Advanced Engineering Informatics最新文献

{"title":"REACT: Runtime-Enabled active collision-avoidance technique for autonomous driving","authors":"Heye Huang ,&nbsp;Hao Cheng ,&nbsp;Zhiyuan Zhou ,&nbsp;Zijin Wang ,&nbsp;Haoran Wang ,&nbsp;Qichao Liu ,&nbsp;Xiaopeng Li","doi":"10.1016/j.aei.2025.104248","DOIUrl":"10.1016/j.aei.2025.104248","url":null,"abstract":"<div><div>Achieving rapid and effective active collision avoidance in dynamic interactive traffic remains a core challenge for autonomous driving. This paper proposes REACT (Runtime-Enabled Active Collision-avoidance Technique), a lightweight, closed-loop framework designed to unify risk assessment with interpretable active control. By leveraging energy transfer principles and human-vehicle–road interaction modeling, REACT dynamically quantifies runtime risk and constructs a spatially continuous risk field. The system incorporates physically grounded safety constraints such as directional risk and traffic rules to identify high-risk zones and generate interpretable avoidance behaviors. A hierarchical warning mechanism and lightweight runtime design ensure real-time responsiveness and onboard deployability. Evaluations across four representative high-risk scenarios including car-following braking, cut-in, rear-approaching, and intersection conflict demonstrate REACT’s capability to accurately identify critical risks and execute proactive avoidance. Its risk estimation aligns closely with human driver cognition (i.e., warning lead time &lt; 0.4  s), achieving 100 % safe avoidance with zero false alarms or missed detections. Furthermore, it exhibits low-latency performance (&lt; 50 ms latency), strong foresight, and generalization. The lightweight architecture achieves state-of-the-art accuracy, highlighting its potential for real-time deployment in safety–critical autonomous systems.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"71 ","pages":"Article 104248"},"PeriodicalIF":9.9,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145792210","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}

{"title":"Automatic operational modal analysis for high arch dams using enhanced SSI-COV with adaptive MVMD and improved FCM clustering algorithm","authors":"Huokun Li ,&nbsp;Siyu Zeng ,&nbsp;Bo Liu ,&nbsp;Gang Wang ,&nbsp;Yiyuan Tang ,&nbsp;Wentao Wang ,&nbsp;Wei Huang","doi":"10.1016/j.aei.2025.104257","DOIUrl":"10.1016/j.aei.2025.104257","url":null,"abstract":"<div><div>Operational modal analysis provides critical value for structural health monitoring of arch dams, where ambient vibration-derived modal parameters characterize operational dynamic properties. This paper presents an automated method for high-fidelity identification of modal parameters. The proposed method enhances the covariance-driven stochastic subspace identification (SSI-COV) algorithm by introducing the adaptive multivariate variational mode decomposition (MVMD), the three-dimensional stabilization diagram, and an improved fuzzy c-means (FCM) clustering algorithm. Initially, the adaptive MVMD algorithm is utilized to decompose the multichannel vibration response signals. The multichannel signals are adaptively denoised by extracting and superposing sensitive signal components. Subsequently, the modal parameters of the high arch dam are obtained using the SSI-COV, and a three-dimensional stabilization diagram is developed for the automatic determination of the model order. Finally, an improved FCM clustering algorithm that integrates the local outlier factor for abnormal poles and the coati optimization algorithm for global optimization of the initial cluster centers is proposed to group physical modes, thereby facilitating automatic modal parameter estimation. Validation employs a four-DOF numerical model, a physical arch dam model, and a prototype arch dam. Results demonstrate effective noise suppression and reliable automatic modal identification under varying water discharge conditions, providing a new idea supporting continuous long-term observation of dynamic characteristics in high arch dams.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"71 ","pages":"Article 104257"},"PeriodicalIF":9.9,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145792211","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}

