【摘要】 城市燃气公司为防止出现储气量过剩或不足等问题,必须要对天然气负荷量进行预测。对此,提出一种基于小波变换和长短期记忆网络的深度学习模型(wavelet_LSTM),对未来一年的天然气日负荷量进行预测。在某市燃气公司真实数据集上的实验结果表明,该模型有较高的准确性,可以较好地满足城市燃气公司的业务需求。更多还原

【Abstract】 In order to prevent problems such as excess or insufficient gas reserves, city gas companies must forecast the natural gas load. This paper proposes a deep learning model based on wavelet transform and long short term memory network(wavelet_LSTM) to predict the daily natural gas load in the coming year. The experimental results on a real data set of a city gas company show that this model has higher accuracy and can better meet the business needs of city gas companies.更多还原