科夫（Markov）链的随机功率多场景预测模型。首先，针对园区新能源及负荷的随机特征，分别采用差分自回归移动平均（ARIMA）模型及Markov链对其建模；其次，针对园区负荷随生产、季节等因素周期性波动的特点，采用后验信息自适应调整Markov概率矩阵以提高其预测精度；然后，为了提高预测时域内多步预测的精度，考虑多步预测场景及其概率提出了一种基于场景树的多场景预测模型，以便更有效地利用Markov概率矩阵；最后，由园区历史功率数据进行了算例分析。结果表明，相比未调整的Markov模型，当自适应调整时间为7 d时，负荷功率的预测误差最低，为0.034 5（标幺值）。相比于常用的极大似然估计法，所提多场景预测模型误差的加权平均值更低。

关键词: 新能源消纳, Markov链, 预测模型, 自适应调整, 分布式能源

Abstract:

To promote the consumption of distributed energy in industrial parks， a multi-scenario prediction model based on Markov chain is proposed for stochastic power prediction. Firstly， ARIMA model and Markov chain are respectively used to build the load model of a park according to the random characteristics of renewable energy and the load. Then， posterior information is used to adjust the Markov probability matrix adaptively to improve its prediction accuracy which is affected by the periodical fluctuation of load varying with production，season or other factors. To improve the accuracy of multi-step prediction in a prediction horizon，a multi-scenario prediction model based on scenario tree is proposed， considering the multi-step prediction scenarios and their probabilities. The model can make more effective use of Markov probability matrix. Finally，a case study is implemented based on historical power data of an industrial park. The results show that compared with the unadjusted Markov model， the adaptive model is of a lower load prediction error，and the prediction error of the one with 7-day adjustment time is as low as 0.034 5 p.u. Compared with the commonly used Maximum Likelihood Estimation，the proposed multi-scenario prediction model is of a lower weighted average error.

Key words: renewable energy consumption, Markov chain, prediction model, adaptive adjustment, distributed energy

中图分类号:

TK01

魏妍萍, 王军, 李南帆, 师长立. 基于Markov链的园区随机功率多场景预测模型[J]. 综合智慧能源, 2023, 45(1): 14-22.

WEI Yanping, WANG Jun, LI Nanfan, SHI Changli. Prediction model of stochastic power in industrial parks based on Markov chain[J]. Integrated Intelligent Energy, 2023, 45(1): 14-22.

