Advanced Engineering Informatics最新文献

Adaptive blind deconvolution via convolutional neural networks for early fault detection in degraded gears under different speeds 基于卷积神经网络的自适应盲反卷积在不同转速下退化齿轮早期故障检测中的应用

IF 9.9 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-10-09 DOI: 10.1016/j.aei.2025.103950

Jiahao Gao , Youren Wang , Jinglin Wang , Yong Shen

{"title":"Adaptive blind deconvolution via convolutional neural networks for early fault detection in degraded gears under different speeds","authors":"Jiahao Gao , Youren Wang , Jinglin Wang , Yong Shen","doi":"10.1016/j.aei.2025.103950","DOIUrl":"10.1016/j.aei.2025.103950","url":null,"abstract":"<div><div>Early fault detection of degraded gears at different speeds is both essential and challenging. Adaptive blind deconvolution methods have shown considerable promise for extracting fault characteristics from complex vibration signals. Their performance depends on accurate cyclic frequency estimation and optimal filter length selection. However, this estimation often fails due to gear meshing shock interference and early weak fault characteristics. Additionally, determining the filter length relies on additional metrics with inefficient search strategies, thereby limiting the overall reliability and efficiency. To address these issues, an adaptive blind deconvolution via convolutional neural network (ABDCNN) is proposed. First, we employ an envelope harmonic product spectrum guided by gear frequency-domain features to reduce interference from noise and meshing shocks, enabling precise estimation of the target cyclic frequency. Then, an attention mechanism is integrated into the convolutional neural network to jointly optimize filter coefficients and length estimation, thereby improving computational efficiency. Simulations and gear contact fatigue experiments demonstrate that ABDCNN enables more efficient detection of early faults across different speeds while maintaining strong interpretability.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"69 ","pages":"Article 103950"},"PeriodicalIF":9.9,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266797","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

NQPSO-SLFN: A Q-Learning enhanced PSO framework with neighborhood rough sets for failure mode recognition of reinforced concrete columns NQPSO-SLFN：一种基于邻域粗糙集的Q-Learning改进PSO框架，用于钢筋混凝土柱的失效模式识别

IF 9.9 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-10-09 DOI: 10.1016/j.aei.2025.103940

Jiyuan Jiang , Liangdong Qu

{"title":"NQPSO-SLFN: A Q-Learning enhanced PSO framework with neighborhood rough sets for failure mode recognition of reinforced concrete columns","authors":"Jiyuan Jiang , Liangdong Qu","doi":"10.1016/j.aei.2025.103940","DOIUrl":"10.1016/j.aei.2025.103940","url":null,"abstract":"<div><div>To address the challenges of data heterogeneity and incompleteness in reinforced concrete (RC) column failure mode prediction, this study proposes a novel Q-learning-enhanced particle swarm optimization (PSO) algorithm integrated with neighborhood rough set theory, termed NQPSO. A single-hidden layer feedforward neural network (SLFN) is employed as the classifier, resulting in a unified framework: NQPSO-SLFN. Firstly, a neighborhood construction scheme is designed to effectively handle incomplete datasets with mixed-type attributes. Secondly, Q-learning is incorporated to dynamically adjust the PSO parameters, thereby improving the algorithm’s convergence behavior. Thirdly, the particle encoding scheme of PSO is redesigned, and a new fitness function is formulated, which jointly considers relevance, redundancy, and interaction among features, as well as classification accuracy. The proposed NQPSO algorithm was validated on several UCI benchmark datasets, where it demonstrated superior feature selection capability. Subsequently, comparative experiments were conducted on a real-world RC column failure mode dataset. The NQPSO-SLFN framework significantly outperforms machine learning methods such as SVM, RF, CatBoost, and GNN, achieving an accuracy of 0.9394, an F1-score of 0.9195, and a kappa of 0.8929. Furthermore, additional experiments were performed to assess the model’s performance using individual features, all features, and the selected feature subset. Results indicate that individual features yield poor performance, while using all features offers a moderate improvement. The selected feature subset delivers the best classification performance, thereby confirming the effectiveness of the proposed feature selection strategy in enhancing accuracy while reducing dimensionality. Overall, the findings underscore the practical applicability and robustness of the proposed NQPSO-SLFN framework in structural engineering tasks involving complex and incomplete data.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"69 ","pages":"Article 103940"},"PeriodicalIF":9.9,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266798","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A spatio-temporal trajectory planning framework for AGVs based on motion primitive and dynamic programming in off-road environments 基于运动原语和动态规划的越野环境agv时空轨迹规划框架

