Advanced Engineering Informatics最新文献

{"title":"MGTN-DSI: A multi-sensor graph transfer network considering dual structural information for fault diagnosis under varying working conditions","authors":"Jianjie Liu ,&nbsp;Xianfeng Yuan ,&nbsp;Xilin Yang ,&nbsp;Weijie Zhu ,&nbsp;Yansong Zhang ,&nbsp;Tianyi Ye ,&nbsp;Xinxin Yao ,&nbsp;Fengyu Zhou","doi":"10.1016/j.aei.2025.103119","DOIUrl":"10.1016/j.aei.2025.103119","url":null,"abstract":"<div><div>In recent years, mechanical systems have increasingly integrated multiple sensors to monitor equipment status more effectively. However, extracting domain-invariant features from multi-sensor data and adapting the diagnostic model to significant variations in operating conditions remain challenging tasks. To address these issues, a novel multi-sensor graph transfer network considering dual structural information (MGTN-DSI) is designed for fault diagnosis under varying working conditions. Firstly, we develop an advanced multi-sensor feature extraction mechanism.Specifically, on the one hand, a multi-sensor collaborative fusion layer is proposed to uncover the intrinsic connections among different sensor data. On the other hand, a relationship graph between data points of the fused high-level features is constructed using a graph convolutional network with constrained filters. Secondly, a joint alignment method using a virtual discriminator is proposed to simultaneously align both subdomain and global distributions. The extensive experiments conducted on a public fault diagnosis dataset and a practical fault diagnosis test platform indicate that the proposed MGTN-DSI has higher accuracy and better generalization ability than other state-of-the-art comparison methods.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103119"},"PeriodicalIF":8.0,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143136519","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":"Graph neural network-assisted evolutionary algorithm for rapid optimization design of shear-wall structures","authors":"Yifan Fei ,&nbsp;Sizhong Qin ,&nbsp;Wenjie Liao ,&nbsp;Hong Guan ,&nbsp;Xinzheng Lu","doi":"10.1016/j.aei.2025.103129","DOIUrl":"10.1016/j.aei.2025.103129","url":null,"abstract":"<div><div>When solving expensive optimization problems (EOPs), e.g., optimization design of shear-wall structures, conventional evolutionary algorithms (EAs) face a challenge of elevated costs related to fitness evaluation. On the other hand, surrogate-assisted evolutionary algorithms (SAEAs) can effectively reduce evaluation costs and are therefore widely used. However, when encountering new structure cases to be optimized, existing SAEAs require the effort of establishing datasets and training machine-learning surrogate models from scratch, which significantly lowers their efficiency. To address this issue, a novel graph neural network (GNN)-assisted evolutionary algorithm (GAEA) is proposed, which features a distinct framework and workflow from existing SAEAs. Then, a powerful GNN surrogate model is proposed, which is based on graph representations and exhibits good generalization abilities. After being trained on a large number of cases of the same type, the model can be applied to various new cases of that particular type. By integrating GNN with EA, GAEA can directly use the trained GNN surrogate model for fitness evaluation when dealing with new structure cases. Furthermore, a distance-based evaluation and updating strategy of surrogate models is innovatively proposed, which can efficiently correct the prediction error of GNN without retraining the model. The proposed GAEA is applied to the optimization design of reinforced concrete (RC) shear-wall structures, which is a typical EOP in the field of structural engineering. Numerical experiments show that: 1) for the same number of fitness evaluations, GAEA can achieve 64.4 % to 92.5 % of the optimization outcomes of typical EA or SAEA, respectively, with only 2.8 % to 33.2 % of their computational time; 2) for the same optimization duration (10 min, 45 min, and 120 min), GAEA’s optimization outcomes are superior to those of typical EA and SAEA, meeting the rapid optimization demands for shear-wall structures in the schematic design phase. The findings of this study can be applied to the rapid optimization of various RC building structures and provide references for more efficient solutions to other EOPs.