METHOD AND SYSTEM FOR PHYSICALLY CONSCIOUS CONTROL OF AN HVAC SYSTEM
Patent Information
- Application Number
- DE602023011836
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2022-11-08
- Filing Date
- 2023-10-26
- Publication Date
- 2026-02-11
- Estimated Expiration
- 2043-10-26
AI Technical Summary
Conventional HVAC systems rely on PID control, which does not account for unmeasurable parameters like wall temperature, leading to significant energy inefficiencies and discomfort due to inaccurate thermal comfort calculations, and existing Physics Informed Neural Networks (PINNs) are challenging to implement for HVAC control due to high-dimensional input spaces and unknown exogenous inputs.
A method and system using a modified PINN that trains on time series data with constant exogenous variables, incorporating governing equations and initial conditions, to predict future system states and generate control signals for HVAC equipment, reducing the need for extensive sensory deployments.
The system achieves up to 16% energy savings and 26% reduction in thermal discomfort by accurately predicting and controlling HVAC operations, outperforming traditional methods and ensuring compliance with thermal comfort constraints.