Search

The traditional method for transformer moisture diagnosis is to establish empirical equations between feature parameters extracted from frequency domain spectroscopy (FDS) and the transformer’s moisture content. However, the established empirical equation may not be applicable to a novel testing environment, resulting in an unreliable evaluation result. In this regard, it is acknowledged that FDS combined with machine learning is more suitable for estimating moisture content in a variety of test environments. Nonetheless, the accuracy of the estimation results obtained using the existing method is limited by the algorithm’s inability to generalize. To address this issue, we propose an AdaBoost algorithm-enhanced back-propagation neural network (BP_AdaBoost). This study creates a database by extracting feature parameters from the FDS that characterize the insulation states of the prepared samples. Then, using the BP_AdaBoost algorithm and the newly constructed database, the moisture estimation models are trained. Finally, the results of the estimation are discussed in terms of laboratory and field transformers. By comparing the proposed BP_AdaBoost algorithm to other intelligence algorithms, it is demonstrated that it not only performs better in generalization, but also maintains a high level of accuracy. [ABSTRACT FROM AUTHOR]

Induction period evaluated from time domain dielectric response of oil‐paper insulation (OPI) contains information related to heat dissipating property of oil. In the present paper, effect of aging sensitive oil properties on the value of induction period is investigated. Three types of dielectric liquids (pure mineral oil, synthetic ester, mixture of synthetic ester, and mineral oil) are considered for the present work. A neural network (NN) is used for modelling the non‐linear relationship between induction period and various oil properties. The trained network is subsequently used to study the influence of various oil properties on induction period. [ABSTRACT FROM AUTHOR]

In the petroleum and coal industries, digital image technology and acoustic emission technology are employed to study rock properties, but both exhibit flaws during data processing. Digital image technology is vulnerable to interference from fractures and scaling, leading to potential loss of image data; while acoustic emission technology is not hindered by these issues, noise from rock destruction can interfere with the electrical signals, causing errors. The monitoring errors of these techniques can undermine the effectiveness of rock damage analysis. To address this issue, this paper focuses on the restoration of image data acquired through digital image technology, leveraging deep learning techniques, and using soft and hard rocks made of similar materials as research subjects, an improved Incremental Transformer image algorithm is employed to repair distorted or missing strain nephograms during uniaxial compression experiments. The concrete implementation entails using a comprehensive training set of strain nephograms derived from digital image technology, fabricating masks for absent image segments, and predicting strain nephograms with full strain detail. Additionally, we adopt deep separable convolutional networks to optimize the algorithm's operational efficiency. Based on this, the analysis of rock damage is conducted using the repaired strain nephograms, achieving a closer correlation with the actual physical processes of rock damage compared to conventional digital image technology and acoustic emission techniques. The improved incremental Transformer algorithm presented in this paper will contribute to enhancing the efficiency of digital image technology in the realm of rock damage, saving time and money, and offering an innovative approach to traditional rock damage analysis. [ABSTRACT FROM AUTHOR]

This paper proposes a ResGA-Net algorithm for cloud top height (CTH) retrieval using FY-4A satellite data. The algorithm utilizes genetic algorithms for data selection and employs a residual module-based neural network for modeling. It takes the spectral channel data from the FY-4A satellite as input features and uses CTH extracted from ground-based millimeter-wave cloud radar reflectivity as the target. By combining the large observation scale of the FY-4A satellite and the high accuracy of ground-based cloud radar observations, the model can generate satellite CTH products with higher precision. To validate the effectiveness of the algorithm, experiments were conducted using data from the Beijing area spanning from January 2020 to January 2022. The experimental results show that the metrics of the proposed ResGA-Net outperform those of various contrastive algorithms, and compared to the original FY-4A CTH product, the RMSE and MAE have decreased by 37.89% and 34.77%, while the PCC and SRCC have increased by 11.17% and 9.47%, respectively, demonstrating the superiority of the proposed method presented in this paper. [ABSTRACT FROM AUTHOR]

With the aim of solving the problem of identifying complex underwater acoustic targets using a single signal feature in the Time–Frequency (TF) feature, this paper designs a method that recognizes the underwater targets based on the Delay-Doppler joint feature. First, this method uses symplectic finite Fourier transform (SFFT) to extract the Delay-Doppler features of underwater acoustic signals, analyzes the Time–Frequency features at the same time, and combines the Delay-Doppler (DD) feature and Time–Frequency feature to form a joint feature (TF-DD). This paper uses three types of convolutional neural networks to verify that TF-DD can effectively improve the accuracy of target recognition. Secondly, this paper designs an object recognition model (TF-DD-CNN) based on joint features as input, which simplifies the neural network's overall structure and improves the model's training efficiency. This research employs ship-radiated noise to validate the efficacy of TF-DD-CNN for target identification. The results demonstrate that the combined characteristic and the TF-DD-CNN model introduced in this study can proficiently detect ships, and the model notably enhances the precision of detection. [ABSTRACT FROM AUTHOR]

In this paper, a control method combined with variable frequency and phase shifting is proposed, the primary purpose of which is to optimize the operating efficiency of the phase‐shift full‐bridge (PSFB) DC‐DC converter. By modeling the losses of the PSFB converter, the optimal switching frequency related to the load conditions is found. A neural network aided variable frequency controller that can work linearly is built to achieve the proposed method. At the same time, the voltage control loop is retained to ensure the stability of the output voltage and the necessary dynamic response of the converter. This paper builds an experimental platform for a 2 kW PSFB DC‐DC converter to verify the effectiveness of the proposed control method. With the control method proposed in this paper, the PSFB converter can always work at the optimum efficiency point under different load conditions. In addition, this paper also analyzes the influence of variable frequency control on PSFB operating states, such as zero‐voltage‐switching (ZVS) constraints, continuous conduction mode (CCM) operating states, and power density, which have an impact on its operating conditions. [ABSTRACT FROM AUTHOR]

The identification of cross-sectional loss in cables due to corrosion is crucial for evaluating the remaining strength of bridge cables. To accurately determine the cross-sectional loss rate, this paper derived a three-dimensional magnetic dipole model for spatial cable damage. The study employed an independently designed self-magnetic flux leakage (SMFL) sensor array to detect corrosion on a bundle of 37 parallel steel wires. The analysis investigated the correlation between corrosion degrees and SMFL signal features. The results show that the spatial magnetic field inversion collected by the sensor array device is more accurate. The cable damage location can be pinpointed by observing abrupt changes in the Bx and Bz curves. Additionally, this paper introduces five corrosion characterization features, all correlated with the cable cross-sectional loss rate. However, recognition stability using a single characteristic value is insufficient. The cable cross-sectional loss rate identification method, utilizing a back propagation neural network in conjunction with multiple characteristic indicators, demonstrates robust quantitative and adaptive capabilities. The maximum relative error of this method is 7.6%, offering a new perspective for future cable damage detection. [ABSTRACT FROM AUTHOR]

Train wheels are crucial components for ensuring the safety of trains. The accurate and fast identification of wheel tread defects is necessary for the timely maintenance of wheels, which is essential for achieving the premise of conditional repair. Image-based detection methods are commonly used for detecting tread defects, but they still have issues with the misdetection of water stains and the leaking of small defects. In this paper, we address the challenges posed by the detection of wheel tread defects by proposing improvements to the YOLOv8 model. Firstly, the impact of water stains on tread defect detection is avoided by optimising the structure of the detection layer. Secondly, an improved SPPCSPC module is introduced to enhance the detection of small targets. Finally, the SIoU loss function is used to accelerate the convergence speed of the network, which ensures defect recognition accuracy with high operational efficiency. Validation was performed on the constructed tread defect dataset. The results demonstrate that the enhanced YOLOv8 model in this paper outperforms the original network and significantly improves the tread defect detection indexes. The average precision, accuracy, and recall reached 96.95%, 96.30%, and 95.31%. [ABSTRACT FROM AUTHOR]

Aiming at the poor transient convergence and real-time performance of the horizontal vibration state variables of the high-speed elevator car system, and the low control accuracy and stability, this paper proposes an adaptive control strategy to ensure the transient response of the elevator car system. First, a prescribed preset transient performance function is introduced into the controller design to control the variation range of the state variables and ensure the steady transient performance of the elevator car; Second, representing the information of observation error/control error through algebraic operations, designing an adaptive law based on e-correction, estimating unknown parameters in the elevator car system, and achieving online parameter updates; Then, using neural networks to learn and compensate for unknown dynamics in the elevator car system, and solving the online estimation problem of neural network weights through adaptive laws, so that the tracking error and weight estimation error converge to a tight set near zero; Finally, using MATLAB/SIMULINK to compare and analyze the four control algorithms of passive control, PID control, adaptive control based on gradient descent method and transient response adaptive control proposed in this paper under two different rail excitations: Random excitation and pulse excitation. The simulation results show that the adaptive control strategy proposed in this paper effectively suppresses the horizontal vibration of the elevator car, makes the state variables have faster convergence speed and smaller convergence error, and ensures the stable and transient performance of the elevator car system. [ABSTRACT FROM AUTHOR]

Quality‐related fault detection has become a hot research topic in recent years. It is not reliable to measure quality‐related relationships only by mutual information among process variables and single‐quality variables. Frequent alarms for quality‐unrelated faults seriously affect the normal operation of industrial production. At the same time, the strong nonlinearity of the process data leads to the difficulty of feature extraction. In this paper, we propose a fault detection algorithm based on nonlinear quality‐related neighborhood embedding neural orthogonal mapping (QR‐NENOM). First, quality‐related and quality‐unrelated variables are selected by Bayesian fusion mutual information, and the weighted method of mutual information is used to enhance the quality‐related information and suppress the quality‐unrelated information. Second, local manifold information is obtained by reconstructing nearest neighbors of process data, and key features are extracted by the nonlinear method composed of neural network and orthogonal mapping. Then, statistical indicators are established to complete fault detection. Finally, the nonlinear feature extraction ability of NENOM is verified by numerical examples, and the QR‐NENOM algorithm proposed in this paper is applied to the penicillin fermentation process. Comparative experiments show that QR‐NENOM has better detection performance for quality‐related faults and fewer alarms for quality‐unrelated faults. [ABSTRACT FROM AUTHOR]

This paper studies the distributed convex optimization of bipartite containment control problem for a class of higher order nonlinear multi-agent systems with uncertain states. For the optimization problem, the penalty function is constructed by summing the local objective function of each agent and combining the penalty term formed by the adjacency matrix. For the unknown nonlinear function and unpredictable states in the system, this paper construct radial basis function Neural-networks and state observer for approaching, respectively. In order to avoid "explosion of complexity," under the framework of Lyapunov function theory, we propose the dynamic surface control (DSC) technology and design the distributed adaptive backstepping neural network controller to ensure all the signals remain semi-global uniformly ultimately bounded in the closed-loop system and all agents can converge to the convex hull containing each boundary trajectory as well as its opposite trajectory different in sign. Simulation results confirm the feasibility of the proposed control method. [ABSTRACT FROM AUTHOR]

In recent years, the proliferation of vehicles on roadways, accompanied by an escalating demand for augmented safety and refined traffic management, has fueled substantial progress in the domains of computer vision and deep learning methodologies. This technological surge has precipitated ground-breaking applications, with real-time vehicle detection emerging as a linchpin. Capability to accurately detect vehicles in a real-time holds profound implications across diverse domains. In the context of autonomous driving, it underpins essential functions such as path planning, collision avoidance, and decision-making, essential for the safe operation of autonomous vehicles. In the arena of traffic monitoring, it empowers transportation authorities to acquire real-time insights into traffic dynamics, congestion patterns, and incident detection, thereby optimizing traffic flow and resource allocation strategies. Furthermore, within surveillance systems, real-time vehicle detection equips security personnel with the means to swiftly identify potential security threats, enhancing overall situational awareness and safety. This paper contributes to this dynamic landscape by introducing an innovative architecture expressly crafted for vehicle detection. Leveraging state-of-the-art computer vision techniques and deep learning methodologies, this architecture addresses the evolving challenges posed by contemporary traffic scenarios. Through the enhancement of real-time vehicle detection capabilities, our aim is to fortify the foundations of safety, efficiency, and security within vehicular contexts. The paper includes a fusion of AlexNet and Darknet accompanied by a comparative analysis. It is expected to deliver unmatched dependability and accuracy in real-time traffic monitoring, setting a new standard for performance in this application. [ABSTRACT FROM AUTHOR]

The research on baggage flow plays a pivotal role in achieving the efficient and intelligent allocation and scheduling of airport service resources, as well as serving as a fundamental element in determining the design, development, and process optimization of airport baggage handling systems. This paper examines baggage checked in by departing passengers at airports. The crrent state of the research on baggage flow demand is first reviewed and analyzed. Then, using examples of objective data, it is concluded that while there is a significant correlation between airport passenger flow and baggage flow, an increase in passenger flow does not necessarily result in a proportional increase in baggage flow. According to the existing research results on the influencing factors of baggage flow sorting and classification, the main influencing factors of baggage flow are divided into two categories: macro-influencing factors and micro-influencing factors. When studying the relationship between the economy and baggage flow, it is recommended to use a comprehensive analysis that includes multiple economic indicators, rather than relying solely on GDP. This paper provides a brief overview of prevalent transportation flow prediction methods, categorizing algorithmic models into three groups: based on mathematical and statistical models, intelligent algorithmic-based models, and combined algorithmic models utilizing artificial neural networks. The structures, strengths, and weaknesses of various transportation flow prediction algorithms are analyzed, as well as their application scenarios. The potential advantages of using artificial neural network-based combined prediction models for baggage flow forecasting are explained. It concludes with an outlook on research regarding the demand for baggage flow. This review may provide further research assistance to scholars in airport management and baggage handling system development. [ABSTRACT FROM AUTHOR]

Cotton is an important economic crop; therefore, enhancing cotton yield and cultivating superior varieties are key research priorities. The seedling stage, a critical phase in cotton production, significantly influences the subsequent growth and yield of the crop. Therefore, breeding experts often choose to measure phenotypic parameters during this period to make breeding decisions. Traditional methods of phenotypic parameter measurement require manual processes, which are not only tedious and inefficient but can also damage the plants. To effectively, rapidly, and accurately extract three-dimensional phenotypic parameters of cotton seedlings, precise segmentation of phenotypic organs must first be achieved. This paper proposes a neural network-based segmentation algorithm for cotton seedling organs, which, compared to the average precision of 75.4% in traditional unsupervised learning, achieves an average precision of 96.67%, demonstrating excellent segmentation performance. The segmented leaf and stem point clouds are used for the calculation of phenotypic parameters such as stem length, leaf length, leaf width, and leaf area. Comparisons with actual measurements yield coefficients of determination R 2 of 91.97%, 90.97%, 92.72%, and 95.44%, respectively. The results indicate that the algorithm proposed in this paper can achieve precise segmentation of stem and leaf organs, and can efficiently and accurately extract three-dimensional phenotypic structural information of cotton seedlings. In summary, this study not only made significant progress in the precise segmentation of cotton seedling organs and the extraction of three-dimensional phenotypic structural information, but the algorithm also demonstrates strong applicability to different varieties of cotton seedlings. This provides new perspectives and methods for plant researchers and breeding experts, contributing to the advancement of the plant phenotypic computation field and bringing new breakthroughs and opportunities to the field of plant science research. [ABSTRACT FROM AUTHOR]

[Display omitted] • Paper chromogenic array (PCA) integrated with machine learning (ML) was developed. • PCA exhibit distinguishable pattern shifts when reacting volatile metabolites. • PCA pattern recognition achieved using a cross-validated neural network. • PCA accurately identify multiplexed pathogens from indigenous microflora. • The nondestructive PCA-ML holds great potential as a smart food safety system. Non-destructive detection of human foodborne pathogens is critical to ensuring food safety and public health. Here, we report a new method using a paper chromogenic array coupled with a machine learning neural network (PCA-NN) to detect viable pathogens in the presence of background microflora and spoilage microbe in seafood via volatile organic compounds sensing. Morganella morganii and Shewanella putrefaciens were used as the model pathogen and spoilage bacteria. The study evaluated microbial detection in monoculture and cocktail multiplex detection. The accuracy of PCA-NN detection was first assessed on standard media and later validated on cod and salmon as real seafood models with pathogenic and spoilage bacteria, as well as background microflora. In this study PCA-NN method successfully identified pathogenic microorganisms from microflora with or without the prevalent spoilage microbe, Shewanella putrefaciens in seafood, with accuracies ranging from 90% to 99%. This approach has the potential to advance smart packaging by achieving nondestructive pathogen surveillance on food without enrichment, incubation, or other sample preparation. [ABSTRACT FROM AUTHOR]

AbstractAs part of the verification, validation, and uncertainty quantification process applied to neutronics deterministic codes, there is a requirement to expand the validation domain, especially to accommodate new third-generation reactors. The objective of the present work is to estimate the numerical biases arising from the several approximations used in deterministic codes and across different points in the phase space. Typically, this is accomplished by comparing the deterministic code to be validated with a Monte Carlo or stochastic reference code (without significant approximations). Since these reference calculations are computationally expensive, this paper proposes an alternative approach for predicting model biases of the APOLLO3® deterministic code for third-generation pressurized water reactors using machine learning algorithms.Three types of metamodels are employed (polynomial regression, kriging, and neural networks). Two scales are investigated, from a single assembly to a cluster of 3 × 3 assemblies [small two-dimensional (2-D) core], with model biases evaluated for APOLLO3 schemes with various levels of accuracy (lattice and core solvers, with high- to low-fidelity approaches). For the small 2-D core, numerical biases are observed for reactivity and power peak, representing both global and local quantities of interest. Throughout the study, the best results are achieved using kriging or neural networks, even if polynomial regression provides satisfactory predictions in some cases. The possibility of predicting biases for different quantities is also introduced.In conclusion, this paper discusses the prospects of extending the applicability of these metamodels to small and large third-generation pressurized water reactor cores, the idea being to potentially use these metamodels to support safety demonstrations for the new reactors in the long term. [ABSTRACT FROM AUTHOR]

Satellite-based navigation systems are widely used in transportation. GNSS signal's strength or quality can easily be degraded by local environments. As a result, the position accuracy of satellite-based navigation systems decreases. In this paper, a novel approach for estimating the positioning error is proposed using ML/DL technique. For learning the relationship between position errors and increased data from GNSS receivers without any prior experience, neural networks have become the machine learning option of choice in the past few years. Signal degradation is best measured by dilution of precision, elevation angles, and carrier-to-noise ratios. To estimate the position error of satellite-based navigation systems, neural networks are trained in this paper. This paper applies a long-short-term memory (LSTM) network to model the temporal correlation of position error measurements. Therefore, neural networks are capable of learning the trend of position errors through training. [ABSTRACT FROM AUTHOR]

When man⁃made or natural disasters occur suddenly, the rapid deployment of search and rescue (SAR) robots is crucial for saving lives. To accomplish rescue tasks, SAR robots need to autonomously plan paths in continuously dynamic and unknown environments to reach the rescue target locations. This paper proposes a sensor configuration scheme for SAR robots, applying a Q-learning algorithm based on Q-table and neural networks to achieve autonomous control of SAR robots. It addresses the challenge of path planning in unknown environments, specifically how to avoid static and dynamic obstacles. Balancing the exploration and exploitation during the training process is one of the challenges in reinforcement learning. This paper introduces a mixed optimization method for dynamically selecting search strategies, building upon greedy search and Boltzmann search. Simulations are conducted using MATLAB, and the results indicate that the proposed method is feasible and effective. SAR robots equipped with the sensor configuration can effectively respond to environmental changes, reaching target locations while successfully avoiding both static and dynamic obstacles. [ABSTRACT FROM AUTHOR]

Recent research has revealed that subtle imperceptible perturbations can deceive well-trained neural network models, leading to inaccurate outcomes. These instances, known as adversarial examples, pose significant threats to the secure application of machine learning techniques in safety-critical systems. In this paper, we delve into the study of one-pixel attacks in deep neural networks, recently reported as a kind of adversarial examples. To identify such one-pixel attacks, most existing methodologies rely on the differential evolution method, which utilizes random selection from the current population to escape local optima. However, the differential evolution technique might waste search time and overlook good solutions if the number of iterations is insufficient. Hence, in this paper, we propose a gradient ascent with momentum approach to efficiently discover good solutions for the one-pixel attack problem. As our method takes a more direct route to the goal compared to existing methods relying on blind random walks, it can effectively identify one-pixel attacks. Our experiments conducted on popular CNNs demonstrate that, in comparison with existing methodologies, our technique can detect one-pixel attacks significantly faster. [ABSTRACT FROM AUTHOR]

This paper proposes an positional encoding-based attention mechanism model which can quantify the temporal correlation of ship maneuvering motion to predict the future ship motion in real sea state. To represent the temporal information of the sequential motion status, the positional encoding consisted by sine and cosine functions of different frequencies is chosen as the input of the model. First, the reasonableness of the improved architecture of the model is validated on the standard turning test datasets of an unmanned surface vehicle. Then, the absolute positional encoding based-scaled-dot product attention mechanism model is compared with other two attention mechanism models with different positional encoding and attention calculation methods and its superiority is verified. As demonstrated by exhaustive experiments, the model has the highest prediction accuracy when the input sequence length equals the output sequence length and the accuracy defined in this paper of the model will drop to less than 90% when the predicted length exceeds 45. Finally, the attention mechanism model is compared with the LSTM model with different lengths of input sequences to demonstrate that the attention mechanism model has a faster training speed when dealing with long sequences. [ABSTRACT FROM AUTHOR]

This paper studies obstacle and collision avoidance strategies for nonlinear second-order multi-agent systems (MAS) formation control. Due to the uncertainties and complexities in nonlinear systems, including external disturbances and communication delays, radial basis function (RBF) neural network control is employed to address the control requirements of nonlinear terms in the system. In addition, as traditional artificial potential fields (APF) based obstacle avoidance algorithms have limitations, this paper applies an improved APF algorithm for collision avoidance and obstacle avoidance in multi-agent systems formation control. The stability and feasibility of the proposed approach are proved based on the Lyapunov stability theory. Simulation experiments further validate the effectiveness of the formation control strategy. [ABSTRACT FROM AUTHOR]

Improving generalization ability in multi-robot formation can reduce repetitive training and calculation. In this paper, we study the multi-robot formation problem with the ability to generalize the target position. Since the generalization ability of neural network is directly proportional to spatial dimension, we adopt the strategy of using different networks to solve different objectives, so that the network learning can focus on the learning of one objective to obtain better performance. In addition, this paper presents a distributed deep reinforcement learning method based on soft actor–critic algorithm for solving multi-robot formation problem. At the same time, the formation evaluation assignment function is designed to adapt to distributed training. Compared with the original algorithm, the improved algorithm can get higher reward cumulative values. The experimental results show that the proposed algorithm can better maintain the desired formation in the moving process, and the rotation design in the reward function makes the multi-robot system have better flexibility in formation. The comparison of control signal curve shows that the proposed algorithm is more stable. At the end of the experiments, the universality of the proposed algorithm in formation maintenance and formation variations is demonstrated. [ABSTRACT FROM AUTHOR]

This paper presents an intelligent healthcare framework by incorporating modern computing technologies like machine learning and deep learning. The sole motivation of this paper is to predict diabetes status intelligently so that patients can be made aware of their diabetes status regularly. A novel ensemble machine learning-based diabetes prediction framework, namely "iDP", has been proposed in this paper. Six different machine learning approaches have been used for the iDP framework, namely random forest, decision tree, neural network, AdaBoost, support vector machine, and XGBoost. Training and testing of ensemble modeling have been performed based on these learning techniques. A comprehensive preliminary data analysis and screening of statistical analysis have been systematically demonstrated in this paper. The recursive feature elimination and fivefold cross-validation methods have been used to select the best model. All experiments have been performed using R-Studio 1.2.5 with the help of the Rattle library 5.4.0. The preliminary data analytic and Pearson correlation coefficient values have been computed to verify the linearity relationship among the data. Results have been reported in eight performance metrics, including accuracy, specificity, sensitivity, area under curve, area under convex hull, minimum error rate, minimum weighted loss, and precision. The proposed iDP framework improves the results on different aspects in order to compare them with individual learning approaches. Despite this, a comparative performance analysis of the iDP framework has also been performed with state-of-the-art prevailing models. As per experiment, the iDP framework gives better results in most cases, like accuracy: 95.26%, AUC: 91.15%, sensitivity: 96.81%, specificity: 97.72% and precision: 90.72%. [ABSTRACT FROM AUTHOR]

This paper presents an artificial intelligence deep learning model for an energy-aware task scheduling algorithm based on learning automata (LA) in the Fog Computing (FC) Applications. FC is a distributed computing model that serves as an intermediate layer between the cloud and Internet of Things (IoT) to improve the quality of service. The IoT is the closest model to the wireless sensor network (WSN). One of its important applications is to create a global approach to health care system infrastructure development that reflects recent advances in WSN. The most influential factor in energy consumption is task scheduling. In this paper, the issue of reducing energy consumption is investigated as an important challenge in the fog environment. Also, an algorithm is presented to solve the task scheduling problem based on LA and measure the makespan (MK) and cost parameters. Then, a new artificial neural network deep model is proposed, based on the presented LA task scheduling fog computing algorithm. The proposed neural model can predict the relation among MK, energy and cost parameters versus VM length for the first time. The proposed model results show that all of the desired parameters can be predicted with high precision. [ABSTRACT FROM AUTHOR]

Gradient stiffened plates are widely used in aerospace, automotive, transportation and other fields for their excellent performance. Aiming at the gradient structure of gradient plate which leads to the problem of excessive computation and low efficiency of homogenization and two-scale optimization design, this paper adopts a machine learning method to build an artificial neural network with microstructure deformation parameters as input and equivalent stiffness coefficients as output, to realize the efficient prediction of equivalent stiffness. In the optimization process, this paper introduces unit design variables characterizing the deformation of the single cell to achieve explicit control of the local deformation of the gradient plate, and introduces mapping function node design variables to ensure that the local deformation of the single cell during the optimization process is consistent with the mapping function, which facilitates the decoupling of the twoscale optimization results. Numerical examples verify the effectiveness and correctness of the proposed method. [ABSTRACT FROM AUTHOR]

Data-driven learning of the fractional discrete-time unified system is studied in this paper. A neural network method is suggested in the parameter estimation of fractional discrete-time chaotic systems. An optimization problem is obtained and the famous Adam algorithm is employed to train the neural network's weights and parameters. The parameter estimation result is compared with that of the stepwise response sensitivity approach (SRSA). This paper provides a high accuracy method for parameter inverse problems. The method also can be applied to data-driven learning of other fractional chaotic systems. [ABSTRACT FROM AUTHOR]

Estimation of vivo muscle forces during human motion is important for understanding human motion control mechanisms and joint mechanics. This paper combined the advantages of the convolutional neural network (CNN) and long-short-term memory (LSTM) and proposed a novel muscle force estimation method based on CNN–LSTM. A wearable sensor system was also developed to collect the angles and angular velocities of the hip, knee, and ankle joints in the sagittal plane during walking, and the collected kinematic data were used as the input for the neural network model. In this paper, the muscle forces calculated using OpenSim based on the Static Optimization (SO) method were used as the standard value to train the neural network model. Four lower limb muscles of the left leg, including gluteus maximus (GM), rectus femoris (RF), gastrocnemius (GAST), and soleus (SOL), were selected as the studying objects in this paper. The experiment results showed that compared to the standard CNN and the standard LSTM, the CNN–LSTM performed better in muscle forces estimation under slow (1.2 m/s), medium (1.5 m/s), and fast walking speeds (1.8 m/s). The average correlation coefficients between true and estimated values of four muscle forces under slow, medium, and fast walking speeds were 0.9801, 0.9829, and 0.9809, respectively. The average correlation coefficients had smaller fluctuations under different walking speeds, which indicated that the model had good robustness. The external testing experiment showed that the CNN–LSTM also had good generalization. The model performed well when the estimated object was not included in the training sample. This article proposed a convenient method for estimating muscle forces, which could provide theoretical assistance for the quantitative analysis of human motion and muscle injury. The method has established the relationship between joint kinematic signals and muscle forces during walking based on a neural network model; compared to the SO method to calculate muscle forces in OpenSim, it is more convenient and efficient in clinical analysis or engineering applications. [ABSTRACT FROM AUTHOR]

Summary: An adaptive optimal controller is proposed in this paper to maximize the captured power of a variable speed wind power system. The proposed controller is a combination of optimal and adaptive control components. The Adaptive Dynamic Programming technique is used to design the optimal control component to overcome the nonlinear problem of system dynamics and ensure stability. While the neural network is used to approximate unknown disturbances and system uncertainties. After that, the adaptive control component fully compensates for the effects of these unknown elements. Neither optimal nor adaptive control components necessitate prior knowledge of system dynamics. Furthermore, the approximation network updates only the weight matrix norm rather than the weight matrix of the neural network in each interval time, which significantly reduces computation. The stability analysis of the closed‐loop system is obtained using Lyapunov stability theory. The correctness and robustness of the control scheme are validated in two different scenarios using MATLAB/Simulink. The presented robust adaptive optimal controller is also compared to other existing controllers to demonstrate its benefits. [ABSTRACT FROM AUTHOR]

This paper proposes a flexible sensor made of fiber Bragg gratings (FBGs) package that can effectively recognize the direction and velocity of relative sliding on its surface. The measurement principle and structural design of the sensor are introduced, and the sensor is analyzed by finite element simulation software to verify that the sensor can detect the shear deformation of relative sliding on its surface. On this basis, an experimental platform is built to test the relative sliding direction and velocity, the feature values are extracted after the initial signal is processed with noise reduction, and the recognition prediction is carried out by using back-propagation neural network (BPNN) and GA-BP neural network optimized by Genetic Algorithm (GA). The experimental results show that GA-BP neural network has better recognition accuracy, and the correlation coefficients R2 in the recognition of the sliding direction and velocity parameters are 0.9851 and 0.9819, respectively, which are able to recognize the sliding sensation parameters effectively. [ABSTRACT FROM AUTHOR]

The precipitation estimation with high spatial and temporal resolution over Tibet is of great significance for monitoring the stability of the plateau ecosystem and early warning of natural disasters. In response to the problem of the unsatisfactory accuracy in traditional real-time precipitation inversion methods and quantitative accuracy estimation products in this region. This paper adopts deep-learning methods to explore the corresponding relationship between precipitation and multi-band observation data from geostationary satellite imagery. The U-High Resolution Network (U-HRNet) as backbone is fully exploited to generate multi-scale features and fused using attention gates to extract high-resolution spatial representations representing precipitation clouds, thereby achieving precipitation estimation of target clouds at different scales. In addition, in response to the characteristics of high precipitation intensity and drastic cloud cluster changes in summer convective weather within a short period of time, a multi-head attention mechanism is introduced to extract temporal information from time series inputs and fuse it with spatial features to improve the accuracy of real-time precipitation estimation. Compared with the operational precipitation products and baseline deep-learning models, the classification metrics such as Probability of detection (POD), False Alarm Ratio (FAR) and Critical success index (CSI) show that the proposed model (U-HRNet+) has better performance in test set and has high generalization ability for estimating short-term heavy rainfall over Tibet and its surrounding areas in summer. Therefore, the proposed model has the potential to serve as a more accurate and reliable satellite-based precipitation estimation product. [ABSTRACT FROM AUTHOR]

The objective of this paper is to conduct a comprehensive investigation into the buckling and post-buckling behavior of hyperelastic beams through both computational and experimental means. Natural rubber is used in the construction of a beam with a square cross-section. To determine the mechanical properties of natural rubber, a uniaxial tensile test is performed in accordance with ASTM D412. In finite element modeling (FEM), the nonlinear behavior of rubber is modeled using hyperelastic theory and the Yeoh strain energy function. The Static-Riks method is also implemented using Abaqus for the analysis of nonlinear buckling. To validate the present investigation results with FEM, an experimental test of digital image correlation (DIC) is conducted. The critical buckling force obtained via numerical methods exhibits an error of nearly 5% when compared to the corresponding results obtained from experimental testing. In order to ascertain the impact of various design parameters on the buckling behavior of the system, a comprehensive parametric analysis has been conducted. The parameters studied include the cross-sectional thickness, length of the structure, eccentric loads, as well as the mechanical properties of the materials used in the system. Consistent with the FEM outcomes, the critical buckling force exhibited by the hyperelastic beam demonstrates a positive correlation with increasing levels of hardness, cross-sectional thickness, and eccentric loads. The buckling behavior of the system is adversely affected by increasing its length. To ultimately validate the precision and reliability of the model, a supervised neural network (NN) learning method is employed. [ABSTRACT FROM AUTHOR]

In this paper, we propose a gate recurrent unit (GRU) method to forecast ionospheric the total electron content (TEC) at different latitudes and longitudes. This method aims to address the issues of low accuracy in traditional neural networks and the challenges faced in training long-term time series. We construct a GRU model using TEC data from the European Orbit Determination Center (CODE) for the years 2016 to 2018, geomagnetic indices, and solar activity indices (16 points along the latitude and longitude lines of 120°E and 30°N). Comparative experiments are carried out with the international reference ionosphere (IRI) model and the long short-term memory (LSTM) model. The results show that the GRU model outperforms other models in terms of accuracy. On geomagnetic quiet days, the 5-day predicted average residual can reach 0.51 TECU, while on storm days it can reach 0.72 TECU. The root mean square error (RMSE) for these days is 0.64 TECU and 1.01 TECU, respectively. The correlation coefficients are 0.98 and 0.93, respectively. The RMSE of the GRU model ranges from 0.60 TECU to 2.62 TECU for both geomagnetic quiet days and storm days. In terms of TEC forecast accuracy in 2018, the RMSE of the GRU model is 1.6 TECU, which is smaller than that of the IRI model and LSTM model. These results demonstrate that the GRU model performs better than the IRI model and LSTM model. [ABSTRACT FROM AUTHOR]

Seabed topography is important for marine geophysics and geodesy. However, conventional seabed topography inversion methods based on gravity data are constrained to a linear fitting, neglecting the impact of nonlinear terms. In this paper, we propose an innovative method for seabed topography inversion by establishing fresh mathematical relationships through machine learning techniques based on the Smith and Sandwell (SAS) method. Utilizing global sea depth data from the National Geophysical Data Center (NGDC) and gravity anomaly data from satellite altimetry, our study employs the SAS method for seabed topography inversion. The improved SAS method, enhanced by the Genetic Algorithm-Backpropagation (GA-BP) algorithm, is specifically applied to invert the Huangyan Seamount Chain topography in the South China Sea. The accuracy is evaluated by seabed topography models, including EOTPO1, GEBCO-2023, and S & S V19.1. The results show the GA-BP method significantly reduces residuals and improves the accuracy estimated by the SAS inversion method. The Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) are decreased by 14.98% and 5.07%, respectively. The proposed method has valuable reference significance for future marine exploration endeavours around the world. [ABSTRACT FROM AUTHOR]

X‐ray computed tomography (CT) is widely used as a non‐destructive inspection technology. However, due to various limitations such as sample dimensions and blocking of in‐situ instruments, projections can be acquired only in a limited‐angle range during CT acquisition. When such projections are reconstructed with conventional algorithms, the details and contours are deteriorated by artefacts due to lack of information at certain view angles. To address this problem, we propose a reconstruction method based on U‐Conv‐Swin‐Net (UCSN) in this paper. Validated through synthetic and experimental data, the proposed method can effectively remove the degrading artefacts and fully restore the sample details and edges. In addition, the UCSN method exhibits superior reconstruction effect for different scanning ranges compared with conventional reconstruction methods. [ABSTRACT FROM AUTHOR]

Support vector machine (SVM) is a supervised machine-learning method which can be used for both classification and regression models. In this paper, we introduce a new model of SVM which any of training samples containing inputs and outputs are considered the random variables with known probability functions. The SVM is first converted into equivalent quadratic programming (QP) formulations in linear and nonlinear cases. An artificial neural network for SVM learning is then proposed. The presented neural network framework guarantees to obtain the optimal solution of the SVM problem. The existence and convergence of the trajectories of the network are studied. The Lyapunov stability for the considered neural network is also shown. The efficiency of the proposed method is shown by four illustrative examples. [ABSTRACT FROM AUTHOR]

Graph anomaly detection aims to identify the nodes that display significantly different behavior from the majority. However, existing methods neglect the combined interaction between the network structure and node attributes, resulting in suboptimal latent representations of nodes due to network noise. In this paper, we introduce a novel approach called adversarial regularized attributed network embedding (ARANE) for graph anomaly detection. ARANE addresses this issue by forcing normal nodes to inhabit a compact manifold in the latent space, taking into account both the network structure and node attributes. It ensures that data points from the normal class, originating from different distributions, are distributed within a single compact latent space, while excluding anomalies from this region. ARANE employs a dual-encoder architecture consisting of an attribute encoder and a structure encoder. The attribute encoder learns node attribute embeddings, while the structure encoder focuses on learning structure embeddings. To obtain high-quality node embeddings for effective anomaly detection, we apply adversarial learning to regularize the learned embeddings separately in both the structure and attribute spaces. Furthermore, we introduce a fusion module that combines the final node embeddings derived from the structure and attribute spaces. These joint embeddings serve as inputs to a dual-decoder for graph reconstruction, where the resulting reconstruction errors are utilized as anomaly scores for anomaly detection. Extensive experiments conducted on real-world attributed networks demonstrate the superior effectiveness of our proposed method compared to state-of-the-art approaches. • We introduce ARANE for one-class graph classification. It uses network structure & node attributes to tightly cluster normal nodes, excluding anomalies. • Our method uses a compact fusion module, capturing interactions between structural & attribute data in networks, for seamless integration. • Extensive experiments on real-world networks prove ARANE's effectiveness as a top solution for one-class graph classification. [ABSTRACT FROM AUTHOR]

