Advanced Engineering Informatics最新文献

{"title":"Continual contrastive reinforcement learning: Towards stronger agent for environment-aware fault diagnosis of aero-engines through long-term optimization under highly imbalance scenarios","authors":"Haoze Wu ,&nbsp;Shisheng Zhong ,&nbsp;Minghang Zhao ,&nbsp;Xuyun Fu ,&nbsp;Yongjian Zhang ,&nbsp;Song Fu","doi":"10.1016/j.aei.2025.103297","DOIUrl":"10.1016/j.aei.2025.103297","url":null,"abstract":"<div><div>Although the stability of aero-engines is high, their failures can lead to catastrophic consequences. Due to the infrequent nature of faults, traditional data-driven fault diagnosis methods rely on limited amounts of historical failure data for training classification models. They cannot update models on time in response to environmental changes and data growth. To address the issue, this paper proposes a new machine learning method, i.e., Continual Contrastive Reinforcement Learning (CCRL), that integrates environmental interaction and continual dynamic evolution for fault diagnosis of aero-engine under conditions of high imbalance and continually growing data. First, the operating environment of the airline is treated as the learning environment for the agent. The aircraft’s flight data is used as the state information provided by the environment, while the failure identification results from ground personnel and experts serve as the labels for this state information. This framework ensures the agent can continually learn in the face of increasing data volumes. Next, a contrastive learning encoder for highly imbalanced scenarios is designed, where a large number of normal samples are used to train an encoder that constructs positive and negative sample pairs with actual data, fine-tuning the encoder to improve its ability to distinguish different faults. Finally, the contrastive learning encoder is embedded into the enhanced learning model, enabling the agent to better perceive environmental changes and diagnose failures under highly imbalanced scenarios. This paper conducts a series of contrastive and ablation experiments using real data, which fully validate the application potential of the proposed method.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103297"},"PeriodicalIF":8.0,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143714650","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":"Physics-informed neural network for load sway prediction in travelling autonomous mobile cranes","authors":"Zhuomin Zhou ,&nbsp;Brandon Johns ,&nbsp;Yihai Fang ,&nbsp;Yu Bai ,&nbsp;Elahe Abdi","doi":"10.1016/j.aei.2025.103269","DOIUrl":"10.1016/j.aei.2025.103269","url":null,"abstract":"<div><div>Excessive load sway is a critical safety concern during crane operations, exposing cranes to risks of instability and collision with surrounding objects. Existing methods for predicting load sway struggle with inefficiency and inaccuracy. Advances in robotics and automation have led to the robotisation of cranes, enhancing both safety and efficiency. This paper proposed a physics-informed neural network (PINN) for predicting the load sway of autonomous mobile cranes (AMCs) in base-moving conditions, and introduced a transfer learning (TL) framework to address complexities in AMC dynamics while reducing the need for extensive training data. Initially trained on numerically simulated data with simplified dynamics, the PINN was subsequently fine-tuned using real-world data, which included realistic dynamic uncertainties and complexities. Numerical simulations and laboratory experiments were conducted to validate the PINN’s performance. The proposed PINN accurately predicted payload motion and maintained robust performance in both numerical simulations and laboratory experiments while exhibiting superior computational efficiency, requiring only 12.5% of the time needed by traditional dynamic models for 1 s prediction windows. Furthermore, it was compared and outperformed other machine learning models, including recurrent neural networks (RNN), long short-term memory (LSTM) networks and multilayer perception (MLP). These findings indicate that the proposed PINN provides a robust and efficient solution for sensorless load sway prediction in crane operations.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103269"},"PeriodicalIF":8.0,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143704916","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":"SLDAE: An interpretable stacked Denoising Auto-Encoders for fan fault diagnosis on steelmaking workshops","authors":"Xiaoqiang Liao ,&nbsp;Dong Wang ,&nbsp;Siqi Qiu ,&nbsp;Min Xia ,&nbsp;Xinguo Ming","doi":"10.1016/j.aei.2025.103260","DOIUrl":"10.1016/j.aei.2025.103260","url":null,"abstract":"<div><div>Fault diagnosis of the fan in steelmaking shops has significant practical value in guaranteeing smooth operation and achieving quality control of steel coils. The existing data-driven systems, mainly deep neural network-based diagnostic models, have achieved some success in recognizing fan faults. However, these models face a significant challenge in reaching reliable diagnostic conclusions with uncertainty due to the absence of interpretable representations between fault labels and features. To address this issue, this paper develops and evaluates an interpretable model, namely Stacked Logic Denoising Auto-Encoders (SLDAE). SLDAE is a flexible neural-symbolic system that extracts confidence rules and Binary Decision Logic (BDL) rules to explain how SDAE conducts feature learning, and conducts uncertain reasoning of diagnostic decision-making. To extract confidence rules, a logic grouping DAE is designed to consider the impact of different literals on neuron activation so as to reduce information loss. To extract BDL rules, we design a novel network structure, Discrete Logic Networks (DLNs), to facilitate extracting implicit relationships between fault features and labels while learning and representing the belief of BDL rules. Experiments verified on two fan datasets indicate that SLDAE can perform quantitative reasoning of uncertain diagnostic logic and exhibits a notable performance in fault recognition.