Skip to content
ADML — Additive Design and Manufacturing Laboratory
← All software
MLACCD
Conformal cooling

MLACCD

Machine Learning Aided Conformal Cooling Design

Built by Zhenyang Gao

Official site →GitHub

Overview

The MLACCD software is designed based on the python tkinter GUI library, and powered by the tensorflow. It contains models of MLACCD implementation, and an expandable design data base. This software provides a user-friendly interface that guide the researchers and industrial users to create their own MLACCD channels with desired conformal cooling topologies to minimize the resulting part surface temperature and increase the part quality.

The functions of the software include:

Features

03
  • 01

    Provide the mesh design of the temperature variance minimized (TVM) cooling channels with zigzag, spiral, and conformal porous structures (CPS) based on the user defined cooling surface in .stl format.

  • 02

    Provide the data of the control lines for MLACCD cooling channels for simulation and research purposes.

  • 03

    A flexible and expandable machine learning design database that can improve the speed of TVM simulations.