HVAC Energy Management via Hybrid Physics-AI Control
Solution Overview
Problem
Conventional HVAC systems lack a comprehensive approach to monitor and control both high and low side equipment as a single system, relying on predefined rules and failing to utilize real-time information for energy-efficient decision-making.
Innovation Solution
An AI-powered platform with a three-pronged control strategy, utilizing a hybrid model combining physics-based simulation and machine learning, processes data from distributed sensors to optimize energy management across HVAC equipment, including chiller, air handling units, and cooling towers, while maintaining indoor comfort.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional predefined rules and separate high/low side control strategies are used, then system complexity is reduced, but energy efficiency and real-time adaptability deteriorate
Solution Approach 1:
The patent merges the previously separate high side and low side control systems into a single integrated control platform. This integration enables unified real-time monitoring and control of all HVAC components, improving adaptability while managing complexity through centralized architecture.
Solution Approach 2:
The system transitions from static predefined rules to dynamic real-time control based on actual sensor data. The control strategies adapt continuously to changing environmental conditions, equipment status, and energy prices, enabling real-time adaptability.
2Productivity
If comprehensive sensor networks and AI processing platforms are deployed, then energy management capability is improved, but device complexity and deployment difficulty increase
Solution Approach 1:
The control platform is designed as a universal system that can manage diverse HVAC equipment types (chillers, AHUs, cooling towers, pumps) through a single interface. This multi-functionality improves energy management capability while simplifying deployment by avoiding equipment-specific control systems.
Solution Approach 2:
The system incorporates automated anomaly detection, predictive maintenance alerts, and self-optimizing control strategies that reduce the need for manual configuration and intervention. This self-service capability enhances energy management while easing deployment burden.
3Loss of energy
If real-time monitoring and control of all HVAC equipment is implemented, then energy consumption is reduced, but system complexity and computational requirements increase
Solution Approach 1:
The system implements continuous feedback loops where sensor data from all HVAC components is constantly monitored, analyzed, and used to adjust control strategies in real-time. This feedback mechanism enables energy optimization through dynamic adjustment rather than complex predictive modeling.
Solution Approach 2:
The control system optimizes energy consumption by dynamically adjusting operational parameters (temperatures, flow rates, fan speeds) based on real-time conditions. These parameter changes enable energy reduction through simple control adjustments rather than complex computational optimizations.
Data Source
AI summary
Disclosed herein is a system and a method for improving the energy management of HVAC equipment. The system comprising: a plurality of sensors distributed in a building for sensing a set of parameters including environmental information, thermal zone information, energy consumption information, operational parameter information and field information from the building; a network for connecting the plurality of sensors; a server includes a hybrid platform with physics based simulation model and machine learning model for processing and controlling the parameters of the HVAC equipment.