Search Results (394)

Magnetotelluric (MT) sounding is a crucial technique in mineral exploration. However, MT data are highly susceptible to various types of noise. Traditional data processing methods, which rely on the assumption of signal stationarity, often result in severe distortion when suppressing non-stationary noise. In this study, we propose a novel, adaptive, and less parameter-dependent signal decomposition method for MT signal denoising, based on time–frequency domain analysis and the application of modal decomposition. The method uses Variational Mode Decomposition (VMD) to adaptively decompose the MT signal into several intrinsic mode functions (IMFs), obtaining the instantaneous time–frequency energy distribution of the signal. Subsequently, robust statistical methods are introduced to extract the independent components of each IMF, thereby identifying signal and noise components within the decomposition results. Synthetic data experiments show that our method accurately separates high-amplitude non-stationary interference. Furthermore, it maintains stable decomposition results under various parameter settings, exhibiting strong robustness and low parameter dependency. When applied to field MT data, the method effectively filters out non-stationary noise, leading to significant improvements in both apparent resistivity and phase curves, indicating its practical value in mineral exploration. Full article

Accurate short-term weather forecasting plays a vital role in disaster response, agriculture, and energy management, where timely and reliable predictions are essential for decision-making. Graph neural networks (GNNs), known for their ability to model complex spatial structures and relational data, have achieved remarkable success in meteorological forecasting by effectively capturing spatial dependencies among distributed weather stations. However, most existing GNN-based approaches rely on pairwise station connections, limiting their capacity to represent higher-order spatial interactions. Moreover, their dependence on supervised learning makes them vulnerable to spatial heterogeneity and temporal non-stationarity. This paper introduces a novel spatial–temporal pretraining framework, Hypergraph-enhanced Meteorological Pretraining (HyMePre), which combines hypergraph neural networks with self-supervised learning to model high-order spatial dependencies and improve generalization across diverse climate regimes. HyMePre employs a two-stage masking strategy, applying spatial and temporal masking separately, to learn disentangled representations from unlabeled meteorological time series. During forecasting, dynamic hypergraphs group stations based on meteorological similarity, explicitly capturing high-order dependencies. Extensive experiments on large-scale reanalysis datasets show that HyMePre outperforms conventional GNN models in predicting temperature, humidity, and wind speed. The integration of pretraining and hypergraph modeling enhances robustness to noisy data and improves generalization to unseen climate patterns, offering a scalable and effective solution for operational weather forecasting. Full article

This study analyzes air pollution time-series big data to assess stationarity, seasonal patterns, and the performance of machine learning models in forecasting PM2.5 concentrations. Fifty-two low-cost sensors (LCS) were deployed across Krakow city and its surroundings (Poland), collecting hourly air quality data and generating nearly 20,000 observations per month. The network captured both spatial and temporal variability. The Kwiatkowski–Phillips–Schmidt–Shin (KPSS) test confirmed trend-based non-stationarity, which was addressed through differencing, revealing distinct daily and 12 h cycles linked to traffic and temperature variations. Additive seasonal decomposition exhibited time-inconsistent residuals, leading to the adoption of multiplicative decomposition, which better captured pollution outliers associated with agricultural burning. Machine learning models—Ridge Regression, XGBoost, and LSTM (Long Short-Term Memory) neural networks—were evaluated under high spatial and temporal variability (winter) and low variability (summer) conditions. Ridge Regression showed the best performance, achieving the highest $R^2$ (0.97 in winter, 0.93 in summer) and the lowest mean squared errors. XGBoost showed strong predictive capabilities but tended to overestimate moderate pollution events, while LSTM systematically underestimated PM2.5 levels in December. The residual analysis confirmed that Ridge Regression provided the most stable predictions, capturing extreme pollution episodes effectively, whereas XGBoost exhibited larger outliers. The study proved the potential of low-cost sensor networks and machine learning in urban air quality forecasting focused on rare smog episodes (RSEs). Full article

