Advanced Engineering Informatics最新文献

{"title":"Noise-robust compound fault feature extraction of variable speed rotating machinery via an amplitude modulation-driven decomposition framework","authors":"Zuhua Jiang,&nbsp;Fucai Li,&nbsp;Mingzhe Li,&nbsp;Xiaolei Xu","doi":"10.1016/j.aei.2025.103501","DOIUrl":"10.1016/j.aei.2025.103501","url":null,"abstract":"<div><div>In practical industrial applications, rotating machinery often operates under conditions with time-varying rotation speed. This paper proposes a novel amplitude modulation-driven decomposition framework named time–frequency amplitude modulation mode decomposition (TFAMMD), aiming to provide a robust solution for feature extraction under variable speed and compound fault scenarios. The framework utilizes adaptive mode segmentation based on spectral trend to divide the signal into physically meaningful modes and reduce redundant components. Subsequently, characteristic harmonic kurtosis is introduced to quantify fault information in each mode without any prior knowledge, after which the fault-related modes are determined automatically through difference ratio. Finally, time–frequency spectral amplitude modulation guided by the estimated instantaneous fault characteristic frequency is employed to realize nonlinear enhancement of fault characteristics in those valuable modes. Validation through simulated analysis demonstrates that the framework can effectively detect and extract multiple fault components in signals under different signal-to-noise ratios and interferences, while its capability of noise suppression also outperforms the existing state-of-the-art methods. Experimental tests on real-world rotating machinery, including bearings and gearbox, further confirm its adaptability and robustness in diagnosing faults under variable speed conditions. These results indicate that the proposed method holds promising potential for industrial fault diagnosis.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"67 ","pages":"Article 103501"},"PeriodicalIF":8.0,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144168134","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":"Principal attributes of wearable warning alarms to promote roadway worker safety","authors":"Daniel Bin Lu,&nbsp;Semiha Ergan","doi":"10.1016/j.aei.2025.103481","DOIUrl":"10.1016/j.aei.2025.103481","url":null,"abstract":"<div><div>In response to a concerning increase in annual worker fatalities within U.S. roadway work zones and a lack of effective worker-centered alert systems in current practice, this study investigates how wearable alarms impact worker reactions (e.g., body movement away from traffic, head turn towards traffic) for their safety from traffic hazards (e.g., speeding, collision vehicles) in unstructured and short-term urban roadway work zones. This study captures human behavioural data in roadway work zones through virtual reality and micro-traffic simulation-based user testing, where varied alarm patterns (e.g., changing modality, duration, repetitions) triggered by traffic hazards are sent to a smartwatch wearable warning device. Through a machine learning-based Shapley value analysis to assess the influence of alarm attributes on roadway worker behaviour, this study identified that sensory modality (i.e., auditory/tactile senses stimulated) and duration (i.e., continuous active time interval) have significant impact on improving workers’ safety in their reactions to traffic hazards. Workers often improved their level of safety in reaction to alarm patterns with a “haptics and sound” modality and a continuous duration of 350 ms. Results identifying modality and duration as principal alarm attributes can inform future research directions towards improving the alarm design of wearable warning devices for roadway workers.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"67 ","pages":"Article 103481"},"PeriodicalIF":8.0,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144168133","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":"Hierarchical attention-based framework for enhanced prediction and optimization of organic and inorganic material synthesis","authors":"Muhammad Munsif,&nbsp;Altaf Hussain,&nbsp;Zulfiqar Ahmad Khan,&nbsp;Min Je Kim,&nbsp;Sung Wook Baik","doi":"10.1016/j.aei.2025.103462","DOIUrl":"10.1016/j.aei.2025.103462","url":null,"abstract":"<div><div>Optimizing the synthesis of organic and inorganic materials, including molybdenum disulfide (MoS<span><math><msub><mrow></mrow><mrow><mtext>2</mtext></mrow></msub></math></span>), and estimating the photoluminescent quantum yield (PLQY) remains a complex and time-intensive challenge with significant applications in high-impact areas such as energy storage, light-emitting devices, and light-filtering materials. Traditional machine learning approaches like XGBoost and support vector machines (SVMs) have shown effectiveness in predicting material properties; however, they often require manual feature engineering and are limited in capturing intricate dependencies across experimental parameters. To address these limitations, this study proposes a unified hierarchical attention transformer network (HATNet) that leverages the multi-head-attention (MHA) mechanism to automatically learn complex interactions within feature spaces, providing a more flexible and powerful alternative for synthesis optimization. Our proposed framework is applied to two key tasks: MoS<span><math><msub><mrow></mrow><mrow><mtext>2</mtext></mrow></msub></math></span> growth status classification and carbon quantum dot (CQD) PLQY estimation. This framework captures high-order feature dependencies in small and large datasets for regression and classification through a shared attention-based encoder. The experimental results demonstrate that HATNet outperforms state-of-the-art methods, achieving higher predictive performance, with a 95% classification accuracy for MoS<span><math><msub><mrow></mrow><mrow><mtext>2</mtext></mrow></msub></math></span> synthesis and a mean squared error (MSE) of 0.003 on inorganic compositions and 0.0219 on organic compositions for carbon quantum yield estimation. These results illustrate HATNet’s adaptability and accuracy in synthesizing advanced materials, highlighting its versatility as a tool for guiding experimental synthesis across various materials in the field of materials science.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"67 ","pages":"Article 103462"},"PeriodicalIF":8.0,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144177627","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":"Multi-scale generative adversarial network for 2D subsurface reconstruction using multi-fidelity geological exploration data","authors":"Xiaoqi Zhou,&nbsp;Peixin Shi","doi":"10.1016/j.aei.2025.103482","DOIUrl":"10.1016/j.aei.2025.103482","url":null,"abstract":"<div><div>Accurate stratigraphic profiling from sparse site exploration data is crucial for geotechnical construction. Traditional methods, including manual delineation, geostatistical inference, and stochastic simulation, often face limitations such as oversimplification, high data demands, or dependence on expert assumptions, while probabilistic approaches require appropriate priors and domain expertise. This study proposes an intelligent computer-aided system for subsurface reconstruction using multi-fidelity geotechnical data using a Generative Adversarial Network (GAN). A multi-scale symmetric architecture is designed for unsupervised learning from a single image, incorporating a confidence-modulated noise map. Multi-fidelity data are fused through hybrid image representation to improve reconstruction accuracy. The GAN is trained on a benchmark geological profile and predicts on incomplete images with applied masks. Model performance is evaluated both qualitatively and quantitatively, with extensive ablation studies analyzing the impact of data fidelity and hyperparameters. Comparative results with state-of-the-art methods validate the effectiveness and efficiency of the proposed framework in integrating observed geological information into deep neural networks for realistic subsurface modeling and practical engineering applications.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103482"},"PeriodicalIF":8.0,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144167986","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":"Kansei-oriented and artificial intelligence-driven framework for mutual aid elderly care service optimization","authors":"An-Jin Shie ,&nbsp;En-Min Xu ,&nbsp;Zhen-Zhen Ye ,&nbsp;Jing-Qi Ruan ,&nbsp;Guanghai Yang ,&nbsp;Ching-Hung Lee","doi":"10.1016/j.aei.2025.103489","DOIUrl":"10.1016/j.aei.2025.103489","url":null,"abstract":"<div><div>Current Mutual Aid Elderly Care (MAEC) services face challenges such as insufficient consideration of the human affective needs of the elderly and limited diversity in service models, leading to poor service quality and low participation. To address these issues, this study integrates Convolutional Neural Network (CNN), Service Blueprint (SB), Fuzzy Failure Mode and Effects Analysis (Fuzzy-FMEA), and the Theory of Inventive Problem Solving (TRIZ) to propose a Kansei-oriented and innovation-driven optimization framework for MAEC services. The framework consists of four phases: Phase 1: Extraction of crucial Kansei words. This study leverages text-mining techniques and CNN to refine Kansei Engineering (KE). By using word frequency statistics and emotional intensity categories derived from CNN-based classification, it comprehensively calculates the importance of each Kansei word, thereby identifying crucial Kansei words. Phase 2: Construction of service failure space. Based on the crucial Kansei words, the study employs SB and Fuzzy-FMEA to diagnose potential service failure nodes within MAEC services. Phase 3: Generation of service innovative optimization solutions. Kansei needs are integrated within TRIZ to propose specific service innovative optimization solutions. Phase 4: Evaluation of service innovative optimization solutions. The interview method and New Service Development Maturity (NSDM) model are used to assess these solutions. Finally, the “Time Banking” elderly care model is used as a case study to validate the feasibility and superiority of the proposed framework. By integrating Kansei needs as implicit human knowledge with AI-driven innovative technologies, this study extends the frontier of human-AI collaboration in the MAEC service innovation.