Editorial: Advanced Anomaly Detection Technologies and Applications in Energy Systems
Tinghui Ouyang, Xun Shen, Yusen He, Zhenhao Tang, Yahui Zhang · Frontiers in Energy Research · 2022
Advanced Anomaly Detection Technologies and Applications in Energy SystemsAnomaly detection is an important topic that has been well-studied in diverse research areas and application domains.It generally involves the detection of abnormal data, unhealthy statuses, and fault diagnosis, and is helpful to guarantee industrial systems' stability, security, and economy.With the development of intelligent industries and sensor systems, large amounts of data become easily available, but there are major challenges to industrial systems' anomaly detection.One typical case is the study on energy-related systems, like thermal energy, renewable energy (e.g., wind energy, photovoltaic), electric vehicles, and so on.These systems involve various data formats and more complex data structures making anomaly data detection a challenge.Currently, under the development of deep learning and big data analytics, many promising results have been achieved in energy systems' anomaly data detection.However, many challenging problems remain unsolved due to the complex nature of energy industries.New techniques and advanced engineering applications of anomaly detection in energy systems still appeal to a wide range of scholars and industries.The objective of this Research Topic is to solicit papers on recent developments in anomaly detection techniques and advances in applications of energy-related systems.The topic can cover techniques related to anomaly detection algorithm development, such as machine learning, data mining, deep learning, graph theory, big data, and so on.Various aspects of energy applications can be addressed, like data cleaning, unhealthy evaluation of energy systems, condition monitoring, and faults diagnosis in energy-related industries.Special attention could be paid to energy-related systems, e.g., wind energy, photovoltaic, thermal energy, electric vehicle (EV) development, and so on.After paper Research Topic and rigorous review, 63 high-quality articles contributed by 327 authors were finally accepted for their contributions to the study of condition monitoring and anomaly detection in power systems, renewable energy systems, and other industrial systems.In the paper Series Arc Fault Diagnosis Based on Variational Mode Decomposition and Random Forest, Zhao et al. proposed a method based on variational mode decomposition and energy entropy to extract the characteristic quantity of series arc faults, and subsequently complete the fault detection.In the paper Sequential Detection of Microgrid Bad Data via a Data-Driven Approach Combining Online Machine Learning with Statistical Analysis, Huang et al. proposed a sequential detection method to detect bad data in Energy Management Systems (EMS).