IF 9.9 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-10-09 DOI: 10.1016/j.aei.2025.103934

Zheng Zang , Xi Zhang , Xiaojie Gong , Jiarui Song , Ruiguang Yu , Jianwei Gong

{"title":"A spatio-temporal trajectory planning framework for AGVs based on motion primitive and dynamic programming in off-road environments","authors":"Zheng Zang , Xi Zhang , Xiaojie Gong , Jiarui Song , Ruiguang Yu , Jianwei Gong","doi":"10.1016/j.aei.2025.103934","DOIUrl":"10.1016/j.aei.2025.103934","url":null,"abstract":"<div><div>Generating safe, smooth, and dynamically feasible trajectories remains a critical yet challenging task for autonomous ground vehicles (AGVs) operating in off-road environments. This paper proposes a spatio-temporal trajectory planning framework to systematically address the challenges of off-road environments and dynamic obstacles for AGVs. The proposed framework consists of reference path planning based on motion primitives (MP) and spatio-temporal trajectory planning based on dynamic programming (DP). At the reference path level, a hierarchical map model is constructed to store terrain elevation information, traversability information, and static obstacle information separately. Based on the hierarchical map data, we establish a vehicle static stability model and a safe feasible region generation model, and employ the MP algorithm to generate the optimal reference path. At the spatio-temporal trajectory planning level, a spatio-temporal sampling space model is constructed to search for reference trajectories, and a DP-based method is designed to select the optimal trajectory. Based on the reference trajectory, a spatio-temporal safety corridor generation method is proposed to iteratively optimize the trajectory solution. Finally, the proposed framework is validated in both simulations and real-vehicle, and experimental results demonstrate that the proposed system can plan a feasible trajectory fast with the constraints from vehicle kinematics, obstacle avoidance and off-road terrains.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"69 ","pages":"Article 103934"},"PeriodicalIF":9.9,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266811","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Ontologies for the generic motor winding process 通用电机绕组过程的本体

IF 9.9 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-10-09 DOI: 10.1016/j.aei.2025.103937

K.S.K.U. Perera, Riku Ala-Laurinaho, Petri Kuosmanen

{"title":"Ontologies for the generic motor winding process","authors":"K.S.K.U. Perera, Riku Ala-Laurinaho, Petri Kuosmanen","doi":"10.1016/j.aei.2025.103937","DOIUrl":"10.1016/j.aei.2025.103937","url":null,"abstract":"<div><div>Integrating information technology (IT) and operational technology (OT) in the manufacturing ecosystem is crucial for improving productivity, efficiency, and situational awareness. However, integrating various IT/OT systems is often time-consuming and expensive. Semantic integration resolves this by unifying data from heterogeneous sources while preserving the contextual meaning of each source. The success of semantic integration requires robust ontologies that describe objects, processes, and relationships for knowledge representation, system integration, and semantic interoperability. Despite the significance of ontologies in many industrial domains, a scientifically defined ontology for the motor manufacturing process is in demand.</div><div>This research addressed this gap by applying the top-down approach with 5M (manpower, machine, method, measurement, and material) methodology to develop a generic motor winding process ontology systematically. Encoded in the Terse RDF Triple Language (TTL), the developed ontology systematically addressed the needs of diverse job roles by incorporating fundamental aspects such as motor types, winding techniques, and thermal classes. The ontology consisted of a clear definition of core classes and their relationships, and outlined the major factors influencing the motor winding process. Finally, validation experiments confirmed the robustness of the ontology through syntax validation, logical validation using “HermiT” reasoning, domain compatibility assessments via competency questions, and SPARQL query execution outputs. The results confirmed the robustness of the ontology and its applicability, offering a framework for semantic interoperability and knowledge representation in the motor winding process.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"69 ","pages":"Article 103937"},"PeriodicalIF":9.9,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266810","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Industrial IoT-driven condition-based maintenance plus for complex system with multiple dependencies 工业物联网驱动的基于状态的维护加上具有多个依赖关系的复杂系统

