Advanced Engineering Informatics最新文献

{"title":"Edge-cloud cooperation driven surface roughness classification method for selective laser melting","authors":"Shuaiyin Ma ,&nbsp;Yuming Huang ,&nbsp;Yang Liu ,&nbsp;Zhiqiang Yan ,&nbsp;Jingxiang Lv ,&nbsp;Wei Cai","doi":"10.1016/j.aei.2025.103473","DOIUrl":"10.1016/j.aei.2025.103473","url":null,"abstract":"<div><div>Additive manufacturing (AM) technology is extensively utilized in aerospace and industrial manufacturing. However, parts built using AM are susceptible to spheroidization, porosity, cracks, and poor surface quality, making it difficult to establish an actionable product quality degree. Hence, developing a reasonable method to equate product quality with a new degree and further analyzing the product quality based on these standards has proven effective for enhancing part quality in AM. To achieve this goal, this paper proposes a surface roughness classification method that utilizes surface roughness analysis and sample enhancement. This method leverages edge cloud cooperation to efficiently analyze and integrate data from different sensors, enabling real-time monitoring and adjustment of the manufacturing process. Subsequently, the quality degree analysis system was developed utilizing matter-element extension cloud model. Furthermore, a bidirectional-gated recurrent unit (Bi-GRU) based model for quality classification and recognition has been established, with Wasserstein generative adversarial network (WGAN) employed for sample enhancement to address the issue of imbalanced column classification and to enhance the accuracy of both classification and recognition. Finally, the results obtained from this case study demonstrate through comparative experiments that the proposed method for classifying surface roughness can accurately identify 98% of prepared samples.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103473"},"PeriodicalIF":8.0,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144123480","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":"Smoothed dynamic scheduling of aircraft engines with off-site warehouse","authors":"Wenzhao Dong ,&nbsp;Gangyan Xu ,&nbsp;Xuanyu Zhang ,&nbsp;Xuan Qiu","doi":"10.1016/j.aei.2025.103451","DOIUrl":"10.1016/j.aei.2025.103451","url":null,"abstract":"<div><div>Efficient aircraft engine scheduling is vital for the reliable operation of flights. However, due to the highly uncertain engine maintenance demands and the special requirement in its logistics operations, airlines are suffering from high cost in maintaining an efficient aircraft engine logistics system. Motivated by real-life problems, this work aims to develop a novel smoothed dynamic scheduling method for aircraft engine to decrease its cost and improve its ability in coping with uncertainties. Specifically, this work coordinates the decisions in different steps and develops an integrated engine scheduling model. Based on which, a dynamic scheduling method is proposed that could cope with both emergency demand and changes of regular demand with integrated rolling-horizon and event-driven method. Besides, the concept of switching cost is introduced to propose smoothed dynamic decisions, which could trade off between plan changes and cost savings. Furthermore, this work investigates the effect of setting an off-site warehouse for the performance of engine scheduling in terms of cost savings. Extensive experiments have been conducted to verify the proposed method, and several managerial implications are also proposed.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103451"},"PeriodicalIF":8.0,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144131264","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":"Intelligent reconstruction of supersonic combustion flow fields using sparse sensor data and heuristic-guided learning","authors":"Xue Deng ,&nbsp;Ye Tian ,&nbsp;Maotao Yang ,&nbsp;Erda Chen ,&nbsp;Jiaweng Deng ,&nbsp;Hua Zhang","doi":"10.1016/j.aei.2025.103486","DOIUrl":"10.1016/j.aei.2025.103486","url":null,"abstract":"<div><div>Reconstructing the global combustion flow state under supersonic combustion conditions using sparse sensor data presents a novel and significant challenge. This challenge is critical for advancing measurement and diagnostic technologies in supersonic vehicles operating in complex and extreme environments. However, the task becomes notably tricky when the number of sensors is minimal. Current end-to-end learning models often face significant generalization issues when reconstructing combustion flow across the entire spatial domain, especially in the context of sparse pressure measurement systems typically used in ground wind tunnel tests. To this end, this study introduces an intelligent reconstruction model for combustion flow that incorporates heuristic-guided learning, enabling accurate flow field reconstruction across the entire spatial domain using sparse pressure sensor measurements. The proposed model operates in two distinct stages. The first phase is a heuristic learning phase, which uses the multi-gradient learning strategy to learn the joint feature distribution of different scales of the flow field in the whole spatial domain based on the sparse pressure measurement data. In the second stage, the guided learning stage, a multi-scale feature fusion mechanism is applied to refine both the content and structural details of the coarse and fine distribution features within the joint feature distribution. The efficacy of the proposed model is validated using a ground test dataset collected at various Mach numbers. Experimental results demonstrate that the model achieves superior performance across various complex and extreme scenarios.