Case Studies in Thermal Engineering · 2021 · 46 citations · 36 references
ComputingComputational EngineeringComputer ScienceComputational Heat TransferHeat TransferMachine LearningDeep LearningHeat ConductionHeat Transfer AnalysisHeat Transfer ModelingRepresentation LearningThermal SimulationConvolutional Neural NetworksNumerical DiscretizationNumerical SimulationSigned Distance Field RepresentationSigned Distance FieldsSigned Distance FunctionsSpatial DiscretizationTemperature Distribution Modeling
Steady‑state heat conduction is governed by Laplace’s equation, with temperature distribution determined by local heat diffusion gradients, yielding strong locality in the solution. The study develops a data‑driven model for rapid prediction of steady‑state heat conduction in hot objects of arbitrary geometry. A convolutional neural network trained on simple geometries, using a signed distance function to encode shape, captures local temperature features and serves as a numerical discretization. After training, the model accurately predicts heat conduction for unseen complex geometries at speeds more than ten times faster than conventional simulation, demonstrating promise for engineering optimization and design.
A data-driven model for rapid prediction of the steady-state heat conduction of a hot object with arbitrary geometry is developed. Mathematically, the steady-state heat conduction can be described by the Laplace's equation, where a heat (spatial) diffusion term dominates the governing equation. As the intensity of the heat diffusion only depends on the gradient of the temperature field, the temperature distribution of the steady-state heat conduction displays strong features of locality. Therefore, a convolution neural network-based data-driven model is proposed, which is good at capturing local features (sub-invariant) thus can be treated as numerical discretization in some sense. Furthermore, a signed distance function (SDF) is proposed to represent the geometry of the problem, which contains more information than the binary image. The hot objects in the training datasets are composed of simple geometries, the geometry is different in size, shape, orientation, and location. After training, the data-driven model can accurately predict steady-state heat conduction of hot objects with complex geometries which have never been seen by the network; and the prediction speed is more than one order faster than numerical simulation. The outstanding performance of the network model indicates the potential of the approach for applications of engineering optimization and design in future.
36
Fully convolutional networks for semantic segmentation
Jonathan Long, Evan Shelhamer, Trevor Darrell · 2015 · 36.9K citations
FlowNet: Learning Optical Flow with Convolutional Networks
Alexey Dosovitskiy, Philipp Fischer, Eddy Ilg et al. · 2015 · 4.2K citations