Search

In the field of prognostics and health management (PHM) for industrial equipment and systems, the estimation of remaining useful life (RUL) constitutes a fundamental task. A reliable and accurate method for estimating the RUL is therefore essential. This paper proposes a dynamic self-adaptive ensemble model, aimed at improving the rolling bearing RUL prediction. This model integrates an adaptive multi-scale feature extractor, a gated convolutional unit (GCU) and temporal convolutional networks (TCN). Through a redesign of the data flow, this model directly incorporates multi-scale comprehensive feature evaluation indicators into the neural network data flow, significantly enhancing the model's feature extraction capabilities. Subsequently, the study extends the traditional TCN by incorporating the GCU module and its gating mechanisms, further strengthening the model's capacity to capture long-term dependencies in sequence tasks. Experimental results demonstrate that, compared to existing state-of-the-art (SOTA) models, our method achieves at least a 10% increase in the prediction accuracy on two public run-to-failure bearing datasets. Beyond the tested datasets, the architecture that directly maps multi-scale evaluation indicators into the structure of neural network data flows also holds potential for broader application across diverse PHM tasks, promising significant improvements in the industrial safety and efficiency. • Multi-scale evaluation module for original features. • Direct mapping from multiple evaluation metrics to feature weights. • The feature selection process is incorporated into the RUL estimation model. • Better ability to capture inherent long-term and localized dependencies. [ABSTRACT FROM AUTHOR]

Most existing Convolutional Neural Networks (CNNs), Transformers, and their variants have limitations in capturing relationships between hyperspectral image (HSI) data, leading to unclear descriptions of region boundaries and limited generalization abilities. While semi-supervised Graph Neural Networks (GNNs) come with higher computational costs. Therefore, this paper proposes a method interacting the frequency and topology information for HSI Classification to address the aforementioned shortcomings, which combines convolution and self-attention to capture both local and global contextual information, thereby enhancing feature representation. Additionally, this method focuses on exploring spectral and topological structure features and enhancing the information exchange and interaction to improve performance. Experimental results demonstrate that this method gains a competitive advantage in HSI classification by proving highly effective in handling spectral ambiguity and material heterogeneity. It also exhibits lower computational costs, making it more feasible and practical compared to most benchmark methods. Our code is available at https://github.com/youngboy03/FTINet. [Display omitted] • Use CIformer, AFLM and ECM to obtain features from different angles in the image. [ABSTRACT FROM AUTHOR]

Mathematical teaching techniques encompass a diverse range of approaches for students to grasp mathematical concepts. Currently, various mathematical teaching techniques exist, and it is crucial to extract personalized teaching techniques based on students' knowledge level. Cognitive Computing systems in the learning domain open a substantial range of possible alternatives, thereby helping both students and teachers. These systems enhance the human cognitive capabilities by developing cognitive assistant, which is a flexible and adaptive learning software and a continuous learning system. In this paper, a personalized cognitive layered framework is proposed to map the student's reasoning level with the mathematical teaching techniques. The analysis is done with real time data collected from the School for Grades 1 to 6 and mathematical teaching techniques extracted from mathematical research articles. The Hypothesis Evidence Extraction (HypoEE) algorithm is proposed to extract optimal teaching techniques. The proposed system follows the pipeline of cognitive computing processes such as extraction of factors of importance, knowledge assertion graph construction, node prioritization, correlation extraction, text summarization, hypothesis identification, evidence extraction and scoring, hypothesis refinement, inference and decision formulation. The system is evaluated using performance metrics and cognitive metrics. The validation is done by three mathematical subject experts. The results showcase a Confidence Weighted Score (CWS) of 0.96, an accuracy of 0.994, 0.92 precision, 0.88 recall, 0.89 F1 score, 0.83 specificity and cognitive metric scores indicating a taskability of 0.89, 0.86 knowledge capacity and 0.89 knowledge utilization. [Display omitted] • A personalized cognitive layered framework for deriving the most effective and personalized mathematical teaching strategy. • A cognitive system driven by evidence-based inference to personalize education. • A text summarization algorithm crafted to produce concise summaries from large volumes of text. • Application of the Hypothesis Evidence Extraction (HypoEE) algorithm to extract optimal mathematical teaching techniques based on individual reasoning levels. • Analyzing performance through expert assessments and cognitive evaluation metrics. [ABSTRACT FROM AUTHOR]

