Engineering Applications of Artificial Intelligence最新文献

{"title":"Review of physics-informed neural networks in hemodynamics","authors":"Xianglong Yu ,&nbsp;Yu Hu ,&nbsp;Rui Guo ,&nbsp;Lei Fan ,&nbsp;Haiyan Ding ,&nbsp;Jingjing Xiao","doi":"10.1016/j.engappai.2025.112834","DOIUrl":"10.1016/j.engappai.2025.112834","url":null,"abstract":"<div><div>The circulatory system sustains physiological function through oxygen transport, nutrient delivery, and waste clearance, all of which rely on efficient blood flow. Accurate characterization and quantification of hemodynamics are essential for the diagnosis and treatment of cardiovascular diseases. However, assessing blood flow in a noninvasive and real-time manner remains a major challenge, as current imaging modalities often suffer from limited spatial and temporal resolution, while traditional computational fluid dynamics algorithms are computationally intensive and sensitive to anatomical and physiological uncertainties. Physics-informed neural networks (PINNs), combining physical laws with data-driven learning, provide a promising framework to connect computational modeling with clinical applications. In this review, we provide a comprehensive overview of recent advances in the application of PINNs to hemodynamics. We introduce theoretical foundations, highlight methodological innovations, and discuss applications in simulating blood flow under physiological and pathological conditions, as well as in estimating clinically relevant hemodynamic parameters. Importantly, our analysis highlights that PINNs achieve comparable accuracy to traditional methods while unlocking novel opportunities for patient-specific diagnosis and risk prediction. We conclude with a discussion of the benefits, current limitations, and future directions of PINNs in cardiovascular research, underscoring the transformative potential to accelerate clinical translation through interdisciplinary collaboration.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112834"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335057","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An improved graph attention network for semantic segmentation of industrial point clouds in automotive battery sealing nail defect detection","authors":"Wei Pan ,&nbsp;Yuhao Wu ,&nbsp;Wenming Tang ,&nbsp;Qinghua Lu ,&nbsp;Yunzhi Zhang","doi":"10.1016/j.engappai.2025.112793","DOIUrl":"10.1016/j.engappai.2025.112793","url":null,"abstract":"<div><div>Accurate defect detection in automotive battery sealing nails is vital for safety and reliability. Traditional methods combine two-dimensional (2D) vision for localization with three-dimensional (3D) vision for measurement, resulting in complex workflows and reduced efficiency. We propose Local Graph Attention for Semantic Segmentation (LGASS), an end-to-end 3D point cloud segmentation model. LGASS processes raw point cloud data from structured-light systems, performing simultaneous defect localization and geometric quantification in a single stage. By leveraging a graph attention mechanism in an encoder–decoder architecture, LGASS captures local geometric features and long-range dependencies, excelling on industrial metallic surfaces. Experiments show LGASS achieves 99.47% Overall Accuracy (OA), 92.37% mean Accuracy (mAcc), and 79.23% mean Intersection over Union (mIoU), offering a robust solution for automated sealing nail inspection.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112793"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145334884","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Detection and localization of false data injection attacks based on multi-scale feature fusion and attention enhancement network in smart grid","authors":"Jian Li,&nbsp;Hanting Lu,&nbsp;Qingyu Su","doi":"10.1016/j.engappai.2025.112787","DOIUrl":"10.1016/j.engappai.2025.112787","url":null,"abstract":"<div><div>This study proposes a novel framework based on the multi-scale feature fusion and attention enhancement network (MSFF-AEN) for detecting and localizing false data injection attacks (FDIAs) in smart grid. The model innovatively designs improved residual block with convolutional block attention module (CBAM) after the second convolutional layer, reducing early noise interference, and enhancing interpretability. It also incorporates a bidirectional long short-term memory network (BiLSTM) and multi-head attention (MHA) to capture temporal features and global dependencies respectively. Additionally, hierarchical feature fusion (HFF) with learnable weights optimizes and integrates multi-scale features, thereby enhancing feature representation and model interpretability. Experimental results on the IEEE 14-bus and IEEE 118-bus systems show that the proposed model outperforms existing conventional models and deep learning methods across multiple evaluation metrics, including accuracy, precision, recall, and F1-score. Particularly, the model performs exceptionally well on the large-scale IEEE 118-bus power system, achieving an accuracy of 98.73%, precision of 98.48%, recall of 97.45%, and F1-score of 97.95%. Furthermore, the model demonstrates strong robustness to various Gaussian noise conditions, maintaining high localization accuracy.