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"67 ","pages":"Article 103489"},"PeriodicalIF":8.0,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144168115","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":"Multi-target prediction and dynamic interpretation method for displacement of arch dam with cracks based on A-DSRSN and SHAP","authors":"Bo Xu ,&nbsp;Hu Zhang ,&nbsp;Chongshi Gu ,&nbsp;Zeyuan Chen ,&nbsp;Hao Gu","doi":"10.1016/j.aei.2025.103467","DOIUrl":"10.1016/j.aei.2025.103467","url":null,"abstract":"<div><div>To address the challenges of accuracy and interpretability in displacement prediction for arch dams with cracks, this study proposes a deep stacked residual shrinkage network based on attention mechanisms (A-DSRSN), combined with the multi-target prediction method leveraging maximum correlated stacking of single-target (MCSST) and the interpretive method of SHapley Additive exPlanations (SHAP), constructing a multi-target prediction and dynamic interpretation method for displacement of arch dams with cracks. Initially, factors influencing displacement are preliminarily selected using hydrostatic-air temperature–time (HT_AT) and hydrostatic-temperature-crack-time (HTCT) models. Elastic net (ElaNet) and principal component analysis (PCA) are subsequently employed for feature selection and extraction. A-DSRSN is then constructed by integrating convolutional block attention modules and residual shrinkage blocks. Using the A-DSRSN model, a multi-target displacement prediction method based on MCSST is established. Furthermore, the integration of the A-DSRSN model with SHAP analysis develops global, local, and dynamic interpretation methods for prediction results. Finally, a case study demonstrates the validity of the proposed method by comparing it with existing baseline approaches. The findings reveal that the A-DSRSN model outperforms baseline methods in prediction accuracy, while the MCSST method enhances overall predictive precision. The interpretation method in this paper reveals the process of model prediction, accurately identifying dominant influencing factors during water level rise and periods of water level and temperature decline. This research provides a novel method for dam displacement prediction, which has significant practical application value for long-term health monitoring and safety management of dams.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103467"},"PeriodicalIF":8.0,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144167987","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":"DeepMonte-Frame: an intelligent workflow for planar steel frame design based on Monte Carlo Tree Search and Feedforward Neural Networks","authors":"Zhexi Yang,&nbsp;Qingxin Yang,&nbsp;Wei-Zhen Lu","doi":"10.1016/j.aei.2025.103510","DOIUrl":"10.1016/j.aei.2025.103510","url":null,"abstract":"<div><div>Optimization of steel frame structures is typically formulated as a large-scale combinatorial problem. Previous research predominantly employs metaheuristic algorithms, which frequently face challenges such as high computational costs, sensitivity to hyperparameters, and reliance on initial solutions. To overcome these limitations, this study proposes a novel optimization workflow termed DeepMonte-Frame, integrating Monte Carlo Tree Search (MCTS) and Feedforward Neural Networks (FNNs). The FNNs rapidly predict structural responses, enhancing both the expansion and rollout phases of the MCTS, thereby significantly improving optimization performance in scenarios with sparse feasible solutions. Ablation experiments demonstrated the essential contribution of FNNs, while comparative evaluations against metaheuristic algorithms proved the superior performance of DeepMonte-Frame. Moreover, the method maintains robust performance when applied to irregular structural configurations, and exhibits remarkable flexibility in various optimization objectives, including cost and carbon footprint reduction. A comprehensive case study further validated the practical applicability of DeepMonte-Frame, achieving a 15.45 % cost reduction while ensuring compliance with engineering standards. The optimized designs can also be automatically transformed into BIM models, facilitating design decision-making and supporting subsequent interdisciplinary collaboration. Overall, DeepMonte-Frame is a highly effective and adaptable approach, significantly outperforming conventional methods and providing innovative insights for future research in structural optimization.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103510"},"PeriodicalIF":8.0,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144167988","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":"AIN-YOLO: A lightweight YOLO network with attention-based InceptionNext and knowledge distillation for underwater object detection","authors":"Xuanting He ,&nbsp;Yue Zhang ,&nbsp;Qiang Zhan","doi":"10.1016/j.aei.2025.103504","DOIUrl":"10.1016/j.aei.2025.103504","url":null,"abstract":"<div><div>Underwater object detection technology based on machine vision is the key to realizie intelligent seafood fishing. Although existing methods based on deep learning have achieved considerable advancement, such key problems as high computation complexity and low detection accuracy still exist. In order to solve these problems, a novel lightweight network named AIN-YOLO is proposed in this paper. Firstly, an attention-based InceptionNext block is designed to replace the original C2f block of YOLOv8 model by introducing a parallel branch architecture with multiple depthwise convolutions to reduce the number of parameters and the complexity of YOLOv8 model. Secondly, a nearly parameter-free improved shuffle attention module is designed to enhance the AIN-YOLO model’s ability to strengthen feature channel interactions and reduce computation resource consumption. Subsequently, a knowledge distillation strategy based on bridging cross-task protocol inconsistency is introduced to train the AIN-YOLO model, facilitating effective knowledge transfer and enhancing detection accuracy. Finally, the proposed model is comprehensively analyzed and compared with six existing state-of-the-art models, and results show the proposed model exhibits better effectiveness and generalizability when detecting objects in complex underwater environment on three publicly available datasets. Moreover, model deployment experiment shows that with fewer parameters and less computation complexity, the proposed model is suitable for deploying on computation resource constrained underwater equipments.