In the field of prognostics and health management (PHM) for industrial equipment and systems, the estimation of remaining useful life (RUL) constitutes a fundamental task. A reliable and accurate method for estimating the RUL is therefore essential. This paper proposes a dynamic self-adaptive ensemble model, aimed at improving the rolling bearing RUL prediction. This model integrates an adaptive multi-scale feature extractor, a gated convolutional unit (GCU) and temporal convolutional networks (TCN). Through a redesign of the data flow, this model directly incorporates multi-scale comprehensive feature evaluation indicators into the neural network data flow, significantly enhancing the model's feature extraction capabilities. Subsequently, the study extends the traditional TCN by incorporating the GCU module and its gating mechanisms, further strengthening the model's capacity to capture long-term dependencies in sequence tasks. Experimental results demonstrate that, compared to existing state-of-the-art (SOTA) models, our method achieves at least a 10% increase in the prediction accuracy on two public run-to-failure bearing datasets. Beyond the tested datasets, the architecture that directly maps multi-scale evaluation indicators into the structure of neural network data flows also holds potential for broader application across diverse PHM tasks, promising significant improvements in the industrial safety and efficiency. • Multi-scale evaluation module for original features. • Direct mapping from multiple evaluation metrics to feature weights. • The feature selection process is incorporated into the RUL estimation model. • Better ability to capture inherent long-term and localized dependencies. [ABSTRACT FROM AUTHOR]

Mathematical teaching techniques encompass a diverse range of approaches for students to grasp mathematical concepts. Currently, various mathematical teaching techniques exist, and it is crucial to extract personalized teaching techniques based on students' knowledge level. Cognitive Computing systems in the learning domain open a substantial range of possible alternatives, thereby helping both students and teachers. These systems enhance the human cognitive capabilities by developing cognitive assistant, which is a flexible and adaptive learning software and a continuous learning system. In this paper, a personalized cognitive layered framework is proposed to map the student's reasoning level with the mathematical teaching techniques. The analysis is done with real time data collected from the School for Grades 1 to 6 and mathematical teaching techniques extracted from mathematical research articles. The Hypothesis Evidence Extraction (HypoEE) algorithm is proposed to extract optimal teaching techniques. The proposed system follows the pipeline of cognitive computing processes such as extraction of factors of importance, knowledge assertion graph construction, node prioritization, correlation extraction, text summarization, hypothesis identification, evidence extraction and scoring, hypothesis refinement, inference and decision formulation. The system is evaluated using performance metrics and cognitive metrics. The validation is done by three mathematical subject experts. The results showcase a Confidence Weighted Score (CWS) of 0.96, an accuracy of 0.994, 0.92 precision, 0.88 recall, 0.89 F1 score, 0.83 specificity and cognitive metric scores indicating a taskability of 0.89, 0.86 knowledge capacity and 0.89 knowledge utilization. [Display omitted] • A personalized cognitive layered framework for deriving the most effective and personalized mathematical teaching strategy. • A cognitive system driven by evidence-based inference to personalize education. • A text summarization algorithm crafted to produce concise summaries from large volumes of text. • Application of the Hypothesis Evidence Extraction (HypoEE) algorithm to extract optimal mathematical teaching techniques based on individual reasoning levels. • Analyzing performance through expert assessments and cognitive evaluation metrics. [ABSTRACT FROM AUTHOR]

