Search Results (18)

The expansion of renewables in modern power systems and the coordinated development of upstream and downstream industrial chains are promoting a shift on the utility side from traditional settlement by energy toward operation driven by data and models. Industrial electricity consumption data exhibit pronounced multi-scale temporal structures and sectoral heterogeneity, which makes unified long-term load and generation forecasting while maintaining accuracy, interpretability, and scalability a challenge. From a modern system identification perspective, this paper proposes a System Identification in Adaptive Bayesian Framework (SIABF) for medium- and long-term industrial load forecasting based on daily freeze electricity time series. By combining daily aggregation of high-frequency data, frequency domain analysis, sparse identification, and long-term extrapolation, we first construct daily freeze series from 15 min measurements, and then we apply discrete Fourier transforms and a spectral complexity index to extract dominant periodic components and build an interpretable sinusoidal basis library. A sparse regression formulation with ${\ell }_1$ regularization is employed to select a compact set of key basis functions, yielding concise representations of sector and enterprise load profiles and naturally supporting multivariate and joint multi-sector modeling. Building on this structure, we implement a state-space-implicit physics-informed Bayesian forecasting model and evaluate it on real data from three representative sectors, namely, steel, photovoltaics, and chemical, using one year of 15 min measurements. Under a one-month-ahead evaluation using one year of 15 min measurements, the proposed framework attains a Mean Absolute Percentage Error (MAPE) of 4.5% for a representative PV-related customer case and achieves low single-digit MAPE for high-inertia sectors, often outperforming classical statistical models, sparse learning baselines, and deep learning architectures. These results should be interpreted as indicative given the limited time span and sample size, and broader multi-year, population-level validation is warranted. Full article

Real-time control of web traffic is a critical issue for network operators and service providers. It helps ensure robust service and avoid service interruptions, which has an important financial impact. However, due to the high speed and volume of actual internet traffic, standard multivariate time series models are inadequate for ensuring efficient real-time traffic management. In this paper we introduce a new model for functional time series analysis, developed by combining a local linear smoothing approach with an $L^1$-robust estimator of the quantile’s derivative. It constitutes an alternative, robust estimator for functional modal regression that is adequate to handle the stochastic volatility of high-frequency of web traffic data. The mathematical support of the new model is established under functional dependent case. The asymptotic analysis emphasizes the functional structure of the data, the functional feature of the model, and the stochastic characteristics of the underlying time-varying process. We evaluate the effectiveness of our proposed model using comprehensive simulations and real-data application. The computational results illustrate the superiority of the nonparametric functional model over the existing conventional methods in web traffic modeling. Full article

Nonintrusive load monitoring (NILM) relies on high-resolution sensor data to disaggregate total building energy into end-use load components, for example HVAC, ventilation, and appliances. On the ADRENALIN corpus, simple NaN handling with forward fill and mean substitution reduced average NMAE from 0.82 to 0.76 for the Bayesian baseline, from 0.71 to 0.64 for BI-LSTM, and from 0.59 to 0.53 for the Time–Frequency Mask (TFM) model, across nine buildings and four temporal resolutions. However, many NILM models still show degraded accuracy due to unresolved data-quality issues, especially missing values, timestamp irregularities, and sensor inconsistencies, a limitation underexplored in current benchmarks. This paper presents a fully automated data-quality assurance pipeline for time-series energy datasets. The pipeline performs multivariate profiling, statistical analysis, and threshold-based diagnostics to compute standardized quality metrics, which are aggregated into an interpretable Building Quality Score (BQS) that predicts NILM performance and supports dataset ranking and selection. Explainability is provided by SHAP and a lightweight large language model, which turns visual diagnostics into concise, actionable narratives. The study evaluates practical quality improvement through systematic handling of missing values, linking metric changes to downstream error reduction. Using random-forest surrogates, SHAP identifies missingness and timestamp irregularity as dominant drivers of error across models. Core contributions include the definition and validation of BQS, an interpretable scoring and explanation framework for time-series quality, and an end-to-end evaluation of how quality diagnostics affect NILM performance at scale. Full article