{"title":"Linking microstructure informatics with characterization knowledge in additively manufactured composites through customized and hybrid vision-language representations for automated qualification","authors":"Mutahar Safdar ,&nbsp;Gentry Wood ,&nbsp;Max Zimmermann ,&nbsp;Guy Lamouche ,&nbsp;Priti Wanjara ,&nbsp;Yaoyao Fiona Zhao","doi":"10.1016/j.aei.2025.104238","DOIUrl":"10.1016/j.aei.2025.104238","url":null,"abstract":"<div><div>Rapid and reliable qualification of advanced materials remains a bottleneck in industrial manufacturing, particularly for heterogeneous structures produced via non-conventional additive manufacturing processes. This study introduces a novel framework that links microstructure informatics with a range of expert characterization knowledge using customized and hybrid vision-language representations (VLRs). By integrating deep semantic segmentation with pre-trained multi-modal models (CLIP and FLAVA), we encode both visual microstructural data and textual expert assessments into shared representations. To overcome limitations in general-purpose embeddings, we developed a customized similarity-based representation that incorporates both positive and negative references from expert-annotated images and their associated textual descriptions. This allowed zero-shot classification of previously unseen microstructures through a net similarity scoring approach. Validation on an additively manufactured metal matrix composite (MMC) dataset demonstrated the framework’s ability to distinguish between acceptable and defective samples across a range of characterization criteria with up to 80% top-5 retrieval accuracy. Comparative analysis revealed that FLAVA model offers higher visual sensitivity and penalized weak similarities with score differences as large as 0.17 relative to CLIP. However, FLAVA’s text encoder exhibited sharp drops in similarity for paraphrased expert descriptions (falling below 0.20), whereas CLIP maintained more stable alignment with textual criteria (0.29–0.36). Z-score normalization adjusted raw unimodal and cross-modal similarity scores based on their local dataset-driven distributions, enabling more effective alignment and classification in the hybrid vision-language framework. The standardized scores provided strong binary classification results across three categories (82% for distribution, 90% for dilution, and 82% for reinforcement). The proposed method enhanced traceability and interpretability in qualification pipelines via human-in-the-loop decision-making without task-specific model retraining. By advancing semantic interoperability between raw data and expert knowledge, this work contributes toward scalable and domain-adaptable qualification strategies in engineering informatics.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"71 ","pages":"Article 104238"},"PeriodicalIF":9.9,"publicationDate":"2025-12-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145792207","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}

{"title":"Integration of LiDAR scan-to-IFC and UWB real-time positioning for automated construction monitoring: a precast module case study","authors":"Maggie Y. Gao,&nbsp;Chengjia Han,&nbsp;Zhen Peng,&nbsp;Yiqing Dong,&nbsp;Robert L.K. Tiong,&nbsp;Yaowen Yang","doi":"10.1016/j.aei.2025.104247","DOIUrl":"10.1016/j.aei.2025.104247","url":null,"abstract":"<div><div>This paper presents a novel framework for construction monitoring, focusing on efficient as-built model registration through the integration of LiDAR scanning and Ultra-Wideband (UWB) positioning technology. The proposed approach leverages UWB positioning data as preliminary spatial references for precise alignment of as-built model for precast module and components from LiDAR point cloud processing. This integrated framework addresses the critical gap in construction monitoring by integrating multiple technologies into a cohesive system, overcoming the limitations of fragmented approaches. During these procedures, the study presents a systematic targeted partial transformation to correct angular misalignments during point cloud registration. This framework employs BIMCrossNet, a custom deep learning architecture specifically designed for point cloud segmentation in construction scenarios. At last, the study enables automated updating of semantic enriched as-built BIM models with real-time validation of component placement, making it particularly valuable for quality control and progress monitoring in modular construction applications.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"71 ","pages":"Article 104247"},"PeriodicalIF":9.9,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145792145","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}

{"title":"Knowledge-data driven model for explosion-induced crack identification in high arch dams","authors":"Junwei Xiong ,&nbsp;Wenbo Lu ,&nbsp;Gaohui Wang ,&nbsp;Yufeng Huang ,&nbsp;Zhidong Liu","doi":"10.1016/j.aei.2025.104129","DOIUrl":"10.1016/j.aei.2025.104129","url":null,"abstract":"<div><div>The identification of explosion-induced damage in high arch dams is a crucial aspect in the anti-explosion safety evaluation of such structures. This study proposes a knowledge-data driven framework for identifying explosion-induced cracks in high arch dams. The model formalizes underlying physical knowledge pertaining to modal evolution law in cracked arch dams and integrates the efficient nonlinear computational capabilities of machine learning, enabling accurate and efficient crack identification. First, quantitative damage identification indices, including modal frequency change and mode localization factor, are introduced based on the theory of mode transition and mode localization. A modal sequence reconstruction method is developed to establish a data preprocessing framework considering modal transition. Building upon this, a modal parameter prediction model, using a human evolutionary optimization algorithm-based support vector regression (HEOA-SVR), replaces the finite element model (FEM) in the sampling process. Additionally, Bayesian theory is integrated to form a knowledge-data driven model for explosion crack identification in high arch dams. Finally, a modal database for explosive-damaged arch dam is established based on one arch dam, and the effectiveness of this damage identification model is validated through the identification of horizontal cracks. The results demonstrate that the identification indices based on modal frequency and beamwise mode localization factor offer superior accuracy and noise immunity for horizontal crack detection, aligning with physical knowledge. In comparison with the methods that neglect mode transition and mode localization, the knowledge-data driven model improves recognition efficiency while maintaining high accuracy and precision.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"71 ","pages":"Article 104129"},"PeriodicalIF":9.9,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145792160","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}

