The efficient condition monitoring and accurate module defect detection in large-scale photovoltaic (PV) farms demand for novel inspection method and analysis tools. This paper presents a deep learning based solution for defect pattern recognition by the use of aerial images obtained from unmanned aerial vehicles (UAVs). The convolutional neural network (CNN) is used in the machine learning process to classify various forms of module defects. Such supervised learning process can extract a range of deep features of operating PV modules. It significantly improves the efficiency and accuracy of asset inspection and health assessment for large-scale PV farms in comparison with the conventional solutions. The proposed algorithmic solution is extensively evaluated from different aspects, and the numerical result clearly demonstrates its effectiveness for efficient defect detection of PV modules.

Because insulators provide electrical insulation and mechanical support for electric transmission lines, these components are of paramount importance to safe and reliable operations of power systems. However, insulators are often considered to be prone to different faults, e.g., bunch-drop, which demands a novel solution for accurate fault-detection and fault-location. Current research efforts have primarily focused on the bunch-drop fault of glass insulators, and the study of ceramic insulators has not been reported to date. To this end, this paper proposes an algorithmic solution for bunch-drop fault detection for both glass and ceramic insulators based on spatial morphological features, which can be integrated into an unmanned aerial vehicle (UAV)-based inspection system. Color models can be established based on the unique color features of both glass and ceramic insulators. Next, the target areas of the insulators can be identified according to the color determination combined with the insulator’s spatial features. The target area is morphologically processed to highlight the fault location, and the rules are established based on the spatial feature differences between the insulators with and without faults. Consequently, the fault location can be accurately identified, and the coordinates can be determined. The performance of the proposed solution is evaluated in comparison with existing solutions. The numerical results demonstrate that the proposed solution can detect bunch-drop faults of insulators with a better than average detection rate. In addition, the performance is assessed and validated in terms of robustness and real-time performance.

Critical industrial infrastructures are currently facing increasing cyberspace threats in their underlying information and communication systems. The advanced monitoring, control, and management functionalities of the industrial systems firmly rely on the reliable and secure operations of the industrial control system (ICS) network. This article characterizes the ICS network traffic and presents a scalable and efficient solution for real-time ICS network traffic anomaly detection, considering various forms of ICS anomaly events. The events due to the cyberattacks, malicious operating behaviors, and network anomalies can be effectively detected without sophisticated computational requirements and retrieval of communication protocols. The proposed hybrid statistical-machine learning model integrates a seasonal autoregressive integration moving average (SARIMA)-based dynamic threshold model and a long short-term memory (LSTM) model to jointly identify the abnormal traffic patterns with low false omission rates. The proposed solution is extensively evaluated at a realistic ICS cyber-physical system (CPS) testbed, and the numerical results confirm its high detection accuracy and low computational complexity.