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103486"},"PeriodicalIF":8.0,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144123477","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":"LIO-BIM – Coupling lidar inertial odometry with building information modeling for robot localization and mapping","authors":"Jan Stührenberg,&nbsp;Kay Smarsly","doi":"10.1016/j.aei.2025.103477","DOIUrl":"10.1016/j.aei.2025.103477","url":null,"abstract":"<div><div>Mobile robots deployed to automate tasks in the construction industry require accurate robot localization and navigation. Building information modeling (BIM) is increasingly prevalent, and BIM models are interpretable by robots to help navigate in or around buildings. Most approaches towards robot localization through BIM models solely rely on maps derived from the BIM models requiring models with a high level of development (LOD) and the accurate modeling of non-structural objects. In practice, however, BIM models often employ a limited LOD and non-structural objects, if modeled, may appear in different locations, which may result in scan-BIM deviations and thus in localization errors. This paper presents the so called “LIO-BIM” framework, which couples lidar inertial odometry (LIO) and BIM for robust mobile robot localization and mapping, using 3D lidar to overcome the issues related to scan-BIM deviations. LIO-BIM builds upon simultaneous localization and mapping techniques and performs scan matching multiple times, i.e. (i) scan matching of the latest lidar scan with lidar scans previously recorded to maintain an accurate map of the environment, and (ii) scan matching of a local map around the robot with a BIM model to enable localization and mapping relative to the BIM model. The maps may be used at run-time, e.g., for construction progress monitoring or quality inspection. The framework, whose code is provided as open source, is implemented on a quadruped robot equipped with a 3D lidar, an inertial measurement unit, and a camera, and it is validated in a cluttered indoor office environment represented by a BIM model. Furthermore, the framework is validated on the ConSLAM dataset showcasing a cluttered construction site environment. As a result, the validation tests demonstrate accurate and robust 3D localization and mapping aligned with BIM models in real-time.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103477"},"PeriodicalIF":8.0,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144123481","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":"Dynamic compensation approach for mitigating vibration interference in 3D point cloud data of electrical equipment","authors":"Yuxiang Leng,&nbsp;Youyuan Wang,&nbsp;Xinyuan Hong","doi":"10.1016/j.aei.2025.103493","DOIUrl":"10.1016/j.aei.2025.103493","url":null,"abstract":"<div><div>The quality and precision of three-dimensional point cloud data for electrical equipment are adversely affected by vibrations caused by various operational factors. In this study, we developed a dynamic compensation method to address vibration-induced interference in point cloud data of electrical equipment. First, we devised a coupled vibration response model to integrate vibration response components into the coordinates, generating a static reference point cloud for subsequent analysis. We compared this static reference point cloud with on-site data and analyzed amplitude offset effects to extract the direction axis and displacement error. Next, we derived a depth curve and its associated index sequence from the identified direction axis, enabling precise characterization of vibration effects. Finally, we employed one-dimensional bilateral filtering and triple exponential smoothing to isolate and predict vibration amplitudes. Using these predictions, we optimized the point cloud coordinates to achieve robust compensation for vibration interference. Experimental results show that our method improves smoothness by over 50% across diverse datasets. For precision assessment, we measured the diameter of electrical equipment cables. Our method reduces the standard deviation of these measurements by 70.3%; by contrast, existing methods achieve 32.4% at most. The proposed method also offers better precision and robustness in visual perception tasks.