User electrical load prediction system and method

The invention discloses a user power consumption load prediction system and method, which solve the defects in the prior art, and comprises the following steps: step 1, collecting historical power consumption data of a user and component attributes of historical dates; step 2, selecting feature data related to power utilization load prediction from historical power utilization data based on feature importance; step 3, constructing a variational automatic encoder model, and training the variational automatic encoder model by using the feature data and component attributes of historical dates; and step 4, predicting the electrical load of the user through the variational automatic encoder model. 本发明公开了一种用户用电负荷预测系统及方法，解决了现有技术的不足，包括以下步骤：步骤1，收集用户历史用电数据以及历史日期的成分属性；步骤2，基于特征重要性在历史用电数据中选择与用电负荷预测相关的特征数据；步骤3，构建变分自动编码器模型，利用特征数据以及历史日期的成分属性对变分自动编码器模型进行训练；步骤4，通过变分自动编码器模型对于用户用电负荷进行预测。