The spread of COVID-19 is influenced by multiple factors, including control policies, virus characteristics, individual behaviors, and environmental conditions, exhibiting highly complex nonlinear dynamic features. The time series of new confirmed cases shows significant nonlinearity and non-stationarity. Traditional prediction methods that rely solely on one-dimensional case data struggle to capture the multi-dimensional features of the data and are limited in handling nonlinear and non-stationary characteristics. Their prediction accuracy and generalization capabilities remain insufficient, and most existing studies focus on single-step forecasting, with limited attention to multi-step prediction. To address these challenges, this paper proposes a multi-module fusion prediction model—TSMixer-BiKSA network—that integrates multi-feature inputs, Variational Mode Decomposition (VMD), and a dual-branch parallel architecture for 1- to 3-day-ahead multi-step forecasting of new COVID-19 cases. First, variables highly correlated with the target sequence are selected through correlation analysis to construct a feature matrix, which serves as one input branch. Simultaneously, the case sequence is decomposed using VMD to extract low-complexity, highly regular multi-scale modal components as the other input branch, enhancing the model’s ability to perceive and represent multi-source information. The two input branches are then processed in parallel by the TSMixer-BiKSA network model. Specifically, the TSMixer module employs a multilayer perceptron (MLP) structure to alternately model along the temporal and feature dimensions, capturing cross-time and cross-variable dependencies. The BiGRU module extracts bidirectional dynamic features of the sequence, improving long-term dependency modeling. The KAN module introduces hierarchical nonlinear transformations to enhance high-order feature interactions. Finally, the SA attention mechanism enables the adaptive weighted fusion of multi-source information, reinforcing inter-module synergy and enhancing the overall feature extraction and representation capability. Experimental results based on COVID-19 case data from Italy and the United States demonstrate that the proposed model significantly outperforms existing mainstream methods across various error metrics, achieving higher prediction accuracy and robustness. Full article

As the global energy system accelerates its transition toward a low-carbon economy, renewable energy sources (RESs), such as wind and photovoltaic power, are rapidly replacing traditional fossil fuels. These RESs are becoming a critical element of deeply decarbonized power systems (DDPSs). However, the inherent non-stationarity, multi-scale volatility, and uncontrollability of RES output significantly increase the risk of source–load imbalance, posing serious challenges to the reliability and economic efficiency of power systems. Scenario generation technology has emerged as a critical tool to quantify uncertainty and support dispatch optimization. Nevertheless, conventional scenario generation methods often fail to produce highly credible wind and solar output scenarios. To address this gap, this paper proposes a novel renewable energy scenario generation method based on a multi-resolution diffusion model. To accurately capture fluctuation characteristics across multiple time scales, we introduce a diffusion model in conjunction with a multi-scale time series decomposition approach, forming a multi-stage diffusion modeling framework capable of representing both long-term trends and short-term fluctuations in RES output. A cascaded conditional diffusion modeling framework is designed, leveraging historical trend information as a conditioning input to enhance the physical consistency of generated scenarios. Furthermore, a forecast-guided fusion strategy is proposed to jointly model long-term and short-term dynamics, thereby improving the generalization capability of long-term scenario generation. Simulation results demonstrate that MDDPM achieves a Wasserstein Distance (WD) of 0.0156 in the wind power scenario, outperforming DDPM (WD = 0.0185) and MC (WD = 0.0305). Additionally, MDDPM improves the Global Coverage Rate (GCR) by 15% compared to MC and other baselines. Full article

