One-dimensional Convolutional Neural Network Research Articles

Zero-shot knowledge transfer for seismic damage diagnosis through multi-channel 1D CNN integrated with autoencoder-based domain adaptation

Accurate and timely structural damage diagnosis is crucial to efficient disaster response and city renovation in post-earthquake events. The scarcity of labeled data hinders the powerful deep learning techniques from in-domain damage detection on target structures. Cross-domain transfer learning has emerged as a captivating strategy through digging knowledge from the abundant source domain to detect the damage in the target domain. However, the heterogeneity among multi-domain structures poses the challenge in seismic damage diagnosis. This study proposes a novel zero-shot knowledge transfer approach for seismic damage diagnosis through multi-channel one-dimensional convolutional neural networks (1D CNN) integrated with deep autoencoder (DAE)-based domain adaptation (DA). The framework consists of three modules, namely, data preprocessor adaptive to seismic vibration signals, DAE-based DA module, and damage diagnosis via multi-channel 1D CNN. The DA module is customized to seamlessly translate the unseen target-domain data to mimic latent representation via a DAE pretrained on the source data, thus realizing rigorous zero-shot learning. Imbalanced data distribution is also considered during the network training and testing. Two representative phases of knowledge transfer are performed to substantiate the knowledge transferability of the proposed method. The first phase involves multi-class damage quantification on two ASCE benchmark models from the simplified model to the refined one, and the second phase conducts binary damage detection on a three-story reinforced frame structure from the finite element numerical model to the laboratory-tested physical model. Both examples show that the proposed method exhibits high prediction accuracy and a lower false negative rate in achieving zero-shot knowledge transfer for cross-domain structural damage diagnosis. With a delicate network design for diverse data, the proposed knowledge transfer framework can be further extended from the present zero-shot approach to the few-shot learning paradigm, thus suggesting a feasible algorithm adaptability and promising engineering applicability.

1D-CNN-based audio tampering detection using ENF signals

The extensive adoption of digital audio recording has revolutionized its application in digital forensics, particularly in civil litigation and criminal prosecution. Electric network frequency (ENF) has emerged as a reliable technique in the field of audio forensics. However, the absence of comprehensive ENF reference datasets limits current ENF-based methods. To address this, this study introduces ATD, a blind audio forensics framework based on a one-dimensional convolutional neural network (1D-CNN) model. ATD can identify phase mutations and waveform discontinuities within the tampered ENF signal, without relying on an ENF reference database. To enhance feature extraction, the framework incorporates characteristics of the fundamental harmonics of ENF signals. In addition, a denoising method termed ENF noise reduction (ENR) based on the variational mode decomposition (VMD) and robust filtering algorithm (RFA) is proposed to reduce the impact of external noise on embedded electric network frequency signals. This study investigates three distinct types of audio tampering—deletion, insertion, and replacement—culminating in the design of binary-class tampering detection scenarios and four-class tampering detection scenarios tailored to these tampering types. ATD achieves a tampering detection accuracy of over 93% in the four-class scenario and exceeds 96% in the binary-class scenario. The effectiveness, efficiency, adaptability, and robustness of ATD in the two and four classification scenarios have been confirmed by extensive experiments.

Physiologic validation of the Compensatory Reserve Metric obtained from pulse oximetry: A step toward advanced medical monitoring on the battlefield.

