Digital twinDigital twin of a mouse integrating neural dynamics, biomechanics, and toucharXiv API — полнотекстовый поиск "digital twin" · yesterdayDigital twinPrototype digital twin of CEBAF injector controlled via EPICSarXiv API — полнотекстовый поиск "digital twin" · 22hDigital twinHow Much Training is Needed with a Digital Twin?arXiv API — полнотекстовый поиск "digital twin" · 2dDigital twinPUR-1 Cyber-Physical Digital TwinarXiv — Computational Engineering, Finance, and Science (cs.CE) · 2dDigital twinForesight session on the future of cyber-physical systems security in RussiaspbIT · 2dForecastingStochastic Nonlinear MPC with Gaussian Mixture Uncertainty PropagationarXiv — Systems and Control (eess.SY) · 2dMarket & productsUS Nurses Protest Against Hospital Contracts with PalantirCNews · 3dModellingDigital Sector Expands AI Practice with Agentic ScenariosCNews · 3dModellingMechanics of Democratic Dominance: System Dynamics for Consent EngineeringarXiv — Physics and Society (physics.soc-ph) · 3dDigital twinRemote-Sensing-Based Channel Modeling for 6G Digital TwinsarXiv API — полнотекстовый поиск "digital twin" · 3dForecastingDistributed MPC for Optimal Consensus of Heterogeneous Multi-Agent SystemsarXiv — Systems and Control (eess.SY) · 3dEnergyInitialization Is Key in Federated Short-Term Load ForecastingarXiv — Systems and Control (eess.SY) · 3d
Initialization Is Key in Federated Short-Term Load Forecasting
US & West ·
The study focuses on federated learning for short-term load forecasting (STLF), addressing data privacy concerns. The authors identify structured heterogeneity in clients' load data: different responses to exogenous factors and distinct temporal load profiles, which degrade forecasting performance in federated learning. To mitigate these issues, they propose two model initialization strategies — global and local — that improve forecasting accuracy.
Why it matters
Relevant for forecasting tasks in the energy sector where distributed data and privacy must be considered, and adaptation to customer heterogeneity is needed.
The article discusses an architecture of a digital twin for nuclear systems, integrating multiple models (physics-based and data-driven) to support decision making, state estimation, predictive control, and real-time data processing. The twin must synchronize with the physical facility faster than its operational cycle.
A new stochastic nonlinear model predictive control method for systems with additive noise is proposed. State distribution is approximated by Gaussian mixture with error bounds in Wasserstein distance. This yields closed-form expressions for expected costs and chance constraints, and the problem is solvable via nonlinear programming with correctness guarantees.
COMSOL is advancing integration of multiphysics simulation with digital twin technologies. This enables more accurate and comprehensive models for industrial and engineering applications.
