基于深度学习的综合能源配电系统负荷分析预测

Load Analysis and Prediction of Integrated Energy Distribution System Based on Deep Learning

为支撑综合能源配电系统的经济调度和优化运行,提出了一种基于深度学习的冷热电多元负荷综合预测方法。首先,使用皮尔逊系数定量计算多元负荷间的相关关系,分析负荷与影响因素间相关性;然后,构建基于卷积神经网络和支持向量回归的深度学习模型,其中卷积神经网络作为特征提取器从输入数据中提取隐含的更具代表性的特征信息,支持向量回归作为预测模型输出预测结果,同时开展缺失数据与离群数据的预处理;最后,应用某综合能源系统的实际数据对算法的有效性进行了验证,比较分析了考虑多元负荷相关性对预测结果的影响。结果表明:所提出的RCNN-SVR模型对冷、热、电负荷均有较好的预测精度。研究成果可为综合能源配电系统的综合负荷预测提供参考。

In order to support the economic dispatch and optimal operation of an integrated energy distribution system, a comprehensive forecasting method of cooling, heating, and power loads based on deep learning is proposed. Firstly, the Pearson coefficient is used to quantitatively calculate the correlation between multiple loads and analyze the correlation between loads and influencing factors. Then, the structure of deep learning model based on convolutional neural network and support vector regression is introduced. The convolutional neural network is used as a feature extractor to extract more representative hidden feature information from input data, and the support vector regression is used as a prediction model to output prediction results. Meanwhile, missing data and outlier data are preprocessed. Finally, the actual data of an integrated energy system are used to verify the effectiveness of the algorithm, and the influences of multi-load correlation on the prediction results are compared and analyzed. The results show that the RCNN-SVR model proposed in this paper has good prediction accuracy for cooling, heating, and power loads. The research results can provide references for comprehensive load forecasting of integrated energy distribution system.