Search Results (61)

Landslide displacement prediction is crucial for disaster mitigation, yet traditional methods often fail to capture the complex, non-stationary spatiotemporal dynamics of slope evolution. This study introduces an enhanced prediction framework that integrates multi-scale signal processing with dynamic, geology-aware graph modeling. The proposed methodology first employs the Maximum Overlap Discrete Wavelet Transform (MODWT) to denoise raw Global Navigation Satellite System (GNSS)-monitored displacement time series data, enhancing the underlying deformation features. Subsequently, a geology-aware graph is constructed, using the temporal correlation of displacement series as a practical proxy for physical relatedness between monitoring nodes. The framework’s core innovation lies in a dynamic graph optimization model with low-rank constraints, which adaptively refines the graph topology to reflect time-varying inter-sensor dependencies driven by factors like mining activities. Experiments conducted on a real-world dataset from an active open-pit mine demonstrate the framework’s superior performance. The DCRNN-proposed model achieved the highest accuracy among eight competing models, recording a Root Mean Square Error (RMSE) of 2.773 mm in the Vertical direction, a 39.1% reduction compared to its baseline. This study validates that the proposed dynamic graph optimization approach provides a robust and significantly more accurate solution for landslide prediction in complex, real-world engineering environments. Full article

Structural damage in bridges is typically a local phenomenon. When a vehicle passes over the damage location, it induces a local response, which is highly sensitive to the damage. The interaction between the bridge and moving vehicle is a non-stationary time-varying process. The local damage can be accurately identified by analyzing the time-varying characteristics of the bridge response subjected to a moving vehicle. Synchrosqueezing transform, a reassignment method used to sharpen time–frequency representations, offers an effective tool to decompose the non-stationary signal into distinct components. This paper proposes a novel method based on multi-synchrosqueenzing transform to extract the time-varying characteristics of the vehicle–bridge interaction systems for bridge structural health monitoring. A vehicle–bridge interaction model is built to simulate the bridge under moving vehicles. Different damage scenarios of concrete bridges have been simulated. The effect of bridge damage parameters, the vehicle speed, the road surface roughness on the time-varying characteristics of the vehicle–bridge interaction system is studied. Numerical and experimental results demonstrate that the proposed method efficiently and accurately extracts the time-varying features of the vehicle–bridge interaction system, which could serve as potential indicators of structural damage in bridges. Full article

Time series data are fundamental for analyzing temporal dynamics and patterns, enabling researchers and practitioners to model, forecast, and support decision-making across a wide range of domains, such as finance, climate science, environmental studies, and signal processing. In the context of high-dimensional time series, the Vector Autoregressive model (VAR) is widely used, wherein each variable is modeled as a linear combination of lagged values of all variables in the system. However, the traditional VAR framework relies on the assumption of stationarity, which states that the autoregressive coefficients remain constant over time. Unfortunately, this assumption often fails in practice, especially in systems subject to structural breaks or evolving temporal dynamics. The Time-Varying Vector Autoregressive (TV-VAR) model has been developed to address this limitation, allowing model parameters to vary over time and thereby offering greater flexibility in capturing non-stationary behavior. In this study, we propose an enhanced modeling approach for the TV-VAR framework by incorporating minimization solvers in generalized additive models and one-sided kernel smoothing techniques. The effectiveness of the proposed methodology is assessed using simulations based on non-homogeneous Markov chains, accompanied by a detailed discussion of its advantages and limitations. Finally, we illustrate the practical utility of our approach using an application to real-world financial data. Full article

