基于数据分解与斑马算法优化的混合核极限学习机月径流预测

李菊; 崔东文

doi:10.11988/ckyyb.20230782

PDF(7174 KB)

长江科学院院报 ›› 2024, Vol. 41 ›› Issue (6) : 42-50. DOI: 10.11988/ckyyb.20230782

水资源

基于数据分解与斑马算法优化的混合核极限学习机月径流预测

李菊¹, 崔东文²

作者信息 +

Monthly Runoff Prediction Using Hybrid Kernel Extreme Learning Machine Based on Data Decomposition and Zebra Algorithm Optimization

LI Ju¹, CUI Dong-Wen²

Author information +

文章历史 +

摘要

为提高月径流预测精度,改进混合核极限学习机(HKELM)预测性能,提出小波包分解(WPT)－斑马优化算法(ZOA)－HKELM组合模型。利用WPT处理月径流时序数据,构建局部高斯径向基核函数和全局多项式核函数相混合的HKELM;通过ZOA优化HKELM超参数(正则化参数、核参数、权重系数),建立WPT-ZOA-HKELM组合模型,并构建WPT-遗传算法(GA)-HKELM、WPT-灰狼优化(GWO)算法-HKELM、WPT-鲸鱼优化算法(WOA)-HKELM、WPT-ZOA-极限学习机(ELM)、WPT-ZOA-最小二乘支持向量机(LSSVM)、ZOA-HKELM作对比模型,通过黑河流域莺落峡、讨赖河水文站月径流时间序列预测实例对各模型进行检验。结果表明:①莺落峡、讨赖河水文站月径流时间序列WPT-ZOA-HKELM模型预测的平均绝对百分比误差分别为1.054%、0.761%,决定系数均达0.999 9,优于其他对比模型,具有更高的预测精度,预测效果更好。②利用ZOA优化HKELM超参数,可提高HKELM预测性能,优化效果优于GWO、WOA、GA。③预测模型能充分发挥WPT、ZOA和HKELM优势,提高月径流预测精度;在相同分解和优化情形下,HKELM的预测性能优于ELM、LSSVM。

Abstract

In order to enhance the precision of monthly runoff forecasts and optimize the prediction performance of the Hybrid Kernel Extreme Learning Machine (HKELM), we propose a synergistic approach integrating Wavelet Packet Decomposition (WPT), the Zebra Optimization Algorithm (ZOA), and HKELM. The approach involves applying WPT to preprocess monthly runoff time series data and constructing a HKELM that combines local Gaussian radial basis function with global polynomial kernel function. By refining HKELM hyperparameters (including regularization parameters, kernel parameters, and weight coefficients) through ZOA, we establish the WPT-ZOA-HKELM model, alongside comparative models such as WPT-Genetic Algorithm (GA)-HKELM, WPT-Grey Wolf Optimization (GWO) algorithm-HKELM, WPT-Whale Optimization (WOA)-HKELM, WPT-ZOA Extreme Learning Machine (ELM), WPT-ZOA Least Squares Support Vector Machine (LSSVM), and ZOA-HKELM. These models are evaluated using monthly runoff time series data from the Yingluoxia and Tuolai River hydrological stations in the Heihe River Basin. Our findings indicate that: (1) The WPT-ZOA-HKELM model achieves average absolute percentage errors of 1.054% and 0.761% respectively, with determination coefficients of 0.999 9, surpassing other comparative models in terms of prediction accuracy and performance. (2) Optimization of HKELM hyperparameters with ZOA enhances predictive performance compared to GWO, WOA, and GA. (3) Through leveraging WPT, ZOA, and HKELM, the prediction model significantly improves monthly runoff forecast accuracy. Under equivalent decomposition and optimization conditions, the predictive performance of HKELM is superior to ELM and LSSVM.