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112787"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145334888","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Long time series prediction of milling force via a hybrid multi neuro-network-based algorithm","authors":"Meng Liu ,&nbsp;Hui Xie ,&nbsp;Xiangkun He ,&nbsp;Wencheng Pan ,&nbsp;Fengling Han ,&nbsp;Guangxian Li ,&nbsp;Songlin Ding","doi":"10.1016/j.engappai.2025.112805","DOIUrl":"10.1016/j.engappai.2025.112805","url":null,"abstract":"<div><div>The application of machine learning and deep learning has significantly improved the accuracy and efficiency of cutting force prediction in machining processes. However, challenges such as short prediction period, degradation in accuracy over time, and the risk of overfitting remains. These limitations collectively hinder the reliability and generalizability of artificial intelligence-based force prediction models. To address these issues, this study proposed a novel hybrid multi-neural-network algorithm that integrates convolutional neural networks, long short-time memory, and residual networks to enhance both the accuracy and duration of cutting force prediction. Prior to model training, raw force signals are pre-processed using particle swarm optimization-based variational mode decomposition to effectively eliminate noise and reduce uncertainty. The training and testing datasets are derived from milling experiments conducted under varying cutting parameters, tool types, and sensor configurations to better emulate real-world industrial conditions. Experimental results demonstrate that the hybrid model model can accurately predict cutting forces over a duration exceeding 1 s. The model's higher mean absolute error under varying test conditions suggests good robustness. The proposed data pre-processing phase contributes to a 6.38 % improvement in prediction accuracy. Furthermore, increasing the hyperparameter “timestep” helps mitigate overfitting, with only a minor trade-off in accuracy (less than 5 %). These findings demonstrate the effectiveness of the hybrid algorithm in addressing key limitations of existing models and highlight its potential for robust and generalizable prediction using AI in manufacturing applications.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112805"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145334891","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing voltage and power prediction in vibrating cylinders using machine learning algorithms: Insights from wind tunnel experiments","authors":"Amir Hossein Rabiee ,&nbsp;Mostafa Esmaeili ,&nbsp;Matin Rajabi","doi":"10.1016/j.engappai.2025.112848","DOIUrl":"10.1016/j.engappai.2025.112848","url":null,"abstract":"<div><div>Flow-induced vibrations (FIV) of circular cylinders offer a promising mechanism for low-power energy harvesting, but accurately predicting the resulting voltage and power is challenging due to the nonlinear nature of fluid–structure interactions. In this study, wind tunnel experiments were conducted to generate three datasets based on different configurations of tandem circular cylinders. The datasets were used to evaluate the performance of three machine learning regression algorithms including Support Vector Regression (SVR), Gradient Boosting Regression (GBR), and Extreme Gradient Boosting (XGBoost), in predicting the root mean square (RMS) voltage and harvested power. Sobol sensitivity analysis was applied to quantify the influence of input parameters. XGBoost showed the best performance, with R² values of 0.91, 0.98, and 0.86 for datasets 1, 2, and 3. Despite using 1000 estimators, the XGBoost model demonstrated efficient training time due to its parallel tree boosting structure and built-in regularization, offering a favorable balance between accuracy and computational complexity. Sensitivity analysis revealed that the displacement between cylinders and upstream cylinder diameter were the most influential parameters depending on the configuration. The results show that machine learning techniques, particularly XGBoost, can successfully model complex nonlinear relationships in FIV-based energy harvesting systems, providing a data-driven tool for improving design and efficiency.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112848"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145334883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"AFEformer: An adaptive frequency enhancement transformer for time series prediction","authors":"Zhiyong An,&nbsp;Lanlan Dong","doi":"10.1016/j.engappai.2025.112736","DOIUrl":"10.1016/j.engappai.2025.112736","url":null,"abstract":"<div><div>Long-term time series forecasting (LTSF), as a key research domain with pervasive applications in real-world scenarios, has garnered sustained interest from both academic and industrial communities. Although transformer-based models have demonstrated high predictive capability in capturing long-term temporal dependencies, most of them directly process raw data in the time domain while ignoring the representation of features in the frequency domain. Additionally, transformer models with frequency domain often learn weights directly but overlook frequency statistics for time series, leading to the impact of low-quality interference frequencies. Moreover, Transformer’s