Acceleration Signals Research Articles

Effect of sensor number and location on accelerometry-based vertical ground reaction force estimation during walking.

Knee osteoarthritis is a major cause of global disability and is a major cost for the healthcare system. Lower extremity loading is a determinant of knee osteoarthritis onset and progression; however, technology that assists rehabilitative clinicians in optimizing key metrics of lower extremity loading is significantly limited. The peak vertical component of the ground reaction force (vGRF) in the first 50% of stance is highly associated with biological and patient-reported outcomes linked to knee osteoarthritis symptoms. Monitoring and maintaining typical vGRF profiles may support healthy gait biomechanics and joint tissue loading to prevent the onset and progression of knee osteoarthritis. Yet, the optimal number of sensors and sensor placements for predicting accurate vGRF from accelerometry remains unknown. Our goals were to: 1) determine how many sensors and what sensor locations yielded the most accurate vGRF loading peak estimates during walking; and 2) characterize how prescribing different loading conditions affected vGRF loading peak estimates. We asked 20 young adult participants to wear 5 accelerometers on their waist, shanks, and feet and walk on a force-instrumented treadmill during control and targeted biofeedback conditions prompting 5% underloading and overloading vGRFs. We trained and tested machine learning models to estimate vGRF from the various sensor accelerometer inputs and identified which combinations were most accurate. We found that a neural network using one accelerometer at the waist yielded the most accurate loading peak vGRF estimates during walking, with average errors of 4.4% body weight. The waist-only configuration was able to distinguish between control and overloading conditions prescribed using biofeedback, matching measured vGRF outcomes. Including foot or shank acceleration signals in the model reduced accuracy, particularly for the overloading condition. Our results suggest that a system designed to monitor changes in walking vGRF or to deploy targeted biofeedback may only need a single accelerometer located at the waist for healthy participants.

Data fusion based on short-term memory Kalman filtering using intermittent-displacement and acceleration signal with a time-varying bias

Accurate displacement information plays a vital role in various applications from structural health monitoring to damage detection in physical systems. Due to practical reasons, displacement sensing is often performed using sensors that provide less accuracy or a low sampling rate. This study proposes an algorithm for displacement signal estimation that fuses high sampling rate acceleration data (which consists of unknown time-varying bias) with low sampling rate displacement data. In contrast to the present literature, the state vector in the proposed state-space formulation consists of short-term memory of the past realizations of displacement and velocity, and bias terms which allow (i) online estimation of the displacements by avoiding multi-rate Kalman filtering, and (ii) estimate time-varying bias in the measured acceleration signal. The state equations comprise the finite difference approximations of the time derivative of velocity and displacement which act as a moving average filter to provide smooth estimates, thus preventing the need for offline smoothing procedures. By inclusion of bias terms in the state vector, the drifts in displacement and velocity caused due to time-varying bias in measured acceleration are easily eliminated. The proposed method was found to provide accurate estimates of displacement, velocity, and time-varying bias in measured acceleration signals when tested on different types of bias, including but not limited to constant, linear, and parabolic variations. Additional tests on three oscillators subjected to three earthquake ground motions showed significantly better state estimates for stiff oscillators and improved estimates for flexible oscillators. Experimental validation on a base-excited two-story frame further affirms the capability of the method to estimate the hidden bias in the measured acceleration and provide an accurate estimate of displacement.

Identification of optimal accelerometer placement on trains for railway switch wear monitoring via multibody simulation

Accelerometers play a crucial role in the railway industry, especially in track monitoring. Traditionally, they are placed on the railway tracks or often on bridges to monitor the health and condition of the infrastructure. Recently, there has been an increased focus on using regular trains to monitor the condition of railway infrastructure. Often, the sensors are placed based on certain assumptions without much scientific evidence or support. This paper utilizes the multibody simulation software GENSYS to identify the optimal placement of accelerometers on a passenger train for monitoring railway switch wear. Switch wear profiles were generated systematically and used as input for the simulations, studying acceleration at a total of 93 locations distributed among the wheelsets, bogies, and carbody. Based on both time and frequency domain analyses, optimal sensor locations were identified, generally close to the first bogie or wheelset at the leading carbody. Accelerations generated by the wheelset passing the switch can also be captured in the carbody, but it is important to note that these are several orders lower in magnitude compared to the acceleration on the wheelset. If accelerometers are to be placed in the carbody, correct sensitivity must be chosen, and a high-pass filter should be applied to capture the acceleration signals associated with switch wear. The study confirms that there is a direct correlation between the depth of switch wear and the magnitude of the acceleration. It remains effective even under various curve radii and train speeds.