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103493"},"PeriodicalIF":8.0,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144123476","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 intelligent traffic anomaly detection system based on self-supervised learning and self-powered hybrid nano-sensor","authors":"Zheng Fang ,&nbsp;Lingji Kong ,&nbsp;Hongyu Chen ,&nbsp;Xinyi Zhao ,&nbsp;Yuan Wang ,&nbsp;Chengliang Fan ,&nbsp;Zutao Zhang ,&nbsp;Luyao Bai","doi":"10.1016/j.aei.2025.103461","DOIUrl":"10.1016/j.aei.2025.103461","url":null,"abstract":"<div><div>Transportation is a foundational element of global economic development, with vehicle anomaly detection playing a pivotal role in ensuring safety and operational efficiency. Conventional anomaly detection techniques predominantly rely on battery-powered sensors and labor-intensive manual data labeling, thereby limiting both scalability and efficiency. This study introduces a time series-oriented self-supervised framework integrated with a triboelectric-electromagnetic nanosensor (TENS) to form an Anomaly Detection System (ADS). ADS represents a paradigm shift by converting vehicle vibrations into electrical energy and enabling self-supervised anomaly detection through the temporal reconstruction of vibrational characteristics. The proposed framework automatically generates pseudo-labeled datasets using a deep autoencoder, which subsequently trains LSTM-based classifiers without the need for manual labeling. Experimental results demonstrate that TENS achieves a peak RMS power density of 81.97 W/m³. As a self-powered sensor, it effectively detects vibrations without external energy inputs, maintains stable features over more than 100,000 cycles, and has the potential to power third-party sensors. In empirical evaluations involving vehicles, Autonomous Rail Rapid Transit (ART), and bicycles, ADS achieved an average anomaly detection accuracy of 97.15 %. Compared to methods employing only unsupervised reconstruction, ADS improved accuracy by 17.81 % to 60.14 % and also surpassed self-supervised approaches based on 1D-CNN. When deployed in vehicular contexts, ADS further demonstrated robust generalization and self-supervised anomaly detection capabilities. The seamless integration of hybrid nanosensor technology with an advanced self-supervised learning framework illustrates how sustainable energy solutions can synergize with cutting-edge artificial intelligence to advance intelligent transportation systems and predictive maintenance strategies.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103461"},"PeriodicalIF":8.0,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144123478","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-efficient and general-purpose hand–eye calibration method for robotic systems using next best view","authors":"Shuming Yi ,&nbsp;Sichao Liu ,&nbsp;Xiaoyu Lin ,&nbsp;Sijie Yan ,&nbsp;Xi Vincent Wang ,&nbsp;Lihui Wang","doi":"10.1016/j.aei.2025.103432","DOIUrl":"10.1016/j.aei.2025.103432","url":null,"abstract":"<div><div>Calibration between robots and cameras is critical in automated robot vision systems. However, conventional manually conducted image-based calibration techniques are often limited by their accuracy sensitivity and poor adaptability to dynamic or unstructured environments. These approaches present challenges for ease of calibration and automatic deployment while being susceptible to rigid assumptions that degrade their performance. To close these limitations, this study proposes a data-efficient vision-driven approach for fast, accurate, and robust hand–eye camera calibration, and it aims to maximise the efficiency of robots in obtaining hand–eye calibration images without compromising accuracy. By analysing the previously captured images, the minimisation of the residual Jacobian matrix is utilised to predict the next optimal pose for robot calibration. A method to adjust the camera poses in dynamic environments is proposed to achieve efficient and robust hand–eye calibration. It requires fewer images, reduces dependence on manual expertise, and ensures repeatability. The proposed method is tested using experiments with actual industrial robots. The results demonstrate that our NBV strategy reduces rotational error by 8.8%, translational error by 26.4%, and the number of sampling frames by 25% compared to artificial sampling. The experimental results show that the average prediction time per frame is 3.26 seconds.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103432"},"PeriodicalIF":8.0,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144123479","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":"Novel data-to-image method for heating ventilation and air conditioning fault detection and diagnosis in the built world","authors":"Jian Bi ,&nbsp;Hua Wang ,&nbsp;Afshin Afshari ,&nbsp;Ke Yan","doi":"10.1016/j.aei.2025.103469","DOIUrl":"10.1016/j.aei.2025.103469","url":null,"abstract":"<div><div>Tabular data obtained from the heating ventilation and air conditioning (HVAC) systems is not directly suitable for analysis with image-based deep learning models such as convolutional neural networks (CNNs), limiting the usage of advanced deep neural networks in the built world. To overcome