Search

This paper presents a theoretical study on the implementation of a self-tuning vibration absorber for tonal vibration control. The study considers a tuneable vibration absorber, composed by a spring-mass-damper system with variable stiffness and damping components, which is attached to a single degree of freedom mechanical system subjected to a harmonic excitation. The online tuning of the absorber stiffness and damping parameters is performed using an extremum seeking algorithm set to minimise the vibration kinetic energy of the hosting system. The paper proposes a two-paths tuning approach, which, starting from initial estimates of the stiffness and damping parameters of the absorber, searches for their optimal values along constant-damping and constant-stiffness paths respectively. The study presents a comprehensive analysis on the implementation of the algorithm at off-resonance and at resonance frequencies of the hosting system. Also, it investigates how the intrinsic parameters of the algorithm influence the convergence to the optimal stiffness and damping values. The proposed tuning methodology does not require a reference signal for the tonal excitation and, thus, can be suitably implemented both for fixed and time-varying tonal disturbances. [ABSTRACT FROM AUTHOR]

The selective delayless subband structure stands out as a promising algorithmic choice for the multi-channel active control of vehicle interior noise, particularly in the context of road noise. This type of algorithm reduces the eigenvalue spread of the autocorrelation matrix of the signal by decomposing the signal into subbands, and the desired subbands are activated selectively, thus achieving a significant performance improvement while consuming less computational resources compared with the traditional fullband algorithms. Nevertheless, the effectiveness of the subband algorithm appears to be closely tied to the configuration of the primary disturbance signal. In cases where the primary disturbance signal encompasses both broadband and tonal components, the performance advantage of the subband algorithm becomes constrained. In this paper, we conduct a thorough comparative analysis of two extensively employed subband algorithms, namely the Morgan and the Milani methods, through simulations employing data obtained from a multi-channel active noise control (ANC) headrest system. The results indicate that the Morgan method outperforms the Milani method when configured optimally. Subsequently, we propose an enhanced version of the subband algorithm based on the Morgan method. The enhanced algorithm incorporates an additional sinusoidal noise canceller (SNC) subsystem and a narrowband active noise control (NANC) subsystem based on local secondary path modeling to address tonal components, while the selective subband structure is employed for controlling broadband components. In addition, the computational complexity of the algorithm is analyzed. The effectiveness of the proposed algorithm is validated through numerous simulations and the ANC tests conducted on an actual vehicle utilizing the multi-channel ANC headrest system. The performance of the proposed algorithm surpasses that of traditional selective subband and fullband algorithms, and balanced binaural noise reduction is achieved. [ABSTRACT FROM AUTHOR]

This paper introduces a novel parameter related to bearing degradation, namely the Equivalent Damaged Volume (EDV). An algorithm capable of extracting EDV values from experimental data is detailed. To this end, the proposed technique relies on the comparison between experimental and numerical signals. The former are the result of an extensive campaign of run-to-failure tests performed on a dedicated test rig. On the other hand, the latter are obtained by means of a lumped parameter model that accounts for the angular extent and the depth of either localized or extended defects. The potential of this novel technique is demonstrated by exploiting the values obtained by the algorithm as inputs for a prognostic model. In particular, the developed model aims at predicting the time until a certain threshold on the equivalent damaged volume is crossed, regardless of the applied load and the shaft rotation speed. The advantages and downsides introduced by the proposed approach are thoroughly discussed by providing examples based on real degradation histories. [ABSTRACT FROM AUTHOR]

• A near-field weighted subspace fitting algorithm is proposed for ultrasonic guided wave multi-damage localization. • A near-field multi-dimensional solution search algorithm based on a niche particle swarm optimization is proposed, which effectively improves the efficiency of damage localization. • Algorithm performance comparison, finite element simulation, real point damage and slot damage experiments verify the effectiveness of the proposed method. The ultrasonic guided wave-based method for multi-damage localization has been widely proposed. However, the precision of this method is directly correlated with both the quantity of sensors employed and the intricacy of the implementation process. This relationship poses a challenge in striking a balance between the accuracy and efficiency. To improve the computational efficiency under the premise of ensuring the accuracy of multi-damage localization, this paper proposes a near-field weighted subspace fitting algorithm based on niche-particle swarm optimization. Firstly, the fitting relationship between the signal subspace of the diffraction wave and the array steering vector is established under the uniform linear array. Secondly, a multi-dimensional solution space search algorithm based on niche-particle swarm optimization is proposed to improve the search efficiency of damage. Finally, the algorithm is verified by performance comparison, finite element simulation and experiment. The results show that compared with the same type of method, the algorithm improves the computational efficiency by nearly threefold under the identifiable multi-damage conditions. Additionally, the angle error is 1 ∼ 6°, and the distance error is 1 ∼ 20 mm. [ABSTRACT FROM AUTHOR]

Operators from various industries have been pushing the adoption of wireless sensing nodes for industrial monitoring, and such efforts have produced sizeable condition monitoring datasets that can be used to build diagnosis algorithms capable of warning maintenance engineers of impending failure or identifying current system health conditions. However, single operators may not have sufficiently large fleets of systems or component units to collect sufficient data to develop data-driven algorithms. One potential solution to overcome the challenge of having limited representative datasets is to merge datasets from multiple operators with the same type of assets. However, directly sharing data across the company's borders yields privacy concerns. Federated learning (FL) has emerged as a promising solution to leverage datasets from multiple operators to train a decentralized asset fault diagnosis model while maintaining data confidentiality. However, the performance of traditional FL algorithms degrades when local clients' datasets are heterogeneous. The dataset heterogeneity is particularly prevalent in fault diagnosis applications due to the high diversity of operating conditions and system configurations. To address this challenge, this paper proposes a novel clustering-based FL algorithm where clients are clustered based on their dataset similarity. Estimating dataset similarity between clients without explicitly sharing data is achieved by training probabilistic deep learning models and having each client examine the predictive uncertainty of the other clients' models on its local dataset. Clients are then clustered for FL based on relative prediction accuracy and uncertainty. Experiments on three bearing fault datasets, two publicly available and one newly collected for this work, show that our algorithm significantly outperforms FedAvg and a cosine similarity-based algorithm by 5.1% and 30.7% on average over the three datasets. Further, using a probabilistic classification model has the additional advantage of accurately quantifying its predictive uncertainty, which we show it does exceptionally well. [ABSTRACT FROM AUTHOR]

Vehicle on-board monitoring methods use responses of running vehicles to identify track irregularity in real time. But the parameters of both the bridges and the vehicles in such processes may have a non-negligible degree of uncertainty or randomness that will inevitably lead to uncertainty in the track irregularity identification. Quantifying such uncertainty is a great challenge as the vehicle-bridge system (VBS) to be treated with is time-dependent and random. This paper proposes an on-board track irregularity identification algorithm that realizes an inverse random dynamic analysis of the VBS to quantify estimation uncertainty by using a Bayesian Kalman filter technology. The proposed algorithm utilizes sigma point sets to simulate the random parameters and the noises, and is therefore capable of implementing high-fidelity probability density propagation in the nonlinear state transfer and observation functions describing the VBS. The proposed algorithm is validated by numerical examples with various running states and parameter uncertainty levels. [ABSTRACT FROM AUTHOR]

• A single-step recursive algorithm of non-viscous damping forces is developed. • Rectangle-impulse responses of kernel functions determine the algorithm parameters. • The algorithm is compatible with the solution schemes of time integration methods. • The compatible algorithm retains the characteristics of time integration methods. The non-viscous damping force features the convolution integral of the velocity and the kernel function, causing the difficulty of the solutions of the system equation of motion due to the nonlocal time dependence. In accordance with the linear time-invariant system theory in signal processing, this paper presents a single-step recursive algorithm of convolution integral for efficiently computing non-viscous damping forces. The algorithm parameters are effectively identified from the discrete responses of the kernel function to the rectangular unit impulse through digital filter technologies. The proposed algorithm is compatible with the solution schemes based on existing time integration methods, which handle the nonlocal dependence using internal recursive support vectors while leaving the order of the equation of motion unchanged. The central difference method, the Newmark- ß method, and the Bathe method are employed to illustrate the use of the algorithm, proving to retain the stability characteristics of time integration methods. The single and multiple DOF examples show that the simulated results based on the compatible solution schemes compare well with those built on other existing methods, verifying high accuracy and efficiency. The proposed algorithm thus extends the application range of existing time integration methods from the viscously damped system to the non-viscously damped one, fully demonstrating its generalizability. [ABSTRACT FROM AUTHOR]

The First Passage Time (FPT) of a given signal is the time required to reach a level X f for the first time, when starting from a level X 0. For a given set (X 0 , X f) , this time can be obtained by moving along the given signal and clocking the time on when passing through level X 0 , then off when passing through level X f for the first time. For random signals, the FPT is a random variable. Several samples of the first passage time are collected while moving along the given signal to compute the corresponding statistical moments and probability density function. The FPT map associates each statistical moment with every couple (X 0 , X f). A map is constructed for every desired moment. A naive implementation of the signal processing method to compute the FPT map would require to move along the signal as many times as the desired pairs (X 0 , X f). Instead, this paper introduces a new and optimized algorithm to establish the FPT maps of any statistical moment and, optionally, of the whole probability distribution. This algorithm combines different features to produce a computationally effective and memory-efficient method. The algorithm estimates the entire map by iterating only once across the signal. The accuracy of the algorithm has been assessed with particular cases where theoretical solutions exist as well as for lab data. Furthermore the algorithm's performance has also been evaluated in terms of computational burden. A specific attention is given to the computation time scaling for the evaluation of an FPT map, with respect to the signal size. • First passage times are determined for signals from experimental or synthetic data. • Both their statistical moments and, optionally, distributions are computed. • The proposed algorithm runs much faster than an elementary implementation. • The crucial detail of the proposed method is the backward processing of the signal. [ABSTRACT FROM AUTHOR]

• An ATFFC scheme is proposed to improve noise robustness and resolution of TFD. • A mode-type discrimination criterion is established to categorize modes into chirp modes or dispersive modes. • The VGNMD can accurately extract chirp modes and dispersive modes simultaneously. • The VGNMD is well-suitable for analyzing signals with discontinuous TF patterns. Recently proposed variational signal decomposition methods like adaptive chirp mode decomposition (ACMD) and generalized dispersive mode decomposition (GDMD) have attracted much attention in various fields. However, these methods are difficult to simultaneously separate chirp components and dispersive components. This paper proposes a variational generalized nonlinear mode decomposition (VGNMD) framework to address this issue. The VGNMD first introduces an adaptive time–frequency fusion and clustering (ATFFC) scheme to improve noise robustness and resolution of time–frequency distribution (TFD) of signal in a noisy environment and to accurately obtain the TFD of each mode. Then, a mode-type discrimination criterion is established to categorize modes into chirp modes or dispersive modes based on their time–frequency (TF) ridges. Finally, with these TF ridges as initial instantaneous frequencies (IFs) or initial group delays (GDs), a variational optimization algorithm is applied to accurately reconstruct the modes and refine their IFs or GDs. Simulated examples and real-life applications to bat echolocation signal analysis and railway wheel/rail fault diagnosis are considered to show the effectiveness of the VGNMD. The results indicate that the proposed approach can accurately extract chirp modes and dispersive modes simultaneously, and is well-suitable for analyzing nonlinear signals with discontinuous TF patterns. [ABSTRACT FROM AUTHOR]

• A new compounding imaging algorithm integrating amplitude and phase information of signal based on Lamb wave phased array is proposed. • Sigmoid function with variable gradient is used for establishing the nonlinear mapping between imaging index and phase difference. • The proposed imaging algorithm is validated on plates of different materials and dimensions with multiple types of defects. • The proposed imaging algorithm can effectively improve the signal-to-noise ratio of the images and suppress artifacts, implementing imaging without threshold. Aimed at improving the imaging performance of Lamb wave phased array based on Delay-and-Sum (DAS) of signal amplitude, a compounding imaging algorithm integrating amplitude and phase information of signal is proposed in this paper. By applying back propagation dispersion compensation, the pulse echoes caused by defects will be in-phase with the excitation signal, implementing signal focusing. In the proposed imaging algorithm, the amplitude imaging index is calculated by classical total focusing method (TFM) and the phase imaging index is calculated by phase coherence factor (PCF), which establishes the nonlinear mapping between imaging index and phase difference by sigmoid function with variable gradient. The final imaging results are the compound of two imaging indexes. The experimental validation is carried out on plates of different materials and dimensions with multiple types of defects based on uniform linear array. Compared with the imaging results of conventional TFM, the proposed imaging algorithm can effectively improve the signal-to-noise ratio (SNR) of the images and suppress artifacts, implementing threshold-free imaging. [ABSTRACT FROM AUTHOR]

[Display omitted] The unstable statuses in machining, such as chatter and severe forced vibration, limit manufacturing productivity and quality. Although there have been plentiful studies on the on-line sensing signals and off-line surface images to identify these statuses qualitatively, the quantification of chatter severity, which is of great significance on machining stability, has been rarely reported. Workpiece surface morphology is the direct result and thus the objective criteria of unstable machining. This paper proposes a new algorithm to quantify the chatter status based on the milled surface image analysis of end-milling. Besides, the proposed method can also identify the stable, forced vibration, and chatter statuses. Specifically, this proposed algorithm extracts two indexes, called Index_Dmean and Index_Emean , via processing the milled surface image. The definitions and calculations of these two indexes are given in Section 2. Index_Dmean is used to differentiate stable and unstable surfaces and quantify the severity of chatter status, and Index_Emean can distinguish between forced vibration and chatter surfaces. Experimental results demonstrate the availability of the proposed algorithm on the chatter quantification and milling instability classification performance. In addition, an applicable on-line vision system with satisfying performance based on the algorithm is developed. [ABSTRACT FROM AUTHOR]

For active control of road noise inside a vehicle, many reference signals are required in adaptive control systems, resulting in slow convergence speeds and heavy computational loads. In this paper, a means of reducing the computational complexity of adaptive control algorithms is investigated based on the fast-converging least-mean-squares Newton (LMS-Newton) algorithm. By analyzing the autocorrelation matrix of reference signals filtered by secondary paths, it is found that the dimensions of the autocorrelation matrix of filtered reference signals do not necessarily have to equal the dimensions of the adaptive control filters, and instead the dimensions can be determined from the correlation between reference signals filtered by secondary paths. Then a computationally efficient segment-update block algorithm is proposed that splits adaptive control filters into several segments and the coefficients of each segment are updated independently. Theoretical analysis shows that the proposed segment-update block algorithm has similar convergence speed and noise-reduction performance as the traditional LMS-Newton algorithm when a segment of appropriate length is chosen, but the computational complexity is reduced dramatically because the inverse autocorrelation matrix is replaced by a smaller matrix. The performance of the proposed algorithm is verified through simulation experiments with different recorded road noises and secondary paths, and the results are compared with those of some typical algorithms. The simulation results showed that the proposed algorithm achieved a similar noise-reduction performance with an approximately 8.9-times reduction of computational complexity compared with the traditional LMS-Newton algorithm. [ABSTRACT FROM AUTHOR]

This paper describes modifications to the Matrix Power Control Algorithm (MPCA) to improve convergence for Random Vibration Control (RVC) testing. In particular, this paper presents Multiple-Input Multiple-Output (MIMO) implementations of MPCA in simulation and experiment. An Euler–Bernoulli beam model was simulated with applied base excitations and the Box Assembly with Removable Component (BARC) was used in experiment to validate results. The Bayesian optimization package Dragonfly was used to optimize control parameters. Additionally, a moving-average was employed and optimized to improve the measured response feedback for MPCA, reduce the number of averages needed to be taken between control updates, and further improve convergence. The key results of this paper show that the performance of MPCA can be improved by tuning the control parameters and by applying an optimized moving-average. Furthermore, it is demonstrated that convergence can be achieved within 12 drive-frames, which greatly enhances vibration control capability. [ABSTRACT FROM AUTHOR]

High-rate systems are defined as engineering systems that undergo accelerations of amplitudes typically greater than 100 g n over less than 100 ms. Examples include adaptive airbag deployment systems, hypersonic vehicles, and active blast mitigation systems. The use of feedback mechanisms in these high-rate applications is often critical in ensuring their continuous operations and safety. Of interest to this paper are algorithms needed to support high-rate structural health monitoring (HRSHM) to empower sub-millisecond decision systems. HRSHM is a complex task because high-rate systems are uniquely characterized by (1) large uncertainties in the external loads, (2) high levels of non-stationarities and heavy disturbances, and (3) unmodeled dynamics generated from changes in system configurations that necessitate careful crafting of adaptive strategies. This paper studies benefits of integrating a data-driven predictive model with a physics-based state observer to reduce latency and convergence time estimating actionable information. The predictive model, constructed with long short-term memory (LSTM) cells, performs multi-step ahead signal prediction acting as the input to the physical model, a model reference adaptive system (MRAS). The MRAS then performs state estimation of the predicted signal rather than the true signal. A comparison study was done between the proposed hybrid algorithm and a physics-based MRAS on a testbed involving a fast-moving boundary condition. Results showed that the hybrid algorithm could perform state estimations with zero timing deadline overshoot and with up to 50% faster convergence time when compared to the MRAS under constant boundary conditions. However, the hybrid generally underperformed the MRAS algorithm in terms of convergence accuracy during motion of the boundary condition by increasing convergence time by 20%, attributable to the lag in learning the new dynamics used in predicting. The performance of the NSE algorithm was also examined on a true high-rate system, where it was shown to be capable of qualitatively tracking actionable information. • A hybrid, deep learning algorithm for real-time state estimation. • Physics-integrated algorithm with state estimation time of less than the desired sampling rate. • An algorithm with sub-millisecond computation capabilities for applications in high-rate systems. • Demonstrated performance using three case studies on experimental datasets acquired on two different testbeds. [ABSTRACT FROM AUTHOR]

Active sound profiling (ASP) is gaining ever-increasing attention owing to its powerful ability to balance the sound pressure levels of certain frequency points or frequency bands of target noise for sound quality enhancement. Most of the current ASP algorithms, however, only focus on narrowband or broadband noise alone and the computational efficiency is very low. To alleviate this problem, a feedforward time–frequency-domain hybrid active sound profiling (TFD-HASP) algorithm is proposed in this paper. The algorithm consists of three subsystems, in which the narrowband ASP (NASP) subsystem utilizes the time-domain internal model FXLMS algorithm along with the local secondary path estimation approach, the broadband ASP (BASP) subsystem is based on the delayless frequency-domain internal model FXLMS algorithm, and the sinusoidal noise canceller (SNC) subsystem is designed to eliminate narrowband interference to the BASP subsystem. In this way, the presented TFD-HASP algorithm not only has good spectral reshaping capability for both narrowband and broadband noise components, but also requires notably low computational effort. A series of simulations are performed to evaluate the performance of the TFD-HASP algorithm. In addition, the comfortable sound quality of a real vehicle interior noise at high speed is designed and realized. Experimental results demonstrate that the proposed algorithm achieves satisfactory active sound profiling for noise with both narrowband and broadband characteristics. [ABSTRACT FROM AUTHOR]

Time reversal (TR) techniques due to their spatial and temporal focusing characteristics, have been increasing applied in damage evaluation. Since probing waves strongly attenuate in concrete material, imaging of damage in concrete using conventional time reversal imaging (TRI) algorithm often requires a large number of deployed transducers to minimize unexpected artifacts. To relieve economic burden and fussy installation induced by massive transducers, this paper develops a small number of transducers-enabled concrete-adapted time reversal imaging (CATRI) algorithm using implantable technology. The CATRI algorithm adopts a compensation factor to overcome wave attenuation and eliminate artifacts during crack imaging process. The implantable technology is achieved by a designed concrete implantable module (CIM) device that can be easily implanted into concrete structures. The performances of the CIM device and CATRI algorithm were numerically and experimentally studied. The results show that the proposed techniques overcome the strong wave attenuation-induced problems and successfully identify crack regions. The comparison of imaging results by using conventional TRI algorithm highlights the superiority of the CATRI algorithm for concrete crack imaging. [ABSTRACT FROM AUTHOR]

Active control strategies have been widely used in micro-vibration isolation with higher performance requirements. Aiming at the problem of insufficient vibration isolation performance due to time delay in feedback control, the ground-based feedforward active control is introduced to compensate in advance for higher pursuit. However, the inaccuracy of the feedforward control reference model severely restricts the vibration isolation performance. With the assistance of RBF neural network algorithm for accurate real-time identification of online model, a feedforward-feedback dual-loop active hybrid control (DAHC) strategy based on the RBF-RLS adaptive algorithm is proposed in this paper. The adaptive feedforward control uses the RBF neural network algorithm to identify the accurate model of the system online, and input it to the transverse filter of the RLS adaptive control algorithm for recursive calculation, which can effectively reduce the system error caused by the inaccuracy of the model. The experimental results show that the accuracy of online model affects the amplitude attenuation performance by 57.1%. DAHC strategy can greatly reduce the resonance peak by 25.02 dB in micro-vibration. Real applicative experiment further proves the effectiveness of our proposed algorithm. [ABSTRACT FROM AUTHOR]

• The inverse system is proposed to eliminate the mutual influence between the AFS and HPS subsystems. • PID and GRNN inverse system are combined to realize the decoupling control of the vehicle chassis system. • Real vehicle tests are conducted to demonstrate the effectiveness of the decoupling control strategy. Aiming at eliminating the mutual interference and coupling between the active front steering (AFS) and hydro-pneumatic suspension (HPS) subsystems of the emergency rescue vehicle, this paper presents a neural network inverse system based decoupling control strategy, which consists of the neural network inverse system and PID feedback control strategy. At first, a three-degrees of freedom dynamics model of the vehicle is constructed, and the reversibility of the vehicle chassis system is analyzed by the Interactor algorithm. Then, the generalized regression neural network (GRNN) model is used to identify the vehicle chassis inverse system, which decouples the multivariable vehicle chassis system into two independent single input and single output systems. Next, this paper proposes a closed-loop compound controller for the single input and single output systems, which is designed based on the PID control strategy and GRNN inverse system to realize the feedback control of the vehicle chassis system, so that the vehicle can track the desired yaw rate and roll angle well. Simulations and experiments are conducted on the self-developed fire rescue prototype vehicle, and the results prove that the proposed decoupling control strategy can not only eliminate coupling between AFS and HPS subsystems of the vehicle chassis, but also make the vehicle have better handling stability than controlled by the integrated control. [ABSTRACT FROM AUTHOR]

This paper proposes an efficient and reliable eigenvalue solution technique for analytical stochastic dynamic stiffness (SDS) formulations of beam built-up structures with parametric uncertainties. The SDS formulations are developed based on frequency-dependent shape functions in conjunction with both random-variable and random-field structural parameters. The overall numerical framework is aimed towards representing the broadband dynamics of structures using very few degrees of freedom. This paper proposes a novel approach combining the Wittrick–Williams(WW) algorithm, the Newton iteration method and numerical perturbation method to extract eigensolutions from SDS formulations. First, the eigenvalues and eigenvectors of the deterministic DS formulations are computed by the WW algorithm and the corresponding mode finding technique, which are used as the initial solution. Then, a numerical perturbation technique based on the inverse iteration and homotopy method is proposed to update the eigenvectors and eigenvalues. The robustness and efficiency of the proposed method are guaranteed through several technique arrangements. Through numerical examples, the proposed method is demonstrated to be robust within the whole frequency range. This method provides an efficient and reliable tool for stochastic analysis of eigenvalue problems relevant to free vibration and buckling analysis of built-up structures. [ABSTRACT FROM AUTHOR]

• A cost function vector of the simultaneous perturbation stochastic approximation algorithm is proposed. • The proposed method reaches a good convergence performance. • The vibration of the inlet pipeline is eliminated by using the proposed method. Active vibration control of stable rotating machinery with a time-variant structure is a research priority in the field. Simultaneous perturbation stochastic approximation (SPSA), a model-free active control strategy, has attracted a lot of attention from researchers. This paper studies the cost function of the SPSA algorithm to improve its performance in controlling periodic vibration. After the weight vector being perturbed by a perturbation vector, the existing cost functions of the SPSA algorithm can reveal the overall energy variation but hardly detail the variation of each harmonic wave. Therefore, the adaptive notch filter with the SPSA algorithm using these cost functions shows a low convergence speed and a bad control effect in active periodic disturbance controlling. In order to detail the energy variation of each harmonic, this paper proposes a cost function using the weighted cross correlation function. When the algorithm calculates the energy of each harmonic wave, the disturbances from other harmonics and background noise are reduced. The interference noises in the estimated gradient of the cost function are reduced indirectly. The algorithm convergence performance improves. To illustrate that, the simulations of controlling harmonic waves and the experiments of eliminating the vibration of the pump inlet pipeline are conducted. The results indicate that the adaptive notch filter with the proposed cost function has reached the fastest convergence speed and the best control effect so far among the use of other cost functions mentioned in this paper. [ABSTRACT FROM AUTHOR]

Active impulsive noise control (AINC) is a specific research interest for improving the performance of traditional algorithms against impulsive environments. By studying existing algorithms, this paper employs a fractional lower order covariance criterion and proposes a filtered-x fractional lower order covariance (FxFLOC) algorithm. After that, the FxFLOC algorithm is equipped with an on-line estimation method using a sampled characteristic function to overcome the dependency on prior knowledge of α -stable distribution. Furthermore, the convergence condition and the influence of hyperparameters are discussed. Extensive simulations are carried out to suggest that the proposed algorithms enhance the stability and accelerate the convergence speed compared with prior algorithms. Also, experiments are conducted to verify the performance of the proposed algorithms. The results show that the proposed algorithms are practical on symmetric α -stable (S α S) impulsive noise attenuation. • An AINC algorithm is proposed using symmetric fractional lower order covariance. • An on-line estimation is applied using a sampled characteristic function method. • Extensive simulations and experiments are carried out to verify the performance. [ABSTRACT FROM AUTHOR]

This paper presents for variable stiffness tuned mass dampers (VS-TMDs) a vibration control algorithm combined with a nonlinear system identification method. The objective of the proposed approach is the control of multistory shear frame structures, which exhibit abrupt stiffness degradations. A stiffness variable control (SVC) algorithm is developed to control the stiffness of the VS-TMD, which is based on the mitigation of the potential energy of the structure. For the identification of the system parameters, an adaptive unscented Kalman filter (A-UKF) scheme is utilized. The paper describes the theoretical background of the proposed approach with its governing equations and conducts performance studies on a semi-actively controlled 2-story shear frame structure. Harmonic, white noise and earthquake excitation scenarios including seismic aftershock effects are investigated. Sensitivity studies are conducted. With the proposed approach, the abrupt stiffness changes of the structure are detected and the damper stiffness is adapted successfully. The results show that the SVC together with the A-UKF clearly outperforms the passive TMD control. [Display omitted] • For the stiffness control of semi-active TMDs, a control algorithm is proposed. • A potential energy based cost function is derived and minimized by gradient descent. • The algorithm is coupled with an adaptive UKF for identification of abrupt changes. • Harmonic, white noise and seismic excitation with aftershocks are investigated. • A superior and robust control performance is observed. [ABSTRACT FROM AUTHOR]

For a structural dynamics engineer assessing a system's vibration characteristics, the frequency response function (FRF) is indispensable, irrespective of whether the setup being tested is experimental, numerical or analytical. This paper outlines the different FRF estimation techniques that have been developed over the years. Algorithms that compute an ordinary least squares (OLS) estimate of the FRF, assuming uncorrelated measurement noise to be present either on the output (H 1) or the input (H 2) signals, have been compared with those that employ total least squares (TLS) equations by making use of an augmented input–output auto and cross power (G F F X / G X F F) matrix at every frequency such as the H v (using eigenvalue decomposition) and H S V D (using singular value decomposition) algorithms. Another FRF estimation method based upon Cholesky decomposition (H C D) is also discussed. Further discussion has been included of TLS algorithms computing the FRFs for one output at a time, as historically presented in the development of the H v algorithm, with a case that evaluates the FRF matrix computed for all the outputs simultaneously. The development of the corresponding coherence functions has been presented, highlighting the method dependent paradigms that led to concepts such as virtual coherence and partial coherence while underscoring their equivalence with ordinary and multiple coherence calculations. It has been shown that some of the existing c o n d i t i o n e d coherence metrics are inconsistent from an input–output standpoint, for which the corrected interpretations have been subsequently described. This article is unique in that no previous work summarizes all of the previously developed FRF estimation and coherence algorithms and no previous paper explains the inconsistencies in the conditioned coherence functions with respect to multiple coherence. • Review and comparison of OLS vs TLS FRF estimation algorithms. • Comparison of single output and multiple output TLS FRF estimation algorithms. • Review of different coherence metrics. • A generalized interpretation of different conditioned coherence metrics. [ABSTRACT FROM AUTHOR]

By facilitating the generation of samples from arbitrary probability distributions, Markov Chain Monte Carlo (MCMC) is, arguably, the tool for the evaluation of Bayesian inference problems that yield non-standard posterior distributions. In recent years, however, it has become apparent that Sequential Monte Carlo (SMC) samplers have the potential to outperform MCMC in several ways. SMC samplers are better suited to highly parallel computing architectures and also feature various tuning parameters that are not available to MCMC. One such parameter – the 'L-kernel' – is a user-defined probability distribution that can be used to influence the efficiency of the sampler. In the current paper, the authors explain how to derive an expression for the L-kernel that minimises the variance of the estimates realised by an SMC sampler. Various approximation methods are then proposed to aid the implementation of the proposed L-kernel. The improved performance of the resulting algorithm is demonstrated in multiple scenarios. For the examples shown in the current paper, the use of an approximately optimal L-kernel has reduced the variance of the SMC estimates by up to 99 % while also reducing the number of times that resampling was required by between 65% and 70%. Python code and code tests accompanying this manuscript are available through the Github repository https://github.com/plgreenLIRU/SMC_approx_optL. [ABSTRACT FROM AUTHOR]

• The mathematical model of multi-coset sampling is integrated into the blade tip timing technique to obtain the spectrum recovery method. • The characteristics of the spectral values of synchronous resonance frequency in the compression model are analyzed. • The best atoms with the same contribution are traced in the reconstruction process. • The SP algorithm is superior to the commonly OMP algorithm for reconstruction and the probability of frequency recognition can be improved. • The validity of what is stated in this article is verified by the blade vibration tests. Long-term high cycle fatigue vibration will lead to the reduction of life of the blade. Blade Tip Timing (BTT) technology is a research hotspot in the non-contact measurement of blades. Due to the limitation of the number of sensors, BTT signals present serious undersampled conditions. Identifying the frequency of blade vibration through the analysis of undersampled signals is the key problem of the current BTT technology. A sub-Nyquist technique, called multi-coset sampling (MCS), which is suitable for the BTT sampling model can be applied. In this paper, the characteristics of BTT signals under blade synchronous resonance are described and the deficiency of this feature in the recovery process using the orthogonal matching pursuit (OMP) algorithm is found. Based on the problem, the subspace pursuit (SP) algorithm adapted to this feature is proposed. The validity of the algorithm is verified by the blade rotation experiment. The advantage of this algorithm is that the synchronous resonance frequency of the blades can be recovered with a higher probability, which provides a new idea for the development of BTT technology. [ABSTRACT FROM AUTHOR]

Classification models are a fundamental component of physical-asset management technologies such as structural health monitoring (SHM) systems and digital twins. Previous work introduced risk-based active learning , an online approach for the development of statistical classifiers that takes into account the decision-support context in which they are applied. Decision-making is considered by preferentially querying data labels according to expected value of perfect information (EVPI). Although several benefits are gained by adopting a risk-based active learning approach, including improved decision-making performance, the algorithms suffer from issues relating to sampling bias as a result of the guided querying process. This sampling bias ultimately manifests as a decline in decision-making performance during the later stages of active learning, which in turn corresponds to lost resource/utility. The current paper proposes two novel approaches to counteract the effects of sampling bias: semi-supervised learning , and discriminative classification models. These approaches are first visualised using a synthetic dataset, then subsequently applied to an experimental case study, specifically, the Z24 Bridge dataset. The semi-supervised learning approach is shown to have variable performance; with robustness to sampling bias dependent on the suitability of the generative distributions selected for the model with respect to each dataset. In contrast, the discriminative classifiers are shown to have excellent robustness to the effects of sampling bias. Moreover, it was found that the number of inspections made during a monitoring campaign, and therefore resource expenditure, could be reduced with the careful selection of the statistical classifiers used within a decision-supporting monitoring system. [ABSTRACT FROM AUTHOR]

Accurate weight information of freight vehicles is of great value for the assessment and maintenance of railway infrastructures. Existing bridge weigh-in-motion (BWIM) methods can introduce errors when the vehicle speed is high so that the dynamic effect cannot be filtered by signal processing. This paper proposes a novel BWIM algorithm for fast-passing railway freight vehicles, in which the dynamic coupling of the VB system is utilized. An extended state-space model with unknown mass parameters for the time-dependent VB system is established. Based on the model, an SPKF algorithm with a high convergence rate for identifying the mass parameters is proposed. Numerical examples of freight vehicles passing a 40 m railway bridge with a speed up to 120 km/h are presented to validate the effectiveness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

• ASRCKF observation is proposed to estimate vehicle states. • The SP algorithm is put forward to calculate the μ max. • ASRCKF can improve the estimation accuracy, especially for vehicle sideslip angle. • Accurate μ max can always be obtained by the proposed algorithm. • Co-simulation and experiments are designed to verify the algorithm validation. It is infeasible to measure vehicle dynamics states (VDS) directly without expensive measurement instruments, especially for the tire-road peak adhesion coefficient (μ max). However, four-wheel-independent-drive-electric-vehicle (4WIDEV) provides convenience for the observation of these dynamic states, because the rotation rate and torque of the in-wheel motor can be acquired directly. Vehicle nonlinear longitudinal-lateral dynamics, the single estimation method for all VDS and the time-varying measurement noise of sensors cause difficulties for the observation. Common the extended-Kalman-filter (EKF) is unsuitable to estimate VDS in strong nonlinear region. This paper propose a longitudinal-lateral cooperative estimation algorithm based on adaptive-square-root-cubature-Kalman-filter (ASRCKF) and partitioned similarity-principle (SP) to estimate the vehicle states and the tire-road peak adhesion coefficient sequentially for 4WIDEV. Firstly, a nonlinear 7-degree-of-freedom (7DOF) vehicle model and magic-formula (MF) tire model are built as the base of the successive estimation scheme. Then, recursive-least-squares (RLS) is adopted to estimate the tire longitudinal force. With the estimated tire longitudinal force, an ASRCKF which can be adjusted adaptively by the feedback dynamics states, is designed for the estimation of vehicle states. Next, the SP algorithm combined with the characteristic of longitudinal-lateral dynamics, which is benefit for μ max estimation when tire dynamics enters the nonlinear region, is proposed. Finally, experiment and simulation results show that excellent performance can be achieved with the proposed estimation method in varying driving conditions. [ABSTRACT FROM AUTHOR]

• The Bayesian inversion for stochastic model updating is investigated. • A novel UQ metric using the Bray-Curtis distance is proposed. • An approximate Bayesian computation framework is developed. • An adaptive binning algorithm is proposed to improve the computation efficiency. • Both static and dynamic cases with explicit and implicit models are considered. In practical engineering, experimental data is not fully in line with the true system response due to various uncertain factors, e.g., parameter imprecision, model uncertainty, and measurement errors. In the presence of mixed sources of aleatory and epistemic uncertainty, stochastic model updating is a powerful tool for model validation and parameter calibration. This paper investigates the use of Bray-Curtis (B-C) distance in stochastic model updating and proposes a Bayesian approach addressing a scenario where the dataset contains multiple outliers. In the proposed method, a B-C distance-based uncertainty quantification metric is employed, that rewards models for which the discrepancy between observations and simulated samples is small while penalizing those which exhibit large differences. To improve the computational efficiency, an adaptive binning algorithm is developed and embedded into the Bayesian approximate computation framework. The merit of this algorithm is that the number of bins is automatically selected according to the difference between the experimental data and the simulated data. The effectiveness and efficiency of the proposed method is verified via two numerical cases and an engineering case from the NASA 2020 UQ challenge. Both static and dynamic cases with explicit and implicit propagation models are considered. [ABSTRACT FROM AUTHOR]

Active noise control (ANC) headphones are commonly used to reduce annoyed noise around users' ears. Most commercial ANC headphones utilize specific filters with pre-trained coefficients due to their fast response and robustness. However, when dealing with primary noises coming from different directions, their noise reduction performance suffers significantly. Hence, we proposed an adaptive gain (AG) algorithm on the fixed filters with multi-reference method, in which the control signal is formed by adaptively weighting and summing the output signals of multiple fixed filters that have been pre-trained from primary noise with different directions of arrival. By combining the adaptive algorithm with the fixed filter approach, the proposed algorithm outperforms the conventional fixed ANC approach in noise reduction performance. This paper provides a theoretical analysis of the proposed algorithm's step-size bound and time constant, demonstrating its robustness and fast convergence behavior. Furthermore, comparative simulations and real-time experiments with commercial ANC headphone and other adaptive algorithms show its efficacy in dealing with multiple noise sources in the actual scenario. [ABSTRACT FROM AUTHOR]

• A cost-efficient LFE-ASQC algorithm with similar or even better control accuracy than the ANE and ASP algorithms is proposed. • The proposed algorithm reduces the computational complexity more than one order of magnitude compared to the ANE or ASP algorithms. • The proposed LSP modeling methods A and B have significantly higher modeling accuracy. • Simulated and experimental studies confirm the equalization capability of the proposed algorithm in both stationary and non-stationary cases. Current active noise equalization or active sound profiling algorithms with a parallel structure are typically used for sound quality improvement of multi-harmonic noise. The computational efficiency and system performance may, however, be severely degraded due to an increase in the length of estimated secondary path or the number of controlled frequencies. To alleviate this problem, a local secondary path estimation and filtered-error structure based active sound quality control (LFE-ASQC) algorithm is proposed in this paper. The algorithm utilizes a set of low-order local secondary path (LSP) models to perform filtering operations and uses the filtered pseudo-error signals instead of usual filtered reference signals to update control filter weights, thus it not only has good spectral reshaping capability but also requires notably low computational effort. Two new LSP modeling methods are also proposed. Moreover, a computational complexity analysis of the proposed LFE-ASQC algorithm is provided. Numerical simulations are conducted to evaluate the accuracy of the LSP modeling methods and the performance of the LFE-ASQC algorithm. In addition, the powerful sound quality of a real vehicle interior noise during acceleration is designed and realized. Experimental results demonstrate that the proposed algorithm achieves satisfactory active sound quality control for stationary and non-stationary multi-tonal noises. [ABSTRACT FROM AUTHOR]

• A hybrid detection method is proposed for assessing delamination in composites. • Two-step method has combined the global frequency and local FBGs for damage detection. • FBGs were distributed near the damage area predicted by frequency index to further predict damage with better accuracy. • Support vector machine and the extreme learning machine were used as inverse algorithms. In this paper, a combined global-local strategy was developed for structural health monitoring (SHM) of fiber reinforced polymer (FRP) composites to assess the internal delamination damage, which has combined vibrational frequency shifts as the global index and the fiber Bragg grating (FBG) wavelength shifts as the local index. The delamination detection was carried out in two steps. The first step was based on the global damage index, which is the changes in multiple modes of frequencies measured by a non-contact scanning laser Doppler vibrometer (SLDV) system. Machine learning (ML) algorithms including support vector machine (SVM) and extreme learning machine (ELM) algorithms were used to initially predict the delamination interface, location, and size through frequency shifts. Numerical and experimental verification results show that SVM has a higher prediction accuracy than ELM when only a small number of samples exist, and the classification of SVM is better than its regression function to be used in the prediction of discrete delamination interfaces. To verify the damaged area predicted by frequency changes and further update the delamination edges with better accuracy, the second step has taken the changes in the wavelength of multiple FBG sensors as the local damage index. Multistage loads were applied at equidistant positions of the FRP specimens to induce deformations and the wavelength shifts in FBGs were used to determine the boundary of delamination. The results showed that delamination in FRP composites can be assessed more precisely using the global-local two-step SHM strategy, which can compensate for the shortcomings of only using vibration-based or FBG-based monitoring techniques. Besides, the SVM algorithm showed excellent predictive performance in both steps and had great potential in delamination damage monitoring. [ABSTRACT FROM AUTHOR]

• Review of current techniques for background estimation of non-stationary signals. • Background estimation of non-stationary signals by TFR for unknown IS. • Improved Background estimation of non-stationary signals by cycle-order processing. The spectrum background, or in short, the background, is associated with the slow variations of the spectrum. The estimated background can be used to pre-whiten the spectrum and to identify significant variations in machine structure. However, the known background estimation techniques have been considered mainly in the context of stationary signals. Thus, the goal of this paper is to present new algorithms for background estimation of non-stationary signals. Based on the examination of available techniques for separating discrete frequencies and the background, we present two new algorithms for non-stationary signals: (i) time–frequency background estimation (TFBE), recommended when the rotation speed is unknown, and (ii) cycle-order background estimation (COBE), recommended when the rotation speed is known. The performance of these two new algorithms is demonstrated using a large data bank composed of simulated vibration signals of rotating components in different rotating speed profiles combined with realistic transfer functions measured on different rotating machines. These vibration signals were designed for a comprehensive investigation of the capabilities and limitations of the new algorithms. The accuracy of the new algorithms to estimate the background depends on the rotating speed variation. Good estimations of the backgrounds were achieved for rotating speed variations up to 10 Hz/s. Furthermore, the abilities of the algorithms are illustrated on real measured data. [ABSTRACT FROM AUTHOR]

• An algorithm for autonomic estimation of the spectrum background. • Theoretical formulation of the relationship of the parameters of ACS and Ceps-Lift. • An algorithm for improved estimation of the background for quasi-stationary signals. • Mapping of a background from the frequency domain to the order domain. • Performance examinations by measured transfer functions and simulated signals. The vibration signals of rotating machinery contain information about the rotating components and the machine's structure. The peaks of a vibration spectrum are related to the vibration signals of the rotating components, and the background may be regarded as the spectrum without the "peaks". In this paper, we extend studies on the estimation of the spectrum background. Two topics are addressed: first, an analysis of known methods and their extension into a new autonomic algorithm (autonomic Ceps-ACS) for background estimation of stationary signals, and second, a new approach for estimation of the background for quasi-stationary signals. The estimation of the background in both cases is based on two current techniques, namely, liftering of low quefrencies in the cepstrum domain (Ceps-Lift) and adaptive clutter separation (ACS). A relationship between the parameters of Ceps-Lift and ACS was formulated, enabling the development of the autonomic Ceps-ACS algorithm. For quasi-stationary signals, both Ceps-Lift and ACS have limited ability to estimate the spectrum background. To address the topic of estimation of the spectrum background of quasi-stationary signals, we proposed a novel approach that uses the background in the order domain to estimate the background in the frequency domain. Experimental measured transfer functions, measured data and simulated vibration signals were used to demonstrate the performance of the algorithms. [ABSTRACT FROM AUTHOR]

• Proposal of a methodology for estimating the period of faulty impulses. • Development of an APMESR algorithm for extracting incipient bearing fault features. • Selection of MODWPT as the linear transformation for APMESR. • Evaluation and comparison of period estimation methodologies and transformations. • Verification of APMESR's effectiveness through several simulations and experiments. Bearing incipient fault feature extraction is crucial and challenging throughout its life cycle. In this paper, an adaptive period matching enhanced sparse representation (APMESR) algorithm is developed to address this issue. First, a novel methodology for estimating the period of faulty impulses is proposed from the perspective of mining the periodicity-related numerical patterns. Second, the period estimation methodology is embedded in a sparse representation model to implement adaptive period matching to form APMESR, which is capable of achieving periodic sparsity. Third, maximal overlap discrete wavelet packet transform is selected as the linear transformation of APMESR for improving its ability to reduce noise and highlight periodic impulse signatures. Furthermore, evaluations and comparisons are conducted using simulations to demonstrate the validity and performance of the proposed period estimation methodology, linear transformation, and APMESR. Experimental results indicate that APMESR can effectively extract incipient bearing fault features and outperforms other well-advanced methods. [ABSTRACT FROM AUTHOR]

• The response surfaces of the extreme values are constructed using the Kriging method. • An adaptive uniform design is proposed to prepare training samples. • An iterative algorithm is developed to obtain the optimal Kriging response surface. • The weighted sampling method is improved to estimate reliability. • The accuracy and efficiency of the proposed method have been validated via three cases. The performance of complex engineering systems often changes over time and space, potentially leading to multiple failure modes. Such time and space variance presents a challenge to reliability analysis and design methods. To address this issue, this paper first analyzes three categories of approaches to four types of time-variant reliability problems, and then develops an adaptive uniform design-based Kriging (AUDK) framework with an improved weighted sampling method (IWSM), termed the AUDK-IWSM. In particular, a Kriging response surface of the extreme function for time and space-variant limit states is constructed, and then an existing weighted sampling method is improved for high precision to prepare the sampling points for reliability analysis on the Kriging surface. To construct the AUDK model and update it with the most probable point, an adaptive iterative algorithm is developed based on the IWSM. Thus, through integrating the reliability index of the generated Kriging response surface, the resultant AUDK-IWSM algorithm improves reliability calculations and enhances computational efficiency of the reliability analysis. The AUDK-IWSM is illustrated, and its effectiveness is verified through four case studies for engineering practice, in comparison with a number of state-of-the-art techniques reported in the literature. [ABSTRACT FROM AUTHOR]

Active magnetic bearings (AMBs) have unique superiority in vibration reduction and nowadays have been widely used in permanent magnet synchronous motor (PMSM). To maximize the advantage of AMBs, an accurate model of the total system is indispensable in controller design. This paper proposes a two-step identification algorithm for an AMB-supported PMSM system, in which the AMB model remains its nonlinear characteristics and the unbalance magnetic pull (UMP) in PMSM is covered. The proposed identification procedure contains two steps: suspension identification for the AMB-rotor system and steady-rotation-identification for external forces. In simulation verification, the algorithm shows robustness to measure noise in sensors and inertia in amplifiers. The algorithm is also carried out on an industrial AMB-supported PMSM test rig. To validate the identification results, the identified values are applied on a Simulink model of the system, from which the response under condition of designed excitations, steady-rotation state at various rotational speeds and a run-up procedure is generated. The simulation results and the experimental measurements show close agreement with each other, which verifies the accuracy of the identification method. • Calculate and verify the parameters in AMB supported PMSM system. • Use response data under real steady-suspension and steady-rotation state. • Consider the nonlinear characteristics of AMB and the UMP forces generated by PMSM. • The identification method is insensitive to measure noise and inertia elements. [ABSTRACT FROM AUTHOR]

• A novel NANC system based on FEWT and recursive third-order Chebyshev expansion is proposed. • The FEWT with order-statistic filtering is applied to stabilize non-stationary noise. • A RTOCN filter is presented for compensating the higher-order nonlinear distortions. • The channel-reduced diagonal structure is designed to optimize the RTOCN filter. • Experiment results indicate that proposed method outperforms conventional methods. Active noise control (ANC) has been widely studied to eliminate low-frequency noise. However, efficiently controlling higher-order distortions of nonlinear primary and secondary path is still a major challenge. To address this, this paper proposes a novel nonlinear ANC algorithm based on the fast empirical wavelet transform and recursive Chebyshev nonlinear filter. In this method, the fast empirical wavelet transform based on order-statistic filtering is first applied to decompose non-stationary noise into empirical mode functions (EMFs) which are more stationary and easy to control. Furthermore, a novel third-order Chebyshev nonlinear filter including the non-recursive subsection and recursive one is proposed to individually expand EMFs for compensating higher-order nonlinear distortions from primary path and secondary path. Moreover, the strategy of channel-reduced diagonal structure is designed to optimise the number of expanded channels, and the computational complexity of proposed algorithm is also investigated. Experimental results demonstrate that proposed method performs the highest noise reduction over the conventional algorithms but also has a fast convergence rate. [ABSTRACT FROM AUTHOR]

• A real-time simultaneous input-state-parameter estimation method is proposed. • The unknown inputs are modeled as modulated colored noise. • The low-frequency drift in the inputs and displacements estimation is eliminated. • The proposed method does not require the stationarity of the input/response. This paper introduces a novel method for real-time simultaneous input-state-parameter estimation with modulated colored noise excitation. In this method, the unknown inputs are modeled as modulated colored noise, which is realistic for real environmental excitations. By considering this input model, the typical problem of low-frequency drifts in the inputs and displacement responses estimation can be eliminated. The proposed approach provides a joint input-state-parameter estimation algorithm in a real-time manner by introducing the corresponding modeling equation for the unknown inputs into the extended Kalman filter. A Bayesian methodology is utilized to recursively update the noise covariance matrices in the filtering so stationarity of the input/response is not required for the proposed algorithm. Examples using truss and bridge models under different stationary scenarios are investigated. Compared with other existing methods that suffer from severe low-frequency drifts using only noisy acceleration measurements, the presented approach achieves substantially stable and accurate estimation for inputs, structural states and parameters. [ABSTRACT FROM AUTHOR]

• A novel adaptive signal decomposition method named RVME is proposed. • Simulated and experimental cases validate the effectiveness of the proposed method. • RVME has higher computational efficiency than VMD. The variational mode extraction (VME) developed on the similar basis of variational mode decomposition (VMD) can effectively separate a specific mode by knowing an approximate center frequency from the multi-component signal. Compared with VMD, VME has made certain progress in improving the extraction accuracy and reducing the computational cost when the aim is to separate a specific mode. As the fault feature signal of rolling bearing is usually a band-limited signal which is compact around the excited resonant frequency, VME becomes a potentially effective tool for extracting fault characteristics of rolling bearing. However, how to adaptively determine the center frequency and the penalty factor are two difficult problems when using VME to separate the desired component. Accordingly, this paper presents the recursive variational mode extraction (RVME), an iterative VME-based signal decomposition algorithm. At each iteration of RVME, the initial center frequency and penalty factor for the reconstruction of a specific sub-component can be adaptively determined according to the dominant frequency of the residual signal of the previous iterative decomposition, which makes RVME an adaptive signal decomposition algorithm. The presented method is applied to simulated and experimental fault signals and compared with other classical fault feature extraction approaches, such as variational mode decomposition (VMD) and spectral kurtosis (SK). The results confirm that the established method can extract the fault features as effective as VMD, but it is significantly better than VMD in terms of computational efficiency. Meanwhile, the proposed method shows a stronger capability of weak bearing fault feature extraction and compound fault feature separation compared with SK. [ABSTRACT FROM AUTHOR]

• The RUL of a wind turbine is predicted driven by its own condition monitoring data. • A generalized degradation model is constructed under continuously varying speeds. • The impacts of varying speeds on degradation rates and observations are considered. • The first degradation time and first predicting time are detected automatically. • The best fitting model is selected adaptively according to the degradation trend. Predictive maintenance is one of the most promising ways to reduce the operation and maintenance (O&M) costs of wind turbines (WTs). Remaining useful life (RUL) prediction is the basis for predictive maintenance decision. Self-data-driven methods predict the RUL of a WT driven by its own condition monitoring data without depending on failure event data. Therefore, they are applicable in industrial cases where no sufficient failure event data is available. One challenging issue for RUL prediction of WTs is that they generally suffer from varying rotating speeds. The speed variation has serious impact on the degradation rates as well as the amplitudes of state observations. To deal with this issue, this paper proposes a self-data-driven RUL prediction method for WTs considering continuously varying speeds. In the method, a generalized cumulative degradation model is constructed to describe the degradation process of WTs under continuously varying speeds. A baseline transformation algorithm is developed to transform health state observations under varying speeds into a baseline speed. A continuous trigging algorithm is employed to determine the first degradation time (FDT) for degradation modeling and the first predicting time (FPT) for RUL prediction. The best fitting model is selected adaptively to keep in line with the degradation trend of interest. The effectiveness of the method is demonstrated using a simulation case study and an industrial case study. [ABSTRACT FROM AUTHOR]

The real-time, and accurate inference of model parameters is of great importance in many scientific and engineering disciplines that use computational models (such as a digital twin) for the analysis and prediction of complex physical processes. However, fast and accurate inference for processes of complex systems cannot easily be achieved in real-time with state-of-the-art methods under noisy real-world conditions with the requirement of a real-time response. The primary reason is that the inference of model parameters with traditional techniques based on optimization or sampling often suffers from computational and statistical challenges, resulting in a trade-off between accuracy and deployment time. In this paper, we propose a novel framework for inference of model parameters based on reinforcement learning. The proposed methodology is demonstrated and evaluated on two different physics-based models of turbofan engines. The experimental results demonstrate that the proposed methodology outperforms all other tested methods in terms of speed and robustness, with high inference accuracy. • Reinforcement learning enables real-time and end-to-end calibration of physics-based models. • The calibration policy is learnt without any labelled data or demonstration. • The constrained Lyapunov-based actor–critic algorithm provides action stability. • Learnt policies are robust to sensor noise and model bias. • The proposed RL-based method outperforms current state-of-the-art methods. [ABSTRACT FROM AUTHOR]

The deployment of systems experiencing high-rate dynamic events, such as hypersonic vehicles, advanced weaponry, and active blast mitigation systems, require high-rate structural health monitoring (HRSHM) capabilities in the sub-millisecond realm to ensure continuous operations and safety. However, the development of high-rate feedback systems is a complex task because these dynamic systems are uniquely characterized by (1) large uncertainties in their external loads, (2) high levels of non-stationarity and heavy disturbance, and (3) unmodeled dynamics from changes in system configuration. In this paper, we present a deep learning algorithm specifically engineered for HRSHM applications. It consists of an ensemble of recurrent neural networks (RNNs) constructed with long short-term memory (LSTM) cells with transfer learning capabilities to cope with the highly limited availability of training data as it is typical for high-rate systems. The use of an ensemble of RNNs empowers multi-rate sampling capability to capture multi-temporal features of the time series, thus enabling modeling of non-stationarities. Also, because the RNNs use short-sequence LSTMs and are arranged in parallel, the computation time is substantially reduced to the sub-millisecond range. The performance of the algorithm is investigated on experimental high-rate dynamic data produced from accelerated drop tower tests and benchmarked against results from a prior investigation using a purely on-the-edge algorithm termed variable input space observer (VIO) and its hybrid architecture (hybrid VIO) that incorporated physical knowledge and is considered as an upper bound on the algorithm performance. Numerical results show that the ensemble of RNNs, composed of five RNNs, significantly outperformed the VIO, with performance close to that of the hybrid VIO when it comes to the estimation error metrics, with an average computation time of 25 μ s per prediction step. Yet, the ensemble of RNNs exhibits chattering for low-amplitude excitations, likely attributable to the behavior of RNNs sampling at small time delays. An examination of the RNN weights and hidden states confirms that the algorithm can capture multi-temporal features, and an investigation with respect to noise in the training dataset showed that the algorithm is robust for signal-to-noise ratios up to 20 dB. • Machine learning-based architecture for non-stationary time-series analysis. • Transfer learning for accelerated training with limited data. • Ensemble-based system estimation using parallel recurrent neural networks integrated with attention mechanism is proposed. • Sib-millisecond estimation of a high-rate system is achieved though the proposed architecture. [ABSTRACT FROM AUTHOR]

• The enhanced sequential sensor placement (ESSP) performs efficient sensor placement • ESSP configurations need not be supersets (or subsets) of any optimal configuration • ESSP can select more than one configuration at placement levels by a novel criterion • The optimal configurations' alternatives by ESSP cater to unexpected site constraints • Prediction error correlation and minimum sensor spacing mitigate sensor redundancy The problem of optimal sensor placement for system identification and damage detection is addressed by the development of a robust method based on Bayesian theory. Information entropy is used as the optimality measure to select the optimal configuration from the candidate configurations for a given number of sensors. The enhanced sequential sensor placement (ESSP) algorithm was developed to efficiently address the computational bottlenecks that may arise when a large number of measurable degrees of freedom (DOFs) are considered for candidate configurations. In this paper, the sensor redundancy problem in finely meshed models was addressed by considering (1) the spatial correction of prediction errors at measurable DOFs and (2) a minimum sensor interval. The proposed ESSP algorithm and the two strategies for handling sensor redundancy were studied by comparing the optimal configurations from the conventional methods and that from the ESSP algorithm for a rail-sleeper-ballast system. Finally, the optimal sensor configurations thus obtained were verified via model updating of an in-situ ballasted track system using measured data from an impact hammer test. The analysis results clearly show improvements in the optimality of the sensor configuration with the proposed method relative to the conventional methods. [ABSTRACT FROM AUTHOR]

• MEMS-INS, GPS and polarization compass are integrated together for navigation. • A multi-rate algorithm is proposed to solve the problem caused by different sampling frequencies. • Singular value decomposition is used to improve the numerical stability of the algorithm. The microelectromechanical system-inertial navigation system (MEMS-INS) is one of the most widely used sensor systems in navigation but must be used with other sensors because of its error accumulation characteristics. The information fusion method for integrated navigation systems based on filtering technology is thus very important. This paper introduces a new adaptive Kalman filter for nonlinear integrated systems. An improved multi-rate strong tracking square-root cubature Kalman filter (MR-STSCKF) for a MEMS-INS/Global Positioning System (GPS)/polarization compass integrated navigation system is proposed. The proposed filter is used to estimate the system covariance adaptively. The proposed approach can overcome the problem of the inconsistency between the sampling frequencies of different sensors while maintaining the high precision of the integrated navigation results. Experimental results demonstrate that the proposed MR-STSCKF algorithm is effective in improving the accuracy of the MEMS-INS/GPS/polarization compass integrated navigation system with a high sampling frequency. [ABSTRACT FROM AUTHOR]

[Display omitted] • Established characteristic feature of accelerated tool wear (ATW). • Developed an algorithm for detection of the signal pattern change caused by ATW regardless its level. • Defined a method to detect accelerated tool wear in turning. • Verified the effectiveness of the approach with two experimental cases. There are many algorithms for the identification of gradual tool wear (GTW) which usually counts in minutes and the detection of catastrophic tool failure (CTF) which counts in milliseconds. However, the tool may lose its cutting ability by accelerated tool wear (ATW), which may last several seconds and cannot be sensed either by GTW or CTF detection algorithms. The paper presents an innovative algorithm for early detection of ATW and CTF. It consists in comparing the waveforms of the cutting force sensor signal in hierarchical time windows. The compared waveforms are independent of the absolute value of the signal. This allows the detection of ATWs of different intensity and duration. Tests proved that successful detection of ATW allows for prevention of CTF. [ABSTRACT FROM AUTHOR]

• Proposal for determining the complementary portion of calculated mode contribution. • Adaptive determination of the middle modes and frequencies to be calculated. • Further adaptively determination of the terms in the partial sum. • An analytical expression that is valid for the entire excitation frequency interval. • Significant savings of computational cost. The computational cost of mid-frequency response could be considerably high for finite element systems because each response evaluation needs the numerical solving of a system of complex equations. The mode superposition method requires to calculate various modes within a range that is several times of the given excitation frequency interval, yet the accuracy of result is still unknown. In this paper, a new and innovative method for calculating the mid-frequency response of a proportional damping system is proposed. First, the middle modes and frequencies to be calculated are adaptively determined. Then, the unknown lower-order mode and higher-order mode contribution is approximated by the partial sum of a convergent power series. By using an iterative algorithm and considering only the left or right end of the excitation frequency interval, the number of terms in the sum is adaptively determined. A remarkable conclusion is established here that the resulting expression of frequency response is perfectly valid for the entire excitation frequency interval. The frequency response analysis can thus be performed by using this expression, and by changing the excitation frequency only. Finally, two numerical examples are presented to illustrate the accuracy and effectiveness of the proposed new method. The result shows that the proposed method can significantly reduce CPU (central processing unit) computational time with respect to the direct method for mid-frequency response analysis. [ABSTRACT FROM AUTHOR]

• Nonconvex overlapping group sparsity(NOGS) is proposed for identifying impact forces. • The NOGS goes beyond the sparsity and considers the group structure of impact forces. • The NOGS uses the nonconvexity to strengthen the sparsity and accuracy of solutions. • Fast nonconvex overlapping group sparsity algorithm is developed for model resolving. • The NOGS sparse regularization method is validated on a stiffened composite laminate. Although extensively studied, impact force identification is still a challenging task. When the location of the impact force is unknown, an under-determined problem is usually required to be tackled. In this paper, a novel impact force identification method based on the nonconvex overlapping group sparsity(NOGS) is proposed, allowing to localize the impact and recover its time history simultaneously from quite limited measurements(i.e., the number of responses is less than the number of potential impact locations). The NOGS not only enriches the prior information by taking the group sparsity structure of impact forces into consideration, but enhances the sparsity and the accuracy of estimated amplitude via its nonconvexity. A new algorithm, named fast nonconvex overlapping group sparsity algorithm(FaNogSa), derived in the light of the Majorize-Minimization(MM) principle is utilized to minimize the nonconvex objective function. Simulations and experiments are both implemented systematically on a stiffened composite structure to validate the proposed method, and two strain gauges are utilized to monitor 54 potential impacts. The corresponding results, comparing to the plain nonconvex(atan) regularization and the standard ℓ 1 -norm regularization, say that the proposed method is able to localize the impact and at the same time recover its time history accurately, while under the same measuring conditions the nonconvex(atan) method and the ℓ 1 -norm method usually fail. [ABSTRACT FROM AUTHOR]

• A new method to mine driving patterns from drivers' behavioral signals was proposed. • The most useful combination of segmentation and clustering methods was obtained. • Our extracted driving patterns were descriptive in the maneuver level. • Our proposed methodology could help effective online prediction of driving behavior. Understanding drivers' behavioral characteristics is critical for the design of decision-making modules in autonomous vehicles (AVs) and advanced driver assistance systems (ADASs). Current relevant studies are mainly based on supervised learning methods which involve extensive human efforts in model development. This paper proposed a framework for automatic descriptive driving pattern extraction from driving sequence data using unsupervised algorithms. Based on the Bayesian multivariate linear regression model, two unsupervised algorithms were utilized to segment driving sequences into fragments. Three extended latent Dirichlet allocation models were applied to cluster the fragments into multiple descriptive driving patterns. The collected driving data from a naturalistic driving experiment was applied to examine the effectiveness of our proposed framework. Results show that the unsupervised segmentation algorithms could help effectively detect the switch characteristics between two continuous driving maneuvers along time, and the clustered patterns could effectively describe the characteristics of each driving maneuver. The proposed unsupervised framework provides an effective and efficient data mining solution to facilitating deep and comprehensive understanding on drivers' behavioral characteristics, which will benefit the development of AVs and ADASs. [ABSTRACT FROM AUTHOR]