摘要
针对传统风速预测模型精度不足的问题,设计一种融合多源数据的风速预测模型,通过整合气象站点数据和风电场传感器数据,构建深度学习神经网络架构,在数据预处理阶段采用小波变换消除噪声影响。在河西走廊某风电场进行实际应用验证,平均绝对百分比误差(Mean Absolute Percentage Error,MAPE)降至4.2%。实验结果表明,多源数据融合方法能有效提升短期风速预测准确性,为风电场发电功率预测提供可靠依据。
Aiming at the insufficient accuracy of traditional wind speed prediction models,a multi-source data fusion wind speed prediction model is designed.The model integrates meteorological station data and wind farm sensor data to construct a deep learning neural network architecture,employing wavelet transform in the data preprocessing stage to eliminate noise interference.Through practical application validation at a wind farm in the Hexi Corridor,the Mean Absolute Percentage Error(MAPE)reduced to 4.2%.Experimental results demonstrate that the multi-source data fusion method can effectively enhance short-term wind speed prediction accuracy,providing reliable basis for wind farm power generation prediction.
作者
赵文瑄
赵立伦
ZHAO Wenxuan;ZHAO Lilun(School of Information Technology and Media,Hexi University,Zhangye,Gansu 734000,China)
出处
《智能物联技术》
2025年第4期139-143,共5页
Technology of Io T& AI
基金
2023年度河西学院校长基金青年科研项目(QN2023013)。
关键词
风速预测
数据融合
深度学习
传感器网络
气象数据
wind speed prediction
data fusion
deep learning
sensor network
meteorological data
