Blog
Step Towards Secure and Reliable Smart Grids in Industry 5.0: A Federated Learning assisted Hybrid Deep Learning Model for Electricity Theft Detection using Smart Meters
Step Towards Secure and Reliable Smart Grids in Industry 5.0: A Federated Learning assisted Hybrid Deep Learning Model for Electricity Theft Detection using Smart Meters.
I am very excited to share our newly published article authored by: Muhammad Hamza Zafar, Syed Muhammad Salman Bukhari, Mohamad Abou Houran, Syed Kumayl Raza Moosavi, Majad Mansoor, Nedaa Al-Tawalbeh, and Filippo Sanfilippo
Abstract
The integration of Smart Grid technology and conceptual Industry 5.0 has paved the way for advanced energy management systems that enhance efficiency and revolutionized the parallel integration of power sources in a sustainable manner. However, this digitization has opened a new stream of the threat and opportunities of electricity theft posing a significant challenge to the security and reliability of Smart Grid networks. In this paper, we propose a secure and reliable theft detection technique using deep federated learning (FL) mechanism. The technique leverages the collaborative power of FL to train a Convolutional Gated Recurrent Unit (ConvGRU) model on distributed data sources without compromising data privacy. The training deep learning model backbone consists of a ConvGRU model that combines convolutional and gated recurrent units to capture spatial and temporal patterns in electricity consumption data. An improvised preprocessing mechanism and hyperparameter tuning is done to facilitate FL mechanism. The halving randomized search algorithm is used for hyperparameters tuning of the ConvGRU model. The impact of hyperparameters involved in the ConvGRU model such as number of layers, filters, kernel size, activation function, pooling, GRU layers, hidden state dimension, learning rate, and the dropout rate is elaborated. The proposed technique achieves promising results, with high accuracy, precision, recall, and F1 score, demonstrating its efficacy in detecting electricity theft in Smart Grid networks. Comparative analysis with existing techniques reveal the superior performance of the deep FL-based ConvGRU model. The findings highlight the potential of this approach in enhancing the security and efficiency of Smart Grid systems while preserving data privacy.
A comprehensive framework for hand gesture recognition using hybrid-metaheuristic algorithms and deep learning models
I'm happy to share our recent paper:
Hassan Mohyuddin, Syed Kumayl Raza Moosavi, Muhammad Hamza Zafar and Filippo Sanfilippo. A comprehensive framework for hand gesture recognition using hybrid-metaheuristic algorithms and deep learning models. Array, 2023.
???? Access the paper here: Read It Now
???? Key Insights:
- We harnessed data from a Leap Motion Controller (LMC) and a unique Spotted Hyena-based Chimp Optimization Algorithm (SSC) to select crucial features and train deep neural networks (DNNs).
- In rigorous comparisons, our SSC algorithm consistently demonstrated remarkable performance, achieving a noteworthy accuracy rate of 98% in gesture recognition, surpassing conventional methods.
- Our method holds the potential to significantly enhance accuracy in various domains that rely on advanced feature extraction techniques.
I extend my gratitude to my fellow authors: Hassan Mohyuddin, Syed Kumayl Raza Moosavi, and Muhammad Hamza Zafar for their invaluable contributions.
Nomination for the prestigious "Machines 2022 Best Paper Award"
Big News! I am thrilled to share some fantastic news with all of you! Our research article, "Teaching Motion Control in Mechatronics Education Using an Open Framework Based on the Elevator Model," authored by Filippo Sanfilippo, Martin Økter, Tine Eie, and Morten Ottestad, and published in Machines 10(10) in 2022, has been nominated for the prestigious "Machines 2022 Best Paper Award"
Filippo Sanfilippo, Martin Økter, Tine Eie and Morten Ottestad. Teaching Motion Control in Mechatronics Education Using an Open Framework Based on the Elevator Model. Machines 10(10), 2022. URL, DOI
In this paper, we explored the exciting field of Mechatronics Education, specifically focusing on motion control. The Elevator Model served as an excellent platform for students and educators to understand complex mechatronics concepts through practical applications. The openness of the framework encourages collaboration and exploration, enabling a deeper comprehension of motion control and its real-world implications.
This nomination for the award is even more rewarding as two of my talented students, Martin Økter and Tine Eie, were involved in this work.
I finally want to acknowledge the wider community's support and encouragement for this work.
Filippo Sanfilippo