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103260"},"PeriodicalIF":8.0,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143705690","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 data-driven metric-based proper orthogonal decomposition method with Shapley Additive Explanations for aerodynamic shape inverse design optimization","authors":"Chenliang Zhang ,&nbsp;Hongbo Chen ,&nbsp;Xiaoyu Xu ,&nbsp;Yanhui Duan ,&nbsp;Guangxue Wang","doi":"10.1016/j.aei.2025.103277","DOIUrl":"10.1016/j.aei.2025.103277","url":null,"abstract":"<div><div>In the present study, an effective optimization framework of aerodynamic shape inverse design is established based on the data-driven metric-based proper orthogonal decomposition (DMPOD) method. This framework employs a DMPOD method that filters superior data sets and POD bases using a data-driven filtering strategy with Shapley Additive Explanations (SHAP) and a modified application criterion for bases. The efficiency of the framework is improved by reduced design variables and narrowed design space which both benefit from the DMPOD method. In the DMPOD method, the data-driven filtering strategy effectively filters superior data sets, addressing the limitations of traditional methods, while geometric approximation-based sample generation method enhances dynamic change capture during the optimization process. In addition, the modified application criterion for bases selects important bases based on their relevance to the objective function and determines the necessary quantity. The effectiveness, efficiency, and robustness of the optimization framework are validated by the inverse design case of the RAE2822 airfoil. The results show that the optimization framework with DMPOD effectively enhances optimization efficiency and robustness, with improvements of 31.32% and 84.89% for superior data set, respectively, compared to the MPOD method, and have a better optimization effect.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103277"},"PeriodicalIF":8.0,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143705691","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":"User needs insights from UGC based on large language model","authors":"Wei Wei,&nbsp;Chenliang Hao,&nbsp;Zixin Wang","doi":"10.1016/j.aei.2025.103268","DOIUrl":"10.1016/j.aei.2025.103268","url":null,"abstract":"<div><div>With limited resources, it is critical for companies to understand and address user needs to gain a competitive edge.The methods that utilize large-scale user-generated content (UGC) produced by the internet can analyze user needs efficiently and accurately. However, these methods have not been extensively studied.This paper proposes a framework based on large language model (LLM) to extract user’s insights into the priority of product attributes. First, product attributes are extracted from user reviews using LLM. Then, the mapping network between user reviews and satisfaction is established through sentiment analysis based on the LLM and Multi-layer Perceptron (MLP) classification. Finally, a comprehensive analysis of product importance is conducted using a proposed quantified IPA-Kano model. An empirical study on smart wearable bands is conducted to offer an intuitive and quantifiable analysis of user attention and satisfaction for each product attribute. The strengths and weaknesses of the products are highlighted, providing valuable insights that can inspire companies to adopt user-centric optimization strategies.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103268"},"PeriodicalIF":8.0,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143705692","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":"Integrating crack pattern entropy measures with synthesized learners for accumulated seismic damage evaluation in reinforced concrete frames","authors":"Mostafa Kaboodkhani,&nbsp;Mohammadjavad Hamidia,&nbsp;Hamid Bayesteh","doi":"10.1016/j.aei.2025.103271","DOIUrl":"10.1016/j.aei.2025.103271","url":null,"abstract":"<div><div>Reliable decision-making for reinforced concrete buildings affected by earthquakes relies on realistically measuring accumulated seismic damage. The limitations and uncertainties of vision-based qualitative subjective inspection highlight the necessity of quantitative assessment approaches. In this paper, an integrated method is developed for post-earthquake inspection of reinforced concrete buildings, using crack texture entropy quantification and the well-known cumulative Park-Ang damage quantification model. A comprehensive database comprising 969 crack textures is collected by the authors from cyclic-tested beam-column sub-assemblages. For beam, column, and joint crack textures, the multi-scale pixel-based Renyi entropy measures are computed through box-counting, and results