IF 9.9 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-10-08 DOI: 10.1016/j.aei.2025.103939

Yaohui Lu , Shaoping Wang , Chao Zhang , Rentong Chen , Hongyan Dui , Yuning Wang

{"title":"Industrial IoT-driven condition-based maintenance plus for complex system with multiple dependencies","authors":"Yaohui Lu , Shaoping Wang , Chao Zhang , Rentong Chen , Hongyan Dui , Yuning Wang","doi":"10.1016/j.aei.2025.103939","DOIUrl":"10.1016/j.aei.2025.103939","url":null,"abstract":"<div><div>Maintenance management based on industrial Internet of Things (IoT) can significantly increase the efficiency of complex system maintenance and enable a transformation from reactive response to proactive prevention. However, conventional condition-based maintenance (CBM) tends to be single-component independent maintenance, which cannot perform collaborative optimization of multi-component maintenance and resource scheduling. In addition, due to shared resources and sequential workflows, the dependencies of the maintenance processes between components makes the existing CBM models not applicable. To solve the aforementioned problem, this paper proposes an industrial IoT-driven condition-based maintenance plus (CBM+) method allowing to perform collaborative optimization of proactive maintenance activities for complex systems. Firstly, the real-time remaining useful life of components is predicted based on degradation data monitored by IoT sensors. Secondly, considering the economic-functional-maintenance process dependencies, a multi-component opportunistic maintenance strategy, based on hierarchical Bayesian networks and multi-layer maintenance process networks, is proposed to increase the collaborative maintenance capacity. Afterwards, leveraging an industrial IoT-enhanced field management approach, the maintenance elements (human, equipment, material, method, and environment) are systematically managed to optimize the maintenance efficiency. Furthermore, an industrial IoT-driven CBM+ optimization model considering multiple dependencies and maintenance elements is developed. Finally, a case study of an industrial IoT-driven hydraulic system is conducted to demonstrate the proposed maintenance strategy.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"69 ","pages":"Article 103939"},"PeriodicalIF":9.9,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266812","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Human-intelligent trajectory optimization for robotic manipulators with hybrid PSO-PS algorithm 基于混合PSO-PS算法的机器人运动轨迹优化

IF 9.9 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-10-08 DOI: 10.1016/j.aei.2025.103941

Mario Peñacoba Yagüe , Jesús Enrique Sierra García , Matilde Santos Peñas

{"title":"Human-intelligent trajectory optimization for robotic manipulators with hybrid PSO-PS algorithm","authors":"Mario Peñacoba Yagüe , Jesús Enrique Sierra García , Matilde Santos Peñas","doi":"10.1016/j.aei.2025.103941","DOIUrl":"10.1016/j.aei.2025.103941","url":null,"abstract":"<div><div>Industry 5.0 is driving a new era in industrial automation, where the collaboration between artificial intelligence (AI) and human supervision enables the development of smarter, more adaptive, and more efficient systems. Robotic trajectory generation is a clear example of this new paradigm. Metaheuristic techniques help automatically generate optimized trajectories, thereby improving operational efficiency. However, optimizing trajectories using AI alone also presents limitations. Starting from random trajectories, the optimization process becomes computationally expensive, especially in complex environments. In this context, initial input from human experts plays a crucial role: expert-defined trajectories provide structured, feasible, and contextual starting points that guide AI more effectively toward high-quality solutions. Therefore, this work proposes a novel human-guided trajectory optimization algorithm. In this way, human knowledge, Particle Swarm Optimization (PSO), and Pattern Search (PS) are efficiently combined. The results demonstrate that this approach significantly improves robotic system performance, achieving cycle time reductions of up to 16.69% compared to expert-defined trajectories. This approach establishes a solid framework for intelligent automation in Industry 5.0, promoting the development of more efficient, sustainable, and adaptive robotic systems.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"69 ","pages":"Article 103941"},"PeriodicalIF":9.9,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266813","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A multi-stage dynamic self-distillation network for industrial defect detection 用于工业缺陷检测的多级动态自蒸馏网络