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103129"},"PeriodicalIF":8.0,"publicationDate":"2025-01-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143136931","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 dual-perspective alignment: A novel hierarchical selective adversarial network for transfer fault diagnosis","authors":"Yansong Zhang ,&nbsp;Xianfeng Yuan ,&nbsp;Xilin Yang ,&nbsp;Xinxin Yao ,&nbsp;Jianjie Liu ,&nbsp;Fengyu Zhou ,&nbsp;Peng Duan","doi":"10.1016/j.aei.2025.103113","DOIUrl":"10.1016/j.aei.2025.103113","url":null,"abstract":"<div><div>Unsupervised domain adaptation has been widely applied in rotating machinery transfer fault diagnosis, aiming to handle the absence of labeled data and distribution shift issues. However, existing methods still face with some challenges. First, they usually assume that all source domain fault samples contribute equally in domain adaptation whether they should be transferred or not, resulting in negative transfer. Second, all network parameters are considered to be the same transferable, disregarding the fact that some being domain-specific and inappropriate for fault feature distributions alignment. To cope with these issues, this paper proposes a novel hierarchical selective adversarial network (HSAN), which enhances diagnosis performance by fine-grained adaptation at both the instance-level and the parameter-level. Specifically, to suppress negative transfer due to fault samples being treated as equally easy-to-transfer, an instance selection mechanism is designed to adaptively determine transferable fault samples with high contribution in adaptation. As for the parameter-level alignment, learnable parameters are divided into domain-agnostic and domain-specific ones by an identification criterion, where the former part is transferable for adaptation, while the latter part is penalized to alleviate adaptation interference. Extensive diagnostic experiments on two public datasets and a practical fault diagnosis test rig fully demonstrate that HSAN achieves superior diagnostic performance, outperforming other state-of-the-art cross-domain methods.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103113"},"PeriodicalIF":8.0,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143136517","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":"Prediction of bearing remaining useful life based on a two-stage updated digital twin","authors":"Deqiang He ,&nbsp;Jiayang Zhao ,&nbsp;Zhenzhen Jin ,&nbsp;Chenggeng Huang ,&nbsp;Fan Zhang ,&nbsp;Jinxin Wu","doi":"10.1016/j.aei.2025.103123","DOIUrl":"10.1016/j.aei.2025.103123","url":null,"abstract":"<div><div>As a pivotal element in industrial production, bearings are vital for the smooth functioning of the system. It is essential to accurately predict the remaining useful life (RUL) of bearings. Yet, the present methods for predicting RUL do not consider the real-time health state of bearing operation, resulting in poor RUL prediction accuracy. This paper proposes a method for bearing RUL prediction, based on a two-stage updating digital twin and a dual-correlation dynamic graph convolutional network (DC-DGCN), to address the aforementioned problems. First, a bearing defect evolution model with outer ring defect expansion characteristics is established, and the initial defect expansion curve is obtained in the first stage using multi-objective optimization. This process achieves real-time interaction between the twin model and the real bearing. Then, the calibrated defects in the second stage are used to further update the full life cycle defect curve. Bi-directional Long Short-Term Memory (Bi-LSTM) is utilized to correlate the vibration characteristics of the real bearing with the twin defects to complete the real-time mapping. Finally, the mapped defects are incorporated into the feature space used for RUL prediction, allowing the proposed DC-DGCN method to extract correlations between physical and digital space features for the final prediction. The suggested method effectively increases the veracity of bearing RUL prediction, as the experimental results prove.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103123"},"PeriodicalIF":8.0,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143136929","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 spatial-temporal neural network based on ResNet-Transformer for predicting railroad broken rails","authors":"Xin Wang,&nbsp;Junyan Dai,&nbsp;Xiang Liu","doi":"10.1016/j.aei.2025.103126","DOIUrl":"10.1016/j.aei.2025.103126","url":null,"abstract":"<div><div>Broken rails are a primary factor considered in railroad capital planning investments. This paper develops a spatial–temporal neural network model based on ResNet-Transformer architecture to predict the occurrence of broken rails one year in advance. The railroad data for this research includes infrastructure data, operational data, condition-related data, and maintenance activities. First, this research captures detailed spatial correlations and temporal dependencies, ensuring that each aspect is considered for its specific impact on rail integrity. Then, utilizing the ResNet architecture, the proposed model captures spatial correlations among static rail characteristics. Subsequently, the Transformer architecture is utilized for effectively handling long-term temporal data patterns and dependencies that reflect dynamic changes over time. An experiment was conducted based on railroad data collected from one major freight railroad covering about 20,000 miles of track spanning seven years, from 2013 to 2021. AUC values of the proposed model for the training, validation, and test set are 0.84, 0.81, and 0.81, respectively, demonstrating that the model has a relatively good performance and generalizes reasonably well to the validation and test set. The results indicate that the proposed model outperforms traditional machine learning approaches such as XGBoost, especially in identifying high-risk segments. When screening 10% of the highest-risk rail segments, the model can capture 41.6% of broken rails, compared to only 33.1% detected by XGBoost and 38.0% detected by ResNet-only model. This enhanced detection capability highlights the model’s effectiveness in utilizing complex pattern recognition across both spatial and temporal data. The proposed spatial–temporal model not only aids in proactive maintenance to improve the safety and reliability of rail transportation but also contributes to more strategic capital planning in the railroad industry.