A geographically and temporally weighted regression (GTWR) model is an effective tool for dealing with spatial heterogeneity and temporal non-stationarity simultaneously. As an important characteristic of spatiotemporal data, spatiotemporal autocorrelation should be considered when constructing spatiotemporally varying coefficient models. The proposed local autoregressive geographically and temporally weighted regression (GTWRLAR) model can simultaneously handle spatiotemporal autocorrelations among response variables and the spatiotemporal heterogeneity of regression relationships. The two-stage weighted least squares (2SLS) estimation can effectively reduce computational complexity. However, the weighted least squares estimation is essentially a Nadaraya–Watson kernel-smoothing approach for nonparametric regression models, and it suffers from a boundary effect. For spatiotemporally varying coefficient models, the three-dimensional spatiotemporal coefficients (longitude, latitude, and time) inherently exhibit larger boundaries than one-dimensional intervals. Therefore, the boundary effect of the 2SLS estimation of GTWRLAR will be more serious. A local–linear geographically and temporally weighted 2SLS (GTWRLAR-L) estimation is proposed to correct the boundary effect in both the spatial and temporal dimensions of GTWRLAR and simultaneously improve parameter estimation accuracy. The simulation experiment shows that the GTWRLAR-L method reduces the root mean square error (RMSE) of parameter estimates compared to the standard GTWRLAR approach. Empirical analyses of carbon emissions in China’s Yellow River Basin (2017–2021) show that GTWRLAR-L enhances the adjusted $R^2$ from 0.888 to 0.893. Full article

To address the challenges of weak fault features and strong non-stationarity in early-stage vibration signals, this study proposes a novel fault diagnosis method combining enhanced variational mode decomposition (VMD) with a structurally improved GoogLeNet. Specifically, an improved wild horse optimizer (IWHO) with tent chaotic mapping is employed to automatically optimize critical VMD parameters, including the number of modes K and the penalty factor α, enabling precise decomposition of non-stationary signals to extract weak fault features. The vibration signal is decomposed, and the top five intrinsic mode functions (IMFs) are selected based on the kurtosis criterion. Time–frequency features are then extracted from these IMFs and input into a modified GoogLeNet classifier. The GoogLeNet structure is improved by replacing standard n × n convolution kernels with cascaded 1 × n and n × 1 kernels, and by substituting the ReLU activation function with a parameterized TReLU function to enhance adaptability and convergence. Experimental results on two public rolling bearing datasets demonstrate that the proposed method effectively handles non-stationary signals, achieving 99.17% accuracy across four fault types and maintaining over 95.80% accuracy under noisy conditions. Full article

Previous work used machine learning (ML) to estimate past and current alfalfa yields and showed that domain adaptation (DA) with data synthesis shows promise in classifying yields as high, medium, or low. The current work uses similar techniques to forecast future alfalfa yields. A novel technique is proposed for forecasting alfalfa time series data that exploits stationarity and predicts differences in yields rather than the yields themselves. This forecasting technique generally provides more accurate forecasts than the established ARIMA family of forecasters for both univariate and multivariate time series. Furthermore, this ML-based technique is potentially easier to use than the ARIMA family of models. Also, previous work is extended by showing that DA with data synthesis also works well for predicting continuous values, not just for classification. The novel scale-invariant tabular synthesizer (SITS) is proposed, and it is competitive with or superior to other established synthesizers in producing data that trains strong models. This synthesis algorithm leads to R scores over 100% higher than an established synthesizer in this domain, while ML-based forecasters beat the ARIMA family with symmetric mean absolute percent error (sMAPE) scores as low as 12.81%. Finally, ML-based forecasting is combined with DA (ForDA) to create a novel pipeline that improves forecast accuracy with sMAPE scores as low as 9.81%. As alfalfa is crucial to the global food supply, and as climate change creates challenges with managing alfalfa, this work hopes to help address those challenges and contribute to the field of ML. Full article