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103504"},"PeriodicalIF":8.0,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144146882","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":"A framework for technology opportunity discovery using GAT-based link prediction and network analysis","authors":"Zhi-Xing Chang ,&nbsp;Wei Guo ,&nbsp;Hong-Yu Shao ,&nbsp;Lei Wang ,&nbsp;Zi-Liang Wang ,&nbsp;Yuan-Rong Zhang","doi":"10.1016/j.aei.2025.103498","DOIUrl":"10.1016/j.aei.2025.103498","url":null,"abstract":"<div><div>The rapid discovery of innovative technology opportunities plays a crucial role in accelerating innovation and gaining a competitive advantage for enterprises, while link prediction has emerged as an effective method for Technology Opportunity Discovery (TOD). However, the link prediction methods currently applied to TOD suffer from the insufficient utilization of feature data, which limits the prediction accuracy. Furthermore, link prediction results encompass abundant innovation insights, but the current analysis, which focuses solely on the forthcoming interaction between two technologies, fails to fully exploit these insights. To address these issues, we developed a novel TOD framework that incorporates a machine learning model and two network analysis methods. Specifically, we developed a GAT-SimLinkPredictor model, which effectively leverages feature data from the co-occurrence network of patent classification codes to enhance link prediction accuracy. Subsequently, we introduced two network analysis methods that assist researchers in TOD from the perspective of technology combinations. We validated our framework on a patent dataset of autonomous driving. The results demonstrate that our model improves prediction accuracy, while the network analysis methods effectively uncover innovation insights. In conclusion, our framework makes a dual contribution by improving link prediction accuracy and introducing novel approaches for researchers to uncover innovation insights.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103498"},"PeriodicalIF":8.0,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144139676","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":"Federated learning system eliminating model drift in distributed edge computing: Theoretical analytics and application on pit engineering state monitoring","authors":"Zhenzhao Xia ,&nbsp;Botao Zhong ,&nbsp;Tonghui Zhao ,&nbsp;Kai Li ,&nbsp;Shuai Zhang","doi":"10.1016/j.aei.2025.103505","DOIUrl":"10.1016/j.aei.2025.103505","url":null,"abstract":"<div><div>In internet of things (IoT) with smart terminals, data is generated at edges in stream. Edge computing, where AI models are conducted on smart terminals, caters to this trend and is increasingly adopted. On levels of IoT, it places high demands on real-time performance and generalization of deployed models. Federated learning (FL) promises to meet these demands because of distributed paradigm. However, two challenges exist: (1) for incremental data on edges, mini batch and central FL training causes higher risks of data leakage; (2) drifts exist between global and local models while more difficult to reconcile in distributed edge computing. To solve challenges mentioned, measures are necessary to protect and coordinate FL process orient edge incremental data in IoT. In our study, we propose a blockchain-assisted federated learning system named BFLIoT. BFLIoT features flat, drift-resistant and safe: (1) BFLIoT decentralizes FL by blockchain and lets FL process operate fully based on edges without cloud participation; (2) BFLIoT utilizes improved federalized proximal (FedProx) algorithm and adjusts training parameters (<span><math><mrow><mi>λ</mi></mrow></math></span> and <span><math><mrow><mi>γ</mi></mrow></math></span>) to fully eliminate drift during aggregation; and (3) privacy budget (<span><math><mrow><mi>δ</mi></mrow></math></span>) for personalized differential privacy (DP) based on <span><math><mrow><mi>λ</mi></mrow></math></span>, <span><math><mrow><mi>γ</mi></mrow></math></span> is employed considering both data protection and model aggregation. We verified BFLIoT on open datasets and applied BFLIoT to pit engineering edge monitoring system (PEEMS) we developed. FL task of predictive algorithm orient monitoring items verifies BFLIoT is effective in solving two problems presented. BFLIoT provides effective and safe strategies maintaining and optimizing AI models orient distributed edge computing.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103505"},"PeriodicalIF":8.0,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144139680","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}