Remaining useful life (RUL) prediction of industrial equipment is critical for achieving safe operations and optimizing predictive maintenance. To tackle the limitations of poor interpretability, inaccurate predictions, and high computational cost in complex system degradation modeling, this paper proposes SiMBA-PINN, a novel fusion framework that synergizes Physics-Informed Neural Network (PINN) with an enhanced state-space model (SiMBA). The framework achieves dynamic fusion of data-driven features and physical laws through a two-branch synergistic mechanism: the temporal modeling branch combines selective state-space SiMBA with Einstein Fast Fourier Transform (EinFFT)-based spectral mixing to efficiently capture cross-sensor temporal dependencies and degradation trends, while the physics-constraint branch embeds automatically differentiable partial differential equation residuals derived from domain-specific degradation mechanisms, enforcing physical consistency through deep hidden physics modeling. Here, the EinFFT-based spectral mixing leverages frequency-domain interactions to effectively blend the spectral components of multivariate time-series data, thereby enhancing the modeling of cross-sensor dependencies. Meanwhile, deep hidden physics modeling integrates physics-informed partial differential equation (PDE) residuals through differentiable operators, aligning the learned representations with domain-specific dynamics via a constraint-driven loss design. Experimental results from the C-MAPSS dataset confirm that the proposed model significantly outperforms PINN-, Mamba- and attention mechanism-based models, achieving State-of-the-Art RMSE on the most challenging FD004 subset. This physics-aware framework achieves deployable and interpretable RUL prediction by balancing accuracy with linear-time complexity. Full article

Climate change presents significant challenges due to the increasing frequency and intensity of extreme weather events. Mexico, with its diverse climate and geographic position, is particularly vulnerable, underscoring the need for robust strategies to predict atmospheric variables. This work presents TAE Predict (Time series Analysis and Ensemble-based Prediction with relevant feature selection) based on relevant feature selection and ensemble models of machine learning. Dimensionality in multivariate time series is reduced through Principal Component Analysis, ensuring interpretability and efficiency. Additionally, data remediation techniques improve data set quality. The ensemble combines Long Short-Term Memory neural networks, Random Forest regression, and Support Vector Machines, optimizing their contributions using heuristic algorithms such as Particle Swarm Optimization. Experimental results from meteorological time series in key Mexican cities demonstrate that the proposed strategy outperforms individual models in accuracy and robustness. This methodology provides a replicable framework for climate variable forecasting, delivering analytical tools that support decision-making in critical sectors, such as agriculture and water resource management. The findings highlight the potential of integrating modern techniques to address complex, high-dimensional problems. By combining advanced prediction models and feature selection strategies, this study advances the reliability of climate forecasts and contributes to the development of effective adaptation and mitigation measures in response to climate change challenges. Full article

Time-series forecasting is a cornerstone of decision making in domains such as finance, energy management, and meteorology, where precise predictions drive both economic and operational efficiency. However, traditional time-domain methods often struggle to capture the intricate symmetries and hierarchical dependencies inherent in complex multivariate time-series data. These methods frequently fail to distinguish between global trends and localized fluctuations, limiting their ability to model the multifaceted temporal dynamics that arise across different time scales. To address these challenges, we propose a novel dual-component framework that explicitly leverages the symmetry between long-term trends and short-term fluctuations. Inspired by the principles of signal decomposition, we partition time-series data into a low-frequency stabilization component and a high-frequency fluctuation component. The stabilization component captures inter-variable relationships and global frequency-domain component dependencies through Fourier-transformed frequency-domain representations, variable-oriented attention mechanisms, and dilated causal convolutions. Meanwhile, the fluctuation component models localized dynamics using a multi-granularity structure and time-step attention mechanisms to enhance the sensitivity and robustness to transient variations. By integrating these complementary perspectives, our approach provides a more holistic representation of time-series dynamics. Comprehensive experiments on benchmark datasets from electricity, transportation, and weather domains demonstrate that our method consistently outperforms state-of-the-art models, achieving superior accuracy. Beyond predictive performance, our framework offers a deeper interpretability of temporal behaviors, highlighting its potential for practical applications in complex systems. This work underscores the importance of symmetry-aware modeling in advancing time-series forecasting methodologies. Full article

