Bearing fault diagnosis is of great importance to decrease the damage risk of rotating machines and further improve economic profits. Recently, machine learning, represented by deep learning, has made great progress in bearing fault diagnosis. However, applying deep learning to such a task still faces major challenges such as effectiveness and interpretability: i) When bearing signals are highly corrupted by noise, the performance of deep learning models drops dramatically; ii) A deep network is notoriously a black box. It is difficult to know how a model classifies faulty signals from the normal and the physics principle behind the classification. To solve these issues, first, we prototype a convolutional network with recently-invented quadratic neurons. This quadratic neuron-empowered network can qualify the noisy bearing data due to the strong feature representation ability of quadratic neurons. Moreover, we independently derive the attention mechanism from a quadratic neuron, referred to as qttention, by factorizing the learned quadratic function in analogue to the attention, making the model made of quadratic neurons inherently interpretable. Experiments on the public and our datasets demonstrate that the proposed network can facilitate effective and interpretable bearing fault diagnosis. Our code is available at https://github.com/asdvfghg/ QCNN_for_bearing_diagnosis.

Motor imagery EEG classification plays a crucial role in non-invasive Brain-Computer Interface (BCI) research. However, the performance of classification is affected by the non-stationarity and individual variations of EEG signals. Simply pooling EEG data with different statistical distributions to train a classification model can severely degrade the generalization performance. To address this issue, the existing methods primarily focus on domain adaptation, which requires access to the test data during training. This is unrealistic and impractical in many EEG application scenarios. In this paper, we propose a novel multi-source domain generalization framework called EEG-DG, which leverages multiple source domains with different statistical distributions to build generalizable models on unseen target EEG data. We optimize both the marginal and conditional distributions to ensure the stability of the joint distribution across source domains and extend it to a multi-source domain generalization framework to achieve domain-invariant feature representation, thereby alleviating calibration efforts. Systematic experiments conducted on a simulative dataset, BCI competition IV 2a, 2b, and OpenBMI datasets, demonstrate the superiority and competitive performance of our proposed framework over other state-of-the-art methods. Specifically, EEG-DG achieves average classification accuracies of 81.79% and 87.12% on datasets IV-2a and IV-2b, respectively, and 78.37% and 76.94% for inter-session and inter-subject evaluations on dataset OpenBMI, which even outperforms some domain adaptation methods. Our code is available at https://github.com/zxchit2022/EEG-DG for evaluation.

Gait phase analysis is widely used in disease diagnosis, rehabilitation training, and other fields by studying the characteristics of human gait. It systematically evaluates human body’s skeletal muscles and nerves with combined disciplines. Microelectromechanical system (MEMS) inertial sensors are extensively used in attitude detection because of its high-precision, portability, and good real-time performance in time-domain analysis. In this article, we presented a multi-degree-of-freedom MEMS gait detection method, which resolved the problems of single sensor and limited gait phase. We designed a sensor-based gait signal acquisition system, in which gait data acquisition program and feature analysis algorithm were compiled to verify the feasibility of the proposed method. We performed coordinate transformation and corrected position information to eliminate the gait phase detection error caused by random noise interference. Acceleration and angular velocity information were collected from 20 experimenters. We applied an adaptive threshold gait phase detection algorithm to classify the gait information collected by single sensor. To improve the results of gait phase classification, we used multisensor redundant measurement to analyze characteristics of five gait phases. The acceleration and angular velocity information collected by the three sensors placed at instep, ankle, and thigh were input into support vector machine (SVM). The classification results of the five gait phases are approximately 90%. Lastly, we built a human body structure model to simulate human motion in real time, realizing the real-time gait phase detection, which proves effectiveness of the proposed algorithm.

Deep learning has achieved remarkable success in the field of bearing fault diagnosis. However, this success comes with larger models and more complex computations, which cannot be transferred into industrial fields requiring models to be of high speed, strong portability, and low power consumption. In this paper, we propose a lightweight and deployable model for bearing fault diagnosis, referred to as BearingPGA-Net, to address these challenges. Firstly, aided by a well-trained large model, we train BearingPGA-Net via decoupled knowledge distillation. Despite its small size, our model demonstrates excellent fault diagnosis performance compared to other lightweight state-of-the-art methods. Secondly, we design an FPGA acceleration scheme for BearingPGA-Net using Verilog. This scheme involves the customized quantization and designing programmable logic gates for each layer of BearingPGA-Net on the FPGA, with an emphasis on parallel computing and module reuse to enhance the computational speed. To the best of our knowledge, this is the first instance of deploying a CNN-based bearing fault diagnosis model on an FPGA. Experimental results reveal that our deployment scheme achieves over 200 times faster diagnosis speed compared to CPU, while achieving a lower-than-0.4% performance drop in terms of F1, Recall, and Precision score on our independently-collected bearing dataset. Our code is available at https://github.com/asdvfghg/BearingPGA-Net.

Gait has been widely used in the fields of elderly care, posture correction, and identity recognition. By analyzing and processing the motion parameters and attitude data collected by the sensor, the current gait was determined. In order to pointedly prevent falling in daily life, we proposed a gait pattern classification method based on multisensor. Falling and the gait patterns (standing, sitting/rising, squatting/rising, walking, and running) that falling is likely to happen in daily life were distinguished. Besides, we integrated pressure signals with acceleration signals to compensate for the insufficient data provided by single sensor that cannot fully reflect the complex human motion and solved the problem that falling detection based on acceleration signals is prone to misclassify because certain postures are of similar acceleration changes with falling. For further data analysis, the collected gait data were then transmitted to upper machine for signal processing through the designed wireless network. Combined with the characteristics of gait patterns, we analyzed the corresponding pressure signals, acceleration signals, and resultant acceleration signals. Wavelet energy entropy features and wavelet packet energy features were subsequently extracted from the collected gait data. Finally, we input the randomly selected test data into the established extreme learning machine (ELM) and $K$ -nearest neighbor (KNN) model to test gait pattern recognition effects. The performance of ELM algorithm was better in terms of processing time and classification results, with the highest average identification accuracy, precision, and recall rate of 0.974, 0.937, and 0.936, respectively. Besides, precision–recall (PR) curve was optimal, with the largest area of 0.973. Our presented algorithm responded rapidly and prevented falling in daily lifetime, which can be applied to health monitoring systems to detect daily activities of the elderly promptly.

Accurate and reliable prediction of bearing remaining useful life (RUL) is crucial to the prognostics and health management of rotation machinery. Despite the rapid progress of data-driven methods, the generalizability of data-driven models remains an open issue to be addressed. In this article, we tackle this challenge by resolving the feature misalignment problem that arises in extracting features from the raw vibration signals. Toward this goal, we introduce a logarithmic cumulative transformation (LCT) operator consisting of cumulative, logarithmic, and another cumulative transformation for feature extraction. In addition, we propose a novel method to estimate the reliability associated with each RUL prediction by integrating a linear regression model and an auxiliary exponential model. The linear regression model rectifies bias from neural network's point predictions while the auxiliary exponential model fits the differential slopes of the linear models and generates the upper and lower bounds for building the reliability indicator. The proposed approach comprised of LCT, an attention GRU-based encoder–decoder network, and reliability evaluation is validated on the FEMETO-ST dataset. Computational results demonstrate the superior performance of the proposed approach several other state-of-the-art methods.