self-attention captures correlations solely within sequences but neglects correlations among different sequences, increasing susceptibility to overfitting. To address these issues, we innovatively design an adaptive frequency enhancement transformer (AFEformer) with temporal external attention for time series forecasting, which focuses on enhancing important frequency domain features to provide more accurate forecasting. Specifically, a frequency domain enhancement module with an adaptive threshold strategy is proposed , using frequency statistics to selectively extract key spectral components and strengthen frequency domain features. Furthermore, the temporal external attention enhancement module with Infinite Norm and dropout layer is presented to explore potential correlations between different sample sequences and mitigate overfitting. Regarding long-term forecasting, comprehensive experiments demonstrate that AFEformer achieves state-of-the-art forecasting performance on nine time series forecasting benchmarks.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112736"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145334914","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A novel road damage detection model with efficient attention and Dynamic Snake Convolution","authors":"Zhen Wang ,&nbsp;Zhengyao Ma ,&nbsp;Zhan Wang ,&nbsp;Shunqi Gao ,&nbsp;Jinjia Peng","doi":"10.1016/j.engappai.2025.112618","DOIUrl":"10.1016/j.engappai.2025.112618","url":null,"abstract":"<div><div>Road damages detection is crucial for ensuring traffic safety, optimizing maintenance costs, and extending the service life of roads. However, it faces three key challenges: (1) damages and background often have similar pixel intensities, making them hard to distinguish; (2) damage types vary greatly in shape and size, increasing the difficulty of robust feature extraction; and (3) road interferences such as water stains, shadows, or markings can easily cause false detections. To address these problems, we propose <strong>B</strong>i-level Routing Attention, <strong>S</strong>nake Convolution, and <strong>W</strong>ise Intersection over Union enhanced <strong>Y</strong>ou <strong>O</strong>nly <strong>L</strong>ook <strong>O</strong>nce version 8 (BSW-YOLO), which integrates three targeted modules into the You Only Look Once version 8 (YOLOv8) framework. First, the Bi-level Routing Attention with DropKey (BRA-DropKey) highlights true damage features and suppresses background noise, solving the similarity in pixel intensities between damages and surroundings. Second, dynamic Snake Convolution (SnakeConv) captures geometric contours for fine-grained features, improving adaptability to diverse shapes and sizes. Third, Wise Intersection over Union (Wise-IoU) loss refines anchor box quality, reducing false detections from road interferences such as water stains and shadows. Experiments conducted on the Road Damage Dataset (RDD) 2022, a benchmark dataset for road damage detection, demonstrate that BSW-YOLO achieves a mean Average Precision at an intersection over union threshold of 0.5 ([email protected]) of 90.5%, significantly outperforming other baseline models and road damage detection methods.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112618"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145334916","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A machine learning–Powered digital twin framework for adaptive management of urban air quality in Chiang Mai, Northern Thailand","authors":"Natthapong Nanthasamroeng ,&nbsp;Peerawat Luesak ,&nbsp;Rapeepan Pitakaso ,&nbsp;Surajet Khonjun ,&nbsp;Ganokgarn Jirasirilerd ,&nbsp;Surasak Matitopanum","doi":"10.1016/j.engappai.2025.112597","DOIUrl":"10.1016/j.engappai.2025.112597","url":null,"abstract":"<div><div>Urban air quality management in topographically and meteorologically complex regions such as Chiang Mai, Northern Thailand, is increasingly challenged by diverse emission sources and the limitations of conventional reactive response systems. This study introduces a multi-layer digital twin framework powered by artificial intelligence (AI), integrating real-time Internet of Things (IoT) sensing, deep learning–based spatiotemporal forecasting, and simulation-driven policy optimization to enable predictive and adaptive air quality control. Specifically, a Temporal–Spatial Graph Neural Network (TS-GNN) is employed to capture nonlinear dependencies across both spatial and temporal dimensions, achieving high predictive accuracy (root mean square error, RMSE = 2.8 μg/m³; coefficient of determination, R² = 0.96). For adaptive intervention planning, a hybrid Generative Adversarial Network–Deep Reinforcement Learning (GAN–DRL) algorithm is implemented—resulting in a 57.1 % reduction in fine particulate matter (PM2.5) concentrations, and outperforming state-of-the-art metaheuristics such as the Coot Optimization Algorithm and Red Deer Optimization. These AI-driven policy interventions are assessed through coupled agent-based modeling (ABM) and computational fluid dynamics (CFD) simulations, enabling high-fidelity, multi-source policy testing under realistic urban dynamics. The proposed framework exhibits strong scalability across spatial units, rapid inference capability, and robustness under high-pollution scenarios. Economic analysis confirms its cost-efficiency and policy feasibility. Seasonal simulations further validate sustained environmental benefits across emissions from agricultural, industrial, and transportation sectors. Overall, this work establishes an AI-enhanced, real-time decision-support paradigm combining digital twin technologies, urban simulation, and adaptive environmental governance—contributing a transferable model for data-driven urban air quality management.