Gas composition and light mode of industrial greenhouses are some of the most determining factors in the process of growing vegetable crops in greenhouse conditions. The intelligentisation of information technologies for monitoring and control based on artificial intelligence methods can increase the efficiency of agrotechnical procedures for greenhouse cultivation. One of the efficient approaches in today's world practice is the development and implementation of trustworthy hybrid decision-support systems based on the techniques of Fuzzy logic and Human-in-the-Loop. The research object is non-stationary processes of complex intelligent transformation of measurement data on the concentration of carbon dioxide and effective energy illumination in the growing zone of industrial greenhouses. The scientific novelty and practical value of the obtained research results consist in creating a computer model for aggregation and intelligent processing of agricultural monitoring data for greenhouses. The developed computer model is fully transformed into peripheral-level software of Internet of Things systems for agricultural purposes. This makes it possible to implement the for-computing architecture of monitoring systems in greenhouses. The obtained research results make it possible to optimise the resources used in growing crops in greenhouse conditions through the implementation of hardware and software intelligent monitoring tools that are adaptive to the types and periods of crop vegetation. The scientific and applied effect of the research is creating the novel approach of development and practical use of intelligent technologies for agrotechnical monitoring by substantiating the methodological provisions for synthesis of structural and algorithmic organisation of the corresponding software and hardware solutions. [Display omitted] • Intelligent software and hardware control of the microclimate of greenhouses with optimised number of system's parameters. • Software consideration of the complex influence of climatic parameters and agricultural and biological factors. • Implementation of the fog-architecture using a budget microprocessor base. • Software adaptive control of the agrotechnical systems. • The two-level control algorithm taking into account internal and external physical and chemical influences. [ABSTRACT FROM AUTHOR]

High-accuracy positioning and identification of transmission line components is the premise for their status detection and fault diagnosis. However, due to the limitations of imbalance object scale and distribution in aerial images, the problem of detecting dense-tiny objects still needs to be solved. Considering the prior knowledge of fixed connection scenes, we proposed a novel method named scene iterative reasoning network (SIRN) based on self-adaptive clustering and local scene knowledge. It consists of the coarse-grained detector (CGD), scene self-adaptive clustering (SSAC) subnetwork, and scene structure iterative reasoning (SSIR) subnetwork. The CGD was proposed to obtain global coarse detection results. Then, the SSAC subnetwork utilized unsupervised self-adaptive clustering to obtain accurate dense regions based on the CGD results. Finally, the SSIR subnetwork was designed to extract and fuse the scene semantic structure information of local regions for realizing accurate dense-tiny object detection. In summary, the SIRN model utilizes an iterative inference strategy to detect dense regions accurately. By leveraging scene structure information, the model effectively transforms the challenge of detecting dense objects into an advantage, resulting in precise detection. The SIRN model is compatible with single-stage and two-stage object detection models, achieving a notable mean average precision (mAP) improvement ranging from 4.4% to 12.2% compared to the baseline models. Compared to other state-of-the-art object detection models, SIRN demonstrates significant advantages in accuracy and error rate metrics. In addition, qualitative and quantitative experiments show that the SIRN model considerably enhanced the detection of dense-tiny objects. The code is available at https://github.com/CharmingWang/SIRN. [ABSTRACT FROM AUTHOR]

Mobile edge computing for autonomous driving needs to manage heterogeneous resources and process large amounts of data or multi-purpose payload. There needs to be deploying, scheduling and migrating tasks on edge nodes to ensure the reliability of tasks or maximize the utilization of resources. However, applying autonomous learning methods on autonomous driving is exceptionally difficult, due to the complexity of multi-dimensional context and the sensitivity to hyperparameters. In this paper, we propose a learning approach to quality-of-service (QoS) prediction of services via multi-dimensional context, and develop a stable approach for service deployment that requires minimal hyperparameter tuning and a modest number of trials to learn multilayer neural network policies. This approach can automatically trades off exploration against exploitation by automatically tuning hyperparameter based on maximum entropy reinforcement learning. We then demonstrate that this approach achieves state-of-the-art performance on Autoware benchmark environments. • This paper is an extension of our icws2017 paper: Xiong, W. , Wu, Z. , Li, B. , & Gu, Q., A Learning Approach to QoS Prediction via Multi-Dimensional Context, 2017 IEEE International Conference on Web Services. We propose a learning approach to quality-of-service (QoS) prediction of services via multi-dimensional context. • We propose a self-adaptive approach for service deployment under Mobile Edge Computing. • We conduct comprehensive experiments on a real-world system, demonstrating the effectiveness of our approach. [ABSTRACT FROM AUTHOR]