IF 9.9 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-10-08 DOI: 10.1016/j.aei.2025.103921

He Zhao , Jinhai Liu , Qiannan Wang , Zhitao Wen , Xiangkai Shen

{"title":"A multi-stage dynamic self-distillation network for industrial defect detection","authors":"He Zhao , Jinhai Liu , Qiannan Wang , Zhitao Wen , Xiangkai Shen","doi":"10.1016/j.aei.2025.103921","DOIUrl":"10.1016/j.aei.2025.103921","url":null,"abstract":"<div><div>Self-distillation methods have made remarkable achievements in the field of object detection. However, in complex industrial defect detection, it usually faces the challenge of establishing a strong teacher network and effectively transferring knowledge. To address the above issues, a multi-stage dynamic self-distillation network (MSDNet) is designed, which can improve the accuracy of defect detection without adding additional parameters and computation time. Firstly, an auxiliary knowledge extraction (AKE) module is proposed to inject the physical characteristics and label attributes of industrial data into the teacher network, which provides powerful auxiliary information to enhance the performance of the teacher network. Secondly, a dynamic multi-dimensional distillation (DMD) module is designed, which enables the student network to learn important knowledge from the teacher network in three dimensions dynamically. Finally, a bi-directional multi-head distillation (BMD) module is designed, which can distill both foreground and background regions simultaneously to enhance the understanding of key knowledge of industrial defects and complex industrial fields. The experimental results show that the proposed method outperforms existing methods in defect detection (mean precision: MFL-DET: 3.1%; NEU-DET: 2.3% ; GC10-DET: 4.5%) and does not require additional parameter calculations.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"69 ","pages":"Article 103921"},"PeriodicalIF":9.9,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266946","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An enhanced strategy for query-based defect detector via adaptive spatial feature Reorganization And cross-stage query Injection 基于自适应空间特征重组和跨阶段查询注入的基于查询的缺陷检测改进策略

IF 9.9 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-10-07 DOI: 10.1016/j.aei.2025.103910

Liangcheng Ma , Haidong Shao , Xiaoru Xu , Xizhi Wu

{"title":"An enhanced strategy for query-based defect detector via adaptive spatial feature Reorganization And cross-stage query Injection","authors":"Liangcheng Ma , Haidong Shao , Xiaoru Xu , Xizhi Wu","doi":"10.1016/j.aei.2025.103910","DOIUrl":"10.1016/j.aei.2025.103910","url":null,"abstract":"<div><div>Detection of surface defects from images is crucial to ensure high quality products in manufacturing applications, where surface detection of small defects plays a vital role and has received much attention in the manufacturing industry. However, existing detection solutions perform unevenly in different small defect scenarios. Therefore, this paper proposes an efficient enhancement strategy (RAI) to enhance the model’s ability to detect small surface defects. It consists of two major parts: (i) the feature information enhancement part (ASFR), which consists of a frequency balance (FB) module, an adaptive dilation convolution kernel (ADCK) module, and a spatial feature reorganization (SFR) module, to progressively enhance the semantic information of small defects; and (ii) the subsequent-stage correction interpretation part, which consists of a cross-stage query injection (CQI) mechanism to correct the training focus imbalances and the cascading errors, and fine-grained interpretation of minor defect features. On the engineering side, we applied the strategy to Deformable-Detection Transformer (DETR), Dynamic Anchor Boxes-DETR, and Adamixer, based on three datasets: a self-constructed bamboo slice defect dataset, a defect dataset from Northeastern University, and huggingface surface defects. The experiments were conducted, and mAP50 was improved by 1.7% to 12.5% on the bamboo slice defect test set, 2.4% to 7.8% on the NEU-DET test set, and 2.8% to 4.0% on the huggingface surface defect test set.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"69 ","pages":"Article 103910"},"PeriodicalIF":9.9,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266815","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A fault diagnosis strategy for gearboxes based on reinforcement learning with balanced reward and cost- equilibrium sampling 基于奖励均衡和成本均衡采样强化学习的齿轮箱故障诊断策略