this challenge, a novel feature association graph (FAG) generation method is proposed to convert HVAC tabular data into images automatically. In FAG, each feature is converted to a grid in the image, with the feature values conveyed through the grid’s grayscale. The proposed method fully considers the correlation between features, allowing the grouping of highly correlated features. This aids the CNN in extracting the feature interactions from the image. Besides, an image grid adaptation algorithm is proposed to resolve coordinate conflicts, ensuring that features are densely distributed in the two-dimensional grid. Three popular deep CNN-based fault detection and diagnosis (FDD) models were implemented and their performance was evaluated. To verify the generalizability of FAG, three real-world HVAC system FDD datasets were used. The results show that the proposed FAG, in combination with CNN models, achieves superior overall FDD performance compared to existing tabular data-to-image methods. The proposed attention residual neural network trained on FAG outperforms traditional machine learning models typically used for processing tabular data.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103469"},"PeriodicalIF":8.0,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144123475","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":"Reinforcement learning control for systems with unknown coupling induced by the compensator","authors":"Saige Cheng,&nbsp;Yonggang Li,&nbsp;Kai Wang,&nbsp;Chunhua Yang","doi":"10.1016/j.aei.2025.103449","DOIUrl":"10.1016/j.aei.2025.103449","url":null,"abstract":"<div><div>Compensator is a primary method for overcoming external disturbances in industrial processes. However, in certain cases, compensator may cause coupling between the controller and the disturbance, thereby exacerbating the disturbance variation. The effects of this coupling are difficult to observe independently, so compensator design is challenging. To this end, a reinforcement learning (RL)-based correlation learning for controller and compensator method is proposed to link the learning processes of the feedback controller and the compensator. Different from existing methods, the compensator design problem for coupled unknown problems is solved with a lower dimensional exploration space. First, the coupling effect is estimated by zero input response and added to the feedback controller as an auxiliary variable to be learned. Then, an augmented system is established using the Lyapunov difference method to describe the relationship between the compensator and the auxiliary variable. Finally, the Actor-Critic RL method is used to solve the remaining compensator design problem under uncertain coupling effects, and the convergence condition of the algorithm is given. The method has been verified on the mechanism model of industrial alumina evaporation process.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103449"},"PeriodicalIF":8.0,"publicationDate":"2025-05-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144116558","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":"Tailored vision-language framework for automated hazard identification and report generation in construction sites","authors":"Qihua Chen,&nbsp;Xianfei Yin","doi":"10.1016/j.aei.2025.103478","DOIUrl":"10.1016/j.aei.2025.103478","url":null,"abstract":"<div><div>Timely, comprehensive, and accurate identification of construction hazards is essential for mitigating the accident risk. Automated hazard identification via computer vision has advanced beyond traditional inspection methods but struggles with the dynamic complexity of construction environments, leading to limitations in identifying various hazard categories and generating detailed hazard reports. To address these issues, this study proposes an innovative framework comprising an advanced Vision-Language Model (VLM)-empowered construction hazard identifier, ChatCH, and an end-to-end method for generating construction hazard reports. A dedicated Construction Hazard Dataset (CHD) containing 1,308 real construction hazard images across 32 fine-grained categories was developed for validation purposes. Experimental results show that ChatCH, fine-tuned with the pre-trained VLM Qwen2-VL-7B, achieves a precision of 89.4%, outperforming the pre-trained Qwen2-VL-7B by 43.5% and the traditional pre-trained VLM CLIP by 83.9%. Additionally, ChatCH demonstrates strong few-shot learning capabilities and robustness. Moreover, the end-to-end method for construction hazard report generation can automatically produce structured and detailed hazard reports. This framework provides an innovative solution for construction safety management, enhancing efficiency, accuracy, and automation in construction hazard identification.</div></div>","PeriodicalId":50941,"journal":{"name":"Advanced Engineering Informatics","volume":"66 ","pages":"Article 103478"},"PeriodicalIF":8.0,"publicationDate":"2025-05-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144106587","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}