Pub Date : 2025-02-01DOI: 10.1016/j.engappai.2024.109838
Abdullah Ayub Khan , Jing Yang , Asif Ali Laghari , Abdullah M. Baqasah , Roobaea Alroobaea , Chin Soon Ku , Roohallah Alizadehsani , U. Rajendra Acharya , Lip Yee Por
The rapid adoption of Augmented Intelligence of Things (AIoT) in enterprise management (EM) presents significant challenges in securely managing and exchanging information. This study introduces a blockchain-based platform, BAIoT-EMS, designed to enhance security and efficiency in AIoT-enabled EM systems. The platform leverages a consortium network and InterPlanetary File Storage (IPFS) for secure storage and transaction management, supported by smart contracts to automate and safeguard processes like device registration. A novel multi-proof-of-work consensus mechanism is implemented to analyze, validate, and verify AIoT transactions while minimizing resource consumption. Simulation results demonstrate a 63.51% improvement in performance and an 11.75% reduction in computational power usage, highlighting the effectiveness of the proposed framework.
{"title":"BAIoT-EMS: Consortium network for small-medium enterprises management system with blockchain and augmented intelligence of things","authors":"Abdullah Ayub Khan , Jing Yang , Asif Ali Laghari , Abdullah M. Baqasah , Roobaea Alroobaea , Chin Soon Ku , Roohallah Alizadehsani , U. Rajendra Acharya , Lip Yee Por","doi":"10.1016/j.engappai.2024.109838","DOIUrl":"10.1016/j.engappai.2024.109838","url":null,"abstract":"<div><div>The rapid adoption of Augmented Intelligence of Things (AIoT) in enterprise management (EM) presents significant challenges in securely managing and exchanging information. This study introduces a blockchain-based platform, BAIoT-EMS, designed to enhance security and efficiency in AIoT-enabled EM systems. The platform leverages a consortium network and InterPlanetary File Storage (IPFS) for secure storage and transaction management, supported by smart contracts to automate and safeguard processes like device registration. A novel multi-proof-of-work consensus mechanism is implemented to analyze, validate, and verify AIoT transactions while minimizing resource consumption. Simulation results demonstrate a 63.51% improvement in performance and an 11.75% reduction in computational power usage, highlighting the effectiveness of the proposed framework.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"141 ","pages":"Article 109838"},"PeriodicalIF":7.5,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143179688","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}
Pub Date : 2025-02-01DOI: 10.1016/j.engappai.2024.109788
Lóránd Vatamány, Siamak Mehrkanoon
Accurate precipitation nowcasting is crucial for applications such as flood prediction, disaster management, agriculture optimization, and transportation management. While many studies have approached this task using sequence-to-sequence models, most focus on single regions, ignoring correlations between disjoint areas. We reformulate precipitation nowcasting as a spatiotemporal graph sequence problem. Specifically, we propose Graph Dual-stream Convolutional Attention Fusion, a novel extension of the graph attention network. Our model’s dual-stream design employs distinct attention mechanisms for spatial and temporal interactions, capturing their unique dynamics. A gated fusion module integrates both streams, leveraging spatial and temporal information for improved predictive accuracy. Additionally, our framework enhances graph attention by directly processing three-dimensional tensors within graph nodes, removing the need for reshaping. This capability enables handling complex, high-dimensional data and exploiting higher-order correlations between data dimensions. Depthwise-separable convolutions are also incorporated to refine local feature extraction and efficiently manage high-dimensional inputs. We evaluate our model using seven years of precipitation data from Copernicus Climate Change Services, covering Europe and neighboring regions. Experimental results demonstrate superior performance of our approach compared to other models. Moreover, visualizations of seasonal spatial and temporal attention scores provide insights into the most significant connections between regions and time steps.
