A Parallel Short-Term Wind Power Forecasting Model Based onImproved Crested Porcupine Optimizer for VMD
[1]JIANG Jiandong,WANG Yulong,LIU Mingyu,et al.A Parallel Short-Term Wind Power Forecasting Model Based onImproved Crested Porcupine Optimizer for VMD[J].Journal of Zhengzhou University (Engineering Science),2027,48(XX):1-8.[doi:10.13705/j.issn.1671-6833.2026.03.007]
Copy
Journal of Zhengzhou University (Engineering Science)[ISSN
1671-6833/CN
41-1339/T] Volume:
48
Number of periods:
2027 XX
Page number:
1-8
Column:
Public date:
2027-12-10
- Title:
- A Parallel Short-Term Wind Power Forecasting Model Based onImproved Crested Porcupine Optimizer for VMD
- Keywords:
- golden sine algorithm; lens imaging reverse learning; improved crested porcupine optimizer; VMD; in; former; BiLSTM; parallel forecasting
- CLC:
- TU528. 1
- Abstract:
- To address the challenges of significant noise interference in wind power sequences, sensitivity to decomposition parameters, and limited temporal feature extraction in single prediction models, this paper proposes a hybrid short-term wind power forecasting model that integrates an improved golden sine lens opposition-based crestedporcupine optimizer (GSLOCPO) , variational mode decomposition (VMD) , and a parallel Informer-BiLSTM prediction framework. First, the GSLOCPO algorithm is enhanced by incorporating a golden sine strategy and lens imaging opposition-based learning, enabling dynamic optimization of VMD parameters using envelope entropy as the fitness function to effectively mitigate mode mixing. Next, a sliding window strategy is employed for dynamic decomposition of the wind power time series, extracting multi-scale intrinsic mode functions ( IMFs) to separate noisefrom trend features. Subsequently, a parallel Informer-BiLSTM prediction structure is constructed, where the Informer leverages a ProbSparse attention mechanism to capture long-range global dependencies, while the BiLSTMnetwork explores local temporal dynamics in both forward and backward directions. Parallel computation is adoptedto improve prediction efficiency. Finally, a fully connected layer adaptively fuses the outputs, optimizing featureweight distribution. Experimental results demonstrate that the proposed GSLOCPO-VMD-Informer-BiLSTM modelsignificantly outperforms conventional methods in both accuracy and stability, providing a novel solution for shortterm wind power forecasting.
- References:
-
[1] Sun Rongfu, Zhang Tao, He Qing, et al. Review on keytechnologies and applications in wind power forecasting[ J] . High Voltage Engineering, 2021, 47 ( 4) : 1129 -1143. [孙荣富, 张涛, 和青, 等. 风电功率预测关键技术及 应 用 综 述 [ J] . 高 电 压 技 术, 2021, 47 ( 4) :1129-1143. ]
[2] Liu Weipeng, Li Changgang, Liu Yutian, et al. Predic tive control of wind turbine for load reduction during ram ping events[ J] . International Journal of Electrical Power& Energy Systems, 2017, 93: 135-145.
[3] Han Zifen, Jing Qianming, Zhang Yankai, et al. Reviewof wind power forecasting methods and new trends [ J] .Power System Protection and Control, 2019, 47 ( 24 ) :178-187. [韩自奋, 景乾明, 张彦凯, 等. 风电预测方法与新趋势综述[ J] . 电力系统保护与控制, 2019,47(24) : 178-187. ]
[4] Jiang Jiandong, Sun Shukai, Dong Cun, et al. Researchstatus of mid-long term wind power generation forecasting[ J] . High Voltage Engineering, 2022, 48 ( 2 ) : 409 -419. [蒋建东, 孙书凯, 董存, 等. 风电中长期电量预测研究现状[ J] . 高电压技术, 2022, 48( 2) : 409-419. ]
[5] Kusiak A, Zhang Zijun, Verma A. Prediction, opera tions, and condition monitoring in wind energy[ J] . Ener gy, 2013, 60: 1-12.
[6] Tang Qingwei, Xiang Yue, Dai Jiakun, et al. Wind farmpower transfer forecasting method based on CNN-LSTM[J]. Advanced Engineering Sciences, 2024, 56(2): 91-99. [唐清苇, 向月, 代佳琨, 等. 基于 CNN-LSTM 的风电场发电功率迁移预测方法[ J] . 工程科学与技术,2024, 56(2) : 91-99. ]
[7] Liu Yeqi, Gong Chuanyang, Yang Ling, et al. DSTP RNN: a dual-stage two-phase attention-based recurrentneural network for long-term and multivariate time seriesprediction[ J] . Expert Systems with Applications, 2020,143: 113082.
