Neural network integrated with symbolic regression for multiaxial fatigue life prediction

International Journal of Fatigue, Journal Year: 2024, Volume and Issue: 188, P. 108535 - 108535

Published: July 29, 2024

Language: Английский

---

Heteroscedastic Gaussian Process Regression for material structure–property relationship modeling

Computer Methods in Applied Mechanics and Engineering, Journal Year: 2024, Volume and Issue: 431, P. 117326 - 117326

Published: Aug. 26, 2024

Language: Английский

---

An interpretable and reliable framework for alloy discovery in thermomechanical processing

Materials Today Communications, Journal Year: 2025, Volume and Issue: unknown, P. 112134 - 112134

Published: March 1, 2025

Language: Английский

---

A TCN-based feature fusion framework for multiaxial fatigue life prediction: Bridging loading dynamics and material characteristics

International Journal of Fatigue, Journal Year: 2025, Volume and Issue: unknown, P. 108915 - 108915

Published: March 1, 2025

Language: Английский

---

Study of Hybrid Machine Learning Multiaxial Low‐Cycle Fatigue Life Prediction Model of CP‐Ti

Fatigue & Fracture of Engineering Materials & Structures, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 24, 2025

ABSTRACT Symmetric and asymmetric multiaxial low‐cycle fatigue tests were conducted on commercially pure titanium under different control modes strain/stress ratios to establish a reliable hybrid physics data‐driven method. Optimized analysis formula–based models are proposed provide physical information first. Based the dataset enhanced by nonlinear variational autoencoder method, VAE‐ANN model is established trained, developed using Pearson correlation coefficient Leaky ReLU activation function. Through series of life prediction validations both symmetric loading conditions, demonstrates excellent accuracy, broad generalization capability, strong compatibility, achieving lowest average absolute relative error 6.76% 22.61% conditions.

Language: Английский

---

A general physics-informed neural network framework for fatigue life prediction of metallic materials

Engineering Fracture Mechanics, Journal Year: 2025, Volume and Issue: unknown, P. 111136 - 111136

Published: April 1, 2025

Language: Английский

---

Neural Network-Based Approach for Failure and Life Prediction of Electronic Components under Accelerated Life Stress

Electronics, Journal Year: 2024, Volume and Issue: 13(8), P. 1512 - 1512

Published: April 16, 2024

Researchers worldwide have been focusing on accurately predicting the remaining useful life of electronic devices to ensure reliability in various industries. This has made possible by advancements artificial intelligence (AI), machine learning, and Internet Things (IoT) technologies. However, forecasting device with minimal data sets, especially industrial applications, remains a challenge. paper aims address this challenge utilizing learning algorithms, specifically BP, XGBOOST, KNN, predict limited data. The dataset components is obtained through simulation for training testing experimental results show that algorithms achieve certain level accuracy, error rates being as follows: BP algorithm, 0.01–0.02%; XGBOOST KNN 0–0.07%. By benchmarking these study demonstrates feasibility deploying models prediction acceptable accuracy loss, highlights potential AI devices.

Language: Английский

---

Multiaxial Fatigue Lifetime Estimation Based on New Equivalent Strain Energy Damage Model under Variable Amplitude Loading

Crystals, Journal Year: 2024, Volume and Issue: 14(9), P. 825 - 825

Published: Sept. 20, 2024

The largest normal stress excursion during contiguous turn time instants of the maximum torsional is presented as an innovative path-independent fatigue damage quantity upon critical plane, which further employed for characterizing under multiaxial loading. Via using von Mises equivalent formula, axial amplitude with value proposed, incorporating range and plane. influence non-proportional cyclic hardening considered within range. Moreover, according to proposed amplitude, energy-based model estimate lifetime In order verify availability approach, empirical results a 7050-T7451 aluminum alloy En15R steel are used, predictions indicated that estimated lives correlate experimentally observed well variable loadings.

Language: Английский

---

A new approach to multiaxial fatigue life prediction: A multi-dimensional multi-scale composite neural network with multi-depth

Engineering Fracture Mechanics, Journal Year: 2024, Volume and Issue: unknown, P. 110501 - 110501

Published: Sept. 1, 2024

Language: Английский

---

Battery Housing for Electric Vehicles, a Durability Assessment Review

Designs, Journal Year: 2024, Volume and Issue: 8(6), P. 113 - 113

Published: Oct. 31, 2024

Recent research emphasizes the growing use of advanced composite materials in modern transportation, highlighting their superior weight-to-strength ratio. These are increasingly replacing steel and aluminium housings to enhance sustainability, improve efficiency, reduce emissions. Considering these advancements, this article reviews recent studies on materials, focusing fatigue life assessment models. models, which include performance degradation, progressive damage, S–N curve essential for ensuring reliability materials. It is noted that damage process complex, as failure can occur matrix, reinforcement, or transitions such interlaminar intralaminar delamination. Additionally, critically examines integration artificial intelligence techniques predicting offering a comprehensive analysis methods used indicate mechanical properties battery shell composites. Incorporating neural networks into significantly enhances prediction reliability. However, model’s accuracy depends heavily data it includes, including material properties, loading conditions, manufacturing processes, help variability ensure precision predictions. This underscores importance continued advancements significant scientific contributions transportation especially context emerging technologies.

Language: Английский

---