Summary This paper addresses the adaptive saturation prescribed performance control problem for air‐breathing hypersonic vehicles with parameter uncertainties and unknown external disturbances. First of all, different from the traditional performance function, a novel class of performance functions is presented without the initial state information, which does not require to reset parameters even if the initial velocity and altitude change. To ensure the successful design of the velocity controller and altitude controller, the constrained errors are transformed into the unconstrained errors by the projection technique. Secondly, the adaptive neural network and the sliding mode control techniques are combined to design the adaptive neural network sliding mode controllers, which guarantee that the velocity and altitude tracking errors can respectively approach the predetermined regions within the identical preset finite time. Finally, the effectiveness of the designed controllers is verified by an example with comparisons. [ABSTRACT FROM AUTHOR]

Summary This article addresses the position control issue of multi‐quadrotor unmanned aerial vehicle (QUAV) formation. Concerning the translational dynamic of a multi‐QUAV system, on the one hand, it is an under‐actuation dynamic; on the other hand, it does not satisfy the matching condition. These features will cause inevitable thorny in the formation position control design. Furthermore, because of the state coupling problem, the formation control of multi‐QUAV system is more challenging and knotty than the control of single QUAV system. To achieve this control, both backstepping technique and neural network (NN) approximation strategy are combined by introducing an intermediary control, where NN is employed to compensate the system uncertainty. However, since the traditional adaptive NN control methods need to train a large number of adaptive parameters for the high approximation accuracy, it will cause the heavy computing burden if traditional adaptive method is used for the QUAV formation control. The proposed adaptive NN strategy in this paper only requires training a scalar adaptive parameter, which is generated from the norm of NN weight vector or matrix, thereby significantly reducing computational burden. Finally, according to Lyapunov stability proof and computer simulation, it is demonstrated that the control tasks can be successfully accomplished. [ABSTRACT FROM AUTHOR]

The straightness control of a fully mechanized working face is the key technology used in intelligent coal mining, so the position control of a hydraulic support multi-cylinder moving system is of great significance. However, due to the harsh environment of coal mines, complex friction, external disturbances, and the coupling relationship between adjacent cylinders, the accuracy of position control is restricted. Therefore, an output cooperative controller is proposed in this paper for a multi-cylinder system. A high-order sliding mode observer is utilized to estimate the system states with the only available output position signal. A neural network disturbance observer is applied to estimate the lump disturbance of the strict-feedback model, including the system uncertainty, disturbance force and the coupling force between adjacent cylinders. Then, continuous motion position tracking simulation is conducted and the estimation performance of the state observer and neural network is analyzed. Furthermore, a multi-cylinder collaborative control test rig is designed, and experiments based on the actual actions of the hydraulic support are conducted. The results show that the proposed output cooperative controller has an excellent position control performance compared with the traditional proportional–integral controller. [ABSTRACT FROM AUTHOR]

This paper presents a First-Order Recurrent Neural Network activated by a wavelet function, in particular a Morlet wavelet, with a fixed set of parameters and capable of identifying multiple chaotic systems. By maintaining a fixed structure for the neural network and using the same activation function, the network can successfully identify the three state variables of several different chaotic systems, including the Chua, PWL-Rössler, Anishchenko–Astakhov, Álvarez-Curiel, Aizawa, and Rucklidge models. The performance of this approach was validated by numerical simulations in which the accuracy of the state estimation was evaluated using the Mean Square Error (MSE) and the coefficient of determination ( r 2 ), which indicates how well the neural network identifies the behavior of the individual oscillators. In contrast to the methods found in the literature, where a neural network is optimized to identify a single system and its application to another model requires recalibration of the neural algorithm parameters, the proposed model uses a fixed set of parameters to efficiently identify seven chaotic systems. These results build on previously published work by the authors and advance the development of robust and generic neural network structures for the identification of multiple chaotic oscillators. [ABSTRACT FROM AUTHOR]

In recent years, the development of neural networks has significantly advanced their application in Synthetic Aperture Radar (SAR) ship target detection for maritime traffic control and ship management. However, traditional neural network architectures are often complex and resource intensive, making them unsuitable for deployment on artificial satellites. To address this issue, this paper proposes a lightweight neural network: the Multi-Scale SAR Ship Detection Network (MSSD-Net). Initially, the MobileOne network module is employed to construct the backbone network for feature extraction from SAR images. Subsequently, a Multi-Scale Coordinate Attention (MSCA) module is designed to enhance the network's capability to process contextual information. This is followed by the integration of features across different scales using an FPN + PAN structure. Lastly, an Anchor-Free approach is utilized for the rapid detection of ship targets. To evaluate the performance of MSSD-Net, we conducted extensive experiments on the Synthetic Aperture Radar Ship Detection Dataset (SSDD) and SAR-Ship-Dataset. Our experimental results demonstrate that MSSD-Net achieves a mean average precision (mAP) of 98.02% on the SSDD while maintaining a compact model size of only 1.635 million parameters. This indicates that MSSD-Net effectively reduces model complexity without compromising its ability to achieve high accuracy in object detection tasks. [ABSTRACT FROM AUTHOR]

By integrating energy-efficient AIoT-based biosensor networks, healthcare systems can now predict COVID-19 outbreaks with unprecedented accuracy and speed, revolutionizing early detection and intervention strategies. Therefore, this paper explores the rapid growth of electronic technology in today's environment, driven by the proliferation of advanced devices capable of monitoring and controlling various healthcare systems. However, these devices' limited resources necessitate optimizing their utilization. To tackle this concern, we propose an enhanced Artificial Intelligence of Things (AIoT) system that utilizes the networking capabilities of IoT biosensors to forecast potential COVID-19 outbreaks. The system aims to efficiently collect data from deployed sensor nodes, enabling accurate predictions of possible disease outbreaks. By collecting and pre-processing diverse parameters from IoT nodes, such as body temperature (measured non-invasively using the open-source thermal camera TermoDeep), population density, age (captured via smartwatches), and blood glucose (collected via the CGM system), we enable the AI system to make accurate predictions. The model's efficacy was evaluated through performance metrics like the confusion matrix, F1 score, precision, and recall, demonstrating the optimal potential of the IoT-based wireless sensor network for predicting COVID-19 outbreaks in healthcare systems. [ABSTRACT FROM AUTHOR]

