A Method for Anomaly Prediction in Power Consumption using Long Short-Term Memory and Negative Selection

To identify and predict anomalous power consumption, this paper proposes a method based on Long Short-Term Memory (LSTM) and Negative Selection technologies that anticipates the occurrence of anomalies in power consumption, and to provide useful information for energy efficiency. Using the proposed method it is possible to anticipate the occurrence of anomalies in power consumption. When applied to the power consumption recorded during 20 weeks of a building the method yielded promising results. Finally, the effectiveness and advantages of this method is demonstrated which it could be directly used for real-time electricity monitoring and anomaly prediction.