Haijie Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Haijie Wang has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
Bo Li
8 shared publicationsFuzhen Xuan
5 shared publications- Bo Li
East China University of Science and Technology
4 shared publications Fu-Zhen Xuan
4 shared publications- Xintian Liu
School of Mechanical and Automotive Engineering Shanghai University of Engineering Science Shanghai China
4 shared publications Yansong Wang
School of Mechanical and Automotive Engineering, Shanghai University of Engineering Science, Shanghai, China
4 shared publications
Latest Publications
- 26 December 2024
- Published by Emerald Publishing in International Journal of Structural Integrity
- Vol. 16 (1) , 104-126
Machine learning-based fatigue life prediction of laser powder bed fusion additively manufactured Hastelloy X via nondestructively detected defects
Purpose: By incorporating the defect feature information, an ML-based linkage between defects and fatigue life unaffected by the time scale is developed, the primary focus is to quantitatively assess and elucidate the impact of different defect features on fatigue life.Design/methodology/approach: A machine learning (ML) framework is proposed to predict the fatigue life of LPBF-built Hastelloy X utilizing microstructural defects identified through nondestructive detection prior to fatigue testing. The proposed method combines nondestructive micro-computerized tomography (micro-CT) technique to comprehensively analyze the size, location, morphology and distribution of the defects.Findings: In the test set, SVM-based fatigue life prediction exhibits the highest accuracy. Regarding the defect information, the defect size significantly affects fatigue life, and the diameter of the circumscribed sphere of the largest defect has a critical effect on fatigue life.Originality/value: This comprehensive approach provides valuable insights into the fatigue mechanism of structural materials in defective states, offering a novel perspective for better understanding the influence of defects on fatigue performance.
- 26 December 2024
- Published by Emerald Publishing in International Journal of Structural Integrity
- Vol. 16 (1) , 104-126
Machine learning-based fatigue life prediction of laser powder bed fusion additively manufactured Hastelloy X via nondestructively detected defects
Purpose: By incorporating the defect feature information, an ML-based linkage between defects and fatigue life unaffected by the time scale is developed, the primary focus is to quantitatively assess and elucidate the impact of different defect features on fatigue life.Design/methodology/approach: A machine learning (ML) framework is proposed to predict the fatigue life of LPBF-built Hastelloy X utilizing microstructural defects identified through nondestructive detection prior to fatigue testing. The proposed method combines nondestructive micro-computerized tomography (micro-CT) technique to comprehensively analyze the size, location, morphology and distribution of the defects.Findings: In the test set, SVM-based fatigue life prediction exhibits the highest accuracy. Regarding the defect information, the defect size significantly affects fatigue life, and the diameter of the circumscribed sphere of the largest defect has a critical effect on fatigue life.Originality/value: This comprehensive approach provides valuable insights into the fatigue mechanism of structural materials in defective states, offering a novel perspective for better understanding the influence of defects on fatigue performance.
- 22 October 2024
- Published by SAGE Publications in Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture
- Vol. 239 (13) , 1897-1908
Predicting the fatigue life of additively manufactured AlSi10Mg alloy using a physics-informed neural network incorporating continuous damage mechanics
The material characteristics of additively manufactured AlSi10Mg alloy, including the random distribution of process-induced micro defects, microstructual anisotropy, and grain morphologies with complex diversity, poses a significant challenge in accurately predicting its fatigue life, limiting its application in the aircraft field. This work herein introduces a novel approach incorporating a physics-informed neural network (PINN), in which an artificial neural network (ANN) is embedded with a damage mechanics model (CDM) and the critical process parameters of laser powder bed fusion (L-PBF) additive manufacturing. The partial differential equations of the updated CDM model are introduced into the training procedures of the PINN to building a loss function, effectively “teaching” the PINN to learn the physical knowledge. A comparison of the fatigue life prediction results of ANN and the proposed PINN models shows that the PINN model outperforms its counterparts with a 38.71% higher prediction accuracy. The effects of L-PBF process parameters on the fatigue life of as-built AlSi10Mg is examined using both ANN and PINN, proving the better predictive performance and data-physics consistency of the PINN model. The scheme of this work can inform the efficient prediction of fatigue life of additively manufactured alloys and also reverse prediction or fast finding of optimized process parameters for additive manufacturing.