Biomimetics is an emerging field that allows mimicry of living organisms in nature to develop different techniques so as to solve hard and complex problems related to optimization. The different techniques developed in this field takes inspiration from biology or nature. Biology acts as a powerful tool for imitating, copying, learning, understanding and inspiring the development of new systems and models. The different techniques discussed in this paper include techniques based on evolutionary algorithms, neural network and swarm intelligence. All these techniques are biologically inspired and provide good accuracy. The accuracy of all these algorithms can be increased by using them in hybrid form with other techniques and using different datasets. The comparative analysis of these techniques is done using advantages, disadvantages and applications of these techniques. [ABSTRACT FROM AUTHOR]

The extraction of weak acoustic signals under strong background noise is of great significance in the applications of target identification and localization. In this paper, the pulse signal with high randomness is set as the weak signal sound source, random noise and sine sound are used as the background noise. Under the condition of a signal-to-noise ratio of −20 dB, combined with blind source separation and neural network methods, the collected observation signals are subjected to weak sound signal separation and recognition research. The optimization method of centralization and scaling processing is used to eliminate the unfavorable influence of the uncertainty of the separated signal amplitude caused by the blind source separation method on the pattern recognition. The recognition result is verified by the combination of "weak impulse acoustic signal" and "random noise signal," and the output vector (0.99 0.01 0.01) approaches (1 0 0), which is recognized as impulse acoustic signal. By combining blind source separation and neural network methods, the separation and identification of weak pulse signals under the condition of a signal-to-noise ratio of −20 dB can be achieved. [ABSTRACT FROM AUTHOR]

This paper investigates the distributed optimization problem for a class of nonlinear uncertain multi-agent systems with unmeasured states, switched parameters, and directed communication topologies changing in the control process. To achieve the goal of optimizing the global objective function, a penalty function is constructed through making up of a sum of local objective functions and integrating consensus conditions of the multi-agent systems to utilize local and neighboring information. Radial basis function neural-networks and neural-networks state observer are applied to approximate the unknown nonlinear functions and obtain the unmeasured states, respectively. To avoid "explosion of complexity" and obtain derivatives for virtual control functions continuously, dynamic surface control technology is proposed to develop a distributed adaptive backstepping neural network control protocol to ensure that all the agents' outputs asymptotically reach consensus to the optimal solution of the global objective function. Simulations demonstrate the effectiveness of the proposed control scheme. [ABSTRACT FROM AUTHOR]

One of the most interesting problems in the recent renaissance of the studies in kernel regression might be whether kernel interpolation can generalize well, since it may help us understand the 'benign overfitting phenomenon' reported in the literature on deep networks. In this paper, under mild conditions, we show that, for any ε > 0 ⁠ , the generalization error of kernel interpolation is lower bounded by Ω (n − ε) ⁠. In other words, the kernel interpolation generalizes poorly for a large class of kernels. As a direct corollary, we can show that overfitted wide neural networks defined on the sphere generalize poorly. [ABSTRACT FROM AUTHOR]

The variable DC link direct torque control (DTC) for induction motors was proposed in this paper utilizing an artificial neural network (ANN). Variable DC-link voltage is utilized as the inverter's controlling input to improve the induction motor's performance. By injecting the dynamically variable dc-link voltage, inverter switching losses are decreased and inverter efficiency is increased. Two significant limitations are the torque ripple in DTC and the speed regulation of induction motors at low speeds. The driveline of an electric vehicle is modeled, and the suggested control is put into practice to enhance low-speed speed regulation, high dynamic performance, and minimal torque ripple. With variable DC-link voltage based on ANN, an improvement in torque and speed is made possible. Additionally, the suggested strategy minimizes the current ripples. To verify the improved efficacy of the electric driveline under various operating scenarios, a proposed control of the electric driveline is implemented using MATLAB SIMULINK, and the results are compared with the conventional scheme. The recommended speeds for the three speed ranges are as follows: 300 rpm for low speed, 900 rpm for medium speed, and 1275 rpm for high speed. The torque ripples decreased from 0.39 to 0.24 in the low-speed region, 0.38 to 0.24 in the medium-speed region, and 0.36 to 0.24 in the high-speed region while the motor was operated at a constant load torque of 2 N-m. The torque ripples decreased from 0.4 to 0.274 in the low speed region, 0.4 to 0.274 in the medium speed zone, and 0.2 to 0.274 in the medium speed region when the motor operated at various speed regions with a constant load. The main effect of running the motor at different load torques, such as 1.5 N-m, 2.5 N-m, and 2.0 N-m, is shown in Table 9 and numerical data of torque ripples is presented in Table 9. [ABSTRACT FROM AUTHOR]

Empirical evidence suggests that for a variety of overparameterized nonlinear models, most notably in neural network training, the growth of the loss around a minimizer strongly impacts its performance. Flat minima—those around which the loss grows slowly—appear to generalize well. This work takes a step towards understanding this phenomenon by focusing on the simplest class of overparameterized nonlinear models: those arising in low-rank matrix recovery. We analyse overparameterized matrix and bilinear sensing, robust principal component analysis, covariance matrix estimation and single hidden layer neural networks with quadratic activation functions. In all cases, we show that flat minima, measured by the trace of the Hessian, exactly recover the ground truth under standard statistical assumptions. For matrix completion, we establish weak recovery, although empirical evidence suggests exact recovery holds here as well. We complete the paper with synthetic experiments that illustrate our findings. [ABSTRACT FROM AUTHOR]

Radiologists and clinicians must automatically examine breast and tumor locations and sizes accurately. In recent years, several neural network‐based feature fusion versions have been created to improve medical image segmentation. Multi‐modal image fusion photos may efficiently identify tumors. This work uses image fusion to identify computed tomography and magnetic resonance imaging alterations. A Gauss‐log ratio operator is recommended for difference image production. The Gauss‐log ratio and log ratio difference image complement the objective of improving the difference map through image fusion. The feature change matrix extracts edge, texture, and intensity from each picture pixel. The final change detection map classifies feature vectors as "changed" or "unchanged" which has been mapped for high‐resolution or low‐resolution pixels. This paper proposes a multi‐feature blocks (MFB) based neural network for multi‐feature fusion. This neural network modeling approach globalizes pixel spatial relationships. MFB‐based feature fusion also aims to capture channel interactions between feature maps. The proposed technique outperforms state‐of‐the‐art approaches which have been discussed in detail in experimental results section. [ABSTRACT FROM AUTHOR]