A measurement method of in-bore projectile motion acceleration based on bottom pressure correction

Due to factors such as high temperatures, elevated pressures, and severe high-frequency shocks in the bore, there is considerable noise interference in acceleration test signals. This makes it challenging to accurately measure projectile motion acceleration using existing methods. To address this issue, we propose an advanced measurement approach that uses bottom pressure correction. Our model suggests a significant correlation between projectile motion acceleration and thrust. By utilizing the correlations between bottom pressure and motion acceleration in both temporal and frequency domains, we can improve the accuracy of acceleration measurements. Based on these insights, we have developed a novel testing system that synchronously measures bottom pressure and acceleration, using bottom pressure as a corrective mechanism for the measured acceleration signals. Empirical results show that the maximum relative error in peak motion acceleration is only 4.86%, demonstrating the effectiveness of our proposed method.

An Accurate Modeling Method of Body-thread of Penetration Fuze in FEM

To address the issue that the influence of body-thread on the acceleration signal characteristics in penetration fuze is ambiguous, an accurate modeling method of body-thread is proposed in the finite element method (FEM). The body-thread element segments are obtained by the three-dimensional entity thread model, the initial layer nodes set and elements on the outer surface of the thread are established, and the initial layer elements are copied, rotated, and translated. Repeat this transformation, and get a complete thread model. This method is applied to simulation analysis of a warhead penetrating 4-layer concrete plates, and the influence of thread connection on acceleration signal characteristics in penetration fuze is studied. By comparing the acceleration signal and frequency spectrum of the proposed thread model with that of the simplified thread model, the amplitude of the acceleration signal of the proposed one is smaller than that of the simplified one in some certain frequency spectrum. It shows that the thread connection structure has with damping effect and effectively reduces the high-frequency interference of acceleration signal in penetration fuze.

Effects of Pre-Stretching Amount on Motion Characteristics and Screening Efficiency of Flip-Flow Screen Plates

This study systematically investigates the impact of pre-stretching amount (PSA) on the acceleration, motion status, and screening efficiency of the Flip-flow screen plate (FFSP). Initially, a nonlinear spring-multi-body model of the FFSP is established. Subsequently, the acceleration signals at the midpoint of the FFSP under various PSAs are measured and analyzed. The nonlinear stiffness coefficient of FFSP along the vertical direction is then determined and it is found that simplifying the nonlinear spring-multi-body system to a nonlinear spring-tri-body system under the experimental conditions can maintain the calculation error of FFSP’s acceleration within 30%. Phase and Poincaré mapping diagrams of the FFSP under different PSAs are subsequently created to illustrate the impact of PSA on the motion status of the FFSP. Finally, screening experiments are performed to study the optimal PSA for a kind of bituminous coal from Shanxi province.

Attitude Calculation Method of Drilling Tools Based on Cross-Correlation Extraction and ASRUKF

As a key component of the measurement while drilling technology, the accuracy of attitude calculation is directly related to the efficiency of resource exploration. To reduce the influence of vibration, rotation, and other disturbances on the attitude sensor during drilling, a method based on cross-correlation extraction and the adaptive square-root unscented Kalman filter (ASRUKF) is proposed to solve the attitude of the drilling tool in this paper. Firstly, the error of the signal collected by the attitude sensor is compensated, and the unscented Kalman filter (UKF) is used for filtering. Then, the effective gravitational acceleration signal is extracted by the cross-correlation method. Finally, an experimental platform for simulating the fully rotating attitude measurement system is established, and the application effects of the UKF and ASRUKF in the attitude calculation are compared. Compared with the UKF, the root mean square error of the inclination angle calculated by the ASRUKF is reduced by 12.9%, and the variance is reduced by 27.3%; the root mean square error of the azimuth angle is reduced by 29.5%, and the variance is reduced by 39.9%. The experimental results show that the attitude calculation method proposed in this paper can stably and effectively improve the accuracy of the attitude calculation of drilling tools.

