Compact Thermal Wall Model for CFD Data Center Simulation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current CFD modeling for data centers is time-consuming and prone to inaccuracies due to the need for fine grids and complex wall models, especially when accounting for solar heat flux and transient thermal conditions, which can lead to significant computational overhead and failure to predict localized temperature hotspots.

Innovation Solution

A compact thermal wall model that idealizes the wall as a two-dimensional plane, capturing three-dimensional heat transfer and solar load effects, integrated with CFD models to simplify the simulation process and improve speed and robustness, allowing for the use of larger cell sizes and reducing computational complexity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If detailed CFD modeling with fine grids is used to accurately capture wall thermal effects and solar heat flux, then temperature prediction accuracy is improved, but computational time and complexity increase significantly

Engineering Contradiction:
Improvetemperature prediction accuracyVSAvoidcomputational time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The wall is segmented into multiple discrete layers (e.g., inner surface layer, center layer, outer surface layer) with associated thermal nodes. This segmentation allows the complex continuous wall structure to be modeled using discrete thermal resistance networks, reducing computational complexity while maintaining accuracy in predicting temperature distributions across the wall assembly.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

A compact thermal wall model acts as an intermediary between the CFD domain and the wall structure. This intermediary model uses simplified thermal resistance networks to represent wall heat transfer, avoiding the need for fine CFD grids within the wall while still capturing the essential thermal effects and solar heat flux interactions.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If fine grids are used within wall structures to resolve thermal gradients, then localized temperature hotspots are detected, but device complexity and computational requirements increase

Engineering Contradiction:
Improvelocalized temperature detectionVSAvoidCFD model complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The thermal wall model introduces an additional dimensional representation by mapping three-dimensional wall geometry onto a two-dimensional plane with discrete thermal nodes. This dimensional transformation allows the model to capture temperature variations and thermal gradients without requiring fine three-dimensional CFD grids within the wall volume.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

Different thermal properties and resistance values are assigned to different layers and nodes within the wall model based on local material characteristics and boundary conditions. This local quality approach enables accurate detection of localized temperature hotspots while keeping the overall model complexity manageable through selective detail placement.

Inventive Principle:
Principle #3Local quality

3Reliability

If complex multi-layer wall models are implemented to account for solar load and transient conditions, then thermal accuracy is improved, but ease of operation and setup are reduced

Engineering Contradiction:
Improvethermal model accuracyVSAvoidmodel setup ease
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The model uses parameter-based thermal resistance values that can be adjusted to represent different wall configurations, materials, and boundary conditions. This parameterization approach maintains high thermal accuracy for transient conditions and solar load while simplifying model setup, as users can modify performance by changing parameters rather than reconfiguring complex geometric models.

Inventive Principle:
Principle #35Parameter changes

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The compact thermal wall model provides accurate temperature predictions comparable to detailed benchmark solutions, simplifies the CFD modeling process, and reduces computational requirements, making it suitable for preliminary design applications and transient conditions.

Implementation Method 1

each node of the plurality of nodes is thermally coupled to an adjacent node by a thermal resistance

Methodology Applied
Scientific EffectThermal conduction: Conduction (thermal)

Implementation Method 2

receiving input data related to an enclosure, the input data including solar intensity data on an exterior of a wall of the enclosure

Methodology Applied
Scientific EffectSolar radiation: Solar Energy

Data Source

PatentEP3555775B1System and method for constructing a compact wall model
Publication Date: 2023.10.11 SCHNEIDER ELECTRIC IT CORP
  • EP3555775B1 patent drawingFigure 1A~1B
  • EP3555775B1 patent drawingFigure 2
  • EP3555775B1 patent drawingFigure 3

AI summary

Systems and methods for a compact wall model are provided. According to one aspect, embodiments herein provide a method that comprises receiving input data related to an enclosure, the input data including solar intensity data on an exterior of a wall of the enclosure, generating, by a processor, a thermal model of a wall of the enclosure based at least in part on the input data, the wall modeled as having a plurality of layers and the thermal model including a plurality of nodes such that each layer of the plurality of layers is associated with at least one node of the plurality of nodes and each node of the plurality of nodes is thermally coupled to an adjacent node by a thermal resistance, solving, by the processor, an energy balance equation for the at least one node to determine a predicted temperature for the at least one node, and output the predicted temperature to a display device.