Physics-Informed Learning with Bayesian Parameter Inversion for Robust Residential Energy and Environmental Optimization
Abstract
Accurate coordination of building energy consumption and indoor environmental quality requires models that can represent coupled physical processes, time-varying operating parameters, and uncertainty in performance optimization. Here, a physics-informed learning framework is developed to integrate differentiable multi-physics modeling, Bayesian parameter inversion, uncertainty propagation, and robust multi-objective optimization for residential energy and environmental performance. A physics-informed neural network couples temperature, humidity, and carbon dioxide dynamics while jointly performing forward prediction and inverse identification. Time-varying infiltration, internal heat sources, and occupant-related source terms are subsequently identified using a hybrid unscented Kalman and particle filtering scheme. Posterior parameter and occupant-preference distributions are propagated into an expectation–conditional-value-at-risk robust optimization framework to coordinate energy consumption, thermal comfort, and indoor air quality. The framework is validated using 122 days of multimodal measurements from a residential building and blind tests across four climate zones. Relative to a long short-term memory baseline, prediction accuracy for temperature, humidity, carbon dioxide, and energy consumption improves by 15.1–19.5%, while time-varying parameter identification improves by 13.8–26.3 percentage points compared with static calibration. Robust optimization increases the…
Read the paper
Related studies
- Preliminary Toxicity Analysis of 3DCRT versus IMRT on the High Dose Arm of the RTOG 0126 Prostate Cancer Trial
- Adaptive Radiotherapy for Head and Neck Cancer: Initial Clinical Outcomes from a Prospective Trial
- American Association of Physicists in Medicine Task Group 263: Standardizing Nomenclatures in Radiation Oncology
- Motion Management Strategies and Technical Issues Associated with Stereotactic Body Radiotherapy of Thoracic and Upper Abdominal Tumors: A Review from NRG Oncology
- High-dose-rate interstitial brachytherapy as a monotherapy for localized prostate cancer: treatment description and preliminary results of a phase I/II clinical trial
- Highly efficient training, refinement, and validation of a knowledge-based plan quality control system for radiotherapy clinical trials
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.