Generation of irregular particle packing with prescribed statistical distribution, spatial arrangement, and volume fraction

Du, Linxiu; Liu, Xinrong; Han, Yafeng; Deng, Zhiyun

doi:10.1016/j.jrmge.2022.03.009

Cited by 11 publications

(9 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It is straightforward to use MDA for conducting overlap detection between convex particles. 31 4. The Gilbert-Johnson-Keerthi (GJK) proposed by Gilbert et al 36 does not need to construct Minkowski difference of two convex polygons explicitly, but check if the origin is enclosed in the Minkowski difference through iterations using support points of the polygons.…”

Section: Overlap Checking Of Two Polygonsmentioning

confidence: 99%

“…It should be noted that Du et al 31 constructed the Minkowski difference between a polygon and a line or two lines in a more complex way. They first introduced an auxiliary point associated with the line to convert the line to a convex polygon, and then constructed the corresponding Minkowski difference.…”

Section: Minkowski Difference and Its Applicationmentioning

confidence: 99%

“…The detailed introduction of the linear algorithm can be found in Han et al 34 The Minkowski difference algorithm (MDA) constructs the Minkowski difference of two convex polygons explicitly first, and then checks whether the origin is in the corresponding Minkowski difference or not. It is straightforward to use MDA for conducting overlap detection between convex particles 31 The Gilbert‐Johnson‐Keerthi (GJK) proposed by Gilbert et al 36 does not need to construct Minkowski difference of two convex polygons explicitly, but check if the origin is enclosed in the Minkowski difference through iterations using support points of the polygons.…”

Section: Advancing Front Approach: the Closed Formmentioning

confidence: 99%

“…For instance, the open form approach generated 1 million convex rectangles, pentagons, hexagons, heptagons, octagons, and mixed polygons in 3.89, 5.14, 5.59, 6.86, 7.11, and 7.42 s, respectively, using an Intel Core i5‐8300H CPU 2.3 GHz laptop. In contrast, Du et al 31 required 1 s to generate 1101 rectangles with an aspect ratio of 1 using MATLAB on an Intel Core i7‐6700k CPU with 16 GB of RAM. Remarkably, our approach generated 1 million rectangles with an aspect ratio of 1 in only 3.89 s, highlighting its superior performance.…”

Section: Examplesmentioning

confidence: 99%

“…For example, Feng et al 30 extended their closed‐form AFA to generate convex polygons and ellipses based on Minkowski Sum. Subsequently, the AFA has been modified and extended to generate convex polygons 27,31 . However, there are three shortcomings: (1) relatively large gaps can be generated near EB, which reduces the quality of particle assemblies; (2) both interior boundary (IB) and EB cannot be well considered; and (3) the one‐sided lifting problem during packing (as will be detailed in Section 4.1) using the open form is not reported.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Minkowski difference‐based advancing front packing technique for generating convex noncircular particles in complex domains

Xia

Gong

et al. 2023

Numerical Meth Engineering

View full text Add to dashboard Cite

In this work, a Minkowski difference‐based advancing front approach is proposed to generate convex and non‐circular particles in a predefined computational domain. Two specific algorithms are developed to handle the contact conformity of generated particles with the boundaries of the computational domain. The first, called the open form, is used to handle the smooth contact of generated particles with (external) boundaries, while the other, called the closed form, is proposed to handle the internal boundaries of a computational domain with a complex cavity. The Gilbert‐Johnson‐Keerthi (GJK) method is used to efficiently solve the contact detection between the newly generated particle at the front and existing particles. Furthermore, the problem of one‐sided particle lifting, which can cause some defects in the packing structure in existing advancing front methods during packing generation, is highlighted and an effective solution is developed. Several examples of increasing complexity are used to demonstrate the efficiency and applicability of the proposed packing generation approach. The numerical results show that the generated packing is not only more uniform, but also achieves a higher packing density than existing advancing front methods.

show abstract

Section: Overlap Checking Of Two Polygonsmentioning

confidence: 99%

Section: Minkowski Difference and Its Applicationmentioning

confidence: 99%