Search

This paper presents a neural network predictive control scheme for studying the coagulation process of wastewater treatment in a paper mill. A multi-layer back-propagation neural network is employed to model the nonlinear relationships between the removal rates of pollutants and the chemical dosages, in order to adapt the system to a variety of operating conditions and acquire a more flexible learning ability. The system includes a neural network emulator of the reaction process, a neural network controller, and an optimization procedure based on a performance function that is used to identify desired control inputs. The gradient descent algorithm method is used to realize the optimization procedure. The results indicate that reasonable forecasting and control performances have been achieved through the developed system. [Copyright &y& Elsevier]

Artificial intelligence (AI)-driven fault diagnosis methods are crucial for ensuring rotating machinery's safety and effective operation. The success of most current methods relies on the assumption that sufficient high-quality labeled datasets can be obtained for model training. However, in real-world industrial scenarios, obtaining such datasets is difficult or nearly impossible, thereby hindering the practical implementation of these methods. The integration of virtual modeling and transfer learning offers a powerful approach to meet the above challenge. Abundant virtual data of different fault categories can be acquired in the virtual space with highly flexible and at a low cost, and transfer learning can enhance the practical utility of these virtual data for contributing to the construction of diagnosis models. Therefore, this paper proposes a digital twin-driven partial domain fault diagnosis method based on unlabeled physical data and labeled virtual data. First, a virtual model of rotating machinery is built to generate labeled virtual fault data with enough fault types. Then, an adversarial transfer learning network is developed to leverage the effective knowledge from the virtual and physical data. Meanwhile, a weighting learning module is introduced to reduce the negative effect caused by the redundant fault categories in the virtual space. Finally, the proposed digital twin-driven transfer learning network is trained with the labeled virtual data and unlabeled physical data. Experiments on a light truck transmission system demonstrate that the proposed method achieves satisfactory diagnostic performance even without labeled physical fault data, contributing to the advancement of AI engineering applications. [ABSTRACT FROM AUTHOR]

A new document management system is proposed in this paper. Its kernel is based on a new set of neuro-fuzzy systems of the ART family: FasArt and RFasArt. The first one, FasArt, is used to support a simple Optical Character Recognition (OCR) that inherits fine properties of ART architectures, such as fast and incremental learning, stability and modularity. On the other hand, RFasArt is a new recurrent version of FasArt which efficiently exploits contextual information in the task of logical labeling. The proposed system is extensively tested in two real-world applications, i.e. E-mail of printed business letter and digital library of scientific papers. Experimental results show logical labeling and OCR rates over 90%. The proposed system is better compared to a previous system proposed by the group, where instead of using contextual information in an integrated way, a postprocessing Viterbi-based model was employed. [Copyright &y& Elsevier]

This study investigates the penetration game strategy of the high dynamic vehicle against high-velocity interceptors in environments with multiple static no-fly zones. The primary issue addressed is the deficiency in control precision and the inadequacy of control margin under conditions of complex multi-constraint coupling. Firstly, an enhanced artificial potential field method is devised for the lateral penetration guidance strategy of high dynamic vehicles, which includes a predictive repulsion potential field, a buffer zone and new potential field functions. This approach not only averts trajectory oscillations caused by heading judgment ambiguity in the tangent direction of the obstacle area, but also significantly mitigates the inherent conflict between obstacle avoidance and target reachability. Secondly, considering the potential failure of the lateral penetration guidance strategy due to the high-velocity maneuvering of interceptors and detection sensor errors of the high dynamic vehicle, this paper initially designs a Kalman filter to denoise the detection information and provide a single-step optimal estimate. Subsequently, a multi-step state predictor based on the Transformer network is proposed, which obtains its future multi-step early warning information from the denoised detection historical data and refines it based on three-dimensional geometry knowledge. Then, the combination of the filtered estimate and the refined early warning information substantially enhances the success rate of the high dynamic vehicle in game confrontations with the high-velocity interceptors. Lastly, the numerical simulations are conducted to verify the effectiveness and performance of the penetration game guidance strategy. • This paper proposes a penetration game strategy for high dynamic vehicles in complex confrontation environments. • The proposed guidance method addresses the issues of insufficient control margin and inadequate control accuracy in the guidance systems. • An enhanced artificial potential field method is used to develop the lateral guidance strategy for high dynamic vehicles. • This paper develops a model-free predictor that combines a Kalman filter with a Transformer to handle time-series flight trajectories. [ABSTRACT FROM AUTHOR]