导出引用

李菊, 崔东文. 基于数据分解与斑马算法优化的混合核极限学习机月径流预测[J]. 长江科学院院报. 2024, 41(6): 42-50 https://doi.org/10.11988/ckyyb.20230782

LI Ju, CUI Dong-Wen. Monthly Runoff Prediction Using Hybrid Kernel Extreme Learning Machine Based on Data Decomposition and Zebra Algorithm Optimization[J]. Journal of Changjiang River Scientific Research Institute. 2024, 41(6): 42-50 https://doi.org/10.11988/ckyyb.20230782

中图分类号： TV124 P333

参考文献

[1] 王应武,白栩嘉,崔东文.基于WPT-ISO-RELM模型的月径流时间序列预测研究[J].水力发电,2024,50(3):12-18,38. (WANG Ying-wu, BAI Xu-jia, CUI Dong-wen. Research on Monthly Runoff Time Series Prediction Based on WPT-ISO-RELM Model[J].Hydroelectric Power, 2024,50(3): 12-18,38.(in Chinese))
[2] 高雪梅,崔东文.WPT-FLA-RELM模型的马鹿塘水电站入库日径流多步预测[J].云南水力发电,2023,39(11):56-62. (GAO Xue-mei, CUI Dong-wen. Multi-step Prediction of Daily Inflow of Malutang Hydropower Station Using WPT-FLA-RELM Model[J].Yunnan Hydroelectric Power, 2023,39 (11):56-62.(in Chinese))
[3] 赵莹玉,彭慧春,李继清.融合改进灰狼算法的机器学习月径流预测方法[J].水力发电学报,2023,42(9):34-45.(ZHAO Ying-yu, PENG Hui-chun, LI Ji-qing. Machine Learning Method for Monthly Runoff Prediction Based on Improved Grey Wolf Algorithm[J]. Journal of Hydroelectric Engineering, 2023, 42(9): 34-45.(in Chinese))
[4] 王文川, 杜玉瑾, 和吉, 等. 基于CEEMDAN-VMD-BP模型的月径流量预测研究[J]. 华北水利水电大学学报(自然科学版), 2023, 44(1): 32-40, 48.(WANG Wen-chuan, DU Yu-jin, HE Ji, et al. Research on Monthly Runoff Prediction Based on CEEMDAN-VMD-BP Model[J]. Journal of North China University of Water Resources and Electric Power (Natural Science Edition), 2023, 44(1): 32-40, 48.(in Chinese))
[5] 刘尚东, 孙东永, 许晶晶, 等. 基于ESMD-NNBR耦合的月径流预测模型[J]. 水文, 2023, 43(1): 52-56. (LIU Shang-dong, SUN Dong-yong, XU Jing-jing, et al. Monthly Runoff Forecasting Based on ESMD-NNBR Hybrid Model[J]. Journal of China Hydrology, 2023, 43(1): 52-56.(in Chinese))
[6] 张璐, 刘真, 李磊, 等. 基于VMD-PSR-BNN模型的月径流预测方法研究[J]. 中国农村水利水电, 2023(4): 105-113. (ZHANG Lu, LIU Zhen, LI Lei, et al. Research on the Monthly Runoff Prediction Method Based on VMD-PSR-BNN Model[J]. China Rural Water and Hydropower, 2023(4): 105-113.(in Chinese))
[7] 孙国梁, 李保健, 徐冬梅, 等. 基于VMD-SSA-LSTM的月径流预测模型及应用[J]. 水电能源科学, 2022, 40(5): 18-21. (SUN Guo-liang, LI Bao-jian, XU Dong-mei, et al. Monthly Runoff Prediction Model Based on VMD-SSA-LSTM[J]. Water Resources and Power, 2022, 40(5): 18-21.(in Chinese))
[8] 梁晓鑫, 崔东文. 基于奇异谱分析与梯度优化算法优化的RVM、SVM月径流预测研究[J]. 人民珠江, 2022, 43(1): 111-118. (LIANG Xiao-xin, CUI Dong-wen. Monthly Runoff Prediction with RVM and SVM Optimized by Singular Spectrum Analysis and Gradient-based Optimization Algorithm[J]. Pearl River, 2022, 43(1): 111-118.(in Chinese))
[9] 王丽丽, 李新, 冉有华, 等. 基于奇异谱分析—灰狼优化—支持向量回归混合模型的黑河正义峡月径流预测[J]. 遥感技术与应用, 2020, 35(2): 355-364. (WANG Li-li, LI Xin, RAN You-hua, et al. Monthly Runoff Prediction of Zhengyixia in the Heihe River Based on Singular Spectrum Analysis-grey Wolf Optimizer-support Vector Regression Hybrid Model[J]. Remote Sensing Technology and Application, 2020, 35(2): 355-364.(in Chinese))
