基于支持向量机的城市燃气日负荷预测方法研究

发布时间：2019-07-14 12:16

【摘要】：天然气是绿色、高效的能源,在工业生产和居民生活中应用日益广泛。天然气日负荷预测有助于科学合理地进行供应调配,预测结果对于实际工作具有重要参考价值。该文采用支持向量机方法对华北某城市的燃气实际日负荷数据进行了分析,建立了城市燃气日负荷预测模型。讨论了影响燃气日负荷变化的若干主要因素及其对燃气负荷预测建模的影响,分析了数据规则化方法对预测模型准确性的影响。该文建立的模型,对于全年负荷的预测误差小于5%;对于供暖期负荷的预测误差约为2%,结果较好。该文对建模影响因素和预测准确性的讨论,对类似问题有一定借鉴意义。
[Abstract]:Natural gas is a green and efficient energy source, which is widely used in industrial production and residents' life. The daily load forecasting of natural gas is helpful to supply and allocate scientifically and reasonably, and the prediction results have important reference value for practical work. In this paper, the actual daily load data of gas in a city in North China are analyzed by using support vector machine (SVM) method, and the daily load forecasting model of urban gas is established. Some main factors affecting the daily load change of gas and their influence on the modeling of gas load forecasting are discussed, and the influence of data regularization method on the accuracy of the prediction model is analyzed. The prediction error of the model in this paper is less than 5% for the annual load and about 2% for the heating period, and the results are good. This paper discusses the influencing factors of modeling and prediction accuracy, and has certain reference significance for similar problems.
【作者单位】：清华大学工程物理系公共安全研究院;中国标准化研究院;
【基金】：国家自然科学基金资助项目(70833003,90924001) 北京市科技专项(Z09050600910902)
【分类号】：TU996.3

【参考文献】