GRU-based power internet of things time series data anomaly detection and recovery method

The invention relates to a GRU-based power Internet of Things time series data anomaly detection and recovery method. The method comprises the following steps: step 1, preprocessing collected power Internet of Things time series data; step 2, predicting the time series data of the electric power Internet of Things preprocessed in the step 1 through a GRU model to obtain prediction data; step 3, based on the predicted value obtained in the step 2, performing anomaly detection through a threshold value set by an improved loss function; and 4, performing data recovery on the abnormal data detected in the step 3. According to the invention, the technical problem of time series data anomaly detection and recovery in the power internet of things can be solved. 本发明涉及一种基于GRU的电力物联网时序数据异常检测及恢复方法，包括以下步骤：步骤1、对采集的电力物联网的时序数据进行预处理；步骤2、将步骤1中预处理之后的电力物联网的时序数据通过GRU模型进行预测得到预测数据；步骤3、基于步骤2获得的预测值，通过改进的损失函数设立的阈值进行异常检测；步骤4、对步骤3检测出异常数据进行数据恢复。本发明能够解决电力物联网中时序数据异常检测和恢复的技术问题。