Scholarlink

<p>Accurate energy consumption forecasting in smart grids requires privacy-preserving learning mechanisms that remain effective under heterogeneous data distributions and support real-time operation. Existing federated learning approaches remain limited by poor performance under data heterogeneity, unvalidated architectural assumptions, and blockchain consensus mechanisms that are too slow for real-time grid operations. This paper presents PureChain, a blockchain-integrated federated learning framework that combines federated averaging, Dirichlet partitioning, LSTM-based forecasting, and a permissioned blockchain for secure client isolation and model rollback. A partitioning strategy is introduced to improve training stability under extreme non-IID conditions (<InlineEquation ID="IEq1"><EquationSource Format="TEX">\(\alpha =0.1\)</EquationSource></InlineEquation>), revealing that the distributional impact of a given Dirichlet parameter is dataset-dependent. To support low-latency smart grid applications, a permissioned blockchain employing proof-of-authority and association (PoA<InlineEquation ID="IEq2"><EquationSource Format="TEX">\(^2\)</EquationSource></InlineEquation>) consensus achieves 2.0&#xa0;s transaction latency and 20.88 TPS, outperforming Hyperledger Fabric and Quorum in the evaluated setting. Experimental results on two energy-consumption datasets show that LSTM consistently outperforms BiLSTM under high data heterogeneity, achieving an average R<InlineEquation ID="IEq3"><EquationSource Format="TEX">\(^2\)</EquationSource></InlineEquation> of 0.9184 across clients at <InlineEquation ID="IEq4"><EquationSource Format="TEX">\(\alpha =0.1\)</EquationSource></InlineEquation>. Smart contract security assessment further yields a threat score of 98.5/100, demonstrating the framework’s suitability for privacy-sensitive smart grid deployments. The contribution lies in the integration and systematic validation of established federated learning, forecasting, and blockchain technologies within a unified smart grid architecture.</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

PureChain web-based energy predictor with federated learning Dirichlet for real-time energy consumption forecasting

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月20日

查看全文
  • Adah Lubwama Nanteza,
  • Love Allen Chijioke Ahakonye,
  • Dong-Seong Kim,
  • Jae Min Lee

摘要

Accurate energy consumption forecasting in smart grids requires privacy-preserving learning mechanisms that remain effective under heterogeneous data distributions and support real-time operation. Existing federated learning approaches remain limited by poor performance under data heterogeneity, unvalidated architectural assumptions, and blockchain consensus mechanisms that are too slow for real-time grid operations. This paper presents PureChain, a blockchain-integrated federated learning framework that combines federated averaging, Dirichlet partitioning, LSTM-based forecasting, and a permissioned blockchain for secure client isolation and model rollback. A partitioning strategy is introduced to improve training stability under extreme non-IID conditions (\(\alpha =0.1\)), revealing that the distributional impact of a given Dirichlet parameter is dataset-dependent. To support low-latency smart grid applications, a permissioned blockchain employing proof-of-authority and association (PoA\(^2\)) consensus achieves 2.0 s transaction latency and 20.88 TPS, outperforming Hyperledger Fabric and Quorum in the evaluated setting. Experimental results on two energy-consumption datasets show that LSTM consistently outperforms BiLSTM under high data heterogeneity, achieving an average R\(^2\) of 0.9184 across clients at \(\alpha =0.1\). Smart contract security assessment further yields a threat score of 98.5/100, demonstrating the framework’s suitability for privacy-sensitive smart grid deployments. The contribution lies in the integration and systematic validation of established federated learning, forecasting, and blockchain technologies within a unified smart grid architecture.