![airfoil generator using machine learning airfoil generator using machine learning](https://d3i71xaburhd42.cloudfront.net/c6ef6396991a4261e9e3ebc5cab07b0331f1a3ad/13-Figure4-1.png)
In this study, printed circuit heat exchanger (PCHE), a potential IHX concept for high-temperature applications, has been investigated for their heat transfer and pressure drop characteristics under high operating temperatures and pressures. Currently, there is no proven high-temperature (750–800 ☌ or higher) compact heat exchanger technology for high-temperature reactor design concepts. In high-temperature gas-cooled reactors, such as a very high temperature reactor (VHTR), an intermediate heat exchanger (IHX) is required to efficiently transfer the core thermal output to a secondary fluid for electricity generation with an indirect power cycle and/or process heat applications.
Airfoil generator using machine learning code#
Testing and thermomechanical modeling is needed to facilitate future code compliance of PCHEs for high pressure and high temperature applications. A thermal stress analysis was conducted and a brief discussion of the status of code cases of PCHEs for nuclear applications is given. Correlations for the Nusselt number and Darcy friction factor were developed that can be useful for thermal hydraulic analyses using system codes. This work has developed a digital framework for the expedient topology design and evaluation of PCHE designs for gas-cooled microreactor applications. A set of optimal designs that maximizes heat transfer and minimizes pressure drop was identified, and a thermal stress analysis was performed on the optimal design.
![airfoil generator using machine learning airfoil generator using machine learning](https://i1.rgstatic.net/publication/321264886_Application_of_Online-Trained_ANN_and_Parallel_Computing_in_2D_Airfoil_Optimization/links/5a1783494585155c26a78975/largepreview.png)
more » The optimal geometry was found across six channel Reynolds numbers ranging from 1000 to 5000 to analyze how varying inlet conditions affects the optimal design. The genetic algorithm identified a set of optimal points on the Pareto front. A response surface approximation was created from the surrogate models and input to a genetic algorithm. Two methods were compared for generating surrogate models – a 5D polynomial and a regression neural network. STAR-CCM+ was used to analyze a simplified two-channel configuration where five parameters were varied – inlet angle, fin scale, extent of staggering, transverse and longitudinal pitches. An optimization procedure was developed that employs computational fluid dynamics for a set of design points identified using Latin hypercube sampling. A framework for topology optimization of airfoil fin PCHEs has been developed that can be readily extended to different fin sizes and shapes, as well as different inlet and operating conditions, materials of construction, and working fluids.
![airfoil generator using machine learning airfoil generator using machine learning](https://www.mdpi.com/symmetry/symmetry-12-00544/article_deploy/html/images/symmetry-12-00544-g001-550.jpg)
The use of airfoil fins offers the potential to reduce pressure drop across the heat exchanger, as compared to other types of channel configurations. An airfoil fin microchannel design, constructed of Alloy 617 with helium as the working fluid, was analyzed and optimized using a design of experiments with artificial intelligence and machine learning techniques. High-performance microchannel heat exchangers are needed to supply heat for power conversion for nuclear microreactors.