This study presents an improved deep-learning model, termed Enhanced Time Series Mixer (E-TSMixer), for the prediction of particulate matter. By analyzing the temporal evolution of PM_2.5 concentrations from multivariate monitoring data, the model demonstrates significant prediction capabilities while maintaining consistency with observed pollutant transport characteristics in the urban boundary layer. In E-TSMixer, a fully connected output layer is proposed to enhance the predictive capability for complex spatiotemporal dependencies. The relevant data on air quality and traffic flow are fused to achieve high-precision predictions of PM_2.5 concentrations through a multivariate time-series forecasting model. An asymmetric penalty mechanism is added to dynamically optimize the loss function. Experimental results indicate that the proposed E-TSMixer model achieves higher accuracy for the prediction of PM_2.5, which significantly outperforms the traditional models. Additionally, an intelligent dual regulation of fixed and dynamic threshold model is introduced and combined with E-TSMixer for the decision-making model of the real-time adjustments of the frequency, routes, and timing of water truck operation in practice. Full article

With the increase in photovoltaic installed capacity year by year, accurate photovoltaic power prediction is of great significance for photovoltaic grid-connected operation and scheduling planning. In order to improve the prediction accuracy, this paper proposes a photovoltaic power prediction combination model based on Pearson Correlation Coefficient (PCC), Complete Ensemble Empirical Mode Decomposition (CEEMDAN), K-means clustering, Variational Mode Decomposition (VMD), Convolutional Neural Network (CNN), and Bidirectional Long Short-Term Memory (BiLSTM). By making full use of the symmetric structure of the BiLSTM algorithm, one part is used to process the data sequence in order, and the other part is used to process the data sequence in reverse order. It captures the characteristics of sequence data by simultaneously processing a ‘symmetric’ information. Firstly, the historical photovoltaic data are preprocessed, and the correlation analysis of meteorological factors is carried out by PCC, and the high correlation factors are extracted to obtain the multivariate time series feature matrix of meteorological factors. Then, the historical photovoltaic power data are decomposed into multiple intrinsic modes and a residual component at one time by CEEMDAN. The high-frequency components are clustered by K-means combined with sample entropy, and the high-frequency components are decomposed and refined by VMD to form a multi-scale characteristic mode matrix. Finally, the obtained features are input into the CNN–BiLSTM model for the final photovoltaic power prediction results. After experimental verification, compared with the traditional single-mode decomposition algorithm (such as CEEMDAN–BiLSTM, VMD–BiLSTM), the combined prediction method proposed reduces MAE by more than 0.016 and RMSE by more than 0.017, which shows excellent accuracy and stability. Full article

Rare events are occurrences that take place with a significantly lower frequency than more common, regular events. These events can be categorized into distinct categories, from frequently rare to extremely rare, based on factors like the distribution of data and significant differences in rarity levels. In manufacturing domains, predicting such events is particularly important, as they lead to unplanned downtime, a shortening of equipment lifespans, and high energy consumption. Usually, the rarity of events is inversely correlated with the maturity of a manufacturing industry. Typically, the rarity of events affects the multivariate data generated within a manufacturing process to be highly imbalanced, which leads to bias in predictive models. This paper evaluates the role of data enrichment techniques combined with supervised machine learning techniques for rare event detection and prediction. We use time series data augmentation and sampling to address the data scarcity, maintaining its patterns, and imputation techniques to handle null values. Evaluating 15 learning models, we find that data enrichment improves the F1 measure by up to 48% in rare event detection and prediction. Our empirical and ablation experiments provide novel insights, and we also investigate model interpretability. Full article