Under complex operating conditions, vibration signals from rotating machinery often exhibit non-stationary characteristics with non-proportional and closely spaced instantaneous frequency (IF) components. Traditional time–frequency analysis (TFA) methods struggle to accurately extract such features due to energy leakage and component mixing. In response to these issues, an enhanced time–frequency analysis approach, termed Local Entropy Selection Scaling–Reassigning Chirplet Transform (LESSRCT), has been developed to improve the representation accuracy for complex non-stationary signals. This approach constructs multi-channel time–frequency representations (TFRs) by introducing multiple scales of chirp rates (CRs) and utilizes a Rényi entropy-based criterion to adaptively select multiple optimal CRs at the same time center, enabling accurate characterization of multiple fundamental components. In addition, a frequency reassignment mechanism is incorporated to enhance energy concentration and suppress spectral diffusion. Extensive validation was conducted on a representative synthetic signal and three categories of real-world data—bat echolocation, inner race bearing faults, and wind turbine gearbox vibrations. In each case, the proposed LESSRCT method was compared against SBCT, GLCT, CWT, SET, EMCT, and STFT. On the synthetic signal, LESSRCT achieved the lowest Rényi entropy of 13.53, which was 19.5% lower than that of SET (16.87) and 35% lower than GLCT (18.36). In the bat signal analysis, LESSRCT reached an entropy of 11.53, substantially outperforming CWT (19.91) and SBCT (15.64). For bearing fault diagnosis signals, LESSRCT consistently achieved lower entropy across varying SNR levels compared to all baseline methods, demonstrating strong noise resilience and robustness. The final case on wind turbine signals demonstrated its robustness and computational efficiency, with a runtime of 1.31 s and excellent resolution. These results confirm that LESSRCT delivers robust, high-resolution TFRs with strong noise resilience and broad applicability. It holds strong potential for precise fault detection and condition monitoring in domains such as aerospace and renewable energy systems. Full article

Blasting-induced seismic waves are typically nonlinear and non-stationary signals. The EMD-Hilbert transform is commonly used for time–frequency analysis of such signals. However, during the empirical mode decomposition (EMD) processing of blasting-induced seismic waves, endpoint effects occur, resulting in varying degrees of divergence in the obtained intrinsic mode function (IMF) components at both ends. The further application of the Hilbert transform to these endpoint-divergent IMFs yield artificial time–frequency analysis results, adversely impacting the assessment of blasting-induced seismic wave hazards. This paper proposes an improved EMD endpoint effect suppression algorithm that considers local endpoint development trends, global time distribution, energy matching, and waveform matching. The method first analyzes global temporal characteristics and endpoint amplitude variations to obtain left and right endpoint extension signal fragment S(t)_L and S(t)_R. Using these as references, the original signal is divided into “b” equal segments S(t)₁, S(t)₂ … S(t)_b. Energy matching and waveform matching functions are then established to identify signal fragments S(t)_i and S(t)_j that match both the energy and waveform characteristics of S(t)_L and S(t)_R. Replacing S(t)_L and S(t)_R with S(t)_i and S(t)_j effectively suppresses the EMD endpoint effects. To verify the algorithm’s effectiveness in suppressing EMD endpoint effects, comparative studies were conducted using simulated signals to compare the proposed method with mirror extension, polynomial fitting, and extreme value extension methods. Three evaluation metrics were utilized: error standard deviation, correlation coefficient, and computation time. The results demonstrate that the proposed algorithm effectively reduces the divergence at the endpoints of the IMFs and yields physically meaningful IMF components. Finally, the method was applied to the analysis of actual blasting seismic signals. It successfully suppressed the endpoint effects of EMD and improved the extraction of time–frequency characteristics from blasting-induced seismic waves. This has significant practical implications for safety assessments of existing structures in areas affected by blasting. Full article

