A Fuzzy Fusion Rotating Machinery Fault Diagnosis Framework Based on the Enhancement Deep Convolutional Neural Networks

Some artificial intelligence algorithms have gained much attention in the rotating machinery fault diagnosis due to their robust nonlinear regression properties. In addition, existing deep learning algorithms are usually dependent on single signal features, which would lead to the loss of some information or incomplete use of the information in the signal. To address this problem, three kinds of popular signal processing methods, including Fast Fourier Transform (FFT), Short-Time Fourier Transform (STFT) and directly slicing one-dimensional data into the two-dimensional matrix, are used to create four different datasets from raw vibration signal as the input data of four enhancement Convolutional Neural Networks (CNN) models. Then, a fuzzy fusion strategy is used to fuse the output of four CNN models that could analyze the importance of each classifier and explore the interaction index between each classifier, which is different from conventional fusion strategies. To show the performance of the proposed model, an artificial fault bearing dataset and a real-world bearing dataset are used to test the feature extraction capability of the model. The good anti-noise and interpretation characteristics of the proposed method are demonstrated as well.

Sleep Apnea Detection Based on Multi-Scale Residual Network

Aiming at the fact that traditional convolutional neural networks cannot effectively extract signal features in complex application scenarios, a sleep apnea (SA) detection method based on multi-scale residual networks is proposed. First, we analyze the physiological mechanism of SA, which uses the RR interval signals and R peak signals derived from the ECG signals as input. Then, a multi-scale residual network is used to extract the characteristics of the original signals in order to obtain sensitive characteristics from various angles. Because the residual structure is used in the model, the problem of model degradation can be avoided. Finally, a fully connected layer is introduced for SA detection. In order to overcome the impact of class imbalance, a focal loss function is introduced to replace the traditional cross-entropy loss function, which makes the model pay more attention to learning difficult samples in the training phase. Experimental results from the Apnea-ECG dataset show that the accuracy, sensitivity and specificity of the proposed multi-scale residual network are 86.0%, 84.1% and 87.1%, respectively. These results indicate that the proposed method not only achieves greater recognition accuracy than other methods, but it also effectively resolves the problem of low sensitivity caused by class imbalance.

A topographical and physiological exploration of C-tactile afferents and their response to menthol and histamine

Unmyelinated tactile (CT) afferents are abundant in arm hairy skin and have been suggested to signal features of social affective touch. Here we recorded from unmyelinated low-threshold mechanosensitive afferents in the peroneal and radial nerves, with the most distal receptive fields located on the proximal phalanx of the third finger for the superficial branch of the radial nerve, and near the lateral malleolus for the peroneal nerve. We found that the physiological properties with regard to conduction velocity and mechanical threshold, as well as their tuning to brush velocity, were similar in CT units across the antebrachial (n=27), radial (n=8) and peroneal nerves (n=4). Moreover, we found that while CT afferents are readily found during microneurography of the arm nerves, they appear to be much more sparse in the lower leg compared to C nociceptors. We continued to explore CT afferents with regard to their chemical sensitivity and found that they could not be activated by topical application to their receptive field of either the cooling agent menthol or the pruritogen histamine. In light of previous studies showing the combined effects that temperature and mechanical stimuli have on these neurons, these findings add to the growing body of research suggesting that CT afferents constitute a unique class of sensory afferents with highly specialized mechanisms for transducing gentle touch.

Machine Algorithm for Heartbeat Monitoring and Arrhythmia Detection Based on ECG Systems

Cardiac arrhythmia is an illness in which a heartbeat is erratic, either too slow or too rapid. It happens as a result of faulty electrical impulses that coordinate the heartbeats. Sudden cardiac death can occur as a result of certain serious arrhythmia disorders. As a result, the primary goal of electrocardiogram (ECG) investigation is to reliably perceive arrhythmias as life-threatening to provide a suitable therapy and save lives. ECG signals are waveforms that denote the electrical movement of the human heart (P, QRS, and T). The duration, structure, and distances between various peaks of each waveform are utilized to identify heart problems. The signals’ autoregressive (AR) analysis is then used to obtain a specific selection of signal features, the parameters of the AR signal model. Groups of retrieved AR characteristics for three various ECG kinds are cleanly separated in the training dataset, providing high connection classification and heart problem diagnosis to each ECG signal within the training dataset. A new technique based on two-event-related moving averages (TERMAs) and fractional Fourier transform (FFT) algorithms is suggested to better evaluate ECG signals. This study could help researchers examine the current state-of-the-art approaches employed in the detection of arrhythmia situations. The characteristic of our suggested machine learning approach is cross-database training and testing with improved characteristics.

Combine Assembly Fault Diagnosis Based on Optimized Multi-scale Reverse Discrete Entropy

An optimized multi-scale reverse discrete entropy (RDE, OMRDE) method for feature extraction is proposed to address the lack of effective feature extraction and detection methods for combining harvester assembly fault inspection. This method is used to extract vibration signal features from the harvester. A fault diagnostic method is designed to verify the efficiency of the associated methods. First, a comparative study of RDE, multi-scale inverse RDE (MRDE), and OMRDE was performed using simulated signals to verify the effectiveness of OMRDE. Second, the FSTPSO–VMD method was used to decompose the vibration signal of the combine harvester assembly fault, and the OMRDE, MRDE, and fuzzy entropy were compared and analyzed. The actual feature extraction effect of the three entropy functions reached the highest classification accuracy (88.5%) after using OMRDE to extract features. Finally, a fusion feature set is constructed to further improve the classification accuracy, and the LSSVM classifier is further optimized through FSTPSO. Analytical results show that the FSTPSO–LSSVM classifier constructed in this work adopts the fused feature with an accuracy of 93%, which is better than other common methods and verifies the validity of the fault diagnostic model. Therefore, the performance of the OMRDE proposed in this work is better than those of MRDE and MRDE. The proposed fault diagnostic model can realize accurate classification of the combine harvester assembly fault detection.

Principal Components Analysis of EEG Signals for Epileptic Patient Identification

According to the behavior of its neuronal connections, it is possible to determine if the brain suffers from abnormalities such as epilepsy. This disease produces seizures and alters the patient’s behavior and lifestyle. Neurologists employ the electroencephalogram (EEG) to diagnose the disease through brain signals. Neurologists visually analyze these signals, recognizing patterns, to identify some indication of brain disorder that allows for the epilepsy diagnosis. This article proposes a study, based on the Fourier analysis, through fast Fourier transformation and principal component analysis, to quantitatively identify patterns to diagnose and differentiate between healthy patients and those with the disease. Subsequently, principal component analysis can be used to classify patients, employing frequency bands as the signal features. Besides, it is made a classification comparison before and after using principal component analysis. The classification is performed via logistic regression, with a reduction from 5 to 4 dimensions, as well as from 8 to 7, achieving an improvement when there are 7 dimensions in the precision, recall, and F1 score metrics. The best results obtained, without PCA are: precision 0.560, recall 0.690, and F1 score 0.620; meanwhile, the best values obtained using PCA are: precision 0.734, recall 0.787, and F1 score 0.776.

Intellectualization of decision making in security systems of protected objects

The paper discusses the prospects for building decision blocks (DB) for modern security systems built on the basis of fiber optic detectors. An algorithm has been developed that makes it possible to train an artificial neural network as part of a DB. Approaches to the formation of a training sample are outlined. To reduce the dimension of the problem of recognizing informative signal features, an extreme filtering algorithm is presented, which makes it possible to increase the speed and accuracy of training.