Yuchu Qin
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
Yuchu Qin has a total of 66 co-authors. Here are the top co-authors by number of shared publications:
- Paul James Scott
EPSRC Future Advanced Metrology Hub, University of Huddersfield, Huddersfield HD1 3DH, UK
19 shared publications - Qunfen Qi
School of Computing and Engineering, University of Huddersfield, Huddersfield, UK
19 shared publications Xiangqian Jiang
19 shared publicationsYanru Zhong
Guilin University of Electronic Technology Guilin China
17 shared publicationsMeifa Huang
Guangxi Key Laboratory of Manufacturing System and Advanced Manufacturing Technology, School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, Guilin, China
13 shared publications- Meifa Huang
School of Mechanical and Electrical Engineering, Guilin University of Electronic Technology, Guilin, China
13 shared publications
Latest Publications
- Open Access
- 17 July 2026
- Published by Elsevier in SSRN Electronic Journal
A multi-modal deep learning model for layerwise surface measurement of material extrusion-based additive manufacturing
Quantitative analysis of layer-wise surface quality is critical for real-time monitoring in extrusion-based additive manufacturing, enabling direct assessment of layer integrity without manual inspection. However, existing research largely focuses on qualitative evaluations or post-process measurements, limiting applicability for real-time optimisation. To address this, we propose a novel multi-modal mid-level fusion framework designed to quantify key surface parameters, specifically Sa for roughness and Sdr extra surface space. Camera images, vibration signals and temperature signals are collected during the printing as the multi-modal raw data. The method operates by integrating spatial surface images with temporally aligned vibration and temperature signals through a spatio-temporal registration encoder, followed by a dual-head architecture that simultaneously performs ordinal classification and numeric regression with a trainable weighted balancing mechanism. This process is strengthened by a combined Consistent Ordinal-Ranked Entropy loss to ensure ordinal consistency and feature discriminability, while deep ensemble estimation provides uncertainty-aware predictions suitable for industrial metrology. Experimental results demonstrate substantial improvements over single-modality and single-task baselines, achieving Sa RMSE of 2.69 (±0.18) and MAE of 1.75 (±0.04), alongside Sdr RMSE of 5.85 (±0.52) and MAE of 3.01 (±0.20) and proving the effectiveness of the multi-modal inputs and dual-head prediction model structures.
- 17 July 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessA multi-modal deep learning model for layerwise surface measurement of material extrusion-based additive manufacturing
Quantitative analysis of layer-wise surface quality is critical for real-time monitoring in extrusion-based additive manufacturing, enabling direct assessment of layer integrity without manual inspection. However, existing research largely focuses on qualitative evaluations or post-process measurements, limiting applicability for real-time optimisation. To address this, we propose a novel multi-modal mid-level fusion framework designed to quantify key surface parameters, specifically Sa for roughness and Sdr extra surface space. Camera images, vibration signals and temperature signals are collected during the printing as the multi-modal raw data. The method operates by integrating spatial surface images with temporally aligned vibration and temperature signals through a spatio-temporal registration encoder, followed by a dual-head architecture that simultaneously performs ordinal classification and numeric regression with a trainable weighted balancing mechanism. This process is strengthened by a combined Consistent Ordinal-Ranked Entropy loss to ensure ordinal consistency and feature discriminability, while deep ensemble estimation provides uncertainty-aware predictions suitable for industrial metrology. Experimental results demonstrate substantial improvements over single-modality and single-task baselines, achieving Sa RMSE of 2.69 (±0.18) and MAE of 1.75 (±0.04), alongside Sdr RMSE of 5.85 (±0.52) and MAE of 3.01 (±0.20) and proving the effectiveness of the multi-modal inputs and dual-head prediction model structures.