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103126"},"PeriodicalIF":8.0,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143136928","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mental workload in worker-drone communication in future construction: Considering coexistence, cooperation, and collaboration interaction levels","authors":"Woei-Chyi Chang,&nbsp;Sogand Hasanzadeh","doi":"10.1016/j.aei.2025.103110","DOIUrl":"10.1016/j.aei.2025.103110","url":null,"abstract":"<div><div>While unmanned aerial vehicles (a.k.a. drones) have been recognized as potential robotic teammates that could be incorporated into the construction industry, communication between workers and drones may impose additional mental demands and workloads that could lead to workers’ mental overload on construction jobsites. To address this concern, this study examines and quantifies workers’ mental demands while communicating with drones at different human-drone interaction levels—coexistence, cooperation, and collaboration. During a futuristic bricklaying experiment wherein workers needed to communicate with drones at different interaction levels, psychophysiological sensors measured electrodermal activity, brain activation, and eye movements to assess whether the respective interactions affected workers’ mental demands. The results indicate that coexistence requires workers’ visual attention, whereas cooperation imposes affective and perceptual demands since workers were frustrated and confused when decoding and responding to messages from the drone. Moreover, higher levels of mental demands were identified in collaborative communications because sharing an object with nearby drones raised workers’ safety concerns. This research contributes to the body of knowledge by demonstrating workers experience varying dimensions of mental demands during communication with drones, and the study suggests strategies to enhance effortless worker-drone communication at coexistence, cooperation, and collaboration levels to improve worker well-being in future construction.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103110"},"PeriodicalIF":8.0,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143136518","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":"Modular Autonomous Electric Vehicle scheduling for demand-responsive transit services with modular charging strategy","authors":"Yun Yuan,&nbsp;Yitong Li,&nbsp;Xin Li","doi":"10.1016/j.aei.2025.103114","DOIUrl":"10.1016/j.aei.2025.103114","url":null,"abstract":"<div><div>Modular Autonomous Electric Vehicles (MAEV) have shown to provide in-motion transfer and flexible capacity to the demand responsive transit (DRT). However, needs-based charging strategy for the MAEV based DRT systems may reduce the utilization of the MVs during the peak hours. To address this issue, this paper proposes a mixed integer linear programming model for optimizing the route and charging planning of the DRT service, where passenger transfers are assigned to schedule partial charging time between service trips. To deal with the hard problem, an adaptive large neighbourhood search algorithm is developed. A case study regarding the real-world parameters and three numerical testing sets is conducted to show the efficiency and effectiveness of the proposed method. Results show the proposed method has 19.62 %, 12.65 % and 26.81 % reductions on the total system cost in comparison to the MAEV based DRT with the needs-based charging strategy, the comparable system considering transferring at a point, and non-transfer DRT, respectively.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103114"},"PeriodicalIF":8.0,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143136932","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 future autonomous tugs: Design and implementation of an intelligent escort control system validated by sea trials","authors":"Jialun Liu ,&nbsp;Chengqi Xu ,&nbsp;Shijie Li ,&nbsp;Zhilin Dong ,&nbsp;Jiayuan Liu ,&nbsp;Yan Zhao ,&nbsp;Bo Sun","doi":"10.1016/j.aei.2025.103116","DOIUrl":"10.1016/j.aei.2025.103116","url":null,"abstract":"<div><div>Tugs are compact yet powerful watercraft that play an important role in port operations. To enhance efficiency and guarantee operational safety, there is a growing need for the development of autonomous tugs. Aiming to assist human operations, this paper proposes the design and implementation of an intelligent escort