Spray Cooling Heat Dissipation Modeling for Uniform Temperature Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting temperature distribution during metal cooling using coolant sprays are inaccurate due to neglecting factors like interference between sprays and non-linear flow parameters, leading to increased defects such as cracks.
Innovation Solution
A method utilizing a machine learning model that incorporates flow velocities in both normal and tangential directions to estimate heat dissipation, allowing for precise calculation of temperature distribution by training on teacher data that includes these parameters.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If conventional cooling methods using spray nozzles are used, then cooling process is simple and easy to implement, but temperature distribution becomes uneven leading to defects such as cracks
Solution Approach 1:
The invention changes the parameters used for heat dissipation calculation from simple water amount density to include flow velocities in both normal and tangential directions. This parameter transformation enables more accurate temperature distribution prediction while maintaining the simplicity of the spray cooling process itself.
Solution Approach 2:
The invention replaces conventional mechanical calculation methods with a machine learning model that automatically processes complex flow parameters. This substitution enables accurate temperature distribution prediction without requiring complex manual calculations or process modifications.
2Device complexity
If water amount density distribution is used for temperature calculation, then calculation is simple, but heat dissipation caused by coolant flow such as interference between sprays and dripping water cannot be accounted for
Solution Approach 1:
The invention adds a new dimension to the calculation by incorporating tangential flow velocity in addition to normal flow velocity. This dimensional expansion allows the model to capture coolant flow effects such as interference between sprays and dripping water that were previously impossible to calculate.
Solution Approach 2:
The invention transforms the calculation parameters from simple water amount density to a comprehensive set including both normal and tangential flow velocities. This parameter change enables the calculation to account for complex heat dissipation phenomena while the machine learning model keeps the overall approach manageable.
3Measurement precision
If flow parameters are used to predict heat dissipation taking into account boiling form, then prediction accuracy improves, but implementation costs become high due to highly nonlinear parameters
Solution Approach 1:
The invention replaces complex mechanical calculation methods with a machine learning model. This substitution handles the highly nonlinear flow parameters automatically, achieving accurate heat dissipation predictions without the high implementation costs and complexity of traditional computational methods.
Solution Approach 2:
The invention uses a machine learning model trained on teacher data to create a simplified copy of the complex physical processes. This copied model reproduces the accurate heat dissipation predictions of complex calculations but with much lower computational cost and simpler implementation.
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
Accurately predicts temperature distribution with high precision, reducing defects in metal materials by optimizing cooling processes.
Implementation Method 1
acquiring information regarding a heat dissipation amount from the object by the coolant by inputting a plurality of flow velocities of the coolant in a normal direction with respect to the object and a plurality of flow velocities of the coolant in a tangential direction with respect to the object to a machine learning model
Implementation Method 2
a heat dissipation amount acquisition step of acquiring information regarding a heat dissipation amount from the object by the coolant
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
A method for estimating a physical quantity distribution of an object is a method for estimating a physical quantity distribution of an object when a coolant is sprayed from a spray nozzle to cool the object, the method including: a heat dissipation amount acquisition step of acquiring information regarding a heat dissipation amount from the object by the coolant by inputting a plurality of flow velocities of the coolant in a normal direction with respect to the object and a plurality of flow velocities of the coolant in a tangential direction with respect to the object to a machine learning model; and a physical quantity distribution calculation step of calculating a physical quantity distribution of at least one of a surface of the object and inside the object, using the heat dissipation amount as a boundary condition.