作者:吴鹏1,2, 王鼎1, 马宇超1, 陈信华3, 沈金阳1, 苏畅宇1, 虞乐水1
作者单位:1. 常州大学机械与轨道交通学院, 江苏 常州 213164;
2. 上海交通大学电子信息与电气工程学院电气工程流动站, 上海 200030;
3. 溧阳市新力机械铸造有限公司, 江苏 溧阳 213300
关键词:风力发电;功率预测;局部最优;ABC-BP;算法模型
摘要:
稳定可控的风力发电系统是风力资源开发利用的核心,提出一种基于人工蜂群-神经网络算法(artificial bee colony-back propagation,ABC-BP)的风力发电机短期功率预测方法,进一步提高短期风电功率预测的准确性。针对现有短期功率预测方法中遇到的收敛速度慢、局部最优等缺陷,结合ABC人工蜂群算法,提出改进ABC-BP算法。在对其数学模型收敛性证明的基础上,采用实际风电数据进行仿真验证,并通过模拟风力发电平台,进行实验,实验结果表明,预测数据达到实验要求,所提出的改进算法模型是可行的。改进ABC-BP算法应用于风力发电机监测系统中,为降低风电系统运维成本、提高监测效率提供一种有效解决方案。
Short-term power prediction method for wind turbines based on ABC-BP
WU Peng1,2, WANG Ding1, MA Yuchao1, CHEN Xinhua3, SHEN Jinyang1, SU Changyu1, YU Leshui1
1. School of Mechanical Engineering and Rail Transit, Changzhou University, Changzhou 213164, China;
2. Electrical Engineering Mobile Station, School of Electronic Information and Electrical Engineering, Shanghai Jiaotong University, Shanghai 200030, China;
3. Liyang Xinli Machinery Casting Co., Liyang 213300, China
Abstract: Stable and controllable wind power generation system is the core of wind resource development and utilization. Artificial bee colony-back propagation (ABC-BP) was proposed for short-term power prediction of wind turbine. It further improves the accuracy of short-term wind power forecast. The modified ABC-BP combined with artificial bee colony is presented for defects of slow convergence speed and local optimum in the existing short-term power prediction methods. The actual wind power data is used for simulation on the basis of the convergence proof of its mathematical model. Experimental analysis was carried out on the simulative wind power generation platform. Experimental results show that the predicted data meet the experimental requirements, and the proposed improved algorithm model is feasible. The improved ABC-BP algorithm is applied to the wind turbine monitoring system. It provides an effective solution for reducing the operation and maintenance cost of wind power system, and it can improve the monitoring efficiency.
Keywords: wind power generation;power prediction;local optimum;ABC-BP;algorithm model
2023, 49(6):82-91 收稿日期: 2022-03-02;收到修改稿日期: 2022-07-21
基金项目: 江苏省产学研合作项目(BY2021221)
作者简介: 吴鹏(1987-),男,江苏常州市人,讲师,博士,主要从事电力系统优化
参考文献
[1] 张颖超, 王雅晨, 邓华, 等. 基于IAFSA-BPNN的短期风电功率预测[J]. 电力系统保护与控制, 2017, 45(7): 58-63
[2] 马煜普, 乔颖, 鲁宗相, 等. 考虑风电特性的中期-日前嵌套式机组组合优化[J]. 电网技术, 2019, 43(11): 3908-3916
[3] 冯双磊, 王伟胜, 刘纯, 等. 基于物理原理的风电场短期风速预测研究[J]. 太阳能学报, 2011, 32(5): 611-616
[4] 陈颖, 孙荣富, 吴志坚, 等. 基于统计升尺度方法的区域风电场群功率预测[J]. 电力系统自动化, 2013, 37(7): 1-5
[5] 刘勇, 杨熙卉, 燕林滋, 欧云. 基于CEEMD双层分解和Granger因果变量选择的风电功率预测[J/OL]. 中国测试: 1-7(2021-12-03)[2022-07-21]. http://kns.cnki.net/kcms/detail/51.1714.TB.20211201.1603.026.html
[6] MA L, LUAN S, JIANG C, et al. A review on the forecasting of wind speed and generated power[J]. Renewable and Sustainable Energy Reviews, 2009, 13(4): 915-920
[7] ZULUAGA C D, ÁLVAREZ M A, GIRALDO E. Short-term wind speed prediction based on robust Kalman filtering: an experimental comparison[J]. Applied Energy, 2015, 156: 321-330
[8] LIU H, TIAN H, PAN D, et al. Forecasting models for wind speed using wavelet, wavelet packet, time series and Artificial Neural Networks[J]. Applied Energy, 2013, 107: 191-208
[9] 农汉彪, 曾巧妮. 基于聚类分析的心电信号基线漂移去除方法[J/OL]. 中国测试: 1-6(2021-11-30)[2022-07-21]. http://kns.cnki.net/kcms/detail/51.1714.TB.20211129.2033.015.html
[10] 王尤嘉, 鲁宗相, 乔颖, 等. 基于特征聚类的区域风电短期功率统计升尺度预测[J]. 电网技术, 2017, 41(5): 1383-1389
[11] GUO Z, ZHAO W, LU H, et al. Multi-step forecasting for wind speed using a modified EMD-based artificial neural network model[J]. Renewable Energy, 2012, 37(1): 241-249
[12] LI L, ZHAO X, TSENG M, et al. Short-term wind power forecasting based on support vector machine with improved dragonfly algorithm[J]. Journal of Cleaner Production, 2020, 242: 118447
[13] 凡航, 张雪敏, 梅生伟, 等. 基于时空神经网络的风电场超短期风速预测模型[J]. 电力系统自动化, 2021, 45(7): 28-35
[14] 周勇良, 余光正, 刘建锋, 等. 基于改进长期循环卷积神经网络的海上风电功率预测[J]. 电力系统自动化, 2021, 45(3): 183-191
[15] 师洪涛, 杨静玲, 丁茂生, 等. 基于小波—BP神经网络的短期风电功率预测方法[J]. 电力系统自动化, 2011, 35(16): 44-48