are then linked to the associated experimentally computed structural accumulated damage. The dissipated energy weight, <em>β</em>, in the Park-Ang damage index is calculated by collapse state response for all specimens. Contrary to simplified speculations and error-prone <em>β</em> equations, this strategy controlled the damage index divergence from 1 at failure. The influential crack image-related measures are detected by correlation analysis and also by non-linear permutation feature importance. Two alternative scenarios are organized for the intelligent inspection of structures according to post-earthquake circumstances based on structural information accessibility/inaccessibility. Soft machine learning-based algorithms, encompassing eight diverse-structured techniques, are synthesized with Aquila optimizer to improve efficiency and generalizability. In the first alternative, the Gaussian process regression provided satisfactory prediction with the coefficient of determination equal to 0.87 despite lacking structural information merely using crack image-based measures. This score reached 0.89 by adding structural parameters to the first alternative inputs in the Categorical Boosting model. The slight improvement in the accuracy of the second alternative demonstrates crack pattern adequacy in the quantitative inspection. Afterward, the seismic damage limits of crack pattern entropy measures are extracted using the model explanation, and the efficiency of the proposed approach is proved by assessing new crack images.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103271"},"PeriodicalIF":8.0,"publicationDate":"2025-03-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143682341","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":"Construction regulatory document digitalization with layout knowledge-informed object detection and semantic text recognition","authors":"Shuyi Wang ,&nbsp;Seonghyeon Moon ,&nbsp;Yuguang Fu ,&nbsp;Jinwoo Kim","doi":"10.1016/j.aei.2025.103278","DOIUrl":"10.1016/j.aei.2025.103278","url":null,"abstract":"<div><div>Construction documents, containing extensive project information, are often stored and shared in unstructured paper formats, leading to inefficiencies in retrieval and transfer among stakeholders. There has been a pressing need for digitalizing construction documents by converting Portable Document Format documents into machine-readable, structured texts. However, current optical character recognition technologies struggle with complex layouts commonly found in construction project documents. To address this issue, we propose a construction document digitalization approach integrated with layout knowledge-informed object detection and semantic text recognition, improving recognition accuracy across various layouts and preserving the structural integrity of texts. Results show that our approach can reduce the average word error rate by 5.6 %p with the assistance of layout knowledge and achieve a structural similarity of 78.8 %, while achieving 87.4 % mAP@50 for layout analysis. These findings highlight the positive impacts of layout knowledge on digitalizing construction documents and underscore the practical viability of our approach.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103278"},"PeriodicalIF":8.0,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143682340","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":"mKGMPP: A multi-layer knowledge graph integration framework and its inference method for manufacturing process planning","authors":"Zechuan Huang ,&nbsp;Xin Guo ,&nbsp;Chong Jiang ,&nbsp;Mingyue Yang ,&nbsp;Hao Xue ,&nbsp;Wu Zhao ,&nbsp;Jie Wang","doi":"10.1016/j.aei.2025.103266","DOIUrl":"10.1016/j.aei.2025.103266","url":null,"abstract":"<div><div>Manufacturing process planning is the process of organizing the production steps based on product design. It aimed at determining the process routes and formulating resource allocation strategies in response to digital model. In the process of connecting product design and manufacturing, designers and manufacturing technicians focus on different aspects of the digital model. This leads to a distortion when manufacturing technicians transform the digital model information into process information. As a result, this results in a deviation in the mapping between design intent and process intent. Such deviations can lead to disconnections in the association of process knowledge, undermine the consistency and traceability between process documents and digital models. Therefore, this study proposes a multi-layer knowledge graph for manufacturing process planning(mKGMPP) and an interactive manufacturing process planning system (IMPP system) driven by digital model and the proposed knowledge framework. The historical process schemes are analyzed using a dual-dimensional approach based on text and digital model. An extraction strategy based on structured data parsing and intelligent agent processing is employed for textual knowledge extraction. For geometric feature knowledge, the OpenCV library is employed, along with Gaussian blur, morphological operations, and the Canny detection algorithm. The intra knowledge of process schemes is integrated using the 4M1E elements, and multi-dimensional relationships between process schemes are established based on TQCSE. The geometric similarity inference module, process scheme inference module, and processing content modification module have been developed, and an interactive interface has been built based on Gradio. A manufacturing process planning of a slender shaft in the aerospace domain validates the rationality of the proposed knowledge organization framework and knowledge inference method.