- Open Access
- 6 May 2026
- Published by Elsevier in SSRN Electronic Journal
Data-Driven Design of High-Entropy Alloys: A Machine Learning-Based Dual-Task Prediction Framework for Phase and Hardness
High-entropy alloys (HEAs) exhibit comprehensive properties such as high strength, outstanding corrosion resistance, and excellent thermal stability due to their multi-component chemical composition, offering broad application prospects in industrial fields. However, for hexa-component HEAs, the combination space based on element concentrations encompasses approximately 9.65×10⁷ possible compositions. Traditional trial-and-error approaches are extremely inefficient in exploring this vast chemical space, severely limiting rational design and performance optimization for such alloys. To address this, this study proposes a data-driven dual-task prediction framework for both ’phase’ and ’hardness’. Focusing on the Al-Co-Cr-Cu-Fe-Ni system, which offers both excellent mechanical potential and lower cost, the framework enables rapid prediction of phase and hardness for a large-scale dataset of HEAs with unknown properties. The framework establishes separate machine learning (ML) models for phase classification and hardness regression, achieving 93% accuracy and a correlation coefficient of 0.949. Interpretability analysis further reveals the dominant roles of key descriptors such as valence electron concentration and atomic size difference, validating the model's physical reliability. Finally, the trained model enabled high-throughput prediction for over 48,000 potential alloys, yielding high-hardness HEAs (e.g., BCC: Al0.350Co0.124Cr0.310Fe0.086Ni0.068Cu0.061, FCC: Al0.102Co0.221Cr0.285Fe0.283Ni0.057Cu0.052, BCC+FCC: Al0.348Co0.065Cr0.236Fe0.088Ni0.056Cu0.207). A comprehensive ”composition-structure-property” correlation map was constructed, identifying key characteristics influencing phase composition and hardness in HEAs. This study establishes an efficient, scalable composition screening workflow capable of rapidly identifying HEAs with ideal phase structures and high hardness potential. It not only provides an efficient, precise computational tool for designing HEAs in the Al-Co-Cr-Cu-Fe-Ni system, but also constructs a collaborative prediction framework and methodology with strong transferability. This approach offers a paradigm reference for performance-driven reverse design in other multi-component alloy systems.
- 6 May 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessData-Driven Design of High-Entropy Alloys: A Machine Learning-Based Dual-Task Prediction Framework for Phase and Hardness
High-entropy alloys (HEAs) exhibit comprehensive properties such as high strength, outstanding corrosion resistance, and excellent thermal stability due to their multi-component chemical composition, offering broad application prospects in industrial fields. However, for hexa-component HEAs, the combination space based on element concentrations encompasses approximately 9.65×10⁷ possible compositions. Traditional trial-and-error approaches are extremely inefficient in exploring this vast chemical space, severely limiting rational design and performance optimization for such alloys. To address this, this study proposes a data-driven dual-task prediction framework for both ’phase’ and ’hardness’. Focusing on the Al-Co-Cr-Cu-Fe-Ni system, which offers both excellent mechanical potential and lower cost, the framework enables rapid prediction of phase and hardness for a large-scale dataset of HEAs with unknown properties. The framework establishes separate machine learning (ML) models for phase classification and hardness regression, achieving 93% accuracy and a correlation coefficient of 0.949. Interpretability analysis further reveals the dominant roles of key descriptors such as valence electron concentration and atomic size difference, validating the model's physical reliability. Finally, the trained model enabled high-throughput prediction for over 48,000 potential alloys, yielding high-hardness HEAs (e.g., BCC: Al0.350Co0.124Cr0.310Fe0.086Ni0.068Cu0.061, FCC: Al0.102Co0.221Cr0.285Fe0.283Ni0.057Cu0.052, BCC+FCC: Al0.348Co0.065Cr0.236Fe0.088Ni0.056Cu0.207). A comprehensive ”composition-structure-property” correlation map was constructed, identifying key characteristics influencing phase composition and hardness in HEAs. This study establishes an efficient, scalable composition screening workflow capable of rapidly identifying HEAs with ideal phase structures and high hardness potential. It not only provides an efficient, precise computational tool for designing HEAs in the Al-Co-Cr-Cu-Fe-Ni system, but also constructs a collaborative prediction framework and methodology with strong transferability. This approach offers a paradigm reference for performance-driven reverse design in other multi-component alloy systems.
- Open Access
- 9 April 2026
- Published by Elsevier in SSRN Electronic Journal
Data-Driven Design of High-Entropy Alloys: A Machine Learning-Based Dual-Task Prediction Framework for Phase and Hardness
High-entropy alloys (HEAs) exhibit comprehensive properties such as high strength, outstanding corrosion resistance, and excellent thermal stability due to their multi-component chemical composition, offering broad application prospects in industrial fields. However, for hexa-component HEAs, the combination space based on element concentrations encompasses approximately 9.65×10⁷ possible compositions. Traditional trial-and-error approaches are extremely inefficient in exploring this vast chemical space, severely limiting rational design and performance optimization for such alloys. To address this, this study proposes a data-driven dual-task prediction framework for both ‘phase’ and ‘hardness’. Focusing on the Al-Co-Cr-Cu-Fe-Ni system, which offers both excellent mechanical potential and lower cost, the framework enables rapid prediction of phase and hardness for a large-scale dataset of HEAs with unknown properties. The framework establishes separate machine learning (ML) models for phase classification and hardness regression, achieving 93% accuracy and a correlation coefficient of 0.949. Interpretability analysis further reveals the dominant roles of key descriptors such as valence electron concentration and atomic size difference, validating the model's physical reliability. Finally, the trained model enabled high-throughput prediction for over 48,000 potential alloys, yielding high-hardness HEAs (e.g., BCC: Al0.350Co0.124Cr0.310Fe0.086Ni0.068Cu0.061, FCC: Al0.102Co0.221Cr0.285Fe0.283Ni0.057Cu0.052, BCC+FCC: Al0.348Co0.065Cr0.236Fe0.088Ni0.056Cu0.207). A comprehensive “composition-structure-property” correlation map was constructed, identifying key characteristics influencing phase composition and hardness in HEAs. This study establishes an efficient, scalable composition screening workflow capable of rapidly identifying HEAs with ideal phase structures and high hardness potential. It not only provides an efficient, precise computational tool for designing HEAs in the Al-Co-Cr-Cu-Fe-Ni system, but also constructs a collaborative prediction framework and methodology with strong transferability. This approach offers a paradigm reference for performance-driven reverse design in other multi-component alloy systems.
- 9 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessData-Driven Design of High-Entropy Alloys: A Machine Learning-Based Dual-Task Prediction Framework for Phase and Hardness
High-entropy alloys (HEAs) exhibit comprehensive properties such as high strength, outstanding corrosion resistance, and excellent thermal stability due to their multi-component chemical composition, offering broad application prospects in industrial fields. However, for hexa-component HEAs, the combination space based on element concentrations encompasses approximately 9.65×10⁷ possible compositions. Traditional trial-and-error approaches are extremely inefficient in exploring this vast chemical space, severely limiting rational design and performance optimization for such alloys. To address this, this study proposes a data-driven dual-task prediction framework for both ‘phase’ and ‘hardness’. Focusing on the Al-Co-Cr-Cu-Fe-Ni system, which offers both excellent mechanical potential and lower cost, the framework enables rapid prediction of phase and hardness for a large-scale dataset of HEAs with unknown properties. The framework establishes separate machine learning (ML) models for phase classification and hardness regression, achieving 93% accuracy and a correlation coefficient of 0.949. Interpretability analysis further reveals the dominant roles of key descriptors such as valence electron concentration and atomic size difference, validating the model's physical reliability. Finally, the trained model enabled high-throughput prediction for over 48,000 potential alloys, yielding high-hardness HEAs (e.g., BCC: Al0.350Co0.124Cr0.310Fe0.086Ni0.068Cu0.061, FCC: Al0.102Co0.221Cr0.285Fe0.283Ni0.057Cu0.052, BCC+FCC: Al0.348Co0.065Cr0.236Fe0.088Ni0.056Cu0.207). A comprehensive “composition-structure-property” correlation map was constructed, identifying key characteristics influencing phase composition and hardness in HEAs. This study establishes an efficient, scalable composition screening workflow capable of rapidly identifying HEAs with ideal phase structures and high hardness potential. It not only provides an efficient, precise computational tool for designing HEAs in the Al-Co-Cr-Cu-Fe-Ni system, but also constructs a collaborative prediction framework and methodology with strong transferability. This approach offers a paradigm reference for performance-driven reverse design in other multi-component alloy systems.
- Open Access
Influence of Adhesive Bonding on the Surface Accuracy of Flat Optics: A Mechanistic Analysis and a Quantitative Approximation
Surface accuracy is a crucial evaluation criterion for the life cycle performance of optical components. Throughout the assembly process, the optical surface undergoes deformation due to applied assembly stresses, causing the actual surface profile to deviate from the intended design. The key to the quantitative optimization of the adhesive bonding assembly process is to elucidate the quantitative coupling mechanism between assembly stress and optical surface deformation and to establish a corresponding quantitative relationship. To address this, a comprehensive study into the optical surface deformation of multi-point adhesive bonded flat optical components is presented. Firstly, the coupled influence mechanisms governing optical surface deformation are analyzed considering both the components properties and assembly parameters, and this mechanistic analysis includes dimensionless modeling, boundary conditions analysis, and reliable design requirement analysis. Secondly, based on the understanding of mechanisms, a quantitative approximation method is developed to predict the deformation patterns within the critical central region (50% aperture) of optical components subjected to multi-point bonding. Finally, the quantitative approximation for the assembly-induced surface deformation is experimentally validated. Through this research, the surface deviation during the multi-point adhesive bonding assembly of flat optical components can be effectively approximated, holding significant importance for further imaging quality prediction and assembly parameters optimization during assembly process, and facilitating a further performance improvement for optical instruments.
Open AccessInfluence of Adhesive Bonding on the Surface Accuracy of Flat Optics: A Mechanistic Analysis and a Quantitative Approximation
Surface accuracy is a crucial evaluation criterion for the life cycle performance of optical components. Throughout the assembly process, the optical surface undergoes deformation due to applied assembly stresses, causing the actual surface profile to deviate from the intended design. The key to the quantitative optimization of the adhesive bonding assembly process is to elucidate the quantitative coupling mechanism between assembly stress and optical surface deformation and to establish a corresponding quantitative relationship. To address this, a comprehensive study into the optical surface deformation of multi-point adhesive bonded flat optical components is presented. Firstly, the coupled influence mechanisms governing optical surface deformation are analyzed considering both the components properties and assembly parameters, and this mechanistic analysis includes dimensionless modeling, boundary conditions analysis, and reliable design requirement analysis. Secondly, based on the understanding of mechanisms, a quantitative approximation method is developed to predict the deformation patterns within the critical central region (50% aperture) of optical components subjected to multi-point bonding. Finally, the quantitative approximation for the assembly-induced surface deformation is experimentally validated. Through this research, the surface deviation during the multi-point adhesive bonding assembly of flat optical components can be effectively approximated, holding significant importance for further imaging quality prediction and assembly parameters optimization during assembly process, and facilitating a further performance improvement for optical instruments.
- Open Access
- 30 December 2024
- Published by Elsevier in Computers in Industry
- Vol. 165
BlurRes-UNet: A novel neural network for automated surface characterisation in metrology
- 30 December 2024
- Published by Elsevier in Computers in Industry
- Vol. 165
Open AccessBlurRes-UNet: A novel neural network for automated surface characterisation in metrology
- Open Access
- 2 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 33-37
Overview of intelligent CMM guidance technology
- 2 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 33-37
Open AccessOverview of intelligent CMM guidance technology
- Open Access
- 24 June 2024
- Published by MDPI AG in Journal of Manufacturing and Materials Processing
- Vol. 8 (4) , 130
Enriching Laser Powder Bed Fusion Part Data Using Category Theory
Laser powder bed fusion (LPBF) is a promising metal additive manufacturing technology for producing functional components. However, there are still a lot of obstacles to overcome before this technology is considered mature and trustworthy for wider industrial applications. One of the biggest obstacles is the difficulty in ensuring the repeatability of process and the reproducibility of products. To tackle this challenge, a prerequisite is to represent and communicate the data from the part realisation process in an unambiguous and rigorous manner. In this paper, a semantically enriched LPBF part data model is developed using a category theory-based modelling approach. Firstly, a set of objects and morphisms are created to construct categories for design, process planning, part build, post-processing, and qualification. Twenty functors are then established to communicate these categories. Finally, an application of the developed model is illustrated via the realisation of an LPBF truncheon.
- 24 June 2024
- Published by MDPI AG in Journal of Manufacturing and Materials Processing
- Vol. 8 (4) , 130
Open AccessEnriching Laser Powder Bed Fusion Part Data Using Category Theory
Laser powder bed fusion (LPBF) is a promising metal additive manufacturing technology for producing functional components. However, there are still a lot of obstacles to overcome before this technology is considered mature and trustworthy for wider industrial applications. One of the biggest obstacles is the difficulty in ensuring the repeatability of process and the reproducibility of products. To tackle this challenge, a prerequisite is to represent and communicate the data from the part realisation process in an unambiguous and rigorous manner. In this paper, a semantically enriched LPBF part data model is developed using a category theory-based modelling approach. Firstly, a set of objects and morphisms are created to construct categories for design, process planning, part build, post-processing, and qualification. Twenty functors are then established to communicate these categories. Finally, an application of the developed model is illustrated via the realisation of an LPBF truncheon.
- Open Access
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 14940-14952
MoCoKGC: Momentum Contrast Entity Encoding for Knowledge Graph Completion
Qingyang Li, Yanru Zhong, Yuchu Qin. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.
- 1 January 2024
- Published by Association for Computational Linguistics (ACL)
- p. 14940-14952
Open AccessMoCoKGC: Momentum Contrast Entity Encoding for Knowledge Graph Completion
Qingyang Li, Yanru Zhong, Yuchu Qin. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.