control system for tugs, including a model-free planning module and an adaptive motion control method. The model-free planning module consists of path planning, collision avoidance, as well as an improved Logical Virtual Ship (LVS) guidance principle. The adaptive escort control method is implemented via the integrated use of a heading controller and a speed controller, both of which are constructed in a human-like strategy while based on the Model-Free Adaptive Control (MFAC) algorithm. Full-scale simulations are carried out on a marine navigation simulator to evaluate the performance of the proposed planning and control strategy. The sea trials of the escort control system are conducted successfully on a real Azimuth Stern Drive (ASD) tug, the “Jin Gang Lun 36”, at the Rizhao marine test site in China. The experimental results indicate that the system performance is comparable to that of a well-trained human operator maneuvering the tug in escort operations. This could mitigate the workload of human operators, offer important implications for the advancement of autonomous navigation, and serve as a key technology for future autonomous ships.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103116"},"PeriodicalIF":8.0,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143136933","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":"IndVisSGG: VLM-based scene graph generation for industrial spatial intelligence","authors":"Zuoxu Wang ,&nbsp;Zhijie Yan ,&nbsp;Shufei Li ,&nbsp;Jihong Liu","doi":"10.1016/j.aei.2024.103107","DOIUrl":"10.1016/j.aei.2024.103107","url":null,"abstract":"<div><div>Industrial spatial intelligence enables robots and machine tools to understand environmental settings and their relationships, allowing them to manipulate target components. A crucial aspect of this process is scene graph generation (SGG). Previous research on SGG primarily focuses on detection and panoptic segmentation of objects, followed by the prediction of their pairwise relationships. However, these approaches struggle with generalization and transferability when encountering new scenarios. To tackle this problem, we propose the <em>Industrial Visual Scene Graph Generation</em> (IndVisSGG) method, which parses spatial and interactive relationships between objects in temporal industrial settings. This approach leverages the capabilities of <em>Vision-Language Models</em> (VLMs) to generate scene graphs quickly and accurately without any additional object annotations. Furthermore, leveraging the IndVisSGG method, we have implemented a meticulous annotation procedure to compile a high-quality <em>industrial scene graph generation</em> (ISG) dataset, comprising 10,000 images of manufacturing and related industrial scenes. Through comparisons with various scene graph generation methods and benchmarks across two other datasets, we have showcased the superiority of the IndVisSGG method and underscored the benefits of the ISG dataset over existing datasets.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103107"},"PeriodicalIF":8.0,"publicationDate":"2025-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143136490","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":"Collaborative path planning of multi-unmanned surface vehicles via multi-stage constrained multi-objective optimization","authors":"Shihong Yin ,&nbsp;Ningjun Xu ,&nbsp;Zhangsong Shi ,&nbsp;Zhengrong Xiang","doi":"10.1016/j.aei.2025.103115","DOIUrl":"10.1016/j.aei.2025.103115","url":null,"abstract":"<div><div>A collaborative path planning algorithm based on a multi-stage constraint processing strategy is proposed for the task of unmanned surface vehicle (USV) cluster operation in complex water environments. The algorithm takes into account the distinct advantages of different USVs, the collaborative task time, and collision avoidance. Firstly, the objectives and constraints of the collaborative path planning problem for the USV cluster are modeled. Next, a path representation method with an adaptive number of waypoints is designed to improve the smoothness of the USV paths. Subsequently, a multi-stage constrained multi-objective optimization (MSCMO) algorithm is proposed to deal with the cooperative time and collision avoidance constraints of the USV cluster through a multi-stage strategy. Finally, eight collaborative operation scenarios for the USV cluster are designed to verify the performance of MSCMO. The simulation results demonstrate that MSCMO outperforms seven state-of-the-art constrained multi-objective algorithms, exhibiting a strong competitive advantage and superior overall performance. MSCMO enables USV clusters to perform collaborative tasks faster, safer, and smoother without violating any maneuvering constraints, while providing a variety of trade-off solutions for decision-makers. The source code is available at <span><span>https://github.com/Shihong-Yin/MSCMO-MUCP</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103115"},"PeriodicalIF":8.0,"publicationDate":"2025-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143136812","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}