Current non-destructive testing (NDT) methods, such as those based on wave velocity measurements, lack the sensitivity necessary to detect early-stage damage in concrete structures. Similarly, common signal processing techniques often assume linearity and stationarity among the signal data. By analyzing wave attenuation measurements using advanced signal processing techniques, mainly Hilbert–Huang transform (HHT), this work aims to enhance the early detection of damage in concrete. This study presents a novel energy-based technique that integrates complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and Hilbert spectrum analysis (HSA), to accurately capture nonlinear and nonstationary signal behaviors. Ultrasonic non-destructive testing was performed in this study on manufactured concrete specimens subjected to micro-damage characterized by internal microcracks smaller than 0.5 mm, induced through controlled freeze–thaw cycles. The recorded signals were decomposed from the time domain using CEEMDAN into frequency-ordered intrinsic mode functions (IMFs). A multi-criteria selection strategy, including damage index evaluation, was employed to identify the most effective IMFs while distinguishing true damage-induced energy loss from spurious nonlinear artifacts or noise. Localized damage was then analyzed in the frequency domain using HSA, achieving an up to 88% reduction in wave energy via Marginal Hilbert Spectrum analysis, compared to 68% using Fourier-based techniques, demonstrating a 20% improvement in sensitivity. The results indicate that the proposed technique enhances early damage detection through wave attenuation analysis and offers a superior ability to handle nonlinear, nonstationary signals. The Hilbert Spectrum provided a higher time-frequency resolution, enabling clearer identification of damage-related features. These findings highlight the potential of CEEMDAN-HSA as a practical, sensitive tool for early-stage microcrack detection in concrete. Full article

We explored the stationarity, trend, and seasonality of the hive weight and in-hive temperature of ten managed honey bee (Apis mellifera) colonies at a research apiary of the Carl Hayden Bee Research Center in Tucson, Arizona, USA. The hives were monitored with electronic scales and in-hive temperature sensors from June to October 2022. The weight and temperature were recorded every five minutes around the clock. The collected data were curated into 2160 timestamped weight and 2160 timestamped temperature observations. We performed a systematic autoregressive integrated moving average (ARIMA) time series analysis to answer three fundamental questions: (a) Does seasonality matter in the ARIMA forecasting of hive weight and in-hive temperature? (b) To what extent do the best forecasters of one hive generalize to other hives? and (c) Which time series type (i.e., hive weight or in-hive temperature) is better predictable? Our principal findings were as follows: (1) The hive weight and in-hive temperature series were not white noise, were not normally distributed, and, for most hives, were not difference- or trend-stationary; (2) Seasonality matters, in that seasonal ARIMA (SARIMA) forecasters outperformed their ARIMA counterparts on the curated dataset; (3) The best hive weight and in-hive temperature forecasters of the ten monitored colonies appeared to be colony-specific; (4) The accuracy of the hive weight forecasts was consistently higher than that of the in-hive temperature forecasts; (5) The weight and temperature forecasts exhibited common qualitative patterns. Full article

Reliable damage detection in operational offshore wind turbines (OWTs) remains challenging due to the inherent non-stationarity of environmental excitations and signal degradation from noise-contaminated partial measurements. To address these limitations, this study proposes a robust damage detection method for OWTs under non-stationary ambient excitations using partial measurements with strong noise resistance. The method is first developed for a scenario with known non-stationary ambient excitations. By reformulating the time-domain equation of motion in terms of non-stationary cross-correlation functions, structural stiffness parameters are estimated using partially measured acceleration responses through the extended Kalman filter (EKF). To account for the more common case of unknown excitations, the method is enhanced via the extended Kalman filter under unknown input (EKF-UI). This improved approach enables the simultaneous identification of the physical parameters of OWTs and unknown non-stationary ambient excitations through the data fusion of partial acceleration and displacement responses. The proposed method is validated through two numerical cases: a frame structure subjected to known non-stationary ground excitation, followed by an OWT tower under unknown non-stationary wind and wave excitations using limited measurements. The numerical results confirm the method’s capability to accurately identify structural damage even under significant noise contamination, demonstrating its practical potential for OWTs’ damage detection applications. 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