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103266"},"PeriodicalIF":8.0,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143682339","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":"Geometric spatial constraints network for slender and tiny surface defect detection","authors":"Chenghan Pu ,&nbsp;Jun Wang ,&nbsp;Yuan Zhang ,&nbsp;Muyuan Niu ,&nbsp;Qiaoyun Wu ,&nbsp;Ziyu Lin","doi":"10.1016/j.aei.2025.103138","DOIUrl":"10.1016/j.aei.2025.103138","url":null,"abstract":"<div><div>Detecting defects on aircraft impeller surfaces is challenging due to the thin and fragile structure of certain defects, as well as their varying scale and geometry. To address these two challenges, we propose the Geometric Spatial Constraints Network (GSCNet) for precise impeller defect detection. First, we develop an automatic image acquisition equipment to capture high-quality data of impeller surface defects. Subsequently, we introduce GSCNet, which comprises two main components: Rich Semantic Information Representation (RSIR) and Spatial Correlation Awareness (SCA) to detect surface defects. Within RSIR, we propose a geometric-constraints-guided, deformable-convolution-based module named Slender Partial Convolution (SPC), along with a Multi-Geometric Feature Fusion (MFF) module. SPC captures the features of tubular structures without redundant information by aligning the convolution kernel shape with slender defects, while MFF facilitates the fusion of various semantic features, thereby enhancing the ability to extract semantic information. In SCA, we introduce a novel attention mechanism that captures inherent spatial correlation to enhance the high-similarity defects classification capability by modeling representative spatial information. Finally, we design a similarity-enhanced loss function to further improve the detection of multiple geometric defects simultaneously, as it alleviates the scale sensitivity of IoU-based loss. Comparative experiments demonstrate that our framework outperforms all representative detection models, achieving 83.2% mAP on the AISD dataset, which surpasses the second-best model by 3.8%. The first set of ablation experiments confirms the effectiveness of each module within the framework. The second set of ablation experiments on the NEU-SEG and MT datasets validates the feature extraction and plug-and-play capability of RSIR. The generalization ability of GSCNet is further demonstrated on the NEU-DET and GC10-DET datasets.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103138"},"PeriodicalIF":8.0,"publicationDate":"2025-03-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143682338","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":"PA-WSDIS: A prior-aware weakly supervised defect instance segmentation model for car body surface","authors":"Yike He ,&nbsp;Yueming Wang ,&nbsp;Weiwei Jiang ,&nbsp;Songyu Hu ,&nbsp;Jianzhong Fu","doi":"10.1016/j.aei.2025.103254","DOIUrl":"10.1016/j.aei.2025.103254","url":null,"abstract":"<div><div>Car body surface defect instance segmentation is essential for ensuring product quality and setting precise size thresholds for defects during product inspection process. However, few defect instance segmentation applications has been found in industrial scenarios until now. This is due in large part to the fact that the pixel-level annotation of defects is cumbersome and labor-intensive. Although various weakly supervised methods have shown promising results, they usually lack the ability to fully explore prior information and the awareness of hierarchical semantic correlations, thereby limiting the defect instance segmentation performance. To address this issue, we propose a novel prior-aware weakly supervised defect instance segmentation (PA-WSDIS) model for car body surface, removing the need for pixel-level labeling. First, we design a box-driven coarse mask generator to obtain coarse masks, which serve as potential proposals for the subsequent refinement process. Then, we propose a boundary guided prior constraint loss, consisting of boundary alignment and pixel-pair similarity mining losses, to fully leverage prior information to enhance the discriminative ability and provide reliable refinement guidance for the model. Finally, we propose a correlative semantic calibration loss, which comprehensively perceives the rich semantic features of different dimensions from both local and global perspectives. With the collaborative constraints of these meticulously designed loss functions, precise instance segmentation results are achieved. Experimental results showcase the outstanding performance of the PA-WSDIS model with an impressive 87.4% <span><math><msubsup><mrow><mi>mAP</mi></mrow><mrow><mn>50</mn></mrow><mrow><mi>mask</mi></mrow></msubsup></math></span>, which is considerably superior to state-of-the-art methods. As far as we know, our proposed method is the first weakly supervised instance segmentation model based on bounding box labels for industrial defect detection tasks.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"65 ","pages":"Article 103254"},"PeriodicalIF":8.0,"publicationDate":"2025-03-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143682410","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}