Current investigation deals with presenting a new hybrid metaheuristic optimization technique which is named as Interactive Fuzzy Search Algorithm (IFSA). Proposed method combines the affirmative features of Integrated Particle Swarm Optimizer (iPSO) and Teaching and Learning Based Optimizer (TLBO) techniques with a fuzzy decision mechanism. Proposed IFSA benefiting its fuzzy module provides a self-adaptive synchronization between local and global search strategies during the optimization process. Since in the proposed approach the tunable parameters are automatically assigned, IFSA acts as an ad-hoc free algorithm. To authenticate the validity of the proposed method, its performance is verified over the number of different types of mathematical, mechanical and structural optimization problems. Achieved outcomes indicate that the introduced algorithm has a high competitive potential on solving different class of optimization problems. [ABSTRACT FROM AUTHOR]

Wireless Sensor Networks (WSNs) are susceptible to selective forwarding attacks, which can lead to reduced network efficiency and compromise the integrity of transmitted data. The indistinguishability between malicious behavior and normal packet loss in harsh environments exacerbates the challenge. To address this, in this paper, we provide a combined detection scheme called LSTM-NV, consisting of a training stage and a detecting stage. During the training stage, we integrate variational mode decomposition (VMD) with a long short term memory (LSTM) model to learn normal nodes' forwarding behavior and then predict errors for each node. During the detecting stage, dynamic thresholds are determined to identify local anomaly points and a novel neighbor voting method is employed to differentiate between malicious and normal nodes. Our scheme demonstrates superior performance with a low average missed detection rate (MDR) of 0.6% and a low average false detection rate (FDR) of 3.3% compared to other effective methods, while also offering lower detection algorithmic complexity. • Utilizing LSTM model to learn from normal data and predict for each node's forwarding behavior. • Adaptive thresholds are employed to identify local anomaly points. • A novel neighbor-voting method has been designed to enable self-adaptive detection. • Time-variant harsh environments within the network are taken into consideration. [ABSTRACT FROM AUTHOR]

Traffic flow data (TFD) prediction is a hot research area in intelligent transportation system. TFD is non-stationary and nonlinear, so it has become a challenge to predict it accurately. In order to improve TFD prediction accuracy, a new TFD prediction model based on cosine similarity variational mode decomposition (CSVMD), extreme learning machine (ELM), and iterative error compensation strategy, named CSVMD-ELM-error, is proposed. To solve mode number K value selection of variational mode decomposition, CSVMD is proposed, which realizes the self-adaptive determination of K value. The idea of CSVMD-ELM-error is roughly as follows. Firstly, CSVMD decomposes TFD into a series of intrinsic mode functions (IMFs), and ELM is established for each IMF component. Then, in order to further improve the prediction accuracy, ELM is used to correct the prediction error of each IMF. Finally, the revised IMF error results and the IMF prediction results are reconstructed to complete the prediction. Four TFDs and nine comparison models are used for simulation experiment, the experimental result shows that CSVMD-ELM-error has the best prediction accuracy and has an effective application in TFD prediction. [Display omitted] • To our knowledge, this is the first work that uses VMD for traffic flow data (TFD) prediction. • Variational mode decomposition based on cosine similarity (CSVMD) is proposed. • This paper innovatively applies the error correction strategy to predict TFD. • Prediction model based on CSVMD-ELM-error is proposed for TFD. [ABSTRACT FROM AUTHOR]

