Calculation of gamma-ray exposure buildup factor based on backpropagation neural network

Runkai Chen; Antonio Cammi; Marcus Seidl; Rafael Macian-Juan; Xiang Wang

doi:10.1016/j.eswa.2021.115004

Calculation of gamma-ray exposure buildup factor based on backpropagation neural network

Runkai Chen, Antonio Cammi, Marcus Seidl, Rafael Macian-Juan, Xiang Wang

Chair of Nuclear Technology

Research output: Contribution to journal › Article › peer-review

8 Scopus citations

Abstract

In this paper, a method based on a backpropagation neural network (BPNN) is proposed to calculate the exposure buildup factor (B_D) of a point isotropic source in an infinite homogeneous medium under arbitrary energy and mean free path (mfp). The results obtained for aluminum, iron, lead, and concrete based on BPNN are compared to ANSI/ANS-6.4.3 standard data, the results calculated by MCNP 5 Monte Carlo code, and a geometric progression (G-P) fitting formula, and show that the B_D calculated by the BPNN model is more consistent with the ANS standard data. This method improves the calculation and fitting effect of B_D compared to other methods. This paper proposes a systematic process combining a Monte Carlo method and BPNN to calculate and predict the B_D of new materials under different energy and mfp, thus replacing the G-P fitting formula and improving calculation accuracy.

Original language	English
Article number	115004
Journal	Expert Systems with Applications
Volume	177
DOIs	https://doi.org/10.1016/j.eswa.2021.115004
State	Published - 1 Sep 2021

Keywords

Backpropagation neural network
Buildup factor
Gamma-ray
Monte Carlo method

Access to Document

10.1016/j.eswa.2021.115004

Cite this

@article{13249cfbbff1418ebedd41628c40cd63,

title = "Calculation of gamma-ray exposure buildup factor based on backpropagation neural network",

abstract = "In this paper, a method based on a backpropagation neural network (BPNN) is proposed to calculate the exposure buildup factor (BD) of a point isotropic source in an infinite homogeneous medium under arbitrary energy and mean free path (mfp). The results obtained for aluminum, iron, lead, and concrete based on BPNN are compared to ANSI/ANS-6.4.3 standard data, the results calculated by MCNP 5 Monte Carlo code, and a geometric progression (G-P) fitting formula, and show that the BD calculated by the BPNN model is more consistent with the ANS standard data. This method improves the calculation and fitting effect of BD compared to other methods. This paper proposes a systematic process combining a Monte Carlo method and BPNN to calculate and predict the BD of new materials under different energy and mfp, thus replacing the G-P fitting formula and improving calculation accuracy.",

keywords = "Backpropagation neural network, Buildup factor, Gamma-ray, Monte Carlo method",

author = "Runkai Chen and Antonio Cammi and Marcus Seidl and Rafael Macian-Juan and Xiang Wang",

note = "Publisher Copyright: {\textcopyright} 2021 Elsevier Ltd",

year = "2021",

month = sep,

day = "1",

doi = "10.1016/j.eswa.2021.115004",

language = "English",

volume = "177",

journal = "Expert Systems with Applications",

issn = "0957-4174",

publisher = "Elsevier Ltd.",

}

TY - JOUR

T1 - Calculation of gamma-ray exposure buildup factor based on backpropagation neural network

AU - Chen, Runkai

AU - Cammi, Antonio

AU - Seidl, Marcus

AU - Macian-Juan, Rafael

AU - Wang, Xiang

PY - 2021/9/1

Y1 - 2021/9/1

N2 - In this paper, a method based on a backpropagation neural network (BPNN) is proposed to calculate the exposure buildup factor (BD) of a point isotropic source in an infinite homogeneous medium under arbitrary energy and mean free path (mfp). The results obtained for aluminum, iron, lead, and concrete based on BPNN are compared to ANSI/ANS-6.4.3 standard data, the results calculated by MCNP 5 Monte Carlo code, and a geometric progression (G-P) fitting formula, and show that the BD calculated by the BPNN model is more consistent with the ANS standard data. This method improves the calculation and fitting effect of BD compared to other methods. This paper proposes a systematic process combining a Monte Carlo method and BPNN to calculate and predict the BD of new materials under different energy and mfp, thus replacing the G-P fitting formula and improving calculation accuracy.

AB - In this paper, a method based on a backpropagation neural network (BPNN) is proposed to calculate the exposure buildup factor (BD) of a point isotropic source in an infinite homogeneous medium under arbitrary energy and mean free path (mfp). The results obtained for aluminum, iron, lead, and concrete based on BPNN are compared to ANSI/ANS-6.4.3 standard data, the results calculated by MCNP 5 Monte Carlo code, and a geometric progression (G-P) fitting formula, and show that the BD calculated by the BPNN model is more consistent with the ANS standard data. This method improves the calculation and fitting effect of BD compared to other methods. This paper proposes a systematic process combining a Monte Carlo method and BPNN to calculate and predict the BD of new materials under different energy and mfp, thus replacing the G-P fitting formula and improving calculation accuracy.

KW - Backpropagation neural network

KW - Buildup factor

KW - Gamma-ray

KW - Monte Carlo method

UR - http://www.scopus.com/inward/record.url?scp=85104674824&partnerID=8YFLogxK

U2 - 10.1016/j.eswa.2021.115004

DO - 10.1016/j.eswa.2021.115004

M3 - Article

AN - SCOPUS:85104674824

SN - 0957-4174

VL - 177

JO - Expert Systems with Applications

JF - Expert Systems with Applications

M1 - 115004

ER -

Calculation of gamma-ray exposure buildup factor based on backpropagation neural network

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this