[10] 徐冬梅, 廖安栋, 王文川. 基于VMD-EEMD-CNN-LSTM混合模型的月径流预测[J]. 水利规划与设计, 2023(7): 57-63. (XU Dong-mei, LIAO An-dong, WANG Wen-chuan. Monthly Runoff Prediction Based on VMD-EEMD-CNN-LSTM Mixed Model[J]. Water Resources Planning and Design, 2023(7): 57-63.(in Chinese))
[11] TROJOVSKÁ E, DEHGHANI M, TROJOVSKY P. Zebra Optimization Algorithm: A New Bio-inspired Optimization Algorithm for Solving Optimization Algorithm[J]. IEEE Access, 2022, 10: 49445-49473.
[12] 范君, 王新, 徐慧. 粒子群优化混合核极限学习机的构造煤厚度预测方法[J]. 计算机应用, 2018, 38(6): 1820-1825, 1830. (FAN Jun, WANG Xin, XU Hui. Prediction Method of Tectonic Coal Thickness Based on Particle Swarm Optimized Hybrid Kernel Extreme Learning Machine[J]. Journal of Computer Applications, 2018, 38(6): 1820-1825, 1830.(in Chinese))
[13] 张婷慧, 宇洁, 叶张林, 等. 结合粒子群寻优的混合核函数极限学习机分类模型[J]. 测绘科学技术学报, 2019, 36(1): 56-61. (ZHANG Ting-hui, YU Jie, YE Zhang-lin, et al. Classification Model Research of Mixed Kernel Extreme Learning Machine Based on Particle Swarm Optimization[J]. Journal of Geomatics Science and Technology, 2019, 36(1): 56-61.(in Chinese))
[14] 李可军,徐延顺,魏本刚,等.基于PSO-HKELM的变压器顶层油温预测模型[J].高电压技术,2018,44(8):2501-2508.(LI Ke-jun,XU Yan-shun,WEI Ben-gang,et al.Prediction Model for Top Oil Temperature of Transformer Based on Hybrid Kernel Extreme Learning Machine Trained and Optimized by Particle Swarm Optimization[J]. High Voltage Engineering, 2018, 44(8): 2501-2508.(in Chinese))
[15] 石祖智, 常峻, 吴斌平, 等. 基于改进混合核极限学习机的坝基注浆量预测代理模型研究[J]. 水利水电技术(中英文), 2021, 52(9): 57-66. (SHI Zu-zhi, CHANG Jun, WU Bin-ping, et al. Study on Surrogate Model of Dam Foundation Grouting Volume Prediction Based on Improved Multiple Kernel Extreme Learning Machine[J]. Water Resources and Hydropower Engineering, 2021, 52(9): 57-66.(in Chinese))
[16] 王瑞,徐新超,逯静.基于麻雀搜索算法优化变分模态分解和混合核极限学习机的短期风电功率预测[J].信息与控制,2023,52(4):444-454.(WANG Rui, XU Xin-chao, LU Jing. Short-term Wind Power Prediction Based on Sparrow Search Algorithm Optimized Variational Mode Decomposition and Hybrid Kernel Extreme Learning Machine[J]. Information and Control, 2023,52(4): 444-454.(in Chinese))
[17] 郭建帅, 崔双喜, 郭建斌, 等. 基于VMD-SSA-HKELM的超短期负荷预测[J]. 国外电子测量技术, 2022, 41(6): 105-111. (GUO Jian-shuai, CUI Shuang-xi, GUO Jian-bin, et al. Ultra-short-term Load Prediction Based on VMD-SSA-HKELM[J]. Foreign Electronic Measurement Technology, 2022, 41(6): 105-111.(in Chinese))
[18] 谢国民, 刘东阳, 刘明. 多策略改进MPA算法与HKELM的变压器故障辨识[J]. 电子测量与仪器学报, 2023, 37(4): 172-182. (XIE Guo-min, LIU Dong-yang, LIU Ming. Transformer Fault Identification Based on Multi-strategy Improved MPA Algorithm and HKELM[J]. Journal of Electronic Measurement and Instrumentation, 2023, 37(4): 172-182.(in Chinese))
[19] 邵良杉,詹小凡.基于IWOA-HKELM的矿井突水水源识别[J].中国安全科学学报,2019,29(9):113-118.(SHAO Liang-shan,ZHAN Xiao-fan.Identification Method of Mine Water Inrush Source Based on IWOA-HKELM[J]. China Safety Science Journal, 2019, 29(9): 113-118.(in Chinese))