The adaptability of the traditional GM (1, 1) model is poor because it is a rigorous homogenous exponent model with a single fixed structure. To improve the adaptability of the traditional grey model, a self-adaptive intelligence grey predictive model with an alterable structure is proposed in this paper. The proposed model has the advantages of adjustable parameters and is characterised by its variable structure as a homogenous/non-homogenous exponent model or as a single-variable linear–auto-regression model. It can be used to automatically compute the relative optimal modelling parameters and adaptively choose a more reasonable model structure based on the real data characteristics of a modelling sequence. Hence, this novel model outperforms traditional grey models with a single fixed structure. To verify its efficiency and applicability, the proposed model was used to simulate China’s electricity consumption from 2001 to 2013 and to forecast it in 2014 using real data; the results indicate that the novel model has better simulative and predictive accuracy than the GM (1, 1) and DGM (1, 1) models. [ABSTRACT FROM AUTHOR]

This paper proposes the design of non-fragile multivariable PI controller for industrial Multi Input Multi Output (MIMO) processes using Diversity Controlled Self-Adaptive Differential Evolution with Local Search (DCSaDE-LS) algorithm. The proposed non-fragile multivariable PI controller design is formulated as a single objective optimization problem with the objective of robustness subject to non-fragileness constraint. Monte Carlo Simulation is used to evaluate the non-fragileness constraint against stochastic uncertainty of controller parameters. Wood and Berry Distillation process and a real time Level-Temperature reactor process model with simple de-coupler are considered as a test systems. For constraint handling, Constraint Domination Principle (CDP) and Penalty Parameter Less (PPL) methods are incorporated in DCSaDE-LS, seperately. Also, the performance of DCSaDE-LS is compared with the recent adaptive DE algorithms namely Ensemble of mutation strategies and control parameters in Differential Evolution (EPSDE), Composite Differential Evolution (CoDE) and SaDE. The robustness and non-fragileness of the controller is analyzed under +40% and −40% variations in nominal process and controller parameters. The results demonstrated that the DCSaDE-LS designed non-fragile multivariable PI controller has good time domain performance, better robustness and non-fragileness than the reported multivariable PI controllers namely Maghade-PI, IMC-PI and NDT-PI. [ABSTRACT FROM AUTHOR]

In this paper, a self-adaptive classification learning hybrid JAYA and Rao-1 algorithm, which is called EHRJAYA, is proposed for solving large-scale numerical problems and real-world complex engineering optimization problems. JAYA algorithm and Rao-1 algorithm are two kinds of algorithms with simple structure and superior performance, which have the characteristics of no particular controlling parameters. In EHRJAYA the evolution strategies of the two algorithms are selected through a random selection mechanism. Then, a novel self-adaptive classification learning strategy is proposed, which fully utilizes information from different individuals. On this basis, two different adaptive coefficients are introduced to guide the population towards the optimal individual and away from the worst individual. Finally, combining the linear population reduction strategy and the dynamic lens opposition-based learning strategy, the convergence speed and ability to jump out of local optimum of the algorithm are greatly improved. To verify the performance of the proposed EHRJAYA, 59 complex functions from the CEC2014 and CEC2017 competitions are solved by EHRJAYA. Then, EHRJAYA and more than 20 algorithms with superior performance jointly solve ten challenging real-world engineering optimization problems. Experimental results show that the proposed EHRJAYA can obtain optimal results with the least computational resources in most cases. Therefore, in the face of these problems, effective solutions can be provided by EHRJAYA. [ABSTRACT FROM AUTHOR]

