Data Center Cooling Model Prediction Using Real-Time Sensor Feedback

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current data center management systems lack the ability to accurately predict cooling performance and airflow rates, leading to inefficiencies in cooling systems and increased energy costs due to the inability to incorporate real-time temperature and airflow measurements into predictive models.

Innovation Solution

A computer-implemented method and system that adjusts cooling models using measured inlet and exhaust air temperature values and airflow percentages to predict future temperature values, incorporating error checking and factors to reduce differences between measured and predicted values, and includes a tool for data center design and management to optimize cooling performance.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional predictive models are used without real-time measurements, then the system is simpler to operate, but the prediction accuracy of temperature and airflow values deteriorates

Engineering Contradiction:
Improveprediction accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system incorporates real-time temperature and airflow measurements from sensors into the predictive model, creating a feedback loop where actual measurements continuously adjust and refine the predicted values. This allows the model to adapt to changing conditions and maintain high prediction accuracy without requiring complete redesign of the system architecture.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent introduces an intermediary layer that bridges the gap between simple predictive models and complex measurement systems. This intermediary processes real-time sensor data and integrates it with the predictive model, enabling accurate predictions without directly exposing the complexity of the measurement and adjustment mechanisms to the user.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If real-time measurements are incorporated into predictive models, then the prediction accuracy improves, but the computational complexity and data processing requirements increase

Engineering Contradiction:
Improveprediction accuracyVSAvoidcomputational complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system implements partial measurement by selecting only the most critical temperature and airflow parameters to measure and incorporate into the model, rather than attempting to measure and process all possible environmental variables. This selective approach maintains prediction accuracy while reducing computational burden.

Inventive Principle:
Principle #16Partial or excessive action

3Productivity

If cooling models are adjusted frequently with real-time data, then the cooling performance is optimized, but the energy consumption for data processing and model adjustment increases

Engineering Contradiction:
Improvecooling performanceVSAvoidenergy consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The system performs model adjustments at periodic intervals rather than continuously, updating the cooling model with real-time measurements at predetermined time intervals. This periodic approach maintains effective cooling performance while significantly reducing the energy consumption associated with constant data processing and model recalculation.

Inventive Principle:
Principle #19Periodic action

4Reliability

If comprehensive error checking is implemented on temperature and airflow values, then the reliability of predictions improves, but the processing time and system complexity increase

Engineering Contradiction:
Improveprediction reliabilityVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The system implements selective error checking by applying validation rules only to critical temperature and airflow parameters that have the greatest impact on prediction reliability. This partial validation approach ensures sufficient reliability without the time penalty of comprehensive checking of all possible parameters.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP2727446B1System and method for measurement aided prediction of temperature and airflow values in a data center
Publication Date: 2017.08.09 SCHNEIDER ELECTRIC IT CORP
  • EP2727446B1 patent drawingFigure 1
  • EP2727446B1 patent drawingFigure 2
  • EP2727446B1 patent drawingFigure 3

AI summary

A system and method for evaluating cooling performance of equipment in a data center is disclosed. In one aspect, a method includes receiving a plurality of measured inlet and exhaust air temperature values for at least one cooling provider and a subset of the plurality of equipment racks, implementing a cooling model. The method also includes adjusting at least one of an ambient air temperature value and each of a plurality of airflow values in the cooling model, and adjusting the cooling model to compensate for the adjusted at least one of the ambient air temperature value and each of the plurality of airflow values in the cooling model, substituting inlet and exhaust air temperature values in the cooling model with measured inlet and exhaust air temperature values, and predicting plurality of inlet and exhaust air temperature values in the cooling model.