图/表 11

图1

图2

图3

图4

表1

图5

图6

图7

表2

表3

表4

参考文献 23

[1]	舒印彪, 陈国平, 贺静波, 等. 构建以新能源为主体的新型电力系统框架研究[J]. 中国工程科学, 2021, 23(6):61-69.
	SHU Yinbiao, CHEN Guoping, HE Jingbo, et al. Building a new electric power system based on new energy sources[J]. Strategic Study of CAE, 2021, 23(6):61-69.
[2]	国家能源局. 构建以新能源为主体的新型电力系统[EB/OL].(2015-02-16)[2022-12-30]. https://www.nea.gov.cn/2022-04/01/c_1310538944.htm.
[3]	肖先勇, 郑子萱. “双碳”目标下新能源为主体的新型电力系统:贡献、关键技术与挑战[J]. 工程科学与技术, 2022, 54(1):47-59.
	XIAO Xianyong, ZHENG Zixuan. New power systems dominated by renewable energy towards the goal of emission peak & carbon neutrality: Contribution, key techniques, and challenges[J]. Advanced Engineering Sciences, 2022, 54(1):47-59.
[4]	文云峰, 杨伟峰, 汪荣华, 等. 构建100％可再生能源电力系统述评与展望[J]. 中国电机工程学报, 2020, 40(6):1843-1855.
	WEN Yunfeng, YANG Weifeng, WANG Ronghua, et al. Review and prospect of toward 100% renewable energy power systems[J]. Proceedings of the CSEE, 2020, 40(6):1843-1855.
[5]	康重庆, 姚良忠. 高比例可再生能源电力系统的关键科学问题与理论研究框架[J]. 电力系统自动化, 2017, 41(9):1-11.
	KANG Chongqing, YAO Liangzhong. Key scientific issues and theoretical research framework for power systems with high proportion of renewable energy[J]. Automation of Electric Power Systems, 2017, 41(9):1-11.
[6]	MIGUEL E, JOANA P, BENJIAMIN M, et al. 100% renewable energy system in Japan: Smoothing and ancillary services[J]. Applied Energy, 2018, 224:698-707. doi: 10.1016/j.apenergy.2018.04.067
[7]	TUMMURU R, MISHRA K, SRINIVAS S. Dynamic energy management of renewable grid integrated hybrid energy storage system[J]. IEEE Transactions on Industrial Electronics, 2015, 62:7728-7737. doi: 10.1109/TIE.2015.2455063
[8]	万灿, 崔文康, 宋永华. 新能源电力系统概率预测:基本概念与数学原理[J]. 中国电机工程学报, 2021, 41(19):6493-6508.
	WAN Can, CUI Wenkang, SONG Yonghua. Theories, methodologies and applications of probabilistic forecasting for power systems with renewable energy sources[J]. Automation of Electric Power Systems, 2021, 41(19):6493-6508.
[9]	MAHMOUD E, MARK S, ELDAR N, et al. A hierarchical two-stage energy management for a home microgrid using model predictive and real-time controllers[J]. Applied Energy, 2020, 269:115118. doi: 10.1016/j.apenergy.2020.115118
[10]	ZHANG Y, CAMPANA E, LUNDBLAD A, et al. Comparative study of hydrogen storage and battery storage in grid connected photovoltaic system: Storage sizing and rule-based operation[J]. Applied Energy, 2017, 201:397-411. doi: 10.1016/j.apenergy.2017.03.123
[11]	WANG Y, SUN Z, CHEN Z. Development of energy management system based on a rule-based power distribution strategy for hybrid power sources[J]. Energy, 2019, 175:1055-1066. doi: 10.1016/j.energy.2019.03.155
[12]	李国庆, 李欣彤, 边竞, 等. 计及光伏-负荷预测不确定性的直流跨省互联电网双级调度策略[J]. 中国电机工程学报, 2021, 41(14):4763-4776.
	LI Guoqing, LI Xintong, BIAN Jing, et al. Two level scheduling strategy for inter-provincial DC power grid considering the uncertainty of PV-load prediction[J]. Proceedings of the CSEE, 2021, 41(14):4763-4776.
[13]	王玥, 张宇帆, 李昭昱, 等. 即插即用能量组织日前负荷概率预测方法[J]. 电网技术, 2019, 43(9):3053-3060.
	WANG Yue, ZHANG Yufan, LI Zhaoyu, et al. Day-ahead probability load forecasting of energy tissues with plug-and-play function[J]. Power System Technology, 2019, 43(9):3053-3060.
