As energy demand and electricity costs continue to rise, consumers are increasingly adopting energy-efficient practices and appliances, underscoring the need for detailed metering options like appliance-level load monitoring. Non-intrusive load monitoring (NILM) is particularly favored for its minimal hardware requirements and enhanced customer experience, especially in residential settings. However, commercial power systems present significant challenges due to greater load diversity and imbalance. To address these challenges, we introduce a novel neural network architecture that combines sequence-to-sequence, WaveNet, and ensembling techniques to identify and classify single-phase and three-phase loads using appliance power signatures in commercial power systems. Our approach, validated over four months, achieved an overall accuracy exceeding 93% for nine devices, including six single-phase and four three-phase loads. The study also highlights the importance of incorporating nonlinear loads, such as two different inverter-type air conditioners, within NILM frameworks to ensure accurate energy monitoring. Additionally, we developed a web-based NILM energy dashboard application that enables users to monitor and evaluate load performance, recognize usage patterns, and receive real-time alerts for potential faults. Our findings demonstrate the significant potential of our approach to enhance energy management and conservation efforts in commercial buildings with diverse and complex load profiles, contributing to more efficient energy use and addressing climate change challenges.

Lithium-ion batteries (LIBs) play a critical role in electric vehicles (EVs) and hybrid electric vehicles (HEVs) and degradation of LIBs influences lifetime, reliability, safety and dependability. The ability to assess and quantify degradation enables assessment of LIB’s true state of health. This paper investigates LIB degradation using a pseudo two-dimensional (P2D) model, particularly focusing on the changes to Electrochemical Impedance spectroscopy (EIS) results due to degradation. Three key degradation mechanism are considered and the impact of State-of-Charge (SoC) and temperature on EIS results are discussed. This paper also identifies the need for a more realistic approach to assess degradation. Simulations are conducted considering four repetitive standard drive cycles (viz., HTDDT, HWFET, US06 and OCTBC) for a vehicle travel distance of 150,000 km for each case. The cycle counting method is used to convert partial SoC variations during a drive cycle to an equivalent full cycle count which is then used within the degradation model to modify the parameters to represent the P2D model. This study demonstrates a robust process for analyzing degradation dynamics. The methodology presented here can guide future researchers with experimental data, enabling validation and refinement of model parameters to advance LIB degradation analysis and improve battery life predictions under operational scenarios.

The integration of Internet of Things (IoT) technologies into islanded microgrids has increased their vulnerability to cyberattacks, particularly those targeting critical components such as power converters within an islanded AC microgrid. This study investigates the impact of False Data Injection (FDI) and Denial of Service (DoS) attacks on various power converters, including DC–DC boost converters, DC–AC converters, battery inverters, and DC–DC buck–boost converters, modeled in MATLAB/Simulink. A dataset of healthy and compromised operational parameters, including voltage and current, was generated under simulated attack conditions. To enhance system resilience, a deep learning-based detection and classification framework was proposed. After evaluating various deep learning models, including Deep Neural Networks (DNNs), Artificial Neural Networks (ANNs), Support Vector Machines (SVMs), Long Short-Term Memory (LSTM), and Feedforward Neural Networks (FNNs), the final system integrates an FNN for rapid attack detection and an LSTM model for accurate classification. Real-time simulation validation demonstrated a detection accuracy of 95% and a classification accuracy of 92%, with minimal computational overhead and fast response times. These findings emphasize the importance of implementing intelligent and efficient cybersecurity measures to ensure the secure and reliable operation of islanded microgrids against evolving cyberattacks.