Dynamic Cooling Unit Staging for Data Center Load Shifts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cooling delivery units in data centers often operate in groups based on ambient air temperature, which may not be the most energy-efficient method for maintaining optimal temperature conditions.
Innovation Solution
A thermal management system with a supervisory system that determines optimal switching thresholds for activating, deactivating, or ramping cooling delivery units based on current conditions to minimize energy consumption while maintaining temperature requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Temperature
If cooling delivery units operate in groups based on ambient air temperature, then temperature control is maintained, but energy consumption increases
Solution Approach 1:
The system dynamically adjusts the operation state of cooling delivery units based on real-time temperature conditions and energy consumption models. The supervisory system continuously optimizes which units are active, their operating capacities, and switching thresholds to minimize energy consumption while maintaining temperature requirements, rather than using fixed group operation modes.
Solution Approach 2:
The system changes operational parameters such as unit activation thresholds, operating capacities, and staging sequences based on environmental conditions and energy consumption models. By adjusting these parameters dynamically, the system achieves optimal balance between temperature control and energy efficiency.
2Reliability
If more cooling delivery units are activated to handle increasing cooling load, then temperature maintenance improves, but energy consumption increases
Solution Approach 1:
The system activates cooling delivery units partially or in sequence rather than all at once. The supervisory system determines optimal switching thresholds and operating capacities, activating units only when and to the extent needed to meet cooling requirements, thereby reducing unnecessary energy consumption while maintaining temperature reliability.
Solution Approach 2:
The system uses pre-generated energy consumption models and cooling capacity models to automatically determine optimal unit activation sequences and operating points. The supervisory system self-optimizes the staging strategy without external intervention, balancing reliability and energy consumption based on current conditions.
3Device complexity
If fixed group operation modes are used for cooling delivery units, then system complexity is reduced, but energy efficiency decreases
Solution Approach 1:
The system performs preliminary generation of cooling capacity models and energy consumption models covering various operating conditions before actual operation. These pre-computed models enable the supervisory system to quickly determine optimal unit staging strategies without complex real-time calculations, thus maintaining low system complexity while achieving high energy efficiency.
Solution Approach 2:
The supervisory system continuously monitors temperature conditions and compares actual performance against pre-generated models to dynamically adjust unit activation and operating parameters. This feedback mechanism enables adaptive optimization of energy efficiency without requiring complex real-time control algorithms.
Data Source
Figure 1A
Figure 1B~1C
Figure 2A
AI summary
A system and method for dynamic unit staging of a group of cooling delivery units mathematically models cooling capacity and energy consumption for each individual unit and for the group as a whole. Unit and group cooling capacity and energy consumption models are stored to memory accessible to a supervisory system along with a control profile providing for unit activation thresholds and allowable modification ranges for said thresholds. During each online operating cycle, the supervisory system determines the required cooling capacity (based on ambient air temperature) to maintain a data center within a required temperature range. Based on the cooling capacity requirement and applicable group energy consumption models, the supervisory system identifies a set of optimal switching thresholds within predetermined threshold modification ranges for unit activation, via which the group of cooling delivery units can maintain the required cooling capacity while minimizing overall energy consumption.