The Compensatory Reserve Metric (CRM) provides a time sensitive indicator of hemodynamic decompensation. However, its in-field utility is limited because of the size and cost-intensive nature of standard vital sign monitors or photoplethysmographic volume-clamp (PPG VC ) devices used to measure arterial waveforms. In this regard, photoplethysmographic measurements obtained from pulse oximetry may serve as a useful, portable alternative. This study aimed to validate CRM values obtained using pulse oximeter (PPG PO ). Forty-nine healthy adults (25 females) underwent a graded lower body negative pressure (LBNP) protocol to simulate hemorrhage. Arterial waveforms were sampled using PPG PO and PPG VC . The CRM was calculated using a one-dimensional convolutional neural network. Cardiac output and stroke volume were measured using PPG VC . A brachial artery catheter was used to measure intra-arterial pressure. A three-lead electrocardiogram was used to measure heart rate. Fixed-effect linear mixed models with repeated measures were used to examine the association between CRM values and physiologic variables. Log-rank analyses were used to examine differences in shock determination during LBNP between monitored hemodynamic parameters. The median LBNP stage reached was 70 mm Hg (range, 45-100 mm Hg). Relative to baseline, at tolerance, there was a 47% ± 12% reduction in stroke volume, 64% ± 27% increase in heart rate, and 21% ± 7% reduction in systolic blood pressure ( p < 0.001 for all). Compensatory Reserve Metric values obtained with both PPG PO and PPG VC were associated with changes in heart rate ( p < 0.001), stroke volume ( p < 0.001), and pulse pressure ( p < 0.001). Furthermore, they provided an earlier detection of hemodynamic shock relative to the traditional metrics of shock index ( p < 0.001 for both), systolic blood pressure ( p < 0.001 for both), and heart rate ( p = 0.001 for both). The CRM obtained from PPG PO provides a valid, time-sensitized prediction of hemodynamic decompensation, opening the door to provide military medical personnel noninvasive in-field advanced capability for early detection of hemorrhage and imminent onset of shock. Diagnostic Tests or Criteria; Level III.

Authentication with a one-dimensional CNN model using EEG-based brain-computer interface

Brain-computer interface (BCI) technology uses electroencephalogram (EEG) signals to create a direct interaction between the human body and its surroundings. Motor imagery (MI) classification using EEG signals is an important application that can help a rehabilitated or motor-impaired stroke patient perform certain tasks. Robust classification of these signals is an important step toward making the use of EEG more practical in many applications and less dependent on trained professionals. Deep learning methods have produced impressive results in BCI in recent years, especially with the availability of large electroencephalography (EEG) data sets. Dealing with EEG-MI signals is difficult because noise and other signal sources can interfere with the electrical amplitude of the brain, and its generalization ability is limited, so it is difficult to improve EEG classifiers. To address these issues, this paper presents a methodology based on one-dimensional convolutional neural networks (1-D CNN) for motor imagery (MI) recognition for the right hand, left hand, feet, and sedentary task. The proposed model is a lightweight model with fewer parameters and has an accuracy of 91.75%. Then, in an innovative exploitation of the four output classes, there is an idea that allows people with disabilities who are deprived of security measures, such as entering a secret code, to use the output classification, such as password codes. It is also an idea for a unique authentication system that is more secure and less vulnerable to theft or the like for a healthy person at the same time.

Raman and Mid-Infrared Spectroscopy Coupled With Machine–Deep Learning for Adulterant Detection in Ground Turmeric

The intricate nature of the global food supply chain and the presence of regulations spanning multiple jurisdictions contribute to an increased likelihood of food adulteration. This underscores the need for effective monitoring methods to guarantee the safety and nutritional quality of our food. In this context, the application of infrared spectroscopy-based techniques emerges as an environmentally friendly, non-invasive, and waste-minimizing solution for authenticating food products. Infrared spectra serve as unique molecular fingerprints, offering a multidimensional representation of how chemical bonds in the material interact with infrared light. Chemometrics, which are primarily linear-based models, play a crucial role in extracting essential information from spectral data, enabling dimensionality reduction, classification, and predictive analysis. Recent progress in the field of big data science and artificial intelligence has brought forth machine learning and deep learning algorithms explicitly designed to uncover features from complex multidimensional data, encompassing both linear and nonlinear relationships. These advancements have the potential to enhance the detection of adulterants in food products. This study assesses the accuracy of various shallow machine learning models and a deep learning model based on a one-dimensional convolutional neural network (1D CNN). The evaluation is conducted using Raman and infrared spectral data obtained from ground turmeric samples that were deliberately adulterated with five distinct substances. The study highlights the improved classification accuracy achieved through the implementation of the 1D CNN model.