This article proposes a multiscale classification framework for detecting voltage disturbances in electrical distribution systems using artificial neural networks (ANNs) combined with the Hilbert–Huang transform (HHT). The framework targets four core power quality (PQ) events defined in the IEEE 1159-2019 standard: normal operation and voltage sag, swell, and interruption. Unlike traditional methods that operate on a fixed disturbance duration, our approach incorporates multiple time scales (0.2 s, 0.4 s, and 0.8 s) to improve detection robustness across varied event lengths, a critical factor in real-world scenarios where disturbance durations are unpredictable. Features are extracted using empirical mode decomposition (EMD) and Hilbert spectral analysis, enabling accurate representation of the signals’ non-stationary and nonlinear characteristics. The ANN is trained using statistical descriptors derived from the first two intrinsic mode functions (IMFs), capturing both amplitude and frequency content. The method was validated in MATLAB on the IEEE 33-bus radial distribution test system using simulated disturbances. The proposed model achieved a classification accuracy of 94.09% and demonstrated consistent performance across all time windows, supporting its suitability for real-time monitoring in smart distribution networks. This study contributes a scalable and adaptable solution for automated PQ event classification under variable conditions. Full article

Accurate mechanical fault diagnosis of On-Load Tap Changers (OLTCs) remains crucial for power system reliability yet faces challenges from vibration signals’ non-stationary characteristics and limitations of conventional methods. This paper develops a hybrid framework combining metaheuristic-optimized decomposition with hierarchical temporal learning. The methodology employs a Subtraction-Average-Based Optimizer (SABO) to adaptively configure Time-Varying Filtered Empirical Mode Decomposition (TVFEMD), effectively resolving mode mixing through optimized parameter selection. The decomposed components undergo dual-stage temporal processing: A Temporal Convolutional Network (TCN) extracts multi-scale dependencies via dilated convolution architecture, followed by Gated Recurrent Unit (GRU) layers capturing dynamic temporal patterns. An experimental platform was established using a KM-type OLTC to acquire vibration signals under typical mechanical faults, subsequently constructing the dataset. Experimental validation demonstrates superior classification accuracy compared to conventional decomposition–classification approaches in distinguishing complex mechanical anomalies, achieving a classification accuracy of 96.38%. The framework achieves significant accuracy improvement over baseline methods while maintaining computational efficiency, validated through comprehensive mechanical fault simulations. This parameter-adaptive methodology demonstrates enhanced stability in signal decomposition and improved temporal feature discernment, proving particularly effective in handling non-stationary vibration signals under real operational conditions. The results establish practical viability for industrial condition monitoring applications through robust feature extraction and reliable fault pattern recognition. Full article

Deep learning-based fault diagnosis methods have gained extensive attention in recent years due to their outstanding performance. The model input can take the form of multiple domains, such as the time domain, frequency domain, and time–frequency domain, with commonalities and differences between them. Fusing multimodal features is crucial for enhancing diagnostic effectiveness. In addition, original signals typically exhibit nonstationary characteristics influenced by varying working conditions. In this paper, a dual-attentive multimodal fusion method combining a multiscale dilated CNN (DAMFM-MD) is proposed for rotating machinery fault diagnosis. Firstly, multimodal data are constructed by combining original signals, FFT-based frequency spectra, and STFT-based time–frequency images. Secondly, a three-branch multiscale CNN is developed for discriminative feature learning to consider nonstationary factors. Finally, a two-stage sequential fusion is designed to achieve multimodal complementary fusion considering the features with commonality and differentiation. The performance of the proposed method was experimentally verified through a series of industrial case analyses. The proposed DAMFM-MD method achieves the best F-score of 99.95%, an accuracy of 99.96%, and a recall of 99.95% across four sub-datasets, with an average fault diagnosis response time per sample of 1.095 milliseconds, outperforming state-of-the-art methods. Full article

The decomposition of a signal is a fundamental tool in many fields of research, including signal processing, geophysics, astrophysics, engineering, medicine, and many more. By breaking down complex signals into simpler oscillatory components, we can enhance the understanding and processing of the data, unveiling hidden information contained in them. Traditional methods, such as Fourier analysis and wavelet transforms, which are effective in handling mono-dimensional stationary signals, struggle with non-stationary datasets and they require the selection of predefined basis functions. In contrast, the empirical mode decomposition (EMD) method and its variants, such as Iterative Filtering (IF), have emerged as effective non-linear approaches, adapting to signals without any need for a priori assumptions. To accelerate these methods, the Fast Iterative Filtering (FIF) algorithm was developed, and further extensions, such as Multivariate FIF (MvFIF) and Multidimensional FIF (FIF2), have been proposed to handle higher-dimensional data. In this work, we introduce the Multidimensional and Multivariate Fast Iterative Filtering (MdMvFIF) technique, an innovative method that extends FIF to handle data that varies simultaneously in space and time, like the ones sampled using sensor arrays. This new algorithm is capable of extracting Intrinsic Mode Functions (IMFs) from complex signals that vary in both space and time, overcoming limitations found in prior methods. The potentiality of the proposed method is demonstrated through applications to artificial and real-life signals, highlighting its versatility and effectiveness in decomposing multidimensional and multivariate non-stationary signals. The MdMvFIF method offers a powerful tool for advanced signal analysis across many scientific and engineering disciplines. Full article

This paper investigates the design and performance of a tunable pendulum energy harvester (TPEH) integrated with cone continuously variable transmission (CVT) to enhance energy harvesting efficiency in broadband and non-stationary vibrational environments. The cone CVT mechanism enables the tunability of the harvester’s natural frequency, allowing it to dynamically adapt and maintain resonance across varying excitation frequencies. A specific focus is placed on the system’s behavior under chirp signal base excitation, which simulates a time-varying frequency environment. Experimental and analytical approaches are employed to evaluate the system’s dynamic response, energy output, and frequency adaptation capabilities. The results demonstrate that the proposed TPEH system achieves significant energy harvesting performance improvements by leveraging the cone CVT to optimize power generation under resonance conditions. The system is also shown to be effective in maintaining stable operation over a wide range of frequencies, demonstrating its versatility for real-world vibrational energy harvesting applications. This research highlights the importance of tunability in energy harvesting systems and the role of mechanical transmission mechanisms in improving adaptability. The proposed design has strong potential for applications in environments with non-stationary vibrations, such as transportation systems, industrial machinery, and infrastructure monitoring. Full article

Researchers have increasingly focused on thermal comfort, examining both individuals’ thermal sensations and the percentage of people dissatisfied with the thermal environment. Most studies rely on the widely used PMV (Predicted Mean Vote) model and the PPD (Predicted Percentage of Dissatisfied) value derived from it, both defined by the ISO 7730:2005 standard. However, previous studies have shown that this standardized method only applies under steady-state conditions, which do not reflect the dynamic nature of everyday environments. As closed-loop control technologies gain prominence in building services, the need to evaluate thermal comfort under time-varying conditions has grown. The standard method does not account for the thermal inertia of the human body, which limits its applicability in such dynamic contexts. In this study, we develop a method to estimate instantaneous thermal sensation under non-stationary conditions by incorporating thermal inertia through signal processing techniques. This approach addresses a well-recognized limitation of the standard PMV–PPD model and provides a way to assess thermal comfort in real time. We collected experimental data using a thermal comfort measurement station, a thermal manikin, and human subjects in a controlled climate chamber. The proposed method enables real-time evaluation of thermal comfort in dynamic environments and offers a foundation for integration into HVAC control and comfort optimization systems. Full article

Effective leak detection and leak size identification are essential for maintaining the operational safety, integrity, and longevity of industrial pipelines. Traditional methods often suffer from high noise sensitivity, limited adaptability to non-stationary signals, and excessive computational costs, which limits their feasibility for real-time monitoring applications. This study presents a novel acoustic emission (AE)-based pipeline monitoring approach, integrating Empirical Wavelet Transform (EWT) for adaptive frequency decomposition with customized one-dimensional DenseNet architecture to achieve precise leak detection and size classification. The methodology begins with EWT-based signal segmentation, which isolates meaningful frequency bands to enhance leak-related feature extraction. To further improve signal quality, adaptive thresholding and denoising techniques are applied, filtering out low-amplitude noise while preserving critical diagnostic information. The denoised signals are processed using a DenseNet-based deep learning model, which combines convolutional layers and densely connected feature propagation to extract fine-grained temporal dependencies, ensuring the accurate classification of leak presence and severity. Experimental validation was conducted on real-world AE data collected under controlled leak and non-leak conditions at varying pressure levels. The proposed model achieved an exceptional leak detection accuracy of 99.76%, demonstrating its ability to reliably differentiate between normal operation and multiple leak severities. This method effectively reduces computational costs while maintaining robust performance across diverse operating environments. Full article

