Reaction diffusion system prediction based on convolutional neural network

Abstract

The reaction-diffusion system is naturally used in chemistry to represent substances reacting and diffusing over the spatial domain. Its solution illustrates the underlying process of a chemical reaction and displays diverse spatial patterns of the substances. Numerical methods like finite element method (FEM) are widely used to derive the approximate solution for the reaction-diffusion system. However, these methods require long computation time and huge computation resources when the system becomes complex. In this paper, we study the physics of a two-dimensional one-component reaction-diffusion system by using machine learning. An encoder-decoder based convolutional neural network (CNN) is designed and trained to directly predict the concentration distribution, bypassing the expensive FEM calculation process. Different simulation parameters, boundary conditions, geometry configurations and time are considered as the input features of the proposed learning model. In particular, the trained CNN model manages to learn the time-dependent behaviour of the reactiondiffusion system through the input time feature. Thus, the model is capable of providing concentration prediction at certain time directly with high test accuracy (mean relative error <3.04%) and 300 times faster than the traditional FEM. Our CNN-based learning model provides a rapid and accurate tool for predicting the concentration distribution of the reaction-diffusion system.

A Cartesian Grid Representation of Left Atrial Appendages for a Deep Learning Estimation of Thrombogenic Risk Predictors

“…In addition to the aforementioned DL models based on DNN, another approach was evaluated. It consisted on a CNN-based model inspired by the work of Li et al [33], that predicted the spatial distribution of a reaction-diffusion system using an encoder-decoder based CNN. In that work, the CNN was used as a surrogate of a finite element method.…”

SimuLearn

“…Data-driven simulation methods have recently been used to accelerate simulation in various ways, such as numerical coarsening [10], subspace dynamics modeling [13], and reaction-diffusion [21]. These approaches use accurate simulators to trade precomputation effort for better runtime performance.…”

Intelligent Calibration of Static FEA Computations Based on Terrestrial Laser Scanning Reference

“…The calibration and optimization from FEA results enable the output stress analysis of human organs into fast speed, reliable, and real-time analysis. Li et al recently designed an encoder-decoder-based CNN-FEA model [32]. Inputting geometry features and FEA boundary conditions result in the prediction of time-dependent concentration distributions.…”