IF 9.9 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-10-07 DOI: 10.1016/j.aei.2025.103946

Xiaofeng Liu, Zheng Zhao, Daiping Wei, Fan Yang, Lin Bo, Jun Luo

{"title":"A fault diagnosis strategy for gearboxes based on reinforcement learning with balanced reward and cost- equilibrium sampling","authors":"Xiaofeng Liu, Zheng Zhao, Daiping Wei, Fan Yang, Lin Bo, Jun Luo","doi":"10.1016/j.aei.2025.103946","DOIUrl":"10.1016/j.aei.2025.103946","url":null,"abstract":"<div><div>To address the challenge of deep reinforcement learning exhibiting low accuracy in data-imbalanced gearbox fault diagnosis, this paper proposes a <em>Reinforcement Learning Model with Cost-sensitive Sampling and Balanced Reward</em> (RLM-CSBR) from the perspectives of feature representation, data utilization, and learning strategy. To tackle the lack of minority-class feature patterns, a multi-level convolutional deep integrated Q-network is constructed to fully explore deep discriminative features from imbalanced data, thereby maximizing feature-perceptive information. To mitigate model training bias, a balanced reward strategy based on the sample missing rate is designed; this strategy not only guides the agent to prioritize the exploration and learning of sample-scarce categories but also ensures the utilization efficiency of majority-class samples. To solve the problem of insufficient model fitting for minority-class samples, a novel cost-equilibrium matrix is incorporated into the prioritized experience replay mechanism, which prioritizes the selection of high-value experiences learned from critical minority-class samples during training.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"69 ","pages":"Article 103946"},"PeriodicalIF":9.9,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A joint collaborative adaptation network for fault diagnosis of rolling bearing under class imbalance and variable operating conditions 一类不平衡和变工况下滚动轴承故障诊断的联合协同自适应网络

IF 9.9 1区工程技术

Advanced Engineering Informatics Pub Date : 2025-10-07 DOI: 10.1016/j.aei.2025.103931

Ye Li , Jingli Yang , Wenmin Wang , Tianyu Gao

{"title":"A joint collaborative adaptation network for fault diagnosis of rolling bearing under class imbalance and variable operating conditions","authors":"Ye Li , Jingli Yang , Wenmin Wang , Tianyu Gao","doi":"10.1016/j.aei.2025.103931","DOIUrl":"10.1016/j.aei.2025.103931","url":null,"abstract":"<div><div>With the dynamic evolution of electromechanical equipment processing tasks, rolling bearing fault diagnosis is often hindered by variable operating conditions and imbalanced fault data, which compromise the recognition of minority fault types and cause significant domain shifts. Multi-source domain adaptation, by integrating data from multiple sources, can alleviate domain shifts and partially mitigate the class imbalance issues, but a dedicated class-aware mechanism is still needed to further enhance performance on minority fault classes. To jointly tackle these challenges, a joint collaborative adaptation network (JCAN) is developed within a multi-source domain adaptation framework that integrates transfer learning, information fusion, and class-aware techniques. Specifically, JCAN extracts domain-invariant features through adversarial training, and enhances sensitivity to underrepresented fault classes by class-aware technique. The adversarial framework comprises a complex convolutional feature extractor and a domain energy discriminator to facilitate cross-domain feature adaptation. Class attention mechanism and class-overlap optimization loss dynamically adjust the focus on imbalanced classes. Moreover, joint domain alignment mechanism minimizes distributional divergence between different domains to ensure consistent feature representation. Further, JCAN integrates multi-source domain information for collaborative decision, where a soft selection-based decision fusion strategy evaluates the source domain contributions, soft attenuating low-contribution sources during information fusion. Experiments on the Paderborn University (PU) and Mechanical Comprehensive Diagnosis Simulation Platform (MCDSP) bearing datasets validate the effectiveness of the proposed JCAN in fault diagnosis tasks under class imbalance and variable operating conditions, as well as the outperformance compared to advanced methods.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"69 ","pages":"Article 103931"},"PeriodicalIF":9.9,"publicationDate":"2025-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0