Image semantic segmentation is one of the key technologies for intelligent systems to understand natural scenes. As one of the important research directions in the field of visual intelligence, this technology has a wide range of application scenarios in the fields of mobile robots, drones, and intelligent driving. However, in practical applications, there may be problems such as inaccurate prediction of semantic labels, loss of segmented objects and background edge information. This paper proposes an improved semantic segmentation network that combines self-attention module and neural architecture search (NAS) method. The method first uses the NAS method to find a semantic segmentation network with multiple resolution branches. During the search process, the searched network structure is adjusted by combining the self-attention module, and then combined with the semantic segmentation networks searched by different branches to integrate into two semantic segmentation network models with different complexity, and finally integrate two network models with different complexity according to the current general teacher–student framework. The input image will first pass through the high complexity model to obtain more accurate parameters, which will affect the training weight of the student network, then pass the image into the low-complexity model to get the final predicted result. The experimental results on the Cityscapes dataset show that the accuracy of the algorithm is 69.8 %, the inference speed is 166.4 FPS, and the actual image segmentation speed is 48/s. It can optimize edge segmentation for better performance in complex scenes and achieve a good balance between real-time performance and accuracy in practical applications. [ABSTRACT FROM AUTHOR]

One of the challenging issues in supply chain management is the coordination of ordering processes, especially in dynamic situations. In recent years, reinforcement learning algorithms are considered to be efficient techniques for solving such problems. In this paper, an agent-based simulation technique has been integrated with a reinforcement learning algorithm and has been applied to model a four-echelon supply chain that faces non-stationary customer demands. This approach leads to the development of a novel and intelligent simulation-based optimization framework, which includes a detailed simulation modeling of supply chain behavior. Finally statistical methods, including the Var technique, are used for the risk evaluation and sensitivity analysis have been provided to support the decision making process under uncertainty. [ABSTRACT FROM AUTHOR]

Abstract: This paper addresses the problem of unsupervised change detection in Synthetic Aperture Radar (SAR) images. Previous approaches have used evolutionary clustering optimization methods, which can suffer from reduced accuracy, because they often use only a single objective function and can easily become trapped at locally optimal values. To overcome these difficulties, we propose a new approach which combines the artificial immune system (AIS) theory with a multi-objective optimization algorithm. First, the self-adaptive artificial immune multi-objective algorithm is adopted to pre-sort the difference image. During this procedure, the difference image is categorized into three classes – changed class, unchanged class and uncertain samples. Second, based on wavelet decomposition to extract features from the difference image, the immune clonal multi-objective clustering algorithm is used to search for the optimal clustering centers of uncertain samples, labeling them as changed or unchanged. Experimental comparisons with four state-of-the-art approaches show that the proposed algorithm can obtain a higher accuracy, is more robust to noise, and finds solutions which are more globally optimal. Additionally, the proposed algorithm can improve the local search ability for the optimal solutions and produces better cluster centers. [Copyright &y& Elsevier]

Feature selection methods have an important role on the readability of data and the reduction of complexity of learning algorithms. In recent years, a variety of efforts are investigated on feature selection problems based on unsupervised viewpoint due to the laborious labeling task on large datasets. In this paper, we propose a novel approach on unsupervised feature selection initiated from the subspace clustering to preserve the similarities by representation learning of low dimensional subspaces among the samples. A self-expressive model is employed to implicitly learn the cluster similarities in an adaptive manner. The proposed method not only maintains the sample similarities through subspace clustering, but it also considers the underlying structure of data based on a regularized regression model. In line with the convergence analysis of the proposed method, the experimental results on benchmark datasets demonstrate the effectiveness of our approach as compared with the state-of-the-art methods. • Propose a novel unsupervised feature selection method based on subspace clustering and adaptive similarity learning. • Introduce a self-expressive model to adaptively and implicitly learn the cluster similarities. • Employ a regularized regression model to obtain the sparse correlation among the features and clusters. • Introduce a unified objective function to consider the main factors in the proposed approach. • The results show that our proposed approach outperformed several state-of-the-art methods. [ABSTRACT FROM AUTHOR]