Forecasting multivariate time series is a pivotal task in controlling multi-sensor systems. The joint forecasting of all channels may be too complex, whereas forecasting the channels independently may cause important spatial inter-dependencies to be overlooked. In this paper, we improve the performance of single-channel forecasting algorithms by designing an interpretable front-end that extracts the spatial–temporal components from the input multivariate time series. Specifically, the multivariate samples are first segmented into equal-sized matrix symbols. The symbols are decomposed into the frequency-separated Intrinsic Mode Functions (IMFs) using a 2D Empirical-Mode Decomposition (EMD). The IMF components in each channel are then forecasted independently using relatively simple univariate predictors (UPs) such as DLinear, FITS, and TCN. The symbol size is determined to maximize the temporal stationarity of the EMD residual trend using Bayesian optimization. In addition, since the overall performance is usually dominated by a few of the weakest predictors, it is shown that the forecasting accuracy can be further improved by reordering the corresponding channels to make more correlated channels more adjacent. However, channel reordering requires retraining the affected predictors. The main advantage of the proposed forecasting framework for multivariate time series is that it retains the interpretability and simplicity of single-channel forecasting methods while improving their accuracy by capturing information about the spatial-channel dependencies. This has been demonstrated numerically assuming a 64-channel EEG dataset. Full article

Iron ore sintering is a critical process in steelmaking, where the produced sinter is the main raw material for blast furnace ironmaking. The quality and yield of sinter ore directly affect the cost and efficiency of iron and steel production. Accurately predicting the burn-through point (BTP) temperature is of paramount importance for controlling quality and yield. Traditional BTP temperature prediction only utilizes data from bellows, neglecting the information contained in sinter images. This study combines color temperature information extracted from the cross-sectional frame at the discharge end with bellows data. Due to the non-stationarity of the BTP temperature, a hybrid prediction model of the BTP temperature integrating bidirectional long short-term memory and extreme gradient boosting is presented. By combining the advantages of deep learning and tree ensemble learning, a hybrid prediction model of the BTP temperature is established using the color temperature information in the cross-sectional frame at the discharge end and time-series data. Experiments were conducted with the actual running data in an iron and steel enterprise and show that the proposed method has higher accuracy than existing methods, achieving an approximately 4.3% improvement in prediction accuracy. The proposed method can provide an effective reference for decision-making and for the optimization of operating parameters in the sintering process. Full article

Underwater sonar target recognition is crucial in fields such as national defense, navigation, and environmental monitoring. However, it faces issues such as the complex characteristics of ship-radiated noise, imbalanced data distribution, non-stationarity, and bottlenecks of existing technologies. This paper proposes the MultiFuseNet-AID network, aiming to address these challenges. The network includes the TriFusion block module, the novel lightweight attention residual network (NLARN), the long- and short-term attention (LSTA) module, and the Mamba module. Through the TriFusion block module, the original, differential, and cumulative signals are processed in parallel, and features such as MFCC, CQT, and Fbank are fused to achieve deep multi-domain feature fusion, thereby enhancing the signal representation ability. The NLARN was optimized based on the ResNet architecture, with the SE attention mechanism embedded. Combined with the long- and short-term attention (LSTA) and the Mamba module, it could capture long-sequence dependencies with an O(N) complexity, completing the optimization of lightweight long sequence modeling. At the same time, with the help of feature fusion, and layer normalization and residual connections of the Mamba module, the adaptability of the model in complex scenarios with imbalanced data and strong noise was enhanced. On the DeepShip and ShipsEar datasets, the recognition rates of this model reached 98.39% and 99.77%, respectively. The number of parameters and the number of floating point operations were significantly lower than those of classical models, and it showed good stability and generalization ability under different sample label ratios. The research shows that the MultiFuseNet-AID network effectively broke through the bottlenecks of existing technologies. However, there is still room for improvement in terms of adaptability to extreme underwater environments, training efficiency, and adaptability to ultra-small devices. It provides a new direction for the development of underwater sonar target recognition technology. Full article