{"title":"Graph Dual-stream Convolutional Attention Fusion for precipitation nowcasting","authors":"Lóránd Vatamány, Siamak Mehrkanoon","doi":"10.1016/j.engappai.2024.109788","DOIUrl":"10.1016/j.engappai.2024.109788","url":null,"abstract":"<div><div>Accurate precipitation nowcasting is crucial for applications such as flood prediction, disaster management, agriculture optimization, and transportation management. While many studies have approached this task using sequence-to-sequence models, most focus on single regions, ignoring correlations between disjoint areas. We reformulate precipitation nowcasting as a spatiotemporal graph sequence problem. Specifically, we propose Graph Dual-stream Convolutional Attention Fusion, a novel extension of the graph attention network. Our model’s dual-stream design employs distinct attention mechanisms for spatial and temporal interactions, capturing their unique dynamics. A gated fusion module integrates both streams, leveraging spatial and temporal information for improved predictive accuracy. Additionally, our framework enhances graph attention by directly processing three-dimensional tensors within graph nodes, removing the need for reshaping. This capability enables handling complex, high-dimensional data and exploiting higher-order correlations between data dimensions. Depthwise-separable convolutions are also incorporated to refine local feature extraction and efficiently manage high-dimensional inputs. We evaluate our model using seven years of precipitation data from Copernicus Climate Change Services, covering Europe and neighboring regions. Experimental results demonstrate superior performance of our approach compared to other models. Moreover, visualizations of seasonal spatial and temporal attention scores provide insights into the most significant connections between regions and time steps.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"141 ","pages":"Article 109788"},"PeriodicalIF":7.5,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143180396","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-02-01DOI: 10.1016/j.engappai.2024.109817
Liming Xu , Yanrong Lei , Jie Shao , Xianhua Zeng , Weisheng Li
Magnetic resonance imaging (MRI) is commonly used both in clinical diagnosis and scientific research. Owing to the high cost, time constraints, and limited application of multi-contrast MRI images obtained from metallic implants, it incurs low throughput and misses a specific modality. Medical image cross-modal synthesis based on Artificial Intelligence (AI) technologies is proposed to synthesize the desired missing modal images. However, it still suffers from low expandability, invisible latent representations, and poor interpretability. We thus propose modal disentanglement generative adversarial networks for bidirectional T1-weighted (T1-w) and T1-weighted (T2-w) medical image synthesis with controllable cross-modal synthesis and disentangled interpretability. Firstly, we construct a cross-modal synthesis model to achieve bidirectional generation between T1-w and T2-w MRI images, which can be easily extended for adaptive modality synthesis without training multiple generators and discriminators. Then, we use an easily trained deep network to disentangle deep representations in latent space and map representations in latent space into pixel space to visualize morphological images and yield multi-contrast MRI images with controllable feature generation. Besides, we construct an easy-to-interpret deep structure by incorporating morphology consistency to preserve edge contours and visualize deep representations in latent space to enable interpretability, which is critical for artificial intelligence oriented to engineering applications and clinical diagnostics. The experiments demonstrate that ours outperforms recent state-of-the-art methods with average improvements of 15.8% structural similarity (SSIM), 12.7% multiscale structural similarity (MSIM), 38.2% peak signal-to-noise ratio (PSNR) and 5.2% visual information fidelity (VIF) on benchmark datasets.