[8] Wei Ao, Mao Dajun, Han Wanli, et al. Short-term loadforecasting based on EMD and long short-term memoryneural networks[ J] . Journal of Engineering for ThermalEnergy and Power, 2020, 35(4) : 203-209. [魏骜, 茅大钧, 韩万里, 等. 基于 EMD 和长短期记忆网络的短期电力负荷预测研究[ J] . 热能动力工程, 2020, 35(4) : 203-209. ]
[9] Jia Shihui, Liu Lifu, Chi Xiaoni, et al. VMD-LSTMshort-term wind speed prediction model based on BWOand WOA[ J] . Journal of Zhengzhou University ( Engi neering Science) , 2025, 46(3) : 59- 66. [ 贾世会, 刘立夫, 迟晓妮, 等. 基于 BWO 和 WOA 的 VMD-LSTM短期风速预测 [ J] . 郑 州 大 学 学 报 ( 工 学 版) , 2025,46(3) : 59-66. ]
[10] Wang Yongsheng, Guan Shijie, Liu Limin, et al. Windpower prediction method based on XGBoost extended fi nancial factor[ J] . Journal of Zhejiang University ( Engi neering Science) , 2023, 57 ( 5) : 1038 - 1049. [ 王 永生, 关世杰, 刘利民, 等. 基于 XGBoost 扩展金融因子的风 电 功 率 预 测 方 法 [ J] . 浙 江 大 学 学 报 ( 工 学版) , 2023, 57(5) : 1038-1049. ]
[11] Li Runjin, Li Lixia. Short-term wind power predictionbased on VMD-CNN-LSTM model[ J] . Journal of Sheny ang Institute of Engineering (Natural Science) , 2024, 20(1) : 6-13. [李润金, 李丽霞. 基于 VMD-CNN-LSTM模型的短 期 风 电 功 率 预 测 [ J] . 沈 阳 工 程 学 院 学 报(自然科学版) , 2024, 20(1) : 6-13. ]
[12] Zhang Zihua, Li Yan, Xu Tianqi, et al. Research onshort-term wind power forecasting based on VMD-CNN LSTM optimized by sparrow algorithm[J]. Electric Drive,2023, 53(5): 77-83. [张子华, 李琰, 徐天奇, 等. 基于麻雀算法优化的 VMD-CNN-LSTM 的短期风电功率研究[J]. 电气传动, 2023, 53(5): 77-83. ]
[13] Ren Xin, Wang Yimei, Wang Hua, et al. Two types ofmedium-long-term wind power forecasting methods basedon improved CNN-GRU and informer[ J] . Modern Electric Power, 2025, 42( 3) : 542- 549. [ 任鑫, 王一妹,王华, 等. 基于改进卷积-门控网络及 Informer 的两种中长期风电功率预测方法 [ J] . 现代电力, 2025, 42(3) : 542-549. ]
[14] Gao Lu, Kong Fanmiao, Zhang Fei, et al. Ultra short term wind power prediction method based on IPSO-BiL STM-AM model[ J] . Smart Power, 2022, 50 ( 4) : 27 -34. [高 鹭, 孔 繁 苗, 张 飞, 等. 基 于 IPSO-BiLSTM AM 模型的超短期风电功率预测方法[ J] . 智慧电力,2022, 50(4) : 27-34. ]
[15] Li Junqing, Geng Jiya, Guo Xiaoyu, et al. A method forwind turbine bearing deterioration index construction andfault diagnosis by combining POA-VMD and autoencoder[J]. Machine Tool & Hydraulics, 2024, 52( 13): 219-226. [李俊卿, 耿继亚, 国晓宇, 等. 一种 POA-VMD和自编码器结合的风电机组轴承劣化指标构建及故障诊断方法[J]. 机床与液压, 2024, 52(13): 219-226. ]
[16] Abdel-Basset M, Mohamed R, Abouhawwash M. CrestedPorcupine Optimizer: a new nature-inspired metaheuristic[ J] . Knowledge-Based Systems, 2024, 284: 111257.
[17] Yu Fei, Li Yuanxiang, Wei Bo, et al. The application ofa novel OBL based on lens imaging principle in PSO[ J].Acta Electronica Sinica, 2014, 42(2): 230-235. [喻飞,李元香, 魏波, 等. 透镜成像反学习策略在粒子群算法中的应用[J]. 电子学报, 2014, 42(2): 230-235. ]
[18] Wang Ning, He Qing. Seagull optimization algorithmcombining golden sine and sigmoid continuity[ J] . Appli cation Research of Computers, 2022, 39(1) : 157-162.
[王宁, 何庆. 融合黄金正弦与 sigmoid 连续化的海鸥优化算法[ J] . 计算机应用研究, 2022, 39( 1) : 157-162. ]
[19] Zhou Haoyi, Zhang Shanghang, Peng Jieqi, et al. In former: beyond efficient transformer for long sequencetime-series forecasting [ J ] . Proceedings of the AAAIConference on Artificial Intelligence, 2021, 35 ( 12 ) :11106-11115.
[20] Wu Zhenlong, Mo Yipeng, Wang Ronghua, et al. Multi unit wind power prediction based on long short-term mem ory and particle swarm optimization [ J ] . Journal ofZhengzhou University ( Engineering Science) , 2024, 45(6) : 114-121. [吴振龙, 莫艺鹏, 王荣花, 等. 基于LSTM 和粒子群算法的多机组风电功率预测 [ J] . 郑州大学学报(工学版) , 2024, 45(6) : 114-121. ]
[21] Zhu Juping, Wei Xia, Xie Lirong, et al. Short-term windpower peediction based on vmd and improved BiLSTM[ J] . Acta Energiae Solaris Sinica, 2024, 45( 6) : 422-428. [朱菊萍, 魏霞, 谢丽蓉, 等. 基于 VMD 和改进BiLSTM 的短 期 风 电 功 率 预 测 [ J ] . 太 阳 能 学 报,2024, 45(6) : 422-428. ]
[22] Wen Bo, Chen Fangfang, Wang Huayu. Short-term windpower prediction based on data feature extraction andSSA-BiLSTM [ J ] . Applied Science and Technology,2023, 50(4) : 71-78. [文博, 陈芳芳, 王华玉. 基于数据 特 征 提 取 和 SSA-BiLSTM 的 短 期 风 电 功 率 预 测[ J] . 应用科技, 2023, 50(4) : 71-78. ]
- Similar References:
Memo
-
Last Update:
2026-06-03
