Wildfire smoke is no longer just an air-quality concern; it’s becoming a real operational factor for modern electric grids. As photovoltaic (PV) systems continue to integrate into distribution networks, feeder performance becomes increasingly dependent on weather and environmental conditions. Under clear skies, PV panels inject substantial power into the grid and reshape how voltages behave…
A Physics-Informed Neural Network (PINN) is a machine learning architecture designed to solve scientific problems by harmonizing data with physical laws. Its core innovation lies in training a neural network not only to fit observed data but also to adhere to known governing equations, typically ordinary or partial differential equations. This is achieved by incorporating…
Competitive electricity markets are inherently complex. They require the integration of technically feasible operations, grounded in physical laws, with economically viable solutions that incentivize investments. These interactions aim to ensure a reliable, efficient, and technologically flexible market design. However, the operation of these markets is neither trivial nor perfect. Over time, they are exposed to…
Photovoltaic (PV) systems are key to the transition toward clean energy, but their real-world performance is shaped by complex environmental interactions. For researchers working on the physics of solar energy systems, understanding how variables like temperature, humidity, aerosols, and wind impact panel efficiency provides a critical foundation for deeper modeling and analysis. This blog post…
This study explores the impact of weather features on short to medium electricity load prediction across diverse geographical locations. Using hourly load data, we evaluated the effectiveness of several feature selection methods, including Mutual Information (MI), Principal Component Analysis (PCA), Lasso, and Heatmap correlation. We benchmarked these feature selection methods with a hybrid deep learning model to investigate the impact of choosing…
In this paper, a new deep learning-based two-stage dataset-clustering/temporal-clustering method is proposed for time aggregation in renewable energy-integrated power systems. In this way, for the first time, the representative period, including one or more representative days, is obtained using a GAN-based model in which the LSTM Network is embedded in both generator and discriminator models.…
The U.S.-Canada Center on Climate-Resilient Western Interconnected Grid brings together leading experts in power engineering, climate, forestry, data analysis, policy and social sciences from a network of 35 partners. This new interdisciplinary center is aimed at fortifying the region’s power infrastructure against the floods, high winds, drought, even cold snaps that are also taking a…
The University of Calgary and the University of Utah will establish and co-lead the U.S.-Canada Center on Climate-Resilient Western Interconnected Grid through $5M funding by the U.S. National Science Foundation (NSF) and $3.75M funding by the Natural Sciences and Engineering Research Council of Canada (NSERC). Commonly known as “the Western Interconnection,” it is one of two…