ОДНОВИМІРНА ЗГОРТКОВА МОДЕЛЬ НЕЙРОННОЇ МЕРЕЖІ ДЛЯ ОБРОБЛЕННЯ СИГНАЛІВ З АМПЛІТУДНОЮ МОДУЛЯЦІЄЮ БАГАТЬОХ СКЛАДОВИХ

The processing of radio signals using artificial neural networks (ANNs) has great potential for research, which can be explained by the adaptability of ANNs to various transmission conditions and the ability to detect abstract patterns of changes in signal parameters. The article reviews the works of other authors devoted to different ways of using ANNs for processing radio signals. Taking into account the information in the reviewed works, the research task was formed, which consists in developing an optimized ANN model for radio signal processing. Signals with amplitude modulation of many components (AMMC) were chosen to form training samples for ANN. The choice of modulation type is justified by greater energy efficiency compared to other widely used digital modulation types, such as quadrature amplitude modulation. Mathematic basis of AMMC signal generation is described. The process of finding the coordinates of three component 8-AMMC signal constellation is explained, the formation of signals in the time plane based on the found coordinates is explained as well as their discretization and the addition of white noise. An iterative algorithm for generating initial data for ANN based on the described ratios is proposed. The general structure of one-dimensional convolutional neural network is considered. Functions of individual neurons, connections between them, the formation of layers and the convolution operation are described mathematically. On the basis of the previously given ratios, a final display of the network was formed. Specific dimensions and activation functions for layers are selected. The use of convolutional layers is justified by time invariance. Based on the reviewed mathematical models, selected activation functions and dimensions, a neural model was formed. The process of validating the effectiveness of the formed neural model is described, which is based on comparing the symbolic error probabilities of the proposed and reference models at different signal-to-noise ratios. The validation results are presented. The advantages of the obtained model over the previously proposed purely recurrent model and the AMMC reference receiver are explained.

Similarity analysis of chloride transport behavior in fly ash concrete under different environments aiding by machine learning method

To forecast chloride transport in site by applying accelerated test results, it is necessary to investigate the similarity of chloride diffusivity under different environments. However, limited by test period and cost expenses, a large amount of tests in field and simulated environment are hard to conduct, and a limit test data will make prediction on chloride diffusivity of concrete less precise. To solve this problem, machine learning methods are explored to predict chloride transport. Based on the existing chloride concentration test data of fly ash concrete in natural and accelerated simulated environment, this paper establishes a chloride concentration predictive model based on the one-dimensional convolutional neural network (1D-CNN) method at first, and verifies the stability and accuracy of the established model. Second, similarity models for concentration ratio and diffusion time ratio based on multivariate nonlinear model and based on theoretical Fick’s second law are constructed, and the model accuracy is also tested. Finally, the chloride concentration data predicted by the 1D-CNN model is applied to the constructed similarity models respectively, and the application results are compared and analyzed to determine the optimal similarity model. Results indicate that the concentration predicted by the 1D-CNN method has a strong robustness and a small prediction error. Besides, based on multivariate nonlinear model, the error of concentration predicted by the similarity models for concentration ratio and diffusion time ratio is less than 16 % and 30 %, respectively. Similarly, according to the theoretical model based on Fick’s second law, the error of concentration predicted by the similarity models for concentration ratio based on the time dependent chloride diffusion coefficient and time dependent peak chloride concentration is less than 21 % and 36 %, respectively. The similarity model for concentration ratio based on multivariate nonlinear model has the best prediction effect.