There are many time series forecasting methods, but there are few research methods for long-term multivariate time series forecasting, which are mainly dominated by a series of forecasting models developed on the basis of a transformer. The aim of this study is to perform forecasting for multivariate time series data and to improve the forecasting accuracy of the model. In the recent past, it has appeared that the prediction effect of linear models surpasses that of the family of self-attention mechanism models, which encourages us to look for new methods to solve the problem of long-term multivariate time series forecasting. In order to overcome the problem that the temporal order of information is easily broken in the self-attention family and that it is difficult to capture information on long-distance data using recurrent neural network models, we propose a matrix attention mechanism, which is able to weight each previous data point equally without breaking the temporal order of the data, so that the overall data information can be fully utilized. We used the matrix attention mechanism as the basic module to construct the frequency domain block and time domain block. Since complex and variable seasonal component features are difficult to capture in the time domain, mapping them to the frequency domain reduces the complexity of the seasonal components themselves and facilitates data feature extraction. Therefore, we use the frequency domain block to extract the seasonal information with high randomness and poor regularity to help the model capture the local dynamics. The time domain block is used to extract the smooth floating trend component information to help the model capture long-term change patterns. This also improves the overall prediction performance of the model. It is experimentally demonstrated that our model achieves the best prediction results on three public datasets and one private dataset. Full article

Local weather forecasts in the Arctic outside of settlements are challenging due to the dearth of ground-level observation stations and high computational costs. During winter, these forecasts are critical to help prepare for potentially hazardous weather conditions, while in spring, these forecasts may be used to determine flood risk during annual snow melt. To this end, a hybrid VMD-WT-InceptionTime model is proposed for multi-horizon multivariate forecasting of remote-region temperatures in Alaska over short-term horizons (the next seven days). First, the Spearman correlation coefficient is employed to analyze the relationship between each input variable and the forecast target temperature. The most output-correlated input sequences are decomposed using variational mode decomposition (VMD) and, ultimately, wavelet transform (WT) to extract time-frequency patterns intrinsic in the raw inputs. The resulting sequences are fed into a deep InceptionTime model for short-term forecasting. This hybrid technique has been developed and evaluated using 35+ years of data from three locations in Alaska. Different experiments and performance benchmarks are conducted using deep learning models (e.g., Time Series Transformers, LSTM, MiniRocket), and statistical and conventional machine learning baselines (e.g., GBDT, SVR, ARIMA). All forecasting performances are assessed using four metrics: the root mean squared error, the mean absolute percentage error, the coefficient of determination, and the mean directional accuracy. Superior forecasting performance is achieved consistently using the proposed hybrid technique. Full article

Background: Optimal sports performance requires a balance between intensive training and adequate rest. IMUs provide objective, quantifiable data to analyze performance dynamics, despite the challenges in quantifying athlete training loads. The ability of AI to analyze complex datasets brings innovation to the monitoring and optimization of athlete training cycles. Traditional techniques rely on subjective assessments to prevent overtraining, which can lead to injury and underperformance. IMUs provide objective, quantitative data on athletes’ physical status during action. AI and machine learning can turn these data into useful insights, enabling data-driven athlete performance management. With IMU-generated multivariate time series data, this paper uses AI to construct a robust model for predicting fatigue and stamina. Materials and Methods: IMUs linked to 19 athletes recorded triaxial acceleration, angular velocity, and magnetic orientation throughout repeated sessions. Standardized training included steady-pace runs and fatigue-inducing techniques. The raw time series data were used to train a supervised ML model based on frequency and time-domain characteristics. The performances of Random Forest, Gradient Boosting Machines, and LSTM networks were compared. A feedback loop adjusted the model in real time based on prediction error and bias estimation. Results: The AI model demonstrated high predictive accuracy for fatigue, showing significant correlations between predicted fatigue levels and observed declