Liqiao Xia
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
Research Subjects
Research Interests
Liqiao Xia has a total of 51 co-authors. Here are the top co-authors by number of shared publications:
- Pai Zheng
Department of Industrial and Systems Engineering The Hong Kong Polytechnic University Hong Kong China
8 shared publications - Chao Liu
College of Engineering and Physical Sciences, Aston University, Birmingham, UK
4 shared publications - Manuel Herrera
School of Engineering, Newcastle University, Newcastle upon Tyne NE1 7RU, UK
2 shared publications Xiangying Zhang
Department of Mechanical Engineering, Zhejiang University, Hangzhou, China
2 shared publications- Xiao Huang
The Hong Kong Polytechnic University, Hong Kong, Hong Kong
2 shared publications - Yongjing Wang
Department of Mechanical Engineering, School of Engineering, University of Birmingham, Birmingham, U.K.
2 shared publications
Latest Publications
- Open Access
- 12 August 2026
- Published by Elsevier in Journal of Manufacturing Systems
- Vol. 88, 1177-1203
Circularity triage: A multi-stage adaptive model for end-of-life products
- 12 August 2026
- Published by Elsevier in Journal of Manufacturing Systems
- Vol. 88, 1177-1203
Open AccessCircularity triage: A multi-stage adaptive model for end-of-life products
- Open Access
- 27 June 2026
- Published by Elsevier in SSRN Electronic Journal
Noise-Resilient and Lightweight: Hypergraph Computation for Wind Turbine Blade Defect Detection
Current wind turbine blade inspection practices rely heavily on inefficient manual review, while automated deep learning models struggle to balance real-time deployment constraints with detection accuracy under complex background interference. To bridge this gap, this paper proposes the Lightweight Hypergraph Computation and Information-Aware Defect Detection Network (LHCI-Net), a novel architecture engineered to achieve an optimal trade-off between precision and computational efficiency. The proposed method introduces a custom lightweight backbone to drastically reduce computational overhead, an information-aware feature interaction module to actively suppress background noise using polarity and frequency-domain perception, and a hypergraph computation feature aggregation module to establish deep, high-order connections between features. The model is optimised for edge deployment using targeted pruning and knowledge distillation techniques. Experimental results on a custom wind turbine blade image dataset and an open-source DTU dataset validate LHCI-Net's superiority. Compared to state-of-the-art models like YOLOv8, YOLOv11, YOLO26, and RT-DETR, LHCI-Net improves mean Average Precision (mAP) by up to 4.53%, while simultaneously reducing computational load by up to 47.0G Floating Point Operations (FLOPs), demonstrating its exceptional robustness and suitability for practical, real-world wind turbine blade inspections.
- 27 June 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessNoise-Resilient and Lightweight: Hypergraph Computation for Wind Turbine Blade Defect Detection
Current wind turbine blade inspection practices rely heavily on inefficient manual review, while automated deep learning models struggle to balance real-time deployment constraints with detection accuracy under complex background interference. To bridge this gap, this paper proposes the Lightweight Hypergraph Computation and Information-Aware Defect Detection Network (LHCI-Net), a novel architecture engineered to achieve an optimal trade-off between precision and computational efficiency. The proposed method introduces a custom lightweight backbone to drastically reduce computational overhead, an information-aware feature interaction module to actively suppress background noise using polarity and frequency-domain perception, and a hypergraph computation feature aggregation module to establish deep, high-order connections between features. The model is optimised for edge deployment using targeted pruning and knowledge distillation techniques. Experimental results on a custom wind turbine blade image dataset and an open-source DTU dataset validate LHCI-Net's superiority. Compared to state-of-the-art models like YOLOv8, YOLOv11, YOLO26, and RT-DETR, LHCI-Net improves mean Average Precision (mAP) by up to 4.53%, while simultaneously reducing computational load by up to 47.0G Floating Point Operations (FLOPs), demonstrating its exceptional robustness and suitability for practical, real-world wind turbine blade inspections.
- 10 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. PP  (00189456) , 1
Multi-level contrastive self-supervised learning with dynamic spectral-temporal embedding for state-of-health estimation of lithium-ion battery with limited labeled data
Insufficient labeled data is a common issue in state-of-health (SOH) estimation of Lithium-Ion battery. Self-supervised learning method provide a feasible direction to solve the problem of labeled data scarcity by setting up pretext task to fully exploit the intrinsic features of unlabeled data. However, the existing self-supervised methods do not fully consider the physical temporal dependence of signal data, and cannot capture the multi-level key features of the signal, resulting in unsatisfactory SOH estimation results. In order to solve this problem, a multi-level contrastive self-supervised learning with dynamic embedding method (MCSSL-DE) is proposed in this paper. This method fully considers the temporal dependence of the sequence and designs a multi-scale dynamic patch embedding method with local dependence. Specifically, the frequency domain representation of the signal is used to guide the temporal signal to obtain a local comprehensive embedding at the subsequence level, thereby capturing point-level unavailable context information. On this basis, a new multi-level feature extraction model for prediction tasks is designed, which adopts a self-supervised learning method combining contrastive learning and mask modeling to learn the intrinsic multi-level features on the easily available unlabeled data. Furthermore, the multi-layer features are adaptively aggregated through the frequency domain information, and the features are input into the projection network to achieve high-precision SOH estimation by using limited labeled data. Finally, the SOH estimation performance of MCSSL-DE has been verified on the lithium-ion battery dataset NASA, and the interpretability analysis and visual analysis of the model are carried out based on the experimental results, which proves the superiority and practicability of our method.
- 10 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. PP  (00189456) , 1
Multi-level contrastive self-supervised learning with dynamic spectral-temporal embedding for state-of-health estimation of lithium-ion battery with limited labeled data
Insufficient labeled data is a common issue in state-of-health (SOH) estimation of Lithium-Ion battery. Self-supervised learning method provide a feasible direction to solve the problem of labeled data scarcity by setting up pretext task to fully exploit the intrinsic features of unlabeled data. However, the existing self-supervised methods do not fully consider the physical temporal dependence of signal data, and cannot capture the multi-level key features of the signal, resulting in unsatisfactory SOH estimation results. In order to solve this problem, a multi-level contrastive self-supervised learning with dynamic embedding method (MCSSL-DE) is proposed in this paper. This method fully considers the temporal dependence of the sequence and designs a multi-scale dynamic patch embedding method with local dependence. Specifically, the frequency domain representation of the signal is used to guide the temporal signal to obtain a local comprehensive embedding at the subsequence level, thereby capturing point-level unavailable context information. On this basis, a new multi-level feature extraction model for prediction tasks is designed, which adopts a self-supervised learning method combining contrastive learning and mask modeling to learn the intrinsic multi-level features on the easily available unlabeled data. Furthermore, the multi-layer features are adaptively aggregated through the frequency domain information, and the features are input into the projection network to achieve high-precision SOH estimation by using limited labeled data. Finally, the SOH estimation performance of MCSSL-DE has been verified on the lithium-ion battery dataset NASA, and the interpretability analysis and visual analysis of the model are carried out based on the experimental results, which proves the superiority and practicability of our method.
- Top Cited Papers
- 8 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 21  (4) , 3117-3126
Leveraging Large Language Models to Empower Bayesian Networks for Reliable Human-Robot Collaborative Disassembly Sequence Planning in Remanufacturing
Human–robot collaborative disassembly (HRCD) is a promising approach in remanufacturing, leveraging robot's efficiency and human's adaptability for disassembling end-of-life (EoL) products. However, HRCD often encounters numerous choices with uncertain outcomes, posing significant challenges. To address this issue, an HRCD sequence planning model is introduced, providing a quantitative analysis of various decisions with explanations. Initially, HRCD constraint graph is constructed for targeted EoL product based on semantic documents. Subsequently, a Dirichlet Bayesian network (DiBN) is employed to generate feasible sequences based on the HRCD constraint graph, effectively quantifying uncertainty. Then, a fine-tuned large language model (LLM) with tailored prompts is utilized to quantitatively analyze DiBN-based sequences. The DiBN is updated with high-performing sequences from LLM, mitigating the limited knowledge about specific EoL products. Furthermore, a generative adversarial network is proposed to integrate the aforementioned modules for effective training. The effectiveness of the proposed method is demonstrated through two HRCD case studies.
- 8 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Industrial Informatics
- Vol. 21  (4) , 3117-3126
Top Cited PapersLeveraging Large Language Models to Empower Bayesian Networks for Reliable Human-Robot Collaborative Disassembly Sequence Planning in Remanufacturing
Human–robot collaborative disassembly (HRCD) is a promising approach in remanufacturing, leveraging robot's efficiency and human's adaptability for disassembling end-of-life (EoL) products. However, HRCD often encounters numerous choices with uncertain outcomes, posing significant challenges. To address this issue, an HRCD sequence planning model is introduced, providing a quantitative analysis of various decisions with explanations. Initially, HRCD constraint graph is constructed for targeted EoL product based on semantic documents. Subsequently, a Dirichlet Bayesian network (DiBN) is employed to generate feasible sequences based on the HRCD constraint graph, effectively quantifying uncertainty. Then, a fine-tuned large language model (LLM) with tailored prompts is utilized to quantitatively analyze DiBN-based sequences. The DiBN is updated with high-performing sequences from LLM, mitigating the limited knowledge about specific EoL products. Furthermore, a generative adversarial network is proposed to integrate the aforementioned modules for effective training. The effectiveness of the proposed method is demonstrated through two HRCD case studies.
- 11 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73  (00189456) , 1-16
GPSC-GAN: A Data Enhanced Model for Intelligent Fault Diagnosis
In manufacturing, accurate fault diagnosis is imperative but frequently impeded by the scarcity of data, which obstructs the development of effective data-driven diagnostic models. Although Generative Adversarial Networks (GANs) are an effective means of increasing data volume, they still face a challenge in concurrently generating high-quality and multi-mode samples for multiple fault categories. To solve this challenge, a novel data enhanced model GPSC (Gradient Penalty Separate Classifier)-GAN based on GAN is proposed in this paper, which is characterizing by fault scenario-agnostic. Firstly, a new separate classifier is developed to integrate into GAN to generate multimode fault samples. Secondly, Wasserstein distance with gradient penalty is introduced into the loss function of the discriminator to handle the optimization problem of the distribution similarity between generated samples and fault samples. Compared to the traditional GAN, the proposed model can more effectively produce generated samples that are aggregated with fault samples, which means the generated samples is high-quality. Meanwhile, experimental results on two different bearing datasets reveal that the generated data by the proposed model is applicable to assist the training of deep learning-based fault diagnosis models with high accuracy, and is also superior to the state-of-art models.
- 11 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73  (00189456) , 1-16
GPSC-GAN: A Data Enhanced Model for Intelligent Fault Diagnosis
In manufacturing, accurate fault diagnosis is imperative but frequently impeded by the scarcity of data, which obstructs the development of effective data-driven diagnostic models. Although Generative Adversarial Networks (GANs) are an effective means of increasing data volume, they still face a challenge in concurrently generating high-quality and multi-mode samples for multiple fault categories. To solve this challenge, a novel data enhanced model GPSC (Gradient Penalty Separate Classifier)-GAN based on GAN is proposed in this paper, which is characterizing by fault scenario-agnostic. Firstly, a new separate classifier is developed to integrate into GAN to generate multimode fault samples. Secondly, Wasserstein distance with gradient penalty is introduced into the loss function of the discriminator to handle the optimization problem of the distribution similarity between generated samples and fault samples. Compared to the traditional GAN, the proposed model can more effectively produce generated samples that are aggregated with fault samples, which means the generated samples is high-quality. Meanwhile, experimental results on two different bearing datasets reveal that the generated data by the proposed model is applicable to assist the training of deep learning-based fault diagnosis models with high accuracy, and is also superior to the state-of-art models.