{"title":"Modal disentangled generative adversarial networks for bidirectional magnetic resonance image synthesis","authors":"Liming Xu , Yanrong Lei , Jie Shao , Xianhua Zeng , Weisheng Li","doi":"10.1016/j.engappai.2024.109817","DOIUrl":"10.1016/j.engappai.2024.109817","url":null,"abstract":"<div><div>Magnetic resonance imaging (MRI) is commonly used both in clinical diagnosis and scientific research. Owing to the high cost, time constraints, and limited application of multi-contrast MRI images obtained from metallic implants, it incurs low throughput and misses a specific modality. Medical image cross-modal synthesis based on Artificial Intelligence (AI) technologies is proposed to synthesize the desired missing modal images. However, it still suffers from low expandability, invisible latent representations, and poor interpretability. We thus propose modal disentanglement generative adversarial networks for bidirectional T1-weighted (T1-w) and T1-weighted (T2-w) medical image synthesis with controllable cross-modal synthesis and disentangled interpretability. Firstly, we construct a cross-modal synthesis model to achieve bidirectional generation between T1-w and T2-w MRI images, which can be easily extended for adaptive modality synthesis without training multiple generators and discriminators. Then, we use an easily trained deep network to disentangle deep representations in latent space and map representations in latent space into pixel space to visualize morphological images and yield multi-contrast MRI images with controllable feature generation. Besides, we construct an easy-to-interpret deep structure by incorporating morphology consistency to preserve edge contours and visualize deep representations in latent space to enable interpretability, which is critical for artificial intelligence oriented to engineering applications and clinical diagnostics. The experiments demonstrate that ours outperforms recent state-of-the-art methods with average improvements of 15.8% structural similarity (SSIM), 12.7% multiscale structural similarity (MSIM), 38.2% peak signal-to-noise ratio (PSNR) and 5.2% visual information fidelity (VIF) on benchmark datasets.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"141 ","pages":"Article 109817"},"PeriodicalIF":7.5,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143177915","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}
Electric vehicles (EVs) powered by high-energy batteries are anticipated to be a primary avenue for achieving energy decarbonization in future societies. However, the high energy density of lithium batteries poses significant safety risks under complex conditions and sudden environmental changes. Thus, safety and fault issues of high-energy batteries during vehicle operation have gained much attention. This study proposes an intelligent fault diagnosis algorithm based on spiking neural networks (SNNs) using a data-driven approach. Time series data are stacked and fed into the SNN for learning, capturing temporal characteristics and forming a stable kernel representation. The predicted output is compared with the actual output to generate a residual signal for fault diagnosis. The algorithm is validated on a battery management experimental platform with injected internal and external short circuit faults. Results show the algorithm effectively and swiftly detects abnormal signals and has some transferability.
{"title":"Data-driven spiking neural networks for intelligent fault detection in vehicle lithium-ion battery systems","authors":"Penghao Wu , Engang Tian , Hongfeng Tao , Yiyang Chen","doi":"10.1016/j.engappai.2024.109756","DOIUrl":"10.1016/j.engappai.2024.109756","url":null,"abstract":"<div><div>Electric vehicles (EVs) powered by high-energy batteries are anticipated to be a primary avenue for achieving energy decarbonization in future societies. However, the high energy density of lithium batteries poses significant safety risks under complex conditions and sudden environmental changes. Thus, safety and fault issues of high-energy batteries during vehicle operation have gained much attention. This study proposes an intelligent fault diagnosis algorithm based on spiking neural networks (SNNs) using a data-driven approach. Time series data are stacked and fed into the SNN for learning, capturing temporal characteristics and forming a stable kernel representation. The predicted output is compared with the actual output to generate a residual signal for fault diagnosis. The algorithm is validated on a battery management experimental platform with injected internal and external short circuit faults. Results show the algorithm effectively and swiftly detects abnormal signals and has some transferability.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"141 ","pages":"Article 109756"},"PeriodicalIF":7.5,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143177931","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}
Pub Date : 2025-02-01DOI: 10.1016/j.engappai.2024.109670
Alejandro Lozano , David Cruz-Ortiz , Mariana Ballesteros , Isaac Chairez
This work presents the design and implementation of a reference path generator for a virtual version of a robotic orthosis based on a semi-parametric model of a neuromusculoskeletal system. The proposed generator is used to regulate the movements of the mentioned virtual orthosis (VO) as a preliminary stage in designing rehabilitation strategies. The generator considers a differential neural network (DNN) identifier, which predicts the angular positions and velocities of specific articulations in the upper limb using the raw electromyographic (EMG) signals as input. The DNN-based model is validated using experimental data from the elbow joint and the Biceps Brachii muscle collected from ten healthy participants. To regulate the movements of the VO, a controller based on the sliding mode considers the motion restrictions of the articulation of the extremity intervened by the orthosis. The proposed controller guarantees that the VO reaches a set point following a reference path related to the rEMG while the motion constraints are satisfied. The functionality of the proposed path generator was tested with a VO device. The results showed that the VO followed the angular movements of the elbow. All these results confirm the applicability of the proposed semi-parametric path generator based on a DNN.
