Chiller System Neural Network Control for Dynamic Energy Optimization

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing chiller systems for data centers and controlled environments face challenges in efficiently maintaining the required cooling capacity while minimizing energy consumption, especially as cooling demands fluctuate with ambient temperature changes.

Innovation Solution

A system of chiller devices with a supervisory control device that dynamically adjusts the outlet temperature and flow rate setpoints of active chiller devices, and activates or deactivates chiller devices to optimize energy consumption while maintaining the target temperature range.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the chiller system increases cooling capacity by ramping up compressors or activating additional compressors, then the required cooling load is met, but energy consumption increases

Engineering Contradiction:
Improvecooling capacityVSAvoidenergy consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system dynamically adjusts the operating parameters (flow rate and outlet temperature) of active chillers based on real-time cooling load requirements and environmental conditions. The supervisory control device continuously monitors system state and optimizes the configuration of active chillers, allowing the system to adapt its cooling capacity dynamically without necessarily activating additional compressors, thereby meeting cooling demands while minimizing energy consumption.

Inventive Principle:
Principle #15Dynamics

2Reliability

If the chiller system operates with multiple active chiller devices, then cooling capacity is increased, but system complexity and energy consumption increase

Engineering Contradiction:
Improvecooling capacityVSAvoidsystem configuration
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The supervisory control device optimizes the operating parameters (flow rate and outlet temperature setpoints) of active chiller devices to match the actual cooling load requirements. By adjusting these parameters dynamically, the system can maintain adequate cooling capacity with an optimized number of active chillers, reducing unnecessary system complexity while still meeting the required cooling demand.

Inventive Principle:
Principle #35Parameter changes

3Ease of operation

If the chiller system uses fixed setpoints for outlet temperature and flow rate, then control is simple, but energy efficiency decreases under varying cooling demands

Engineering Contradiction:
Improvecontrol simplicityVSAvoidenergy efficiency
Core Design Contradiction:
Ease of operationVSUse of energy by moving object

Solution Approach 1:

The supervisory control device implements a feedback mechanism that continuously monitors the environmental state (cooling load, ambient temperature) and the current configuration of active chillers. Based on this feedback, the control device dynamically adjusts the outlet temperature and flow rate setpoints of active chillers to optimize energy efficiency while meeting the required cooling load, eliminating the need for fixed setpoints and improving energy efficiency under varying cooling demands.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP4567532A1Neural network driven management optimization for system of chiller devices
Publication Date: 2025.06.11 VERTIV CORP
  • EP4567532A1 patent drawingFigure 1A~1B
  • EP4567532A1 patent drawingFigure 2
  • EP4567532A1 patent drawingFigure 3

AI summary

A system and method for optimizing management of a system of multiple chiller devices for circulating a chilled medium within an indoor environment determines a current state of the environment (e.g., ambient temperature, temperature of the medium entering and leaving the environment, inlet flow rate of chilled medium, target cooling load) and a current configuration of the chiller system, e.g., flow and outlet temperature setpoints of each active device and total energy consumption of the chiller system. Based on this information the chiller system controller solves (either offline or online) for an optimal chiller system configuration (and steps for achieving this configuration by adjusting chiller device flow and outlet temperature setpoints) for providing the required cooling load to the indoor environment while minimizing total energy consumption across the chiller system. If an optimal solution is found, the controller monitors its implementation.