ECG waveform generation from radar signals: A deep learning perspective

Cardiovascular diagnostics relies heavily on the ECG (ECG), which reveals significant information about heart rhythm and function. Despite their significance, traditional ECG measures employing electrodes have limitations. As a result of extended electrode attachments, patients may experience skin irritation or pain, and motion artifacts may interfere with signal accuracy. Additionally, ECG monitoring usually requires highly trained professionals and specialized equipment, which increases the treatment's complexity and cost. In critical care scenarios, such as continuous monitoring of hospitalized patients, wearable sensors for collecting ECG data may be difficult to use. Although there are issues with ECG, it remains a valuable tool for diagnosing and monitoring cardiac disorders due to its non-invasive nature and the detailed information it provides about the heart. The goal of this study is to present an innovative method for generating continuous ECG waveforms from non-contact radar data by using Deep Learning. The method can eliminate the need for invasive or wearable biosensors and expensive equipment to collect ECGs. In this paper, we propose the MultiResLinkNet, a one-dimensional convolutional neural network (1D CNN) model for generating ECG signals from radar waveforms. With the help of a publicly accessible radar benchmark dataset, an end-to-end DL architecture is trained and assessed. There are six ports of raw radar data in this dataset, along with ground truth physiological signals collected from 30 participants in five distinct scenarios: Resting, Valsalva, Apnea, Tilt-up, and Tilt-down. By using strong temporal and spectral measurements, we assessed our proposed framework's ability to convert ECG data from Radar signals in three distinct scenarios, namely Resting, Valsalva, and Apnea (RVA). ECG segmentation performed better by MultiResLinkNet than by state-of-the-art networks in both combined and individual cases. As a result of the simulations, the resting, valsalva, and RVA scenarios showed the highest average temporal values, respectively: 66.09523 ± 19.33, 60.13625 ± 21.92, and 61.86265 ± 21.37. In addition, it exhibited the highest spectral correlation values (82.4388 ± 18.42 (Resting), 77.05186 ± 23.26 (Valsalva), 74.65785 ± 23.17 (Apnea), and 79.96201 ± 20.82 (RVA)), along with minimal temporal and spectral errors in almost every case. The qualitative evaluation revealed strong similarities between generated and actual ECG waveforms. As a result of our method of forecasting ECG patterns from remote radar data, we can monitor high-risk patients, especially those undergoing surgery.

A double-branch convolutional neural network model for species identification based on multi-modal data

For species identification analysis, methods based on deep learning are becoming prevalent due to their data-driven and task-oriented nature. The most commonly used convolutional neural network (CNN) model has been well applied in Raman spectra recognition. However, when faced with similar molecules or functional groups, the features of overlapping peaks and weak peaks may not be fully extracted using the CNN model, which can potentially hinder accurate species identification. Based on these practical challenges, the fusion of multi-modal data can effectively meet the comprehensive and accurate analysis of actual samples when compared with single-modal data. In this study, we propose a double-branch CNN model by integrating Raman and image multi-modal data, named SI-DBNet. In addition, we have developed a one-dimensional convolutional neural network combining dilated convolutions and efficient channel attention mechanisms for spectral branching. The effectiveness of the model has been demonstrated using the Grad-CAM method to visualize the key regions concerned by the model. When compared to single-modal and multi-modal classification methods, our SI-DBNet model achieved superior performance with a classification accuracy of 98.8%. The proposed method provided a new reference for species identification based on multi-modal data fusion.

Comparing the compensatory reserve metric obtained from invasive arterial measurements and photoplethysmographic volume-clamp during simulated hemorrhage.