- 9 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Reliability
- Vol. 74  (3) , 3897-3910
Graph Embedding-Based Bayesian Network for Fault Isolation in Complex Equipment
Fault isolation, or fault location, aims to identify anomalous components at the start of the maintenance process. However, fault isolation within complex equipment can be challenging due to constraints on the scarcity of labeled data and the intricate interaction among various substructures. To overcome this challenge, an embedding-based Bayesian Network (BN) probability inference is proposed to locate the fault components, where the embedding, derived from semantic meanings, can approximate the actual fault distribution within BN. First, a Fault Graph (FG) is established based on the equipment's mechanical structure and its mechanisms. Then, a Multifield hyperbolic embedding is employed to vectorize the nodes in the FG, thereby preserving the inherent logic maximally. Following this, the FG is transformed into the BN, which facilitates the prediction of the faulty component based on available evidence, using the well-trained graph embedding. An empirical study on oil drilling equipment showcases the graph embedding properties and inference performance of the proposed method by comparing it with other cutting-edge methods and traditional scenarios.
- 9 July 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Reliability
- Vol. 74  (3) , 3897-3910
Graph Embedding-Based Bayesian Network for Fault Isolation in Complex Equipment
Fault isolation, or fault location, aims to identify anomalous components at the start of the maintenance process. However, fault isolation within complex equipment can be challenging due to constraints on the scarcity of labeled data and the intricate interaction among various substructures. To overcome this challenge, an embedding-based Bayesian Network (BN) probability inference is proposed to locate the fault components, where the embedding, derived from semantic meanings, can approximate the actual fault distribution within BN. First, a Fault Graph (FG) is established based on the equipment's mechanical structure and its mechanisms. Then, a Multifield hyperbolic embedding is employed to vectorize the nodes in the FG, thereby preserving the inherent logic maximally. Following this, the FG is transformed into the BN, which facilitates the prediction of the faulty component based on available evidence, using the well-trained graph embedding. An empirical study on oil drilling equipment showcases the graph embedding properties and inference performance of the proposed method by comparing it with other cutting-edge methods and traditional scenarios.
- 10 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Services Computing
- Vol. 17  (5) , 2139-2150
Cross-Edge Orchestration of Serverless Functions With Probabilistic Caching
Serverless edge computing adopts an event-based paradigm that provides back-end services and dynamically provisions resources as needed, resulting in efficient resource utilization. To improve the end-to-end latency and revenue, service providers need to optimize the number and placement of serverless containers while considering the system cost (i.e., latency cost and container running cost) incurred by the provisioning. The particular reason for this circumstance is that frequently creating and destroying containers not only increases the system cost but also degrades the time responsiveness due to the cold-start process. Function caching is a common approach to mitigate the coldstart issue. However, function caching requires extra hardware resources and hence incurs extra system costs. Furthermore, the dynamic and bursty nature of serverless invocations remains an under-explored area. Hence, it is vitally important for service providers to conduct a context-aware request distribution and container caching policy for serverless edge computing. In this paper, we study the request distribution and container caching problem in serverless edge computing. We prove the proposed problem is NP-hard and hence difficult to find a global optimal solution. We jointly consider the distributed and resourceconstrained nature of edge computing and propose an optimized request distribution algorithm that adapts to the dynamics of serverless invocations with a theoretical performance guarantee. Also, we propose a context-aware probabilistic caching policy that incorporates a number of characteristics of serverless invocations. Via simulation and implementation results, we demonstrate the superiority of the proposed algorithm by outperforming existing caching policies in terms of the overall system cost and cold-start frequency by up to 62.1% and 69.1%, respectively.