{"title":"Upper limb musculoskeletal model as path generator for control a virtual orthosis: A dynamic neural network approach","authors":"Alejandro Lozano , David Cruz-Ortiz , Mariana Ballesteros , Isaac Chairez","doi":"10.1016/j.engappai.2024.109670","DOIUrl":"10.1016/j.engappai.2024.109670","url":null,"abstract":"<div><div>This work presents the design and implementation of a reference path generator for a virtual version of a robotic orthosis based on a semi-parametric model of a neuromusculoskeletal system. The proposed generator is used to regulate the movements of the mentioned virtual orthosis (VO) as a preliminary stage in designing rehabilitation strategies. The generator considers a differential neural network (DNN) identifier, which predicts the angular positions and velocities of specific articulations in the upper limb using the raw electromyographic (EMG) signals as input. The DNN-based model is validated using experimental data from the elbow joint and the Biceps Brachii muscle collected from ten healthy participants. To regulate the movements of the VO, a controller based on the sliding mode considers the motion restrictions of the articulation of the extremity intervened by the orthosis. The proposed controller guarantees that the VO reaches a set point following a reference path related to the rEMG while the motion constraints are satisfied. The functionality of the proposed path generator was tested with a VO device. The results showed that the VO followed the angular movements of the elbow. All these results confirm the applicability of the proposed semi-parametric path generator based on a DNN.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"141 ","pages":"Article 109670"},"PeriodicalIF":7.5,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143178515","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}
Pub Date : 2025-02-01DOI: 10.1016/j.engappai.2024.109787
Xiaochi Ma , Zongxin Huo , Jian Lu , Yiik Diew Wong
Freeway crash risk prediction is a critical component of traffic safety management, yet existing crash risk models often fail to capture complex driving behaviors and lack interpretability. This study introduces a novel freeway crash risk prediction framework based on the Deep Forest (DF) algorithm, considering the detailed risky driving behavior data. The DF model integrates multi-grained scanning and cascade forest layers, enabling it to capture the complex relationship between risky driving behavior features. SHapley Additive Explanations (SHAP) are applied to interpret the model's predictions, including both SHAP summary and interaction results. Additionally, ablation studies are conducted to evaluate the contributions of key components like multi-grained scanning and cascade structures to the model's performance. The experimental results demonstrate that the DF model outperforms traditional machine learning models. The DF model achieves an area under the receiver operating characteristic curve of 0.825, with a balanced Sensitivity of 0.75 and Specificity of 0.816, surpassing other models. The ablation studies show that removing multi-grained scanning, cascade layers, or completely random tree forest leads to performance declines, confirming the importance of each component. The SHAP analysis highlights that sharp acceleration and braking behaviors have the most significant impact on crash risk, offering clear, interpretable insights into how driving behaviors contribute to risk. Overall, the DF model's superior performance and SHAP-based interpretability provide a powerful tool for traffic safety management. These findings emphasize the value of incorporating both driving behavior intensity and model interpretability into crash risk prediction, offering practical applications for reducing crash rates.
{"title":"Deep Forest with SHapley additive explanations on detailed risky driving behavior data for freeway crash risk prediction","authors":"Xiaochi Ma , Zongxin Huo , Jian Lu , Yiik Diew Wong","doi":"10.1016/j.engappai.2024.109787","DOIUrl":"10.1016/j.engappai.2024.109787","url":null,"abstract":"<div><div>Freeway crash risk prediction is a critical component of traffic safety management, yet existing crash risk models often fail to capture complex driving behaviors and lack interpretability. This study introduces a novel freeway crash risk prediction framework based on the Deep Forest (DF) algorithm, considering the detailed risky driving behavior data. The DF model integrates multi-grained scanning and cascade forest layers, enabling it to capture the complex relationship between risky driving behavior features. SHapley Additive Explanations (SHAP) are applied to interpret the model's predictions, including both SHAP summary and interaction results. Additionally, ablation studies are conducted to evaluate the contributions of key components like multi-grained scanning and cascade structures to the model's performance. The experimental results demonstrate that the DF model outperforms traditional machine learning models. The DF model achieves an area under the receiver operating characteristic curve of 0.825, with a balanced Sensitivity of 0.75 and Specificity of 0.816, surpassing other models. The ablation studies show that removing multi-grained scanning, cascade layers, or completely random tree forest leads to performance declines, confirming the importance of each component. The SHAP analysis highlights that sharp acceleration and braking behaviors have the most significant impact on crash risk, offering clear, interpretable insights into how driving behaviors contribute to risk. Overall, the DF model's superior performance and SHAP-based interpretability provide a powerful tool for traffic safety management. These findings emphasize the value of incorporating both driving behavior intensity and model interpretability into crash risk prediction, offering practical applications for reducing crash rates.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"141 ","pages":"Article 109787"},"PeriodicalIF":7.5,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143179353","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}