Natural gas pipeline leakage is a common safety hazard, which can have a great impact on the economy and the environment. This paper proposed a novel manifold learning-enabled feature extraction method for natural gas pipeline leakage diagnosis. Firstly, the natural gas pipeline working condition signal is decomposed and denoised by Variational mode decomposition (VMD). Secondly, the denoised pipeline signals were constructed into a form expressed by the Symmetric positive definite matrix (SPD) using the VMD reconstruction technique, and the geodesic distance measurement method was applied to the SPD matrix to make the data located on the SPD manifold. Then feature extraction is carried out by Local linear embedding (LLE) method based on asymmetric distance. Finally, pattern recognition of the features extracted in this paper by Support vector machine (SVM) can achieve 100% recognition accuracy. By enabling faster and more accurate leak detection, the method minimizes gas loss, as well as mitigating the environmental risks caused by this potent greenhouse gas. [ABSTRACT FROM AUTHOR]

Visual localization is the front-end part of visual simultaneous localization and mapping (VSLAM), and also the basis of environmental perception. Accurate visual localization can improve the map construction of complex dynamic environment, which can make robots and other carriers intelligent. To solve the visual localization problem in complex dynamic environment and overcome the localization error caused by the interference of moving targets, this paper proposes an efficient point and line feature combination method to locate the key features of static and dynamic regions. Firstly, the method uses batch frames to solve the motion compensation in dynamic environment, and optimizes the re-projection error of batch frames to locate low-speed and high-speed moving targets. Then a method of dividing the key feature regions of moving objects is proposed, and the dynamic gradient descent function is introduced to detect the point features of the key feature regions of moving objects. Finally, the edge of the dynamic target is expanded, and the key part of the static region is located by line features. The comparison results show that the point and line feature combination method proposed in this paper can be effectively applied to low-speed and high-speed dynamic scenes, and can accurately locate dynamic objects, with fast real-time performance. [ABSTRACT FROM AUTHOR]

The existing learning-based image dehazing methods usually adopt the encoder–decoder architecture with convolutional neural networks to estimate latent haze-free images from hazy images. However, the limited receptive field of convolutional neural networks may not effectively capture structure-level information, causing the model to be unable to the haze density. To solve this problem, this paper proposes a bi-decoder structure with a dense non-pooling encoder to enhance the structural features that are closely related to the haze density. Compared with conventional methods, the main advantage of our method is the integration of an additional coarse decoder in the encoder–decoder architecture, where a hybrid feature convolution (HFC) block is utilized to enlarge the receptive field to extract the structure of the image. Besides the difference in the network, the inputs of the fine and coarse decoders are non-pooling and pooling respectively. Moreover, a multi-scale feature attention (MSFA) module is proposed to selectively enhance the haze-relevant feature outputs of fine and coarse decoders. Experimental results on synthetic and real-world datasets demonstrate that the proposed method outperforms most state-of-the-art methods in terms of image quality and quantitative metrics. Especially in the NH-HAZE dataset, its PSNR exceeds other methods by more than 2.13 dB. In the end, this paper applies this dehazing technology to object detection. The code of this paper and data are available online at https://github.com/Qiaoyu-K/Bi-Decoder-Dehazing. [ABSTRACT FROM AUTHOR]