[20] 李新华,崔东文.基于WPD-RSA-ELM模型的水文时间序列多步预测[J].水利水电技术(中英文),2022,53(11):69-77.(LI Xin-hua,CUI Dong-wen.Multi-step Prediction of Hydrological Time Series Based on WPD-RSA-ELM Model[J]. Water Resources and Hydropower Engineering, 2022, 53(11): 69-77.(in Chinese))
[21] 崔东文, 袁树堂. 基于WPD-AHA-ELM模型的水质时间序列多步预测[J]. 三峡大学学报(自然科学版), 2023, 45(1): 6-13. (CUI Dong-wen, YUAN Shu-tang. Multi-step Prediction of Water Quality Time Series Based on WPD-AHA-ELM Model[J]. Journal of China Three Gorges University (Natural Sciences), 2023, 45(1): 6-13.(in Chinese))
[22] 雷庆文, 高培强, 李建林. 时序分解和CNN-LSTM相融合的月径流预报模型[J]. 长江科学院院报, 2023, 40(6): 49-54. (LEI Qing-wen, GAO Pei-qiang, LI Jian-lin. A Monthly Runoff Forecast Model Combining Time Series Decomposition and CNN-LSTM[J]. Journal of Changjiang River Scientific Research Institute, 2023, 40(6): 49-54.(in Chinese))
[23] 崔东文,李代华.基坑变形预测的改进供需优化算法-指数幂乘积模型[J].水利水电科技进展,2020,40(4):43-50.(CUI Dong-wen,LI Dai-hua.Improved Supply-demand-based Optimization Algorithm-exponential Power Product Model in Foundation Pit Deformation Prediction[J]. Advances in Science and Technology of Water Resources, 2020, 40(4): 43-50.(in Chinese))
[24] 崔东文, 金波. 花授粉算法-BP神经网络模型及其在月径流预报中的应用[J]. 人民珠江, 2016, 37(4): 36-40. (CUI Dong-wen, JIN Bo. Flower Pollination Algorithm-BP Neural Network Model and Its Application in the Prediction of Monthly Runoff[J]. Pearl River, 2016, 37(4): 36-40.(in Chinese))
[25] 崔东文.几种智能算法与支持向量机融合模型在中长期月径流预测中的应用[J].华北水利水电大学学报(自然科学版),2016,37(5):51-57.(CUI Dong-wen.Application of Several Intelligent Algorithms and Support Vector Machine Fusion Model in Medium and Long Term Runoff Forecasting[J]. Journal of North China University of Water Resources and Electric Power (Natural Science Edition), 2016, 37(5): 51-57.(in Chinese))
[26] 崔东文. 基于WPD-MRFO-ESN模型的水库来水量时间序列预测[J]. 华北水利水电大学学报(自然科学版), 2022, 43(6): 10-17. (CUI Dong-wen. Prediction of Reservoir Inflow Time Series Based on WPD-MRFO-ESN Model[J]. Journal of North China University of Water Resources and Electric Power (Natural Science Edition), 2022, 43(6): 10-17.(in Chinese))
[27] 杨琼波, 崔东文. WPD-RSO-ESN和SSA-RSO-ESN模型在径流时间序列预测中应用比较[J]. 中国农村水利水电, 2022(2): 61-67, 75. (YANG Qiong-bo, CUI Dong-wen. Application Comparison of WPD-RSO-ESN and SSA-RSO-ESN Models in Runoff Time Series Forecasting[J]. China Rural Water and Hydropower, 2022(2): 61-67, 75.(in Chinese))
[28] 岳兆新, 艾萍, 熊传圣, 等. 基于改进深度信念网络模型的中长期径流预测[J]. 水力发电学报, 2020, 39(10): 33-46. (YUE Zhao-xin, AI Ping, XIONG Chuan-sheng, et al. Mid-and Long-term Runoff Forecasting Based on Improved Deep Belief Networks Model[J]. Journal of Hydroelectric Engineering, 2020, 39(10): 33-46.(in Chinese))
