AI Building EnergyEvidence Atlas
Home / Topics
Research theme

Occupancy, HVAC Operation and Predictive Control

These papers integrate prediction with operational control. Yu et al. jointly forecast occupancy and AC behavior for occupant-centered office control, while Yang et al. forecast stochastic disturbances inside a hybrid MPC framework for high-inertia systems. Both demonstrate that predictive value is realized when forecasts are embedded in control decisions.

Research framework for Transformer-based occupancy and air-conditioning behavior prediction.
Figure 1. Research framework linking multi-sensor Transformer prediction with AC schedule optimization and energy-consumption simulation.
Figure excerpt from Yu et al. (2025), Building Simulation. DOI 10.1007/s12273-025-1356-1
Source-attributed paper figure
Model predictive control workflow with grey-box building dynamics and neural disturbance prediction.
Figure 7. Flowchart of the MPC framework integrating a TABS grey-box model with stochastic disturbance prediction.
Figure excerpt from Yang et al. (2025), Building Simulation. DOI 10.1007/s12273-025-1338-3
Source-attributed paper figure

Questions addressed

  • How can occupancy prediction reduce HVAC energy use?
  • How can neural forecasts improve model predictive control?
  • What methods account for stochastic building disturbances?
Source figures

Methods and applications represented in this theme

Research framework for Transformer-based occupancy and air-conditioning behavior prediction.
Figure 1. Research framework linking multi-sensor Transformer prediction with AC schedule optimization and energy-consumption simulation.
Figure excerpt from Yu et al. (2025), Building Simulation. DOI 10.1007/s12273-025-1356-1
Source-attributed paper figure
Model predictive control workflow with grey-box building dynamics and neural disturbance prediction.
Figure 7. Flowchart of the MPC framework integrating a TABS grey-box model with stochastic disturbance prediction.
Figure excerpt from Yang et al. (2025), Building Simulation. DOI 10.1007/s12273-025-1338-3
Source-attributed paper figure

Comparison of selected papers

PaperResearch problemMethodReported evidence
Yu et al. (2025)Occupancy and air-conditioning switching behavior jointly shape office energy use, but are often modeled separately.A Transformer-based multi-feature, multi-step model uses environmental and behavioral sensor data to predict occupancy and AC on/off behavior, followed by control-oriented energy simulation.The reported model reaches R² = 0.917 for occupancy and 0.988 accuracy for AC behavior; the proposed strategy saves up to 62% relative to conventional operation.
Yang et al. (2025)High-inertia thermally activated building systems are difficult to control when key disturbances such as occupancy and solar irradiance are stochastic or unmeasured.A hybrid MPC framework combines grey-box system modeling with neural-network disturbance forecasting and compares no-, single-, and multi-disturbance prediction strategies.Relative to conventional MPC, disturbance-aware MPC reduces temperature violations by 20%–42%, cuts cost by 6%, and improves energy flexibility by 3.1%–8.6%.

Paper pages