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112597"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Three-dimensional geometric deep learning for reaction prediction with equivariant graph transformer","authors":"Zhouxiang Wang ,&nbsp;Haicheng Yi ,&nbsp;Zhuhong You ,&nbsp;Qiangguo Jin","doi":"10.1016/j.engappai.2025.112850","DOIUrl":"10.1016/j.engappai.2025.112850","url":null,"abstract":"<div><div>Organic synthesis, a critical process in drug and material development, often involves complex reactions that can be time-consuming and costly to explore experimentally. Recent advances in machine learning have shown promise in predicting reaction outcomes, but challenges remain in capturing the full complexity of molecular interactions, particularly in three-dimensional space. To this end, we propose an Equivariant Graph Transformer (termed EGT) that predicts organic reactions by learning the three-dimensional (3D) geometric characteristics of molecules. We employed the equivariant graph neural network to extract geometric spatial information and a pairwise distance fed to position embedding to capture long-range interactions, to finely delineate the spatial structure of chemical molecules, making stereochemical information of reactions learnable. To benchmark our model's performance, we conducted reaction prediction experiments on the USPTO_STEREO and USPTO_FULL datasets as well as retrosynthesis prediction on the USPTO_50k and USPTO_MIT datasets. In addition, we conducted case studies focusing on synthesis planning and reaction prediction, and compared the results with those of human evaluations. The proposed EGT model has outperforms all existing algorithms with a Top-1 accuracy of 79.4 % for forward reaction prediction on the USPTO_STEREO dataset, and excels in predicting both forward reactions and retrosynthesis. Moreover, we demonstrated the model's capability to conduct forward total synthesis planning, showcasing its reliability and accuracy in achieving high Top-1 predictions. Molecular 3D geometry learning positions our model as a leading tool in the field of organic synthesis, paving the way for more efficient and accurate drug development.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112850"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335060","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A risk assessment framework for online transactions via Graph Neural Networks and efficient probabilistic prediction","authors":"Jicai Chang ,&nbsp;Xuejing Fu ,&nbsp;Zhen Chen ,&nbsp;Li Pan ,&nbsp;Shijun Liu","doi":"10.1016/j.engappai.2025.112766","DOIUrl":"10.1016/j.engappai.2025.112766","url":null,"abstract":"<div><div>Transactions are integral to daily life, but the occurrence of abnormal behaviors can lead to significant risks. Online transaction risk is characterized by the accumulation of abnormal behaviors, where their frequency surpasses a predefined threshold, resulting in measurable probabilities and consequences. Therefore, the assessment of online transaction risk heavily depends on probabilistic predictions of the accumulated frequency of abnormal behaviors, presenting two major challenges. Firstly, abnormal behaviors across different instances (e.g., behavior types, product categories, regions, and platforms) exhibit temporal correlations, such as co-occurrence and concomitance, which most probabilistic models fail to identify and utilize effectively. Additionally, these models do not fully address the real-time demands. To address these challenges, we propose a novel risk assessment framework based on Graph Neural Network (GNN) and probabilistic prediction, named GNN-Probformer. The framework uses Dynamic Time Warping to capture temporal correlations between abnormal behavior frequency sequences and constructs a graph structure through clustering. It then employs Graph Neural Networks to aggregate features and learn representations through a novel embedding module. A sparse self-attention mechanism and an efficient encoder–decoder architecture are incorporated to further enhance performance, while probabilistic predictions are generated through Monte Carlo sampling and cumulative distribution functions. Experimental results on a real-world dataset demonstrate that GNN-Probformer achieves substantial performance gains, with a 15% reduction in normalized deviation. At the 90th percentile, it further reduces normalized quantile loss by 15% and improves the F1-score by 16%, while also reducing training time and inference time by 47% and 38%, respectively.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"163 ","pages":"Article 112766"},"PeriodicalIF":8.0,"publicationDate":"2025-10-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145335264","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}