Anomaly detection for electric energy consumption in smart farms

Abstract

Electric energy prediction has been a subject of extensive research, spanning various methodologies including traditional statistical methods, conventional machine learning techniques, deep learning (DL) methods, and hybrid DL approaches. This article introduces Electricity Talk, an Internet of Things (IoT) platform tailored for smart farms. By amalgamating artificial intelligence (AI) mechanisms with farming IoT devices, Electricity Talk facilitates electric energy prediction and anomaly detection. The primary objective of this project is to develop Electricity Talk, an innovative IoT platform for smart farms. This platform aims to integrate AI mechanisms with farming IoT devices to enhance electric energy prediction accuracy and facilitate anomaly detection. Specifically, the project seeks to leverage additional information from IoT switch statuses within smart farms and employ a novel random walk model for post-processing to improve the performance of electric energy prediction.