Yuanye Zhou , Hongqiang Wang , Borun Wu , LiGe Wang , Xizhong Chen
{"title":"Physics informed neural network model for multi-particle interaction forces","authors":"Yuanye Zhou , Hongqiang Wang , Borun Wu , LiGe Wang , Xizhong Chen","doi":"10.1016/j.partic.2024.11.002","DOIUrl":null,"url":null,"abstract":"<div><div>The discrete element method (DEM) model calculates interaction forces between each pair of particles. However, it becomes computational expensive especially when the number of particles is large. In this study, a novel artificial neural network (ANN) model is proposed to replace the model of interaction forces between multiple particles in DEM including contact force and electrostatic force. The ANN model combines the residual network (ResNet) with the physics informed neural network (PINN). The physical loss term is derived from the Newton's third law about internal forces in multi-particle system. The performance of the ANN model is evaluated based on the DEM simulation data of 100, 200, and 300-particle system in a wall-bounded 2D swirling flow. It is found that the computing time is reduced nearly an order of magnitude (7–10 times) compared with the DEM model. In addition, the accuracy of the ANN model achieves the <span><math><mrow><msup><mi>R</mi><mn>2</mn></msup><mo>></mo><mn>0.93</mn></mrow></math></span> with only <span><math><mrow><mo>≤</mo><mn>2</mn><mo>%</mo></mrow></math></span> particles are not well predicted.</div></div>","PeriodicalId":401,"journal":{"name":"Particuology","volume":"96 ","pages":"Pages 126-138"},"PeriodicalIF":4.1000,"publicationDate":"2024-11-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Particuology","FirstCategoryId":"88","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S167420012400227X","RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ENGINEERING, CHEMICAL","Score":null,"Total":0}
引用次数: 0
Abstract
The discrete element method (DEM) model calculates interaction forces between each pair of particles. However, it becomes computational expensive especially when the number of particles is large. In this study, a novel artificial neural network (ANN) model is proposed to replace the model of interaction forces between multiple particles in DEM including contact force and electrostatic force. The ANN model combines the residual network (ResNet) with the physics informed neural network (PINN). The physical loss term is derived from the Newton's third law about internal forces in multi-particle system. The performance of the ANN model is evaluated based on the DEM simulation data of 100, 200, and 300-particle system in a wall-bounded 2D swirling flow. It is found that the computing time is reduced nearly an order of magnitude (7–10 times) compared with the DEM model. In addition, the accuracy of the ANN model achieves the with only particles are not well predicted.
期刊介绍:
The word ‘particuology’ was coined to parallel the discipline for the science and technology of particles.
Particuology is an interdisciplinary journal that publishes frontier research articles and critical reviews on the discovery, formulation and engineering of particulate materials, processes and systems. It especially welcomes contributions utilising advanced theoretical, modelling and measurement methods to enable the discovery and creation of new particulate materials, and the manufacturing of functional particulate-based products, such as sensors.
Papers are handled by Thematic Editors who oversee contributions from specific subject fields. These fields are classified into: Particle Synthesis and Modification; Particle Characterization and Measurement; Granular Systems and Bulk Solids Technology; Fluidization and Particle-Fluid Systems; Aerosols; and Applications of Particle Technology.
Key topics concerning the creation and processing of particulates include:
-Modelling and simulation of particle formation, collective behaviour of particles and systems for particle production over a broad spectrum of length scales
-Mining of experimental data for particle synthesis and surface properties to facilitate the creation of new materials and processes
-Particle design and preparation including controlled response and sensing functionalities in formation, delivery systems and biological systems, etc.
-Experimental and computational methods for visualization and analysis of particulate system.
These topics are broadly relevant to the production of materials, pharmaceuticals and food, and to the conversion of energy resources to fuels and protection of the environment.