Experimental Analysis of Smart Drilling for the Furniture Industry in the Era of Industry 4.0.

The fact is that hundreds of holes are drilled in the assembly process of furniture sets, so intelligent drilling is a key element in maximizing efficiency. Increasing the feed rate or the cutting speed in materials characterized by a higher machinability index is necessary. Smart drilling, that is, the real-time adjustment of the cutting parameters, requires the evolution of cutting process variables. In addition, it is necessary to control and adjust the processing parameters in real time. Machinability is one of the most important technological properties in the machining process, enabling the determination of the material's susceptibility to machining. One of the machinability indicators is the unit cutting resistance. This article proposes a method of material identification using the short-time Fourier transform in order to automatically adjust cutting parameters during drilling based on force signals, cutting torque and acceleration signals. In the tests, four types of wood-based materials were used as the processed material: medium-density fiberboard, chipboard, plywood board and high-pressure laminate. Holes with a diameter of 10 mm were drilled in the test materials, with variable feed rate, cutting speed and thickness of cutting layer. An innovative method for determining the value of unit cutting resistance was proposed. The results obtained were used to determine the machinability index. Based on the test results, it was shown that both the selected signal measures in the time and frequency domains and the unit cutting resistance are constant for a given material of a workpiece and do not depend on the drilling process parameters. In this article, the methodology is proposed, which can be used as an intelligent technique to support the drilling process to detect the material being machined using data from sensors installed on the machine tool. The work proposes the fundamentals for material identification based on the analysis of force signals and the magnitude of force derivatives. The proposed methodology shows effectiveness, which proves that it can be used in intelligent drilling processes. Hybrid wood-based material structures consisting of different materials are becoming more and more common in building structures for strength, economic and environmental reasons. Due to the difference in the machinability of interconnected materials, cutting parameters must be optimized in real time during machining. Currently, with the rapid development of Industry 4.0, the on-line identification of parameters is becoming necessary to improve the process flow in industrial reality. The proposed methodology can be used as an intelligent technique to support the drilling process in order to detect the material being processed using data from sensors installed on the machine tool.

ДВОКАНАЛЬНИЙ ГРАВІМЕТР АВІАЦІЙНОЇ ГРАВІМЕТРИЧНОЇ СИСТЕМИ

A two-gyro gravimeter of the aviation gravimetric system is proposed and investigated, which provides higher measurement accuracy than known gravimeters, due to the elimination of errors from the cross angular velocities of the base and the angular velocity of the Earth's rotation and the measurement of the full vector of the acceleration of gravity. To measure the acceleration of gravity, an AGS is proposed, which has greater accuracy and speed than known and consists of a three-stage gyroscope located in the inner and outer frames, equipped with interframe correction systems, which include an angle sensor (DC) located on the axis of the inner frame of the gyroscope and a torque sensor (DM) connected to its output is located on the axis of the outer frame. DM, located on the axis of the inner frame, is connected to the DC output. An additional, identical to the first, three-stage gyroscope, the rotor of which rotates in the opposite direction from the main gyroscope, is introduced in the AGS under consideration. The additional gyroscope of the AGS is also provided with similar correction systems, which consist of a DC located on the axis of the inner frame, and a DM located on the axis of the outer frame connected to its output, a DC located on the axis of the outer frame, to the output of which a DM located on the axis is connected inner frame The centers of gravity of two identical (main and additional) gyroscopes are shifted by the same distance in one direction along the axes of rotation of the rotors of the gyroscopes relative to the axes of the outer frames. The vectors of the kinetic moments of the two gyroscopes are oppositely directed. In the two-channel gravimeter, two output signals of linear acceleration are formed as the sum of signals from DC of two gyroscopes relative to one z-axis and as the sum of signals from DC of two gyroscopes relative to the second x-axis. The output signals are fed to the computer. Both gyroscopes of the system's gyrogravimeter are mounted on a platform whose angular position is controlled by a motor (DV) installed on the x-axis and a DV on the z-axis. Both motors control the angular position of the platform by signals. The computer also receives signals from the system for determining navigation parameters and from the height meter.