- 10 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Services Computing
- Vol. 17  (5) , 2139-2150
Cross-Edge Orchestration of Serverless Functions With Probabilistic Caching
Serverless edge computing adopts an event-based paradigm that provides back-end services and dynamically provisions resources as needed, resulting in efficient resource utilization. To improve the end-to-end latency and revenue, service providers need to optimize the number and placement of serverless containers while considering the system cost (i.e., latency cost and container running cost) incurred by the provisioning. The particular reason for this circumstance is that frequently creating and destroying containers not only increases the system cost but also degrades the time responsiveness due to the cold-start process. Function caching is a common approach to mitigate the coldstart issue. However, function caching requires extra hardware resources and hence incurs extra system costs. Furthermore, the dynamic and bursty nature of serverless invocations remains an under-explored area. Hence, it is vitally important for service providers to conduct a context-aware request distribution and container caching policy for serverless edge computing. In this paper, we study the request distribution and container caching problem in serverless edge computing. We prove the proposed problem is NP-hard and hence difficult to find a global optimal solution. We jointly consider the distributed and resourceconstrained nature of edge computing and propose an optimized request distribution algorithm that adapts to the dynamics of serverless invocations with a theoretical performance guarantee. Also, we propose a context-aware probabilistic caching policy that incorporates a number of characteristics of serverless invocations. Via simulation and implementation results, we demonstrate the superiority of the proposed algorithm by outperforming existing caching policies in terms of the overall system cost and cold-start frequency by up to 62.1% and 69.1%, respectively.
- 5 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 29  (4) , 2637-2648
Histogram-Based Gradient Boosting Tree: A Federated Learning Approach for Collaborative Fault Diagnosis
Data-driven approaches have been widely applied in fault diagnosis. However, having sufficient labels to effectively develop fault diagnosis models in real-world manufacturing scenarios has remained a challenge. The key issue is to break down data silos and expand the source of data availability while safeguarding data privacy. To address abovementioned gap, a federated learning model using histogram-based gradient boosting tree, is proposed for collaborative fault diagnosis under privacy-preserving. During model training process, only an update of relevant local model parameters is communicated, thus protecting data privacy while contributing to the increased accuracy of the global model. Meanwhile, a data compensation mechanism is introduced to alleviate the disparity problem in data volume across participants. Two case studies are carried out and further compared with traditional situations to reveal the effectiveness of the proposed method. It is envisioned that this developed method can be useful for industrial practitioners to integrate their data for model training while maintaining data privacy, and promoting collaboration in digital manufacturing.
- 5 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 29  (4) , 2637-2648
Histogram-Based Gradient Boosting Tree: A Federated Learning Approach for Collaborative Fault Diagnosis
Data-driven approaches have been widely applied in fault diagnosis. However, having sufficient labels to effectively develop fault diagnosis models in real-world manufacturing scenarios has remained a challenge. The key issue is to break down data silos and expand the source of data availability while safeguarding data privacy. To address abovementioned gap, a federated learning model using histogram-based gradient boosting tree, is proposed for collaborative fault diagnosis under privacy-preserving. During model training process, only an update of relevant local model parameters is communicated, thus protecting data privacy while contributing to the increased accuracy of the global model. Meanwhile, a data compensation mechanism is introduced to alleviate the disparity problem in data volume across participants. Two case studies are carried out and further compared with traditional situations to reveal the effectiveness of the proposed method. It is envisioned that this developed method can be useful for industrial practitioners to integrate their data for model training while maintaining data privacy, and promoting collaboration in digital manufacturing.
- 2 December 2023
- Published by Elsevier in Expert Systems with Applications
- Vol. 243
Establishing a dynamic and static knowledge model of the manufacturing cell management system: An active push approach
- 2 December 2023
- Published by Elsevier in Expert Systems with Applications
- Vol. 243
Establishing a dynamic and static knowledge model of the manufacturing cell management system: An active push approach