{"title":"Remaining useful life prediction based on self-attention mechanism -sequential variational autoencoder: From a semi-supervised perspective","authors":"Jiusi Zhang ,&nbsp;Kai Chen ,&nbsp;Fan Wu ,&nbsp;Quan Qian ,&nbsp;Tenglong Huang ,&nbsp;Yuhua Cheng ,&nbsp;Shen Yin","doi":"10.1016/j.aei.2025.104242","DOIUrl":"10.1016/j.aei.2025.104242","url":null,"abstract":"<div><div>Remaining useful life (RUL) prediction is a crucial task in maintaining operational safety and dependability of complex systems. Considering that conventional data-driven RUL prediction approaches demand massive labeled samples for supervised model training, it has notable limitations in making full use of unlabeled degradation data. Furthermore, current deep networks require in-depth research work in terms of interpretability and uncertainty. In this sense, an RUL prediction approach with the self-attention mechanism-sequential variational autoencoder (SAM-SVAE) is proposed from a semi-supervised perspective. Specifically, considering a dynamic serialization modeling, this paper designs a self-attention mechanism network to focus on key parts of the input time window. On this basis, this paper explores the correspondence between a Bayesian deep probability generation network and the state space model in control theory, which approximates the RUL prediction’s density function for uncertainty assessment. Moreover, this paper proposes an SAM-SVAE from a semi-supervised perspective, which can learn valuable feature representations from a large amount of unlabeled degradation data, from which the interpretability is provided through the analyze of latent space. Experimental validation of the presented SAM-SVAE utilizes the aircraft turbofan engine dataset from NASA Prediction Center.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"71 ","pages":"Article 104242"},"PeriodicalIF":9.9,"publicationDate":"2025-12-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145792208","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}

{"title":"Review of condition monitoring approaches for ball screws","authors":"Himanshu Gupta ,&nbsp;Tauheed Mian ,&nbsp;Pradeep Kundu","doi":"10.1016/j.aei.2025.104206","DOIUrl":"10.1016/j.aei.2025.104206","url":null,"abstract":"<div><div>Ball screws are critical components primarily used in feed drives for machine tools and manufacturing systems, as well as in electromechanical actuators (EMAs). These are linear motion systems, enabling the conversion of rotary motion into precise linear motion. The performance and efficiency of these motion systems heavily rely on the safe operation of the ball screws. Their failures can result in costly downtime or catastrophic consequences. Condition monitoring (CM) approaches have been developed to detect these failures on time and estimate their remaining useful life (RUL). Although these approaches have been extensively reviewed for components like bearings, gears, and motors, a comprehensive evaluation explicitly focused on ball screw systems is still lacking. This paper addresses this gap by presenting an in-depth review of existing approaches for ball screw CM. It examines the different ball screw failure modes and physics-based, data-driven, and hybrid approaches for their CM. Further, the challenges associated with the existing approaches are discussed, along with potential solutions and future research directions.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"71 ","pages":"Article 104206"},"PeriodicalIF":9.9,"publicationDate":"2025-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145792143","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}