Study on Gesture Recognition Method with Two-Stream Residual Network Fusing sEMG Signals and Acceleration Signals.

Currently, surface EMG signals have a wide range of applications in human-computer interaction systems. However, selecting features for gesture recognition models based on traditional machine learning can be challenging and may not yield satisfactory results. Considering the strong nonlinear generalization ability of neural networks, this paper proposes a two-stream residual network model with an attention mechanism for gesture recognition. One branch processes surface EMG signals, while the other processes hand acceleration signals. Segmented networks are utilized to fully extract the physiological and kinematic features of the hand. To enhance the model's capacity to learn crucial information, we introduce an attention mechanism after global average pooling. This mechanism strengthens relevant features and weakens irrelevant ones. Finally, the deep features obtained from the two branches of learning are fused to further improve the accuracy of multi-gesture recognition. The experiments conducted on the NinaPro DB2 public dataset resulted in a recognition accuracy of 88.25% for 49 gestures. This demonstrates that our network model can effectively capture gesture features, enhancing accuracy and robustness across various gestures. This approach to multi-source information fusion is expected to provide more accurate and real-time commands for exoskeleton robots and myoelectric prosthetic control systems, thereby enhancing the user experience and the naturalness of robot operation.

A proposed tacho-less order tracking technique suitable for railway vehicle condition monitoring applications

To perform a condition monitoring for rotating machines with variable running speed, order tracking is considered as an effective technique suitable for their non-stationary vibration signals. A tachometer signal is necessary to perform a computed order tracking (COT) technique; however, in some applications it's not possible to add tachometer hardware due to its complexity or shortage of power. To improve the applicability of order tracking techniques, some tacho-less order tracking techniques have been introduced in the last few years. These techniques are working off-line by inspecting the vibration signals in time-frequency domain to extract the machine rotating speed. In this paper, a new tacho-less technique suitable for continuous structural health monitoring applications of railway vehicles, is proposed. In the presented approach, a coarse angular speed of the railway axle is estimated firstly using estimation train speed. Then this coarse angular speed is used to improve the applicability of the traditional tacho-less techniques to be able to work online. The train speed is estimated from two acceleration signals obtained from vertical vibration of axle-boxes from same side of train. The sequential properties of the two signals are used to estimate the delay time of one signal with respect to the other one by means of cross-correlation function. By knowing the horizontal distance between the two sensors collected the vibration signals, the train speed can be estimated. This modification of tacho-less technique improves the effectiveness and robustness of the traditional maxima tracking strategy to be performed automatically. Thereafter, signal processing technique based on the computed order tracking as well as averaging technique can be applied to extract the useful information from noisy and non-stationary signals. This technique has been proposed in a framework of a project investigating crack detection using vibration measurements, therefore, vibration signals of cracked axle have been processed using tacho-based method and using the proposed tacho-less technique and a good agreement has been found between the two techniques.

Wearable Sensor‐Based Multi‐modal Fusion Network for Automated Gait Dysfunction Assessment in Children with Cerebral Palsy

Gait, fundamental to human movement, becomes compromised in cerebral palsy (CP), a childhood‐onset central nervous system motor disorder. Precise assessment of patients’ gait is crucial for tailored rehabilitation interventions. Currently, clinical scales assessing CP gait dysfunction mostly, while valuable, rely on subjective clinician observations. To enhance objectivity and efficiency in CP diagnosis and rehabilitation, there is a need for more objective assessment procedures. This study introduces a multi‐modal and multi‐scale feature fusion (MMFF) framework, a new framework for automating gait dysfunction assessment in children with CP. By utilizing surface electromyography and acceleration signals recorded during children's walking, MMFF generates a feature vector enriched with adaptively refined feature maps, cross‐mode correlations, and both local and global information. Validation of MMFF's effectiveness is evident through an accomplished classification accuracy of 99.13%. The mean values for precision, recall, and F1‐score in Gross Motor Function Classification System (GMFCS)‐1, GMFCS‐2, and GMFCS‐3, reaching 99.00%, 99.00%, and 98.33%, respectively, further reflect the accuracy of functional assessments at each level. This study underscores MMFF's potential as an objective, streamlined tool for clinicians, promising improved gait assessment and well‐informed rehabilitation strategies for children with CP.

