Scholarlink

<p>The Smart Grid (SG), as the next generation of efficient and intelligent electric power transmission networks, is driven by the Internet of Things (IoT). However, Smart Meters (SMs) within the SG environment often rely on unsecured public channels to exchange data and services, posing significant security risks. This manuscript proposes Blockchain Technology for Digital Twin Security in Smart Grids using Interpretable Generalized Additive Neural Networks (BDTS-SG-IGANN) to address these challenges. The methodology begins with data collection from the N-BaIoT dataset, followed by pre-processing using the Constrained Normalized Subband Adaptive Filter (CNSAF) to enhance data quality and normalization. The pre-processed data is then secured through blockchain technology utilizing the Proof-of-Monitoring (PoM) consensus mechanism for authenticating data sources. Subsequently, secured data is analyzed using an Interpretable Generalized Additive Neural Network (IGANN) for intrusion detection. To optimize IGANN weight parameters, the Leaf-in-Wind Optimization (LWO) algorithm is employed. The proposed BDTS-SG-IGANN technique is put into practice using Python and evaluated using performance metrics like Precision, Recall, Accuracy, F1 Score, Mean Square Error (MSE), and Specificity. Experimental results demonstrate that BDTS-SG-IGANN outperforms existing methods, achieving 29.36%, 26.42%, and 23.27% higher accuracy and 18.36%, 14.42% and 12.27% lower compared to digital twin-based reinforcement learning for extracting network structures and load patterns in planning and operation of distribution systems (DTRL-ENS-CNN), Digital twin-driven SDN for smart grid: A deep learning integrated blockchain for cyber security (DTD-IBCS-SGN) and Blockchain-Based Cyber-Physical Security for Electrical Vehicle Aided Smart Grid Ecosystem (BCPS-EVSGE-SDN). This study establishes BDTS-SG-IGANN as an innovative and effective framework for securing digital twins in SG environments.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Blockchain technology for digital twin security in smart grids using interpretable generalized additive neural networks

  • 期刊论文

  • 发表日期：

    2025年5月9日

查看全文
  • Ravi Kumar Suggala,
  • Jaydip Kumar,
  • Prateek Jain,
  • Boddepalli Kiran Kumar

摘要

The Smart Grid (SG), as the next generation of efficient and intelligent electric power transmission networks, is driven by the Internet of Things (IoT). However, Smart Meters (SMs) within the SG environment often rely on unsecured public channels to exchange data and services, posing significant security risks. This manuscript proposes Blockchain Technology for Digital Twin Security in Smart Grids using Interpretable Generalized Additive Neural Networks (BDTS-SG-IGANN) to address these challenges. The methodology begins with data collection from the N-BaIoT dataset, followed by pre-processing using the Constrained Normalized Subband Adaptive Filter (CNSAF) to enhance data quality and normalization. The pre-processed data is then secured through blockchain technology utilizing the Proof-of-Monitoring (PoM) consensus mechanism for authenticating data sources. Subsequently, secured data is analyzed using an Interpretable Generalized Additive Neural Network (IGANN) for intrusion detection. To optimize IGANN weight parameters, the Leaf-in-Wind Optimization (LWO) algorithm is employed. The proposed BDTS-SG-IGANN technique is put into practice using Python and evaluated using performance metrics like Precision, Recall, Accuracy, F1 Score, Mean Square Error (MSE), and Specificity. Experimental results demonstrate that BDTS-SG-IGANN outperforms existing methods, achieving 29.36%, 26.42%, and 23.27% higher accuracy and 18.36%, 14.42% and 12.27% lower compared to digital twin-based reinforcement learning for extracting network structures and load patterns in planning and operation of distribution systems (DTRL-ENS-CNN), Digital twin-driven SDN for smart grid: A deep learning integrated blockchain for cyber security (DTD-IBCS-SGN) and Blockchain-Based Cyber-Physical Security for Electrical Vehicle Aided Smart Grid Ecosystem (BCPS-EVSGE-SDN). This study establishes BDTS-SG-IGANN as an innovative and effective framework for securing digital twins in SG environments.