{"title":"An hourglass-shaped digital twin framework for industrial computed tomography","authors":"Zhiyu Gao ,&nbsp;Haibin Lan ,&nbsp;Changsheng Zhang ,&nbsp;Qianni Wang ,&nbsp;Shengping Yuan ,&nbsp;Wenlei Xiao ,&nbsp;Gang Zhao ,&nbsp;Jian Fu","doi":"10.1016/j.aei.2025.104251","DOIUrl":"10.1016/j.aei.2025.104251","url":null,"abstract":"<div><div>Industrial computed tomography (ICT) is one of the most advanced non-destructive testing, evaluation and characterization techniques, used for revealing internal structures, analyzing material compositions, and identifying defects within objects. However, in practical engineering applications, it has encountered challenges such as complex operation procedures and parameter settings, high costs, as well as artifacts and measurement errors. Digital twin (DT), by creating a virtual environment that closely mirrors and interacts with the physical space, offers an innovative solution to these challenges. In this paper, an hourglass-shaped DT framework for ICT is presented. By establishing both static and dynamic connections between the physical and virtual spaces, the ICT system is twinned across three dimensions: entities, behaviors, and data. This integration enables the applications and functions layer to improve the efficiency and accuracy of the ICT process. The implementation of the DT framework within the ICT system is explained in detail, and a case study of a workpiece’s ICT inspection process is used to validate its feasibility and effectiveness. This work has been helpful in advancing the digitization of the industrial inspection field under Industry 4.0.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"71 ","pages":"Article 104251"},"PeriodicalIF":9.9,"publicationDate":"2025-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145792147","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}

{"title":"Towards transparent object detection models for construction sites: explainable AI and error classification","authors":"Junghoon Kim ,&nbsp;Yue Gong ,&nbsp;Seokho Chi ,&nbsp;Jung In Kim ,&nbsp;JoonOh Seo","doi":"10.1016/j.aei.2025.104245","DOIUrl":"10.1016/j.aei.2025.104245","url":null,"abstract":"<div><div>Construction site monitoring is essential for ensuring projects are executed as planned and achieving goals in productivity, safety, and quality. However, traditional manual monitoring methods are time-consuming, error-prone, and lack scalability. Deep learning-based object detection offers a promising alternative, but its “black-box” nature hinders understanding of detection failures. This study proposes a Grad-CAM-based explainable AI framework to diagnose and classify detection errors systematically. The framework consists of three main processes: (1) defining major types of detection errors, (2) collecting failed images for each error type, and (3) developing a machine learning-based classification model using Grad-CAM features and detection metrics. Unlike previous approaches that relied on qualitative interpretations, this study converts Grad-CAM heatmaps into quantitative features (e.g., GT influence ratio, activation-to-box distance, cluster counts), enabling automated error classification. Errors were categorized into abnormal viewpoint, small size, occlusion, complex background, and lighting variation, achieving 94% classification accuracy on synthetic data, 85% on real images, and 88% on AI-generated data. This framework enhances transparency and interpretability while supporting model optimization and adaptive deployment for real-world construction site applications.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"71 ","pages":"Article 104245"},"PeriodicalIF":9.9,"publicationDate":"2025-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145792164","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}

{"title":"DD-DETR: A dual-decoder DETR with information interaction and competitive learning for blade surface defect detection","authors":"Xiongfeng Shi,&nbsp;Lai Zou,&nbsp;Kefei Qian,&nbsp;Xin Liu","doi":"10.1016/j.aei.2025.104234","DOIUrl":"10.1016/j.aei.2025.104234","url":null,"abstract":"<div><div>Aero-engine blades operate under prolonged high-temperature and high-pressure conditions, making them prone to surface defects. If not detected and repaired promptly, these defects may lead to significant safety hazards. To address the challenge of efficient and accurate defect detection, especially in the absence of publicly available datasets, we construct a dedicated dataset featuring two types of repairable surface defects, scratches and pits. Based on this dataset, we propose DD-DETR, a high-precision detection algorithm built upon the DETR architecture. DD-DETR introduces a novel Dual-Decoder architecture with parallel decoders that share the same structure and parameters. We also develop an information interaction mechanism and a competitive learning mechanism, optimizing information flow, enhancing decoder potential, and maximizing feature information utilization. Furthermore, the Dense One-to-One (Dense O2O) matching strategy is employed to increase positive samples without adding extra parameters, while Matchability-Aware Loss (MAL) improves training robustness and matching quality across samples of varying difficulty. Experimental results on our constructed dataset show DD-DETR achieves 93.4 % mAP₅₀ and 70.7 % mAP, significantly outperforming other DETR-based models. These results demonstrate DD-DETR’s effectiveness and practical value in aero-engine blade surface defect detection.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"71 ","pages":"Article 104234"},"PeriodicalIF":9.9,"publicationDate":"2025-12-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145792162","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}