- Open Access
- 17 August 2023
- Published by MDPI AG in Mathematics
- Vol. 11 (16) , 3561
An Improved Interval-Valued Hesitant Fuzzy Weighted Geometric Operator for Multi-Criterion Decision-Making
In this paper, an improved interval-valued hesitant fuzzy weighted geometric (IIVHFWG) operator for multi-criterion decision-making is proposed. This operator is free of the limitations of the existing interval-valued hesitant fuzzy weighted average operator, interval-valued hesitant fuzzy weighted geometric operator, generalized interval-valued hesitant fuzzy weighted geometric operator, interval-valued hesitant fuzzy Hammer weighted average operator, and interval-valued hesitant fuzzy Hammer weighted geometric operator, which are prone to being influenced by extreme values. Based on the proposed IIVHFWG operator, a new method to solve the multi-criterion decisionmaking problems with interval-valued hesitant fuzzy elements is presented. Several numerical examples together with comparisons are introduced to demonstrate the effectiveness and advantages of this method.
- 17 August 2023
- Published by MDPI AG in Mathematics
- Vol. 11 (16) , 3561
Open AccessAn Improved Interval-Valued Hesitant Fuzzy Weighted Geometric Operator for Multi-Criterion Decision-Making
In this paper, an improved interval-valued hesitant fuzzy weighted geometric (IIVHFWG) operator for multi-criterion decision-making is proposed. This operator is free of the limitations of the existing interval-valued hesitant fuzzy weighted average operator, interval-valued hesitant fuzzy weighted geometric operator, generalized interval-valued hesitant fuzzy weighted geometric operator, interval-valued hesitant fuzzy Hammer weighted average operator, and interval-valued hesitant fuzzy Hammer weighted geometric operator, which are prone to being influenced by extreme values. Based on the proposed IIVHFWG operator, a new method to solve the multi-criterion decisionmaking problems with interval-valued hesitant fuzzy elements is presented. Several numerical examples together with comparisons are introduced to demonstrate the effectiveness and advantages of this method.
- Open Access
A Method for Predicting Surface Finish of Polylactic Acid Parts Printed Using Fused Deposition Modeling
Accurately predicting the surface finish of fused deposition modeling (FDM) parts is an important task for the engineering application of FDM technology. So far, many prediction models have been proposed by establishing a mapping relationship between printing parameters and surface roughness. Each model can work well in its specific context; however, existing prediction models cannot meet the requirements of multi-factor and multi-category prediction of surface finish and cope with imbalanced data. Aiming at these issues, a prediction method based on a combination of the adaptive particle swarm optimization and K-nearest neighbor (APSO-KNN) algorithms is proposed in this paper. Seven input variables, including nozzle diameter, layer thickness, number of perimeters, flow rate, print speed, nozzle temperature, and build orientation, are considered. The printing values of each specimen are determined using an L27 Taguchi experimental design. A total of 27 specimens are printed and experimental data for the 27 specimens are used for model training and validation. The results indicate that the proposed method can achieve a minimum classification error of 0.01 after two iterations, with a maximum accuracy of 99.0%, and high model training efficiency. It can meet the requirements of predicting surface finish for FDM parts with multiple factors and categories and can handle imbalanced data. In addition, the high accuracy demonstrates the potential of this method for predicting surface finish, and its application in actual industrial manufacturing.
Open AccessA Method for Predicting Surface Finish of Polylactic Acid Parts Printed Using Fused Deposition Modeling
Accurately predicting the surface finish of fused deposition modeling (FDM) parts is an important task for the engineering application of FDM technology. So far, many prediction models have been proposed by establishing a mapping relationship between printing parameters and surface roughness. Each model can work well in its specific context; however, existing prediction models cannot meet the requirements of multi-factor and multi-category prediction of surface finish and cope with imbalanced data. Aiming at these issues, a prediction method based on a combination of the adaptive particle swarm optimization and K-nearest neighbor (APSO-KNN) algorithms is proposed in this paper. Seven input variables, including nozzle diameter, layer thickness, number of perimeters, flow rate, print speed, nozzle temperature, and build orientation, are considered. The printing values of each specimen are determined using an L27 Taguchi experimental design. A total of 27 specimens are printed and experimental data for the 27 specimens are used for model training and validation. The results indicate that the proposed method can achieve a minimum classification error of 0.01 after two iterations, with a maximum accuracy of 99.0%, and high model training efficiency. It can meet the requirements of predicting surface finish for FDM parts with multiple factors and categories and can handle imbalanced data. In addition, the high accuracy demonstrates the potential of this method for predicting surface finish, and its application in actual industrial manufacturing.