[14]	林文智, 杨苹, 陈芯羽, 等. 计及需求响应不确定性的园区综合能源系统日前经济优化调度[J]. 电力建设, 2021, 42(12):9-20. doi: 10.12204/j.issn.1000-7229.2021.12.002
	LIN Wenzhi, YANG Ping, CHEN Xinyu, et al. Day-ahead optimal economic dispatch of park integrated energy system considering uncertainty of demand response[J]. Electric Power Construction, 2021, 42(12):9-20. doi: 10.12204/j.issn.1000-7229.2021.12.002
[15]	周坤, 许云飞, 崔昊杨, 等. 基于预测控制的多种新能源互补电力系统动态调度模型[J]. 现代电力, 2021, 38(3):248-257.
	ZHOU Kun, XU Yunfei, CUI Haoyang, et al. A predictive control based dynamic dispatch model for complementary power system containing multi-renewable energy sources[J]. Modern Electric Power, 2021, 38(3):248-257.
[16]	高东学, 李文启, 李程昊, 等. 基于负荷预测的储能功率分配优化策略[J]. 可再生能源, 2021, 39(4):554-560.
	GAO Dongxue, LI Wenqi, LI Chenghao, et al. Optimization strategy of energy storage power allocation based on load forecasting[J]. Renewable Energy Resources, 2021, 39(4):554-560.
[17]	汪希玥, 徐箭, 廖思阳, 等. 考虑虚拟储能特性的柔性负荷调控裕度评估方法[J]. 华电技术, 2021, 43(9):37-45.
	WANG Xiyue, XU Jian, LIAO Siyang, et al. Flexible load regulation margin evaluation method considering virtual energy storage characteristics[J]. Huadian Technology, 2021, 43(9):37-45.
[18]	刘静, 史梦鸽, 胡永锋. 含电池储能系统的智能楼宇多阶段能量管理策略[J]. 综合智慧能源, 2022, 44(3):29-37. doi: 10.3969/j.issn.2097-0706.2022.03.005
	LIU Jing, SHI Mengge, HU Yongfeng. Multi-stage energy management strategy for smart buildings with BESS[J]. Integrated Intelligent Energy, 2022, 44(3):29-37. doi: 10.3969/j.issn.2097-0706.2022.03.005
[19]	熊图, 赵宏伟, 蔡智洋, 等. 动态组合深度学习模型在短期负荷及光伏功率预测中的应用[J]. 可再生能源, 2020, 38(4):458-463.
	XIONG Tu, ZHAO Hongwei, CAI Zhiyang, et al. Load and PV output forecasting based on dynamic ensemble deep learning model[J]. Renewable Energy Resources, 2020, 38(4):458-463.
[20]	李旭炯, 孙林花, 杨郭明. 基于改进粒子群算法的光伏系统附加向心属性最大功率跟踪研究[J]. 综合智慧能源, 2022, 44(3):70-76. doi: 10.3969/j.issn.2097-0706.2022.03.011
	LI Xujiong, SUN Linhua, YANG Guoming. MPPT for PV systems appended with centripetal attribute based on improved PSO algorithm[J]. Integrated Intelligent Energy, 2022, 44(3):70-76. doi: 10.3969/j.issn.2097-0706.2022.03.011
[21]	张孝顺, 谭恬, 蒙蝶, 等. 基于光伏系统的动态代理模型最大功率点跟踪算法研究[J]. 华电技术, 2021, 43(8):1-10.
	ZHANG Xiaoshun, TAN Tian, MENG Die, et al. Study on dynamic surrogate model for MPPT of PV systems[J]. Huadian Technology, 2021, 43(8):1-10.
[22]	HE J, SHI C, WEI T, et al. Hierarchical optimal energy management strategy of hybrid energy storage considering uncertainty for a 100% clean energy town[J]. Journal of Energy Storage, 2021, 41:102917. doi: 10.1016/j.est.2021.102917
[23]	KUSKA N, HEITSCH H, ROMISCH W. Scenario reduction and scenario tree construction for power management problems[C]// 2003 IEEE Bologna Power Tech Conference Proceedings, 2003, 3:7-13.

参数	值
自适应调节参数η	1.2
离散状态数s	30
新能源预测误差的期望μ	0
新能源预测误差的标准差σ	0.05
负荷的历史数据周期T_day/d	30
自适应时间t_day/d	7
非叶节点的子节点数α	2
预测时域/预测步数N_p	4
预测周期Δt/h	0.5

项目	R_RMSE（标幺值）
光伏功率（06:00 —18:00）	0.065 6
风机功率	0.061 4
负荷功率	0.034 5

自适应时间t_day/d	R_RMSE(标幺值)	误差降低/%
无	0.037 1	—
3	0.035 6	3.98
5	0.034 9	5.91
7	0.034 5	7.51
10	0.035 1	5.39
14	0.035 3	4.81

项目	R_RMSE（标幺值）		精度提升/%
项目	场景树法	MLE法	精度提升/%
1步	0.030 2	0.031 9	5.32
2步	0.031 8	0.034 0	6.41
3步	0.033 0	0.036 1	8.58
4步	0.034 5	0.039 2	11.90