The non-stationary characteristics of the vibration signals of rolling bearings will be aggravated under variable speed conditions. Meanwhile, multichannel signals can provide a more comprehensive characterization of state information, providing multiple sources of information that facilitate information fusion and enhancement. However, traditional adaptive signal decomposition methods generally assume that the frequency information is constant and stationary, and it is difficult to achieve a unified decomposition when dealing with multichannel time-varying signals. Therefore, the intention of this paper is to propose a multichannel signal adaptive decomposition method applicable to variable speed conditions. Specifically, this paper takes advantage of the strong adaptability and robustness of symplectic geometric mode decomposition (SGMD). To improve its applicability to multichannel time-varying signals at variable rotational speeds, a generalized multivariate symplectic sparsest united decomposition (GMSSUD) method is proposed. In GMSSUD, firstly, the completely adaptive projection (CAP) method is employed to achieve a unified representation of the multichannel signals. Then, the generalized demodulation method is introduced to stabilize the signal and subsequently reduce the noise through component screening and reconstruction. Finally, with the new proposed operator as the optimization objective, the constructed sparse filter parameters are optimized to achieve the frequency band segmentation. The analysis results demonstrate that the GMSSUD method possesses higher decomposition precision for multichannel signals with variable speeds and also has a stronger diagnosis ability for variable-speed bearing faults. Full article

This paper introduces a novel robust Kalman filter designed to leverage symmetrical properties within the Pearson Type VII-Inverse Wishart (PVIW) distribution, enhancing state estimation accuracy in the presence of time-varying biases and non-stationary heavy-tailed (NSHT) noise. The filter includes a shape parameter from the normal distribution and an extra variable from the Gamma distribution, which are used to symmetrically adjust the average and variation measures of the data to fit better under difficult noise conditions. To deal with unknown noise that changes over time, the filter uses the Inverse Wishart distribution to model and estimate the scale matrix deviations, making it easier to adapt to changes. The filter also uses a technique called Variational Bayesian to estimate both the state and the parameters at the same time. The results from simulations show that this new filter greatly improves the accuracy and strength of the estimation compared to the usual Kalman filters that assume a normal distribution, especially when there is non-stationary heavy-tailed noise. The main objective is to improve estimation in signal processing and control systems where heavy-tailed noise is prevalent. Full article

Aiming to address the difficulty in extracting the early weak fault features of bearings under complex operating conditions, a fault diagnosis method is proposed based on the adaptive fusion of time-varying filtering empirical mode decomposition (TVF-EMD) modal components and singular value decomposition (SVD) noise reduction. First, the snake optimization (SO) technique is used to optimize the TVF-EMD algorithm in order to determine the optimal parameters that match the input signal. Then, the bearing signal is divided into a number of intrinsic mode functions (IMFs) using TVF-EMD in order to reduce the nonlinearity and non-stationary characteristics of the fault signal. An index for the envelope fault information energy ratio (EFIER) is created to overcome the drawback of there being too many IMF components after TVF-EMD decomposition. The IMF components are ranked in descending order according to the EFIER, and they are fused according to the maximum principle of the energy ratio of envelope fault information until the optimal fusion component is determined. Finally, the fault feature is extracted when the optimal fusion component is denoised using SVD. Two measured bearing fault signals and simulation signals are used to validate the performance of the proposed method. The experimental findings demonstrate that the approach has good sensitive feature screening, fusion, and noise reduction capabilities. The proposed method can more precisely extract the early fault features of bearings and accurately identify fault types. Full article