Image super-resolution reconstruction (SR) is an important ill-posed problem in low-level vision, which aims to reconstruct high-resolution images from low-resolution images. Although current state-of-the-art methods exhibit impressive performance, their recovery of image detail information and edge information is still unsatisfactory. To address this problem, this paper proposes a shifted window Transformer (Swin Transformer) with an efficient channel attention network (S-ECAN), which combines the attention based on convolutional neural networks and the self-attention of the Swin Transformer to combine the advantages of both and focuses on learning high-frequency features of images. In addition, to solve the problem of Convolutional Neural Network (CNN) based channel attention consumes a large number of parameters to achieve good performance, this paper proposes the Efficient Channel Attention Block (ECAB), which only involves a handful of parameters while bringing clear performance gain. Extensive experimental validation shows that the proposed model can recover more high-frequency details and texture information. The model is validated on Set5, Set14, B100, Urban100, and Manga109 datasets, where it outperforms the state-of-the-art methods by 0.03–0.13 dB, 0.04–0.09 dB, 0.01–0.06 dB, 0.13–0.20 dB, and 0.06–0.17 dB respectively in terms of objective metrics. Ultimately, the substantial performance gains and enhanced visual results over prior arts validate the effectiveness and competitiveness of our proposed approach, which achieves an improved performance-complexity trade-off. [Display omitted] [ABSTRACT FROM AUTHOR]

Computer vision technology has attracted significant interest in the condition monitoring (CM) community due to its potential to automate visual inspection and analysis of structures and components. By facilitating the processing and interpretation of visual information, including images and video data, computer vision holds promise for CM applications. However, it is essential to distinguish computer vision from non-contact CM techniques regarding their underlying principles and methods. While computer vision enables non-contact, remote monitoring, and condition assessment with minimal disruption to daily operations, it is distinct from non-contact CM techniques, which utilize various sensors to assess the condition of assets without physical contact or interference. Building upon the potential of computer vision technology, this survey paper presents a comprehensive overview of the current state-of-the-art CM methods based on computer vision and deep learning (DL) techniques, focusing on their application in monitoring synthetic fiber ropes (SFRs). SFRs are a viable alternative to steel wire ropes for underwater equipment and cranes that handle heavy loads. This is due to their high resistance to frictional wear, high tensile strength, lightweight, and flexibility. New materials, technologies, and processes for CM are being developed to meet the growing demand for SFRs. The paper explores ongoing research in applications that monitor the wear and aging of materials, as well as estimate their remaining useful life. The survey briefly discusses the traditional non-destructive testing and machine learning (ML) methods for CM applications. More importantly, DL-based methods, including supervised, unsupervised, semi-supervised, and self-supervised methods, are discussed in detail, together with the use of deep generative models and the recently developed diffusion models in the generation of synthetic datasets. Furthermore, the paper addresses the difficulties present in DL-based CM applications, including the scarcity of labeled data and the complexity and variety of the models used. The article ends by discussing the benefits of employing DL-based visual methods to understand SFR degradation processes, particularly in monitoring and maintenance. • This paper surveys visual data-based DL techniques for CM of SFRs. • Reviews defect detection DL models: supervised, unsupervised, semi, and self supervised. • Describes DGMs and diffusion models for generating synthetic data. • Presents TL techniques using pre-trained models for CM tasks. • Addresses challenges and opportunities for vision-based CM applications. [ABSTRACT FROM AUTHOR]