- 22 October 2024
- Published by SAGE Publications in Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture
- Vol. 239 (13) , 1897-1908
Predicting the fatigue life of additively manufactured AlSi10Mg alloy using a physics-informed neural network incorporating continuous damage mechanics
The material characteristics of additively manufactured AlSi10Mg alloy, including the random distribution of process-induced micro defects, microstructual anisotropy, and grain morphologies with complex diversity, poses a significant challenge in accurately predicting its fatigue life, limiting its application in the aircraft field. This work herein introduces a novel approach incorporating a physics-informed neural network (PINN), in which an artificial neural network (ANN) is embedded with a damage mechanics model (CDM) and the critical process parameters of laser powder bed fusion (L-PBF) additive manufacturing. The partial differential equations of the updated CDM model are introduced into the training procedures of the PINN to building a loss function, effectively “teaching” the PINN to learn the physical knowledge. A comparison of the fatigue life prediction results of ANN and the proposed PINN models shows that the PINN model outperforms its counterparts with a 38.71% higher prediction accuracy. The effects of L-PBF process parameters on the fatigue life of as-built AlSi10Mg is examined using both ANN and PINN, proving the better predictive performance and data-physics consistency of the PINN model. The scheme of this work can inform the efficient prediction of fatigue life of additively manufactured alloys and also reverse prediction or fast finding of optimized process parameters for additive manufacturing.
- Open Access
- 7 July 2024
- Published by Taylor & Francis in Virtual and Physical Prototyping
- Vol. 19 (1)
Multi-physics information-integrated neural network for fatigue life prediction of additively manufactured Hastelloy X superalloy
- 7 July 2024
- Published by Taylor & Francis in Virtual and Physical Prototyping
- Vol. 19 (1)
Open AccessMulti-physics information-integrated neural network for fatigue life prediction of additively manufactured Hastelloy X superalloy
- Top Cited Papers
- 29 March 2023
- Published by Elsevier in Engineering Fracture Mechanics
- Vol. 284
Machine learning-based fatigue life prediction of metal materials: Perspectives of physics-informed and data-driven hybrid methods
- 29 March 2023
- Published by Elsevier in Engineering Fracture Mechanics
- Vol. 284
Top Cited PapersMachine learning-based fatigue life prediction of metal materials: Perspectives of physics-informed and data-driven hybrid methods
- 15 September 2022
- Published by SAGE Publications in Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science
- Vol. 237 (3) , 732-747
Structural damage detection of adaptive tangent-based unsaturated bistable stochastic resonance
The output saturation of classical bistable stochastic resonance (CBSR) limits its ability to enhance weak signal characteristics. To solve this problem, an adaptive tangent-based unsaturated bistable stochastic resonance (TUBSR) method is proposed. We construct a tangent-based piecewise unsaturated bistable potential model, and select the optimal system parameters with genetic algorithm (GA) to adaptively adjust the structure of potential model, so as to improve the output saturation and obtain a larger output signal-to-noise ratio (SNR). The simulated signals are used to verify that the TUBSR method can effectively improve the detection ability of weak signals. Finally, the method is applied to the damage detection of bearing inner race and the crack detection of curved plate. The results demonstrate that compared with CBSR and other improved bistable SR method, the TUBSR method has better detection ability for weak signals.
- 15 September 2022
- Published by SAGE Publications in Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science
- Vol. 237 (3) , 732-747
Structural damage detection of adaptive tangent-based unsaturated bistable stochastic resonance
The output saturation of classical bistable stochastic resonance (CBSR) limits its ability to enhance weak signal characteristics. To solve this problem, an adaptive tangent-based unsaturated bistable stochastic resonance (TUBSR) method is proposed. We construct a tangent-based piecewise unsaturated bistable potential model, and select the optimal system parameters with genetic algorithm (GA) to adaptively adjust the structure of potential model, so as to improve the output saturation and obtain a larger output signal-to-noise ratio (SNR). The simulated signals are used to verify that the TUBSR method can effectively improve the detection ability of weak signals. Finally, the method is applied to the damage detection of bearing inner race and the crack detection of curved plate. The results demonstrate that compared with CBSR and other improved bistable SR method, the TUBSR method has better detection ability for weak signals.