The compensatory reserve metric (CRM) is a novel tool to predict cardiovascular decompensation during hemorrhage. The CRM is traditionally computed using waveforms obtained from photoplethysmographic volume-clamp (PPGVC), yet invasive arterial pressures may be uniquely available. We aimed to examine the level of agreement of CRM values computed from invasive arterial-derived waveforms and values computed from PPGVC-derived waveforms. Sixty-nine participants underwent graded lower body negative pressure to simulate hemorrhage. Waveform measurements from a brachial arterial catheter and PPGVC finger-cuff were collected. A PPGVC brachial waveform was reconstructed from the PPGVC finger waveform. Thereafter, CRM values were computed using a deep one-dimensional convolutional neural network for each of the following source waveforms; (1) invasive arterial, (2) PPGVC brachial, and (3) PPGVC finger. Bland-Altman analyses were used to determine the level of agreement between invasive arterial CRM values and PPGVC CRM values, with results presented as the Mean Bias [95% Limits of Agreement]. The mean bias between invasive arterial- and PPGVC brachial CRM values at rest, an applied pressure of -45mmHg, and at tolerance was 6% [-17%, 29%], 1% [-28%, 30%], and 0% [-25%, 25%], respectively. Additionally, the mean bias between invasive arterial- and PPGVC finger CRM values at rest, applied pressure of -45mmHg, and tolerance was 2% [-22%, 26%], 8% [-19%, 35%], and 5% [-15%, 25%], respectively. There is generally good agreement between CRM values obtained from invasive arterial waveforms and values obtained from PPGVC waveforms. Invasive arterial waveforms may serve as an alternative for computation of the CRM.

Gait Impairment Analysis Using Silhouette Sinogram Signals and Assisted Knowledge Learning.

The analysis of body motion is a valuable tool in the assessment and diagnosis of gait impairments, particularly those related to neurological disorders. In this study, we propose a novel automated system leveraging artificial intelligence for efficiently analyzing gait impairment from video-recorded images. The proposed methodology encompasses three key aspects. First, we generate a novel one-dimensional representation of each silhouette image, termed a silhouette sinogram, by computing the distance and angle between the centroid and each detected boundary points. This process enables us to effectively utilize relative variations in motion at different angles to detect gait patterns. Second, a one-dimensional convolutional neural network (1D CNN) model is developed and trained by incorporating the consecutive silhouette sinogram signals of silhouette frames to capture spatiotemporal information via assisted knowledge learning. This process allows the network to capture a broader context and temporal dependencies within the gait cycle, enabling a more accurate diagnosis of gait abnormalities. This study conducts training and an evaluation utilizing the publicly accessible INIT GAIT database. Finally, two evaluation schemes are employed: one leveraging individual silhouette frames and the other operating at the subject level, utilizing a majority voting technique. The outcomes of the proposed method showed superior enhancements in gait impairment recognition, with overall F1-scores of 100%, 90.62%, and 77.32% when evaluated based on sinogram signals, and 100%, 100%, and 83.33% when evaluated based on the subject level, for cases involving two, four, and six gait abnormalities, respectively. In conclusion, by comparing the observed locomotor function to a conventional gait pattern often seen in healthy individuals, the recommended approach allows for a quantitative and non-invasive evaluation of locomotion.

Deep learning algorithm development for early detection of Botrytis cinerea infected strawberry fruit using hyperspectral fluorescence imaging

Botrytis cinerea is a strawberry disease that causes economic loss worldwide. If a disease outbreak occurs during storage or transportation, it can spread rapidly to neighboring objects; thus, there is a need to develop early diagnostic techniques to prevent it. In this study, we developed a method to rapidly and nondestructively determine the infection stage in strawberry fruit using hyperspectral fluorescence imaging. ‘Keumsil’ cultivar strawberries were used, and hyperspectral fluorescence images were acquired over 144 h in control and inoculation groups. Strawberries were categorized into four infection stages based on visible mold spores: healthy, asymptomatic, infected, and after-infected. Hyperspectral fluorescence spectra were extracted to develop a one-dimensional convolutional neural network (1D-CNN) model based on partial least squares-discriminant analysis (PLS-DA), VGG-19, and ResNet-50; data augmentation techniques and six spectral preprocessing techniques were applied to the datasets. The application of data augmentation techniques improved the performances of the PLS-DA and 1D-CNN models in determining the infection stage. The performance of the ResNet-50-based 1D-CNN model with mean normalization data and data augmentation technique was the best, with 96.88% precision, 96.87% recall, 96.85% F1-score, and 96.86% accuracy. The results of this study showed that it is possible to determine the infection stage of Botrytis cinerea on strawberry fruit using hyperspectral fluorescence imaging and 1D-CNN techniques. This technology is expected to be applied for the early detection of Botrytis cinerea in strawberry growth, postharvest sorting and packing, and distribution stages.