Influence of microtextured parameters of dry gas sealing rings on tribological performance

PurposeThis study aims to research the influence mechanism of microtextured geometric parameters of dry gas seal end face on the tribological behavior under dry frictional conditions.Design/methodology/approachThe microtexture was processed using laser processing, while the diamond-like carbon (DLC) film was applied through magnetron sputtering; the experimental platform of friction vibration was established, the frictional and vibrational properties of different geometric parameters were tested; the data signals of vibrational acceleration and frictional torque were collected and processed using data acquisition instrument. The entropy characteristic parameters of 3D vibrational acceleration were extracted based on wavelet packet decomposition method. The end-face topography was measured with ST400 three-dimensional noncontact surface topography instrument.FindingsThe geometry of pits plays a key role in influencing friction performance; the permutation entropy and fuzzy entropy of the vibration acceleration signal changed with variations in microtextured parameters. A textured surface with appropriately size parameters can trap debris, enhance the dynamic pressure effect, reduce impact between the friction interfaces and improve the frictional vibrational performance. In this research, microtextured surface with Φ150 µm-10% and Φ200 µm-5% can effectively reduce friction and vibration between the end faces of a dry gas seal.Originality/valueDLC film improves the hardness of seal ring end face, and microtexture improves the dynamic effect; the tribological behavior monitoring can be realized by analyzing the characteristics of vibration acceleration sensitive parameter with friction state. The findings will provide a basis for further research in the field of tribology and the microtexture optimization of dry gas seal ring end face.Peer reviewThe peer review history for this article is available at: https://publons.com/publon/10.1108/ILT-12-2023-0389/

Assessing the Acceptance of a Mid-Air Gesture Syntax for Smart Space Interaction: An Empirical Study

Mid-gesture interfaces have become popular for specific scenarios, such as interactions with augmented reality via head-mounted displays, specific controls over smartphones, or gaming platforms. This article explores the use of a location-aware mid-air gesture-based command triplet syntax to interact with a smart space. The syntax, inspired by human language, is built as a vocative case with an imperative structure. In a sentence like “Light, please switch on!”, the object being activated is invoked via making a gesture that mimics its initial letter/acronym (vocative, coincident with the sentence’s elliptical subject). A geometrical or directional gesture then identifies the action (imperative verb) and may include an object feature or a second object with which to network (complement), which also represented by the initial or acronym letter. Technically, an interpreter relying on a trainable multidevice gesture recognition layer makes the pair/triplet syntax decoding possible. The recognition layer works on acceleration and position input signals from graspable (smartphone) and free-hand devices (smartwatch and external depth cameras), as well as a specific compiler. On a specific deployment at a Living Lab facility, the syntax has been instantiated via the use of a lexicon derived from English (with respect to the initial letters and acronyms). A within-subject analysis with twelve users has enabled the analysis of the syntax acceptance (in terms of usability, gesture agreement for actions over objects, and social acceptance) and technology preference of the gesture syntax within its three device implementations (graspable, wearable, and device-free ones). Participants express consensus regarding the simplicity of learning the syntax and its potential effectiveness in managing smart resources. Socially, participants favoured the Watch for outdoor activities and the Phone for home and work settings, underscoring the importance of social context in technology design. The Phone emerged as the preferred option for gesture recognition due to its efficiency and familiarity. The system, which can be adapted to different sensing technologies, addresses the scalability concerns (as it can be easily extended for new objects and actions) and allows for personalised interaction.

