In the context of discrete-event systems (DES), the terms detection and diagnosis refer to two distinct stages of handling faults and anomalies. Both steps are critical for ensuring the reliable and safe operation of ...
详细信息
In the context of discrete-event systems (DES), the terms detection and diagnosis refer to two distinct stages of handling faults and anomalies. Both steps are critical for ensuring the reliable and safe operation of complex systems. In this paper, we propose the use of autoencoders for fault detection in an automated production system with sensors and actuators delivering discrete binary signals that can be modeled as DES. We train an autoencoder exclusively on data representing normal behavior. The model learns to encode typical patterns and reconstruct input data with low loss. A predetermined threshold, determined by the characteristics of the training data, is set for the reconstruction error. During normal behavior, the autoencoder is expected to achieve low reconstruction error below this threshold. When a fault occurs, the autoencoder strives to accurately reconstruct faulty data, leading to a higher error. The detection of a reconstruction error exceeding the threshold signals a potential fault in the system. The results of applying our method to the Factory IO software sorting system demonstrate the significant contribution and the interest of this method for detecting faults.
By using machine learning algorithms, banks and other lending institutions can construct intelligent risk control models for loan businesses, which helps to overcome the disadvantages of traditional evaluation methods...
详细信息
By using machine learning algorithms, banks and other lending institutions can construct intelligent risk control models for loan businesses, which helps to overcome the disadvantages of traditional evaluation methods, such as low efficiency and excessive reliance on the subjective judgment of auditors. However, in the practical evaluation process, it is inevitable to encounter data with missing credit characteristics. Therefore, filling in the missing characteristics is crucial for the training process of those machine learning algorithms, especially when applied to rural banks with little credit data. In this work, we proposed an autoencoder-based algorithm that can use the correlation between data to restore the missing data items in the features. Also, we selected several open-source datasets (German Credit Data, Give Me Some Credit on the Kaggle platform, etc.) as the training and test dataset to verify the algorithm. The comparison results show that our model outperforms the others, although the performance of the autoencoder-based feature restorer decreases significantly when the feature missing ratio exceeds 70%.
Accurately detecting defects while reconstructing a high-quality normal background in surface defect detection using unsupervised methods remains a significant challenge. This study proposes an unsupervised method tha...
详细信息
Accurately detecting defects while reconstructing a high-quality normal background in surface defect detection using unsupervised methods remains a significant challenge. This study proposes an unsupervised method that effectively addresses this challenge by achieving both accurate defect detection and a high-quality normal background reconstruction without noise. We propose an adaptive weighted structural similarity (AW-SSIM) loss for focused feature learning. AW-SSIM improves structural similarity (SSIM) loss by assigning different weights to its sub-functions of luminance, contrast, and structure based on their relative importance for a specific training sample. Moreover, it dynamically adjusts the Gaussian window's standard deviation (sigma) during loss calculation to balance noise reduction and detail preservation. An artificial defect generation algorithm (ADGA) is proposed to generate an artificial defect closely resembling real ones. We use a two-stage training strategy. In the first stage, the model trains only on normal samples using AW-SSIM loss, allowing it to learn robust representations of normal features. In the second stage of training, the weights obtained from the first stage are used to train the model on both normal and artificially defective training samples. Additionally, the second stage employs a combined learned Perceptual Image Patch Similarity (LPIPS) and AW-SSIM loss. The combined loss helps the model in achieving high-quality normal background reconstruction while maintaining accurate defect detection. Extensive experimental results demonstrate that our proposed method achieves a state-of-the-art defect detection accuracy. The proposed method achieved an average area under the receiver operating characteristic curve (AuROC) of 97.69% on six samples from the MVTec anomaly detection dataset.
Mode shape is a dynamic characteristic that plays an important role in civil engineering. In this paper, an approach to predict the mode shape of a bridge is proposed using a convolutional neural network (CNN) and an ...
详细信息
Mode shape is a dynamic characteristic that plays an important role in civil engineering. In this paper, an approach to predict the mode shape of a bridge is proposed using a convolutional neural network (CNN) and an autoencoder. First, a large mode shape database of a bridge is established by the finite element method for training networks. Second, a mode shape tensor is formed based on the mode-shape results. Then, an autoencoder is trained to encode the tensor to a three-dimensional latent-space representation and restore it from the representations. The CNN can output the representation directly rather than the mode shape to reduce the training difficulty and improve the accuracy. The CNN takes 18 bridge design parameters and an original shape tensor, which is constructed based on 16 geometric parameters. An evaluation of the test set shows that the approach can predict the first three order mode shapes well, with the accuracy of 0.92, 0.83 and 0.79, while performs poorly in the fourth and fifth orders, with the accuracy of 0.68 and 0.63. In addition, the spatial distribution of the latent space representation is explored. The necessity of an autoencoder and the original shape tensor is demonstrated.
To addresses the challenges of data scarcity and weak computational capabilities of edge devices in the practical application of non-intrusive load monitoring (NILM) of power systems, we propose a semi-federated learn...
详细信息
This paper presents an anomaly detection approach with non-invasive on-chip temperature sensing for hardware Trojan detection, which is coupled with a proposed anomaly detection technique using an autoencoder-based ma...
详细信息
autoencoder (AE) networks are utilized in novelty detection, classification, and deep clustering tasks to learn feature representation. While AEs have demonstrated promising performance in various applications, we obs...
详细信息
The identification of pulsar candidates is a crucial step in radio astronomy research. With the continuous improvement of modern radio telescope equipment and the increasing scale of pulsar sky survey, a pulsar survey...
详细信息
The accurate monitoring of abnormal production conditions in cement process is the basis of intelligent control, which is of great significance to improve the intelligent level of cement production. In this paper, an ...
详细信息
Human Activity Recognition (HAR) with inertial sensors is one of the most active research fields. Various machine learning algorithms have been proposed in HAR for classifying human activities. However, these methods ...
详细信息
暂无评论