The Tactile Internet of Things (TIoT) is transforming the landscape of real-time applications by enabling haptic interactions and immersive experiences. This paper explores the potential of TIoT applications in critical sectors such as healthcare and manufacturing, emphasizing the necessity of ultra-reliable, low-latency communication. Conventional network infrastructures fall short of meeting these demands, necessitating innovative solutions such as Network Slicing (NS) to customize the network according to user activities. One of the key challenges addressed in this research is the allocation of resources for tactile data, which requires specialized solutions to prevent performance degradation in shared environments. Additionally, the paper proposes a solution that includes in-content awareness, enabling precise resource allocation based on the user's intent and requirements. Dynamic resource scaling, proactive resource allocation, and optimized bandwidth usage are essential components of the proposed architecture, guaranteeing responsive and efficient user experiences. Furthermore, the research introduces an end-to-end network slicing (NS) solution, emphasizing the importance of considering all components of the TIoT chain to prevent bottlenecks. Machine learning plays a crucial role in translating TIoT service profiles into specific requirements that are in line with the evolving needs of TIoT. To overcome the limitations of deep learning (DL), federated learning (FL) emerges as a groundbreaking approach, enabling collaborative model training without compromising data privacy. The paper explores the potential of FL and addresses its limitations within a centralized framework. It advocates for a novel architecture that integrates blockchain technology, Software-Defined Networking (SDN), Network Function Virtualization (NFV), and Multi-Access Edge Computing (MEC) to enhance FL in TIoT applications. The study investigates the performance of lightweight deep learning methods used as local models in federated learning for TIoT applications. The research also analyzes various FL algorithms from different perspectives, considering various local models contributing to the global model. Additionally, the study evaluates how the selected FL algorithms and DL local models collaborate, providing valuable insights into the performance and efficiency of the proposed architecture. These advancements aim to revolutionize the applications of TIoT and usher in a new era of intelligent, context-aware, and efficient communication in 6G networks. [ABSTRACT FROM AUTHOR]

Agent negotiation in multi-agent systems has been extensively studied, focusing on both theoretical and applied research. However, a limited number of studies have considered proposing an offer-generating strategy for agents to propose offers during the negotiation process in the multiple-negotiation situation where interdependency exist between a mixture of discrete issues and continuous issues across different negotiations. Especially, considering the above common real-life situation, there is little work of proposing such a strategy which is able to generate an approximately Pareto optimal solution. To address such challenges, this paper targets at multiple-negotiation scenarios involving interdependency between mixed types of issues across different negotiations. The contributions of this paper are threefold. Firstly, this paper addresses the research gap in mixed-type of issues in multiple negotiations. Secondly, the paper introduces a formalized negotiation model for multiple-negotiation scenarios, addressing both discrete and continuous issues, enabling automatic agents to obtain goal-aligned offers effectively. Thirdly, this paper introduces a Hybrid of PSO (Particle Swarm Optimization) and GA (Genetic Algorithm) Algorithm (i.e., named as HPGA in this paper) as an offer-generating strategy to assist agents in achieving approximately Pareto optimization in multiple-negotiation scenarios. To support those claims, this paper presents an overall modeling framework, introduces the proposed offer-generation strategy, conducts a series of experiments to demonstrate the superiority of the proposed approach in this paper, and presents two realistic case studies. Overall, this research expands upon existing studies in agent-based negotiation by addressing the overlooked aspects of mixed types of issues and issue interdependency across multiple negotiations. The proposed modeling approach and offer-generation strategy contribute to the advancement of negotiation techniques in multi-agent systems. • Fill the research gap of multi-negotiations with mixed-types of issues. • Propose a unified modeling for multi-issue negotiations scenario. • Propose an algorithm named HPGA as the offer-generating strategy. • Help agents to achieve approximately Pareto optimal by applying the proposed method. [ABSTRACT FROM AUTHOR]