Non-invasive monitoring of microalgae cultivations using hyperspectral imager

High expectations are placed on microalgae as a sustainable source of valuable biomolecules. Robust methods to control microalgae cultivation processes are needed to enhance their efficiency and, thereafter, increase the profitability of microalgae-based products. To meet this need, a non-invasive monitoring method based on a hyperspectral imager was developed for laboratory scale and afterwards tested on industrial scale cultivations. In the laboratory experiments, reference data for microalgal biomass concentration was gathered to construct 1) a vegetation index-based linear regression model and 2) a one-dimensional convolutional neural network model to resolve microalgae biomass concentration from the spectral images. The two modelling approaches were compared. The mean absolute percentage error (MAPE) for the index-based model was 15–24%, with the standard deviation (SD) of 13-18 for the different species. MAPE for the convolutional neural network was 11–26% (SD = 10–22). Both models predicted the biomass well. The convolutional neural network could also classify the monocultures of green algae by species (accuracy of 97–99%). The index-based model was fast to construct and easy to interpret. The index-based monitoring was also tested in an industrial setup demonstrating a promising ability to retrieve microalgae-biomass-based signals in different cultivation systems.

Soil data augmentation and model construction based on spectral difference and content difference

Soil analysis makes for developing precision agriculture and monitoring land quality, while the models available for spectroscopy-based chemometrics are constrained by limited samples from small areas. The paper proposed sample expansion and model construction based on spectral difference and content difference, realizing data augmentation and deep learning applied to original samples with limited numbers. The spectral subtraction based on maximum or minimum values exploited the maximum or minimum values to acquire the spectral difference and content difference, which provided a new data form for model construction. Keeping enhanced samples whose spectral difference and content difference were all zero was useful for improving model performance. Augmentation of all data or training data based on maximum or minimum values-based spectral subtraction, which sorted the contents and made them the maximum or minimum values in sequence, achieved sample expansion by the spectral difference and content difference. The model utilized the random vector functional link (RVFL) network, extreme learning machine (ELM), and one-dimensional convolutional neural network (1D CNN), which could predict the content of new samples through ensemble averaging when predicting content difference. The experimental result showed the model of the spectral subtraction based on maximum or minimum values had a similar performance to that of the original samples. Augmentation of all data improved model performance by only RVFL and ELM. Augmentation of training data verified 1D CNN was better than RVFL and ELM. The paper implements a new data augmentation method and applies CNN to original samples with inadequate numbers, which lays the foundation for an improved model and applying spectral preprocessing.

Grounding Fault Diagnosis of Running Rails Based on a Multi-scale One-Dimensional Convolutional Neural Network in a DC Subway System

Running rails are the return conductors of the traction current in DC subway systems, which should be insulated from the earth. Due to the large length of the line, as well as the humidity and metal dust in the tunnel, grounding fault of the running rails usually occurs, which will increase stray current (SC) leakage and endanger power supply safety. In this paper, a method of grounding fault diagnosis of running rails based on a multi-scale one-dimensional convolutional neural network (MS-1DCNN) is proposed. Firstly, a platform for the dynamic distribution of SC and rail potential (RP) with grounding faults existing in the running rails is established, which generates the dynamic RP data with various grounding faults. Secondly, a grounding fault diagnostic method of running rails based on MS-1DCNN is proposed, so as to realize the effective identification of ground fault types in subway systems. Thirdly, with the proposed diagnostic method, the datasets under two operating conditions of a single train and two trains are tested; a comparison test between MS-1DCNN and the 1D convolutional neural network (1DCNN) is carried out, and the effectiveness of the proposed method is verified. Results demonstrate that the proposed model can significantly improve the ground fault diagnostic accuracy of running rails. The dynamic RP simulation platform for trains established in this paper lays a theoretical foundation for the grounding fault research of running rail. Moreover, the deep learning method is proposed for the first time to diagnose the grounding fault of running rail, and the high diagnostic accuracy is obtained, which is of great significance for the safe and stable operation of the subway line.