[29] 陈金红,崔东文.基于深度学习神经网络超参数优化的入库径流预测方法研究:以云南省暮底河水库为例[J].三峡大学学报(自然科学版),2023,45(4):25-32.(CHEN Jin-hong, CUI Dong-wen. Study on the Method of Runoff Inflow Prediction Based on Deep Learning Neural Network Hyperparametric Optimization—A Case Study of the Mudihe Reservoir in Yunnan Province[J]. Journal of China Three Gorges University (Natural Sciences), 2023, 45(4): 25-32.(in Chinese))
[30] 李杰,崔东文.新型群体智能算法优化BIGRU/BILSTM的水资源空间均衡评价[J].中国农村水利水电,2023(11):1-9.(LI Jie, CUI Dong-wen. Optimization of Spatial Equilibrium Evaluation of Water Resources for BIGRU/BILSTM Using a New Group Intelligence Algorithm[J]. China Rural Water and Hydropower, 2023(11): 1-9.(in Chinese))
[31] 陈金红, 崔东文. 基于小波包分解的GJO-XGBoost水面蒸发量预测[J]. 三峡大学学报(自然科学版), 2023, 45(3): 1-7. (CHEN Jin-hong, CUI Dong-wen. GJO-XGBoost Prediction of Water Surface Evaporation Based on Wavelet Packet Decomposition[J]. Journal of China Three Gorges University (Natural Sciences), 2023, 45(3): 1-7.(in Chinese))
[32] 李杰,崔东文.若干新型群体智能算法优化高斯过程回归的年降水量预测[J].节水灌溉,2023(7):96-103,109.(LI Jie,CUI Dong-wen.Some New Swarm Intelligence Algorithms for Optimization of Annual Precipitation Prediction Based on Gaussian Process Regression[J]. Water Saving Irrigation,2023(7):96-103, 109.(in Chinese))
[33] 杨琼波, 崔东文. WPD-COA-ELM模型在汛期月降水量时间序列预测中的应用研究[J]. 水文, 2023, 43(1): 17-23. (YANG Qiong-bo, CUI Dong-wen. Application of WPD-COA-ELM Model in the Prediction of Time Series of Monthly Precipitation in Flood Season[J]. Journal of China Hydrology, 2023, 43(1): 17-23.(in Chinese))
[34] 许建伟, 崔东文. WPT-HPO-ELM径流多步预报模型研究[J]. 水资源与水工程学报, 2022, 33(6): 69-76. (XU Jian-wei, CUI Dong-wen. WPT-HPO-ELM Multi-step Runoff Forecast Model[J]. Journal of Water Resources and Water Engineering, 2022, 33(6): 69-76.(in Chinese))
[35] 梁晓鑫, 崔东文. 基于WPD-AGTO-DELM模型的年径流时间序列预测[J]. 三峡大学学报(自然科学版), 2022, 44(5): 14-20. (LIANG Xiao-xin, CUI Dong-wen. Prediction of Annual Runoff Time Series Based on WPD-AGTO-DELM Model[J]. Journal of China Three Gorges University (Natural Sciences),2022,44(5):14-20.(in Chinese))
[36] 杨琼波, 崔东文. 基于小波包分解的AJS-GMDH月径流时间序列预测研究[J]. 水力发电, 2022, 48(6): 45-51. (YANG Qiong-bo, CUI Dong-wen. Research on AJS-GMDH Monthly Runoff Time Series Forecast Based on Wavelet Packet Decomposition[J]. Water Power, 2022, 48(6): 45-51.(in Chinese))
[37] 崔东文, 杨琼波. 基于TSA-DRNN模型的年径流预测研究[J]. 华北水利水电大学学报(自然科学版), 2021, 42(6): 35-41. (CUI Dong-wen, YANG Qiong-bo. Research on Annual Runoff Prediction Based on TSA-DRNN Model[J]. Journal of North China University of Water Resources and Electric Power (Natural Science Edition), 2021, 42(6): 35-41.(in Chinese))
[38] 张代凤,崔东文.三种新型群体智能算法优化正则化极限学习机的三峡入库日径流预测[J/OL].长江科学院院报,[2024-04-30].http://kns.cnki.net/kcms/detail/42.1171.TV.20230829.1119.002.html.(ZHANG Dai-feng, CUI Dong-wen. Optimizing Regularized Extreme Learning Machines for Three Gorges Reservoir Inflow Forecasting using Three Novel Swarm Intelligence Algorithms[J/OL]. Journal of Changjiang River Scientific Research Institute, [2024-04-30]. (in Chinese))