In response to the challenges of low clarity and insufficient training samples in underwater concrete crack detection, this paper proposes an improved unsupervised learning method for the underwater concrete crack image enhancement (increase image quality) and augmentation (increase in number of images). Detecting structural defects underwater is vital for ensuring the proper functioning of underwater structures. However, the harsh underwater environment often leads to low-resolution images of concrete cracks, which in turn diminishes detection accuracy. Additionally, the challenges associated with underwater image collection make it difficult to gather an ample number of samples for training deep learning models to effectively detect these defects. Therefore, this paper proposes an unsupervised learning model that can simultaneously enhance and augment underwater concrete crack images in order to achieve better detection results. For the enhancement of underwater concrete crack images, the proposed method significantly improves the recognizability of images in turbid water environments and exhibits significant superiority compared to other similar methods, the values of the three evaluation indicators decreased by 45.2%, 40.4%, and 69.1%, respectively. Regarding the augmentation of underwater concrete crack images, the proposed method can translate images from clear water and waterless environments to muddy water environments. Compared to other methods, improved image quality by at least 61.2%, the proposed method generates images with better authenticity. This validates that the proposed cross domain translation strategy can simultaneously enhancing and augmenting underwater concrete crack images. • An unsupervised learning model is used to improving quality number of underwater carack images. • The residual network is used to extract local features of low-resolution crack images. • The Swin Transformer is used to obtain local and global features of high-resolution crack images. [ABSTRACT FROM AUTHOR]

With the rapid development of smart cities, urban infrastructure systems produce massive data that reflect their real-time operational conditions. These data provide insights for system monitoring and operation, and many existing studies develop various machine learning methods to understand recurrent system conditions. However, the extreme operational conditions, which could cause system failures, are not well explored. Importantly, methods for the recurrent conditions may not be suitable for modeling the failures. To fill this gap, this paper proposes a novel task of failure prediction, which aims to predict system failures before they happen. To solve this task, a generalized model that integrates survival analysis and the temporal convolutional networks, which is called TCNSurv in this paper, is developed to predict the distribution of system failure time. The model mainly contains three components: a data processing module, a time series module, and a survival analysis module. Specifically, the time series module employs Temporal Convolutional Networks to enable the modeling of temporal dependencies in time series data, and the survival analysis module explicitly formulates the probability of system failures. The proposed model is validated on three real-world datasets: vibration, traffic, and electricity, and results show that the developed model outperforms state-of-the-art regression-based models, survival analysis-based models, as well as integrated models. The research outcomes could help to understand the failure patterns of urban infrastructure systems and to develop early warning systems for smart cities. [ABSTRACT FROM AUTHOR]

Despite the widespread application of data-centric techniques in Geotechnical Engineering, there is a rising need for building trust in the artificial intelligence (AI)-driven safety assessment of road embankments due to its so-called "black-box" nature. In addition, from the lens of limit equilibrium approaches, e.g., Bishop, Fellenius, Janbu and Morgenstern–Price, and finite element method, it is essential to carefully examine the interplay of both topological and physical/mechanical properties during the safety factor (FoS) predictions. First, aside from having conventional geotechnical inputs for soil in core and foundation and the height of embankments, this paper codifies geometric features innovatively. The number of slope types with different ratios including 1:1, 1.5:1 and 2:1 as well as the number of berms is introduced. Second, a pool of 19 machine learning (ML) techniques is effortlessly trained on the dataset using an automated ML (AutoML) pipeline to identify the most optimized ML algorithm. Finally, to achieve post-hoc interpretability for the internal mechanism of the input–output relationship unbiasedly, a game-theory-based explainable AI (XAI) method called Shapley additive explanations (SHAP) values is applied. SHAP-aided importance analysis provides human-interpretable insights and indicates height, California bearing ratio, slope type 2:1 and cohesion as the most influential parameters. Exclusively, analyzing hazardous embankments by classifying main and joint contributors exhibits a complex and highly variable influence on the FoS. This paper harnesses the power of XAI tools to enhance reliability and transparency for the rapid FoS prediction of slopes. It targets geotechnical researchers, practitioners, decision-makers, and the general public for the first time. • Shedding light on the AI-aided slope stability analysis with SHAP values. • Coupling AutoML and XAI for reliable and easy-to-use FoS prediction of embankments. • Defining novel geometric parameters to capture their impact on geotechnical ones. • Categorizing unstable embankments from the global and local XAI perspectives. • Analyzing hazardous embankments by classifying main and joint contributors. [ABSTRACT FROM AUTHOR]