Pub Date : 2025-02-01DOI: 10.1016/j.engappai.2024.109768
Ran Tai, Dantong Ouyang, Weiting Liu, Luyu Jiang, Liming Zhang
Model-based diagnosis (MBD) with multiple abnormal observations utilizes inconsistencies between actual and expected observations of systems to localize system faults. Current state-of-the-art algorithms still require solvers to consider a substantial number of suspected faulty components. To address this challenge, we introduce the Dual Principles with Decision Node (DPDN) algorithm. The first part of DPDN consists of two innovative principles: the Output Dependency Judgment Principle (ODJP) and the Deep Propagation Dependency Principle (DPDP). These principles are designed to identify and categorize a larger portion of components as ‘normal’, thereby broadening the idea of filtered nodes. By increasing the number of components classified as ‘normal’, the diagnostic process becomes more efficient as fewer components need to be diagnosed. The second part of DPDN integrates a newly defined decision node guided by its Decision Node Principle (DNP). This decision node, along with its corresponding principle, further bolsters the diagnostic process by classifying additional components as normal. With DPDN, complex real-world systems can reduce the number of components considered during the diagnostic process by eliminating those that are functioning normally, thereby decreasing the time required to obtain diagnoses. We conduct comparative experiments to evaluate the efficiency of our algorithm against other prevalent methods. Empirical results distinctly underscore DPDN’s superior performance in relation to other state-of-the-art algorithms.
{"title":"A novel approach to model-based diagnosis with multiple observations","authors":"Ran Tai, Dantong Ouyang, Weiting Liu, Luyu Jiang, Liming Zhang","doi":"10.1016/j.engappai.2024.109768","DOIUrl":"10.1016/j.engappai.2024.109768","url":null,"abstract":"<div><div>Model-based diagnosis (MBD) with multiple abnormal observations utilizes inconsistencies between actual and expected observations of systems to localize system faults. Current state-of-the-art algorithms still require solvers to consider a substantial number of suspected faulty components. To address this challenge, we introduce the Dual Principles with Decision Node (DPDN) algorithm. The first part of DPDN consists of two innovative principles: the Output Dependency Judgment Principle (ODJP) and the Deep Propagation Dependency Principle (DPDP). These principles are designed to identify and categorize a larger portion of components as ‘normal’, thereby broadening the idea of filtered nodes. By increasing the number of components classified as ‘normal’, the diagnostic process becomes more efficient as fewer components need to be diagnosed. The second part of DPDN integrates a newly defined decision node guided by its Decision Node Principle (DNP). This decision node, along with its corresponding principle, further bolsters the diagnostic process by classifying additional components as normal. With DPDN, complex real-world systems can reduce the number of components considered during the diagnostic process by eliminating those that are functioning normally, thereby decreasing the time required to obtain diagnoses. We conduct comparative experiments to evaluate the efficiency of our algorithm against other prevalent methods. Empirical results distinctly underscore DPDN’s superior performance in relation to other state-of-the-art algorithms.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"141 ","pages":"Article 109768"},"PeriodicalIF":7.5,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143179676","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}
Pub Date : 2025-02-01DOI: 10.1016/j.engappai.2024.109785
Li Zhang , Lin Fan , Jianjun Liu , Dingyu Jiao , Yuxuan He , Jing Zhou , Karine Zeitouni , Huai Su , Jinjun Zhang
Considering that the existing methods lack spatial and temporal information mining of pipeline multidimensional operation data, it is unable to accurately recognize the unsteady operation conditions among pipeline stations. In this study, a dynamic graph convolutional network classification model is proposed for the recognition of unsteady operating states in multi-product pipeline systems. Firstly, dynamic graph convolutional network of multi-pipeline system (DPipeNet) is constructed based on the visibility graph algorithm, mutual information and long and short-term memory network model. Secondly, static graph convolutional network of multi-pipeline system (SPipeNet) is constructed by using the real geographic location information of each station of multi-pipeline. Then, the input subgraph of the graph convolutional network is used to construct the multi-pipeline system operational state relationship network (OSRN), and the vulnerable state nodes of the system are evaluated using complex network centrality metrics. Finally, the proposed model is applied to real operational data of a multi-pipeline system in China. The results show that in the two-classification scenario, both DPipeNet and SPipeNet have higher accuracies, but DPipeNet has a lower missed rate. In the multi-classification scenario, DPipeNet has the highest precision, which can reach more than 85%, and the recall rate is improved by 13%–25% compared with the neural network models in recent literature and SPipeNet. In the vulnerability analysis scenario, the intermediate station pump startup/stoppage of multi-pipeline has higher vulnerability. The proposed method also provides decision support for managers in pipeline system operation and maintenance management.