A novel dust collection system fault diagnosis based on the noise emission of solenoid pulse valve

The solenoid pulse valve (SPV) is a key component of dust collector. The frequent switch of SPV accelerates its damage, subsequently decrease operation efficiency of dust collector. The traditional manual inspection methods lack reliability and efficiency, while vibration-based condition monitoring is costly and sophisticated in acceleration sensor distribution and signal processing. This paper proposes a fault diagnosis approach based on SPV noise. The average coherence coefficient of SPV noise and vibration signals is above 0.8 and close to 1, which demonstrates that there is a close relationship between SPV noise and its structural vibration. Furthermore, the validation experiments indicate that the SPV fault can be clearly distinguished by its noise signal, confirms the effectiveness of proposed approach. By utilizing this noise monitoring method, the number of sensors can be reduced by about 95 %, the work intensity of the staff is greatly reduced, and production efficiency and safety are guaranteed.

Damage Detection in Bridge Structures through Compressed Sensing of Crowdsourced Smartphone Data

Traditional bridge health monitoring methods that necessitate sensor installation are not only costly but also time-consuming. In contrast, utilizing smartphone data collected from vehicles as they traverse bridges offers an efficient and cost-effective alternative. This paper introduces a cutting-edge damage detection framework for indirect monitoring of bridge structures, leveraging a substantial volume of acceleration data collected from smartphones in vehicles passing over the bridge. Our innovative approach addresses the challenge of collecting and transmitting high-frequency data while preserving smartphone battery life and data plans through the integration of compressed sensing (CS) into the crowdsensing-based monitoring framework. CS employs random sampling and signal recovery from a significantly reduced number of samples compared to the requirements of the Nyquist–Shannon sampling theorem. In the proposed framework, acceleration signals from vehicles are initially acquired using smartphone sensors, undergo compression, and are then transmitted for signal reconstruction. Subsequently, feature extraction and dimensionality reduction are performed using Mel-frequency cepstral coefficients and principal component analysis. Damage indexes are computed based on the dissimilarity between probability distribution functions utilizing the Wasserstein distance metric. The efficacy of the proposed methodology in bridge monitoring has been substantiated through the utilization of numerical models and a lab-scale bridge. Furthermore, the feasibility of implementing the framework in a real-world application has been investigated, leveraging the smartphone data from 102 vehicle trips on the Golden Gate Bridge. The results demonstrate that damage detection using the reconstructed signals obtained through compressed sensing achieves comparable performance to that obtained with the original data sampled at the Nyquist measurement sampling rate. However, it is observed that to retain severity information within the signals for accurate damage severity identification, the compression level should be limited to 20%. These findings affirm that compressed sensing significantly reduces the data collection requirements for crowdsensing-based monitoring applications, without compromising the accuracy of damage detection while preserving essential damage-sensitive information within the dataset.

An invertible hysteresis model for magnetorheological damper with improved adaption capability in frequency and amplitude

It is found during the tests that the damping characteristics of the magnetorheological (MR) damper vary with the excitation amplitude and frequency. However, the existing MR damper models are not able to accommodate the change of excitation amplitude and frequency, which will lead to significant modeling errors. To deal with this problem, this paper analyzes the experimental data and obtains the regularity of the damping characteristics varying with the excitation. Subsequently, an excitation-adaptive MR damper model is constructed based on the hyperbolic tangent function. The proposed model is not only able to adapt to the change of excitation amplitude and frequency but also able to inverse, which is essential for MR damper controller construction. The fitting results show that compared with the existing models, the three normalized errors of the proposed model are improved from 22.61%, 13.96%, and 19.42%–6.30%, 3.81%, and 6.97%, respectively, indicating that the model excitation adaptivity is significantly improved. Furthermore, this study also proposed a damper controller based on the new model, and the simulation results verify the effectiveness of the controller. The proposed model brings the acceleration signal into the model to improve the model adaptivity, which introduces a novel approach to enhance the adaptivity of MR damper models.