DSTN: Dynamic Spatio-Temporal Network for Early Fault Warning in Chemical Processes

Multivariate time series prediction, especially in early fault warning for chemical processes, poses significant challenges. The advent of graph neural network (GNN) method has made breakthroughs in this domain by enabling the processing of topological data. However, the traditional methods suffer from the issue of over-smoothing and inability to capture intricate multi-scale spatio-temporal dependencies. Additionally, the existing graph structures fall short in describing the complex spatial relationships among multi-stage sensors, impeding their adaptability to dynamically evolving chain reaction scenarios. To alleviate these limitations, a novel Dynamic spatio-Temporal Network for early fault warning in chemical processes, named DSTN for short, is proposed in this paper. We extract the spatial and temporal features of the time series by the designed dynamic GNN and the improved Transformer network. Then, we integrate the spatio-temporal features through the residual network. DSTN has the following advantages: (1) A one-dimensional convolutional neural network is seamlessly incorporated into the Transformer architecture for bolstering its capacity to discern both global and local features within time series. (2) The continuous sliding window and mutual information methods are employed to construct a dynamic topology graph, and a K-order adjacency matrix is designed to rectify the inefficiencies in learning weights associated with convolution kernel parameters. (3) Multiple spatio-temporal modules interconnected via residual connection to adaptively fuse multi-scale features. Experimental results demonstrate that our proposed DSTN method outperforms existing methods in terms of both performance and interpretability in early fault warning of chemical processes.

Wavy-attention network for real-time cyber-attack detection in a small modular pressurized water reactor digital control system

Global interest in advanced reactors has been reignited by recent investments in small modular reactors and micro-reactor design. The use of digital devices is essential for meeting the size and modularity requirements of small modular reactor controls. By fully digitizing the small modular reactor control systems, critical information can be obtained to optimize control, reduce costs, and extend the reactor’s lifetime. However, the potential for cyber-attacks on digital devices leaves digital control systems vulnerable. To address this risk, this study presents a novel wavy-attention network for sensor attack detection in nuclear plants. The wavy-attention network comprises stacks of batch-normalized, dilated, one-dimensional convolution neural networks, and sequential self-attention modules, superior to conventional single-layer networks on sequence classification tasks. To evaluate the proposed wavy-attention network architecture, the International Atomic Energy Agency’s Asherah Nuclear Simulator and a false data injection toolbox found in the literature, both implemented in MATLAB/SIMULINK, are utilized. This approach leverages changes in process measurements to identify and classify cyber-attacks on priority signals using the proposed wavy-attention network. Three false data injection attacks are simulated on the simulator’s pressure, temperature, and level sensors to obtain representative process measurements. The wavy-attention network is trained and validated with normal and compromised process variables obtained from the simulator. The performance of the wavy-attention network to discriminate between the reactor states using the test set shows 99% accuracy, as opposed to other baseline models such as vanilla convolution neural networks, long short-term memory networks, and bi-directional long short-term memory networks with 90%, 77%, and 91% accuracy, respectively. An ablation study is also conducted to test the contribution of each component of the proposed architecture. The theoretical framework of the proposed wavy-attention network and its implementation for nuclear reactor digital sensor attack detection are discussed in this paper.