{"title":"A dynamic graph convolutional network-based framework for the unsteady operating states recognition of multi-product pipeline systems","authors":"Li Zhang , Lin Fan , Jianjun Liu , Dingyu Jiao , Yuxuan He , Jing Zhou , Karine Zeitouni , Huai Su , Jinjun Zhang","doi":"10.1016/j.engappai.2024.109785","DOIUrl":"10.1016/j.engappai.2024.109785","url":null,"abstract":"<div><div>Considering that the existing methods lack spatial and temporal information mining of pipeline multidimensional operation data, it is unable to accurately recognize the unsteady operation conditions among pipeline stations. In this study, a dynamic graph convolutional network classification model is proposed for the recognition of unsteady operating states in multi-product pipeline systems. Firstly, dynamic graph convolutional network of multi-pipeline system (DPipeNet) is constructed based on the visibility graph algorithm, mutual information and long and short-term memory network model. Secondly, static graph convolutional network of multi-pipeline system (SPipeNet) is constructed by using the real geographic location information of each station of multi-pipeline. Then, the input subgraph of the graph convolutional network is used to construct the multi-pipeline system operational state relationship network (OSRN), and the vulnerable state nodes of the system are evaluated using complex network centrality metrics. Finally, the proposed model is applied to real operational data of a multi-pipeline system in China. The results show that in the two-classification scenario, both DPipeNet and SPipeNet have higher accuracies, but DPipeNet has a lower missed rate. In the multi-classification scenario, DPipeNet has the highest precision, which can reach more than 85%, and the recall rate is improved by 13%–25% compared with the neural network models in recent literature and SPipeNet. In the vulnerability analysis scenario, the intermediate station pump startup/stoppage of multi-pipeline has higher vulnerability. The proposed method also provides decision support for managers in pipeline system operation and maintenance management.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"141 ","pages":"Article 109785"},"PeriodicalIF":7.5,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143179677","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}
Pub Date : 2025-02-01DOI: 10.1016/j.engappai.2024.109751
Abderrahim Ezzine, Naouar Laaidi, Hassan Satori
In this study, we explored the impact of traffic noise on speech recognition performance for the Moroccan Fessi dialect at varying noise levels. To achieve this, we developed a highly configurable, open-source speech recognition platform, designed to identify optimal parameters for handling the dialect’s unique characteristics. Our approach employs a stochastic hidden Markov model combined with Gaussian mixture models, specifically tailored for the low-resource Moroccan dialect. The novelty of this approach lies in its ability to enable parameter learning and adaptation with minimal prior knowledge of the dialect’s linguistic rules. We further analyzed the noise performance of the system in relation to the syllabic structure of words, assessing its effect on automatic speech recognition accuracy. Experimental results show a substantial degradation in recognition performance under noisy conditions, with a word error rate of 96.95% in noisy environments compared to 8.14% in clean settings. Additionally, we observed that specific syllables in the Moroccan dialect significantly influence the recognition system’s performance.