USING A WIRELESS SENSOR NODE TO MEASURE THE ACCELERATION OF THE SWING ROCKING MACHINE NODES

Technical diagnosis of the drive of well rod pump units (WRPU) remains an urgent task, despite the significant amount of research in this field. One of the areas of diagnosis is the measurement and control of the angular velocity of the rocking machine elements, which requires the provision of synchronous registration and control of acceleration signals at many points of the control object. The authors proposed to implement such a system based on the wireless sensor module Steval-MKSBOX1V1 of STMicroelectronics. The results of the experimental approbation of the use of the specified module showed the effectiveness of its use for solving diagnostic tasks, since the module contains a significant set of sensors in combination with a unit for processing, storing and transmitting information via a wireless Bluetooth interface. An additional advantage of using the selected measuring module is the possibility of remote reprogramming of the module's software to perform the tasks of controlling the kinematic parameters of the installation in the process of conducting research directly at the technological facility. The use of three coordinate accelerometers as part of the measuring module allows to significantly increase the reliability of control in the diagnostic process.

Transferable non-invasive modal fusion-transformer (NIMFT) for end-to-end hand gesture recognition.

Objective.Recent studies have shown that integrating inertial measurement unit (IMU) signals with surface electromyographic (sEMG) can greatly improve hand gesture recognition (HGR) performance in applications such as prosthetic control and rehabilitation training. However, current deep learning models for multimodal HGR encounter difficulties in invasive modal fusion, complex feature extraction from heterogeneous signals, and limited inter-subject model generalization. To address these challenges, this study aims to develop an end-to-end and inter-subject transferable model that utilizes non-invasively fused sEMG and acceleration (ACC) data.Approach.The proposed non-invasive modal fusion-transformer (NIMFT) model utilizes 1D-convolutional neural networks-based patch embedding for local information extraction and employs a multi-head cross-attention (MCA) mechanism to non-invasively integrate sEMG and ACC signals, stabilizing the variability induced by sEMG. The proposed architecture undergoes detailed ablation studies after hyperparameter tuning. Transfer learning is employed by fine-tuning a pre-trained model on new subject and a comparative analysis is performed between the fine-tuning and subject-specific model. Additionally, the performance of NIMFT is compared to state-of-the-art fusion models.Main results.The NIMFT model achieved recognition accuracies of 93.91%, 91.02%, and 95.56% on the three action sets in the Ninapro DB2 dataset. The proposed embedding method and MCA outperformed the traditional invasive modal fusion transformer by 2.01% (embedding) and 1.23% (fusion), respectively. In comparison to subject-specific models, the fine-tuning model exhibited the highest average accuracy improvement of 2.26%, achieving a final accuracy of 96.13%. Moreover, the NIMFT model demonstrated superiority in terms of accuracy, recall, precision, and F1-score compared to the latest modal fusion models with similar model scale.Significance.The NIMFT is a novel end-to-end HGR model, utilizes a non-invasive MCA mechanism to integrate long-range intermodal information effectively. Compared to recent modal fusion models, it demonstrates superior performance in inter-subject experiments and offers higher training efficiency and accuracy levels through transfer learning than subject-specific approaches.

Torsional and lateral vibration analysis of wind turbine generator bearing outer ring fault considering unbalanced magnetic pull

Rolling bearings play a critical role in wind turbine generators, operating continuously in an electromagnetic excitation environment. Their fault vibration characteristics are affected by unbalanced magnetic pull (UMP), which has received little attention in previous studies. Moreover, torsional vibration can provide advantages in bearing fault diagnosis, since it has a good signal-to-noise ratio (SNR) and it can be easily measured by the encoder. However, torsional vibration is rarely considered in the existing model of wind turbine generator. Accordingly, this paper proposes a dynamic model with 8 degrees of freedom to analyze the bearing outer ring fault characteristics of wind turbine generator under the influence of UMP, while also taking into account the torsional vibration in the model. The validity of the model is verified through experiments conducted on a test rig. The results displays modulation of the ball pass frequency outer (BPFO) harmonics with both UMP frequency and rotation frequency in lateral vibration, and this phenomenon can also be observed in instantaneous angular speed (IAS) and instantaneous angular acceleration (IAA) signals that represent torsional vibration. Besides, the lateral vibration spectrum reveals two distinct resonance bands, and simulation analysis suggested that the rotor and bearings give rise to low-frequency and high-frequency resonance, respectively. This study provides a reference for bearing faults diagnosis in wind turbine generator.