Journal

Topics: Computer Security . Side-channel Data Analysis . Microarchitectural Attack . ML/DL-based Anomaly Detector

1. Debopriya Roy Dipta, Berk Gulmezoglu, “MAD-EN: Microarchitectural Attack Detection through System- wide Energy Consumption”, IEEE Transactions on Information Forensics and Security, May 2023. Available at: https://doi.org/10.1109/TIFS.2023.3272748

Research key words: anomaly detection, micro-architectural attacks, convolutional neural networks, energy consumption

In this study, we introduce MAD-EN dynamic detection tool that leverages system-wide energy consumption traces collected from a generic Intel RAPL tool to detect ongoing anomalies in a system.

Our purpose is to distinguish benign applications and microarchitectural attacks based on their fingerprint on CPU power consumption. For this purpose, we designed MAD-EN with offline and online phases. In the offline phase, a diverse set of microarchitectural attacks and benign applications is run on the test setup while the system-wide power consumption traces are collected. Next, the collected traces are utilized to train an Anomaly Detector (AD) model to be used in the online phase to detect ongoing attacks. Furthermore, an additional DL model, namely Attack Recognizer (AR), is created with solely microarchitectural attacks to classify the suspicious activity.

For details, check the following GitHub repository:

https://github.com/Diptakuet/MAD-EN-Microarchitectural-Attack-Detection

Code Ocean: https://codeocean.com/capsule/9860734/tree

Proposed Methodology

2. A. A. Mamun, M. Sohel, N. Mohammad, M. S. Haque Sunny, D. R. Dipta and E. Hossain, "A Comprehensive Review of the Load Forecasting Techniques Using Single and Hybrid Predictive Models," in IEEE Access, vol. 8, pp. 134911-134939, 2020, doi: 10.1109/ACCESS.2020.3010702.

Available at : https://ieeexplore.ieee.org/document/9144528

Research key words:

Load forecasting, predictive models, machine learning, support vector machines, artificial neural networks, computational Intelligence, power industry, smart grid

Load forecasting is a pivotal part of the power utility companies. To provide load-shedding free and uninterrupted power to the consumer, decision-makers in the utility sector must forecast the future demand for electricity with a minimum amount of error percentage. Load prediction with less percentage of error can save millions of dollars to the utility companies. There are numerous Machine Learning (ML) techniques to amicably forecast the demand of electricity among which the hybrid models show the best result. Two or more than two predictive models are amalgamated to design a hybrid model, each of which provides improved performances by the merit of individual algorithms. This paper reviews the current stateof-the-art of electric load forecasting technologies and presents recent works pertaining to the combination of different ML algorithms into two or more methods for the construction of hybrid models. A comprehensive study of each single and multiple load forecasting model is performed with an in-depth analysis of their advantages, disadvantages, and functions. A comparison between their performance in terms of Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE) values are developed with pertinent literature of several models to aid the researchers with the selection of suitable models for load prediction.

Information flow presented in the paper, with the explanation of single and hybrid models for

load forecasting

Pictorial representation of notable hybrid models with two methods based on Support Vector Machine that are explained in the subsection

Pictorial representation of notable hybrid models with two methods based on Artificial Neural Network that are explained in the subsection

Topics: Application of ML/DL . Signal Processing . Renewable Energy

3. E. Hossain, S. Roy, N. Mohammad, N. Nawar, and D. R. Dipta, “Metrics and enhancement strategies for grid resilience and reliability during natural disasters,” Applied Energy, vol. 290, p. 116709, 2021. Available at: https://www.sciencedirect.com/science/article/abs/pii/S0306261921002294

Research key words:

Grid resilience, Grid reliability, Resilience and reliability metrics, Resilience enhancement strategy, Reliability enhancement strategy, Natural disasters

The rise in power shutdowns triggered by severe weather due to deteriorating climate change has expedited the research in enhancing community resilience. Several researchers and policy-makers have contributed to the characterization and parameterization of energy resilience and reliability in particular, which requires accumulated and coordinated studies to underline the outcomes and reflect those in future works on grid resilience and reliability enhancement. The concept of both the resilience and reliability of the grid systems should be defined and distinguished so that the systems can be clearly comprehended, assessed, and operated to maintain flawless operation and ensure environmental sustainability. This paper meets the mentioned objectives to discuss grid resilience and reliability, their quantification metrics, and their enhancement techniques in detail. The paper also categorizes the United States into four tiers based on grid reliability and grid resilience using Monte Carlo Simulations and the discussed metrics. Two novel terminologies named resilience risk factor and grid infrastructure density are propounded in this work, which will serve as vital parameters to determine grid resilience.

Highlights:

Impact of natural disasters on the electricity grid is studied.
Grid resilience and reliability metrics and enhancement strategies are described.
Grid resilience and reliability of the United States are assessed.
The United States map is categorized based on grid resilience and reliability.
Resilience risk factor and grid infrastructure density are newly coined terms.

Relationship between resilience and reliability. Resilience focuses on high impact events and reliability deals with low impact events.

Resilience trapezoid with different stages during a natural disaster. The green box marked area represents the resilience trapezoid, the smaller the area gets; the more resilient grid will be.

Assessing the United States grid reliability

Impact of smart grid and microgrid in overall fault control exercises. Blue boxes represent that time is the same, red box represents additional time, yellow boxes represent longer time duration, and the green boxes represent shorter required time duration.