Similarity indicator and CG-CGAN prediction model for remaining useful life of rolling bearings

To tackle the challenges of performing early fault warning and improving the prediction accuracy for the remaining useful life (RUL) of rolling bearings, this paper proposes a similarity health indicator and a predictive model of CG-conditional generative adversarial network (CGAN), which relies on a CGAN that combines one-dimensional convolutional neural network (CNN) with a bidirectional gate recurrent unit (Bi-GRU). This framework provides a comprehensive theoretical foundation for RUL prediction of rolling bearings. The similarity health indicator allows for early fault warning of rolling bearings without expert knowledge. Within the CGAN framework, the inclusion of constraints guides the generation of samples in a more targeted manner. Additionally, the proposed CG-CGAN model incorporates Bi-GRU to consider both forward and backward information, thus improving the precision of RUL forecasting. Firstly, the similarity indicator between the vibration signals of the rolling bearing over its full life span and the standard vibration signals (healthy status) is calculated. This indicator helps to determine the early deterioration points of the rolling bearings. Secondly, the feature matrix composed of traditional health indicators and similarity health indicator, is utilized to train and test the proposed CG-CGAN model for RUL prediction. Finally, to corroborate the efficacy of the proposed method, two sets of real experiment data of rolling bearing accelerated life from the Intelligent Maintenance Systems (IMS) are utilized. Experimental findings substantiate that the proposed similarity health indicator offers early fault alerts and precisely delineates the performance diminution of the rolling bearing. Furthermore, the put-forward CG-CGAN model achieves high-precision RUL prediction of rolling bearing.

Deep learning neural networks with input processing for vibration-based bearing fault diagnosis under imbalanced data conditions

Deep learning (DL) networks, such as convolutional neural networks (CNNs) and long short-term memory (LSTM), have gained popularity for bearing fault diagnosis utilizing raw vibration signals. However, their accuracy and stability are compromised when facing imbalanced real-world datasets. This research investigates the impact of imbalanced datasets and explores the potential of signal processing techniques on network inputs compared to the direct use of raw vibration signals. The DL techniques studied include LSTM, one-dimensional CNN, and two-dimensional (2D) CNN, and a novel hybrid 2DCNNLSTM algorithm, incorporating signal processing methods such as Fourier transform and continuous wavelet transform while maintaining nearly equal parameters and the same base architecture. The proposed hybrid 2DCNNLSTM algorithm combines the strengths of LSTM and CNN, allowing for improved bearing diagnosis by capturing both spatial and temporal information in vibration signals. The proposed 2DCNNLSTM algorithm also considers multi-channel input augmenting raw vibration signal, mean, and variance channels to extract meaningful features and enhance classification efficiency. The publicly available Case Western Reserve University benchmark-bearing test rig dataset with ten fault classes, the Paderborn University dataset with three fault classes, and NASA Centre for Intelligent Maintenance Systems bearing datasets with five fault classes are utilized to test the proposed deep learning networks’ accuracy, effectiveness, robustness, and stability. The studies reveal that the hybrid 2DCNNLSTM-based networks outperform both CNN and LSTM networks, even without input processing. Further, utilizing multi-channel input by augmenting the 2D raw signal with mean and variance value channels proves to be more efficient in handling imbalanced and complex datasets while employing a 2DCNNLSTM-based network.

Communication behavior recognition using CNN-based signal analysis

This article explores the technology of recognizing non-cooperative communication behavior, with a specific emphasis on analyzing communication station signals. Conventional techniques for analyzing signal data frames to determine their identity, while precise, do not have the ability to operate in real-time. In order to tackle this issue, we developed a pragmatic architecture for recognizing communication behavior and a system based on polling. The method utilizes a one-dimensional convolutional neural network (CNN) to segment data, hence improving its ability to recognize various communication activities. The study assesses the reliability of CNN in several real-world scenarios, examining its accuracy in the presence of noise interference, varying lengths of interception signals, interferences at different frequency points, and dynamic changes in outpost locations. The experimental results confirm the efficacy and dependability of the convolutional neural network in recognizing communication behavior in various contexts.