{"title":"Automatic speech recognition for Moroccan dialect in noisy traffic environments","authors":"Abderrahim Ezzine, Naouar Laaidi, Hassan Satori","doi":"10.1016/j.engappai.2024.109751","DOIUrl":"10.1016/j.engappai.2024.109751","url":null,"abstract":"<div><div>In this study, we explored the impact of traffic noise on speech recognition performance for the Moroccan Fessi dialect at varying noise levels. To achieve this, we developed a highly configurable, open-source speech recognition platform, designed to identify optimal parameters for handling the dialect’s unique characteristics. Our approach employs a stochastic hidden Markov model combined with Gaussian mixture models, specifically tailored for the low-resource Moroccan dialect. The novelty of this approach lies in its ability to enable parameter learning and adaptation with minimal prior knowledge of the dialect’s linguistic rules. We further analyzed the noise performance of the system in relation to the syllabic structure of words, assessing its effect on automatic speech recognition accuracy. Experimental results show a substantial degradation in recognition performance under noisy conditions, with a word error rate of 96.95% in noisy environments compared to 8.14% in clean settings. Additionally, we observed that specific syllables in the Moroccan dialect significantly influence the recognition system’s performance.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"141 ","pages":"Article 109751"},"PeriodicalIF":7.5,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143179678","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}
Pub Date : 2025-02-01DOI: 10.1016/j.engappai.2024.109790
Nan Yang, Yong Wang, Lei Zhang, Bin Jiang
Point cloud data provides rich three-dimensional spatial information. Accurate three-dimensional point cloud semantic segmentation algorithms enhance environmental understanding and perception, with wide-ranging applications in autonomous driving and scene analysis. However, Graph Neural Networks often struggle to retain semantic relationships among neighboring points during feature extraction, potentially leading to the omission of critical features during aggregation. To address these challenges, we propose a novel network, the Feature-Enhanced Residual Attention Network. This network includes an innovative graph convolution module, the Neighborhood-Enhanced Convolutional Aggregation Module, which utilizes K-Nearest Neighbor and Dilated K-Nearest Neighbor techniques to construct diverse dynamic graphs and aggregate features, thereby prioritizing essential information. This approach significantly enhances the expressiveness and generalization capabilities of the network. Additionally, we introduce a new spatial attention module designed to capture semantic relationships among points. Experimental results demonstrate that the Feature-Enhanced Residual Attention Network outperforms benchmark models, achieving an average intersection ratio of 61.3% and an overall accuracy of 86.7% on the Stanford Large-Scale Three-dimensional Indoor Spaces dataset, thereby significantly improving semantic segmentation performance.
{"title":"Point cloud semantic segmentation network based on graph convolution and attention mechanism","authors":"Nan Yang, Yong Wang, Lei Zhang, Bin Jiang","doi":"10.1016/j.engappai.2024.109790","DOIUrl":"10.1016/j.engappai.2024.109790","url":null,"abstract":"<div><div>Point cloud data provides rich three-dimensional spatial information. Accurate three-dimensional point cloud semantic segmentation algorithms enhance environmental understanding and perception, with wide-ranging applications in autonomous driving and scene analysis. However, Graph Neural Networks often struggle to retain semantic relationships among neighboring points during feature extraction, potentially leading to the omission of critical features during aggregation. To address these challenges, we propose a novel network, the Feature-Enhanced Residual Attention Network. This network includes an innovative graph convolution module, the Neighborhood-Enhanced Convolutional Aggregation Module, which utilizes K-Nearest Neighbor and Dilated K-Nearest Neighbor techniques to construct diverse dynamic graphs and aggregate features, thereby prioritizing essential information. This approach significantly enhances the expressiveness and generalization capabilities of the network. Additionally, we introduce a new spatial attention module designed to capture semantic relationships among points. Experimental results demonstrate that the Feature-Enhanced Residual Attention Network outperforms benchmark models, achieving an average intersection ratio of 61.3% and an overall accuracy of 86.7% on the Stanford Large-Scale Three-dimensional Indoor Spaces dataset, thereby significantly improving semantic segmentation performance.</div></div>","PeriodicalId":50523,"journal":{"name":"Engineering Applications of Artificial Intelligence","volume":"141 ","pages":"Article 109790"},"PeriodicalIF":7.5,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143177930","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}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号