LOW-CYCLE FATIGUE RELIABILITY ANALYSIS OF CURVIC COUPLINGS ON AN ARTIFICIAL NEURAL NETWORK SURROGATE MODEL

Hua Yuan; Rongqiao Wang; Xi Liu; Gaoxiang Chen; Dianyin Hu

doi:10.1049/icp.2025.3405

LOW-CYCLE FATIGUE RELIABILITY ANALYSIS OF CURVIC COUPLINGS ON AN ARTIFICIAL NEURAL NETWORK SURROGATE MODEL

Hua Yuan
, Rongqiao Wang
, Xi Liu^*
, Gaoxiang Chen
, Dianyin Hu

^*Corresponding author for this work

Beihang University
United Research Center of Mid-Small Aero-Engine
Beijing Key Laboratory of Aero-Engine Structure and Strength

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Due to the structural complexity of curvic couplings, conducting extensive physical experiments poses considerable challenges. Consequently, an artificial neural network (ANN) surrogate model is proposed. Using both standard and notched specimens, a fatigue life model for low-cycle loading is formulated within the framework of critical plane theory. The maximum error in the Smith-Watson-Topper (SWT) parameter within a 1 mm region is constrained to 7.52%. A parameterized model is established to decompose the typical geometry of the curvic coupling structure, followed by finite element analysis to extract the SWT damage parameter. By using the created life model, the extracted SWT values are used to forecast the fatigue life of the root. The ANN surrogate model is trained with key structural parameters as inputs and the predicted coupling life as output. Employing quasi-Monte Carlo sampling, reliability analysis is conducted. Finally, a physical simulation specimen is designed and tested to verify the consistency of the SWT parameter within the critical distance. The experimental results lie within the ±2σ range of the predicted life distribution.

Original language	English
Title of host publication	15th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, QR2MSE 2025
Publisher	Institution of Engineering and Technology
Pages	169-177
Number of pages	9
Volume	2025
Edition	35
ISBN (Electronic)	9781807050207, 9781807050344, 9781807050351, 9781807050375, 9781837242634, 9781837242900, 9781837242917, 9781837243143, 9781837243150, 9781837243167, 9781837243235, 9781837243341, 9781837243358, 9781837245277, 9781837246847, 9781837246854, 9781837247004, 9781837247011, 9781837247028, 9781837247035, 9781837247042, 9781837247059, 9781837247257, 9781837247264, 9781837247271, 9781837247295, 9781837247325, 9781837247332, 9781837249916
DOIs	https://doi.org/10.1049/icp.2025.3405
State	Published - 1 Dec 2025
Event	15th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, QR2MSE 2025 - Hohhot, China Duration: 23 Jul 2025 → 26 Jul 2025

Conference

Conference	15th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, QR2MSE 2025
Country/Territory	China
City	Hohhot
Period	23/07/25 → 26/07/25

Keywords

ARTIFICIAL NEURAL NETWORK
CURVIC COUPLINGS
QUASI-MONTE CARLO SAMPLINGS
RELIABILITY
SURROGATE SPECIMENS

Access to Document

10.1049/icp.2025.3405

Cite this

Yuan, H., Wang, R., Liu, X., Chen, G., & Hu, D. (2025). LOW-CYCLE FATIGUE RELIABILITY ANALYSIS OF CURVIC COUPLINGS ON AN ARTIFICIAL NEURAL NETWORK SURROGATE MODEL. In 15th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, QR2MSE 2025 (35 ed., Vol. 2025, pp. 169-177). Institution of Engineering and Technology. https://doi.org/10.1049/icp.2025.3405

@inproceedings{28b12046fb6649df94a353a316c0cada,

title = "LOW-CYCLE FATIGUE RELIABILITY ANALYSIS OF CURVIC COUPLINGS ON AN ARTIFICIAL NEURAL NETWORK SURROGATE MODEL",

abstract = "Due to the structural complexity of curvic couplings, conducting extensive physical experiments poses considerable challenges. Consequently, an artificial neural network (ANN) surrogate model is proposed. Using both standard and notched specimens, a fatigue life model for low-cycle loading is formulated within the framework of critical plane theory. The maximum error in the Smith-Watson-Topper (SWT) parameter within a 1 mm region is constrained to 7.52\%. A parameterized model is established to decompose the typical geometry of the curvic coupling structure, followed by finite element analysis to extract the SWT damage parameter. By using the created life model, the extracted SWT values are used to forecast the fatigue life of the root. The ANN surrogate model is trained with key structural parameters as inputs and the predicted coupling life as output. Employing quasi-Monte Carlo sampling, reliability analysis is conducted. Finally, a physical simulation specimen is designed and tested to verify the consistency of the SWT parameter within the critical distance. The experimental results lie within the ±2σ range of the predicted life distribution.",

keywords = "ARTIFICIAL NEURAL NETWORK, CURVIC COUPLINGS, QUASI-MONTE CARLO SAMPLINGS, RELIABILITY, SURROGATE SPECIMENS",

author = "Hua Yuan and Rongqiao Wang and Xi Liu and Gaoxiang Chen and Dianyin Hu",

note = "Publisher Copyright: {\textcopyright} The Institution of Engineering \& Technology 2025.; 15th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, QR2MSE 2025 ; Conference date: 23-07-2025 Through 26-07-2025",

year = "2025",

month = dec,

day = "1",

doi = "10.1049/icp.2025.3405",

language = "英语",

volume = "2025",

pages = "169--177",

booktitle = "15th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, QR2MSE 2025",

publisher = "Institution of Engineering and Technology",

address = "英国",

edition = "35",

}

Yuan, H, Wang, R, Liu, X, Chen, G & Hu, D 2025, LOW-CYCLE FATIGUE RELIABILITY ANALYSIS OF CURVIC COUPLINGS ON AN ARTIFICIAL NEURAL NETWORK SURROGATE MODEL. in 15th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, QR2MSE 2025. 35 edn, vol. 2025, Institution of Engineering and Technology, pp. 169-177, 15th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, QR2MSE 2025, Hohhot, China, 23/07/25. https://doi.org/10.1049/icp.2025.3405

LOW-CYCLE FATIGUE RELIABILITY ANALYSIS OF CURVIC COUPLINGS ON AN ARTIFICIAL NEURAL NETWORK SURROGATE MODEL. / Yuan, Hua; Wang, Rongqiao; Liu, Xi et al.
15th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, QR2MSE 2025. Vol. 2025 35. ed. Institution of Engineering and Technology, 2025. p. 169-177.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - LOW-CYCLE FATIGUE RELIABILITY ANALYSIS OF CURVIC COUPLINGS ON AN ARTIFICIAL NEURAL NETWORK SURROGATE MODEL

AU - Yuan, Hua

AU - Wang, Rongqiao

AU - Liu, Xi

AU - Chen, Gaoxiang

AU - Hu, Dianyin

N1 - Publisher Copyright: © The Institution of Engineering & Technology 2025.

PY - 2025/12/1

Y1 - 2025/12/1

N2 - Due to the structural complexity of curvic couplings, conducting extensive physical experiments poses considerable challenges. Consequently, an artificial neural network (ANN) surrogate model is proposed. Using both standard and notched specimens, a fatigue life model for low-cycle loading is formulated within the framework of critical plane theory. The maximum error in the Smith-Watson-Topper (SWT) parameter within a 1 mm region is constrained to 7.52%. A parameterized model is established to decompose the typical geometry of the curvic coupling structure, followed by finite element analysis to extract the SWT damage parameter. By using the created life model, the extracted SWT values are used to forecast the fatigue life of the root. The ANN surrogate model is trained with key structural parameters as inputs and the predicted coupling life as output. Employing quasi-Monte Carlo sampling, reliability analysis is conducted. Finally, a physical simulation specimen is designed and tested to verify the consistency of the SWT parameter within the critical distance. The experimental results lie within the ±2σ range of the predicted life distribution.

AB - Due to the structural complexity of curvic couplings, conducting extensive physical experiments poses considerable challenges. Consequently, an artificial neural network (ANN) surrogate model is proposed. Using both standard and notched specimens, a fatigue life model for low-cycle loading is formulated within the framework of critical plane theory. The maximum error in the Smith-Watson-Topper (SWT) parameter within a 1 mm region is constrained to 7.52%. A parameterized model is established to decompose the typical geometry of the curvic coupling structure, followed by finite element analysis to extract the SWT damage parameter. By using the created life model, the extracted SWT values are used to forecast the fatigue life of the root. The ANN surrogate model is trained with key structural parameters as inputs and the predicted coupling life as output. Employing quasi-Monte Carlo sampling, reliability analysis is conducted. Finally, a physical simulation specimen is designed and tested to verify the consistency of the SWT parameter within the critical distance. The experimental results lie within the ±2σ range of the predicted life distribution.

KW - ARTIFICIAL NEURAL NETWORK

KW - CURVIC COUPLINGS

KW - QUASI-MONTE CARLO SAMPLINGS

KW - RELIABILITY

KW - SURROGATE SPECIMENS

UR - https://www.scopus.com/pages/publications/105038522474

U2 - 10.1049/icp.2025.3405

DO - 10.1049/icp.2025.3405

M3 - 会议稿件

AN - SCOPUS:105038522474

VL - 2025

SP - 169

EP - 177

BT - 15th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, QR2MSE 2025

PB - Institution of Engineering and Technology

T2 - 15th International Conference on Quality, Reliability, Risk, Maintenance, and Safety Engineering, QR2MSE 2025

Y2 - 23 July 2025 through 26 July 2025

ER -

LOW-CYCLE FATIGUE RELIABILITY ANALYSIS OF CURVIC COUPLINGS ON AN ARTIFICIAL NEURAL NETWORK SURROGATE MODEL

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this