• Energy Research
• PK close
• ET close

The electrical power infrastructure of the modern world is advanced, efficient, and robust, yet power outages still occur. In addition to affecting millions of people around the world, these outage events cost billions of dollars to the global economy. In this paper, the revenue loss borne by electricity-supplying companies in the United States due to power outage events is estimated and predicted. Various factors responsible for power outages are considered in order to present an exploratory data analysis at the U.S. level, followed by the top ten affected states, which bear over 85% of the total revenue loss. The loss is computed using historic observational data of electricity usage patterns and the tariff offered by the energy suppliers. The study is supplemented with reliable and publicly available records, including electricity usage patterns, the consumer category distribution, climatological annotations, population density, socio-economic indicators and land area. Machine learning techniques are used to predict the revenue loss for future outage events, as well as to characterize the key parameters for efficient prediction and their partial dependence. The results show that the revenue loss is a function of several parameters, including residential sales, percentage of industrial customer, time-period of the year, and economic indicators. This study may help energy suppliers make risk-informed decisions, while developing revenue generation strategies as well as identifying safer investment avenues for long-term returns.

The deployment of photovoltaic (PV) cells as a renewable energy resource has been boosted recently, which enhanced the need to develop an automatic and swift fault detection system for PV cells. Prior to isolation for repair or replacement, it is critical to judge the level of the fault that occurred in the PV cell. The aim of this research study is the fault‐level grading of PV cells employing deep neural network models. The experiment is carried out using a publically available dataset of 2,624 electroluminescence images of PV cells, which are labeled with four distinct defect probabilities defined as the defect levels. The deep architectures of the classical artificial neural networks are developed while employing hand‐crafted texture features extracted from the EL image data. Moreover, optimized architectures of the convolutional neural network are developed with a specific emphasis on lightweight models for real‐time processing. The experiments are performed for two‐way binary classification and multiclass classification. For the first binary categorization, the proposed CNN model outperformed the state‐of‐the‐art solution with a margin of 1.3% in accuracy with a significant 50% less computational complexity. In the second binary classification task, the CPU‐based proposed model outperformed the GPU‐based solution with a margin of 0.9% accuracy with an 8× lighter architecture. Finally, the multiclass categorization of PV cells is performed and the state‐of‐the‐art results with 83.5% accuracy are achieved. The proposed models offer a lightweight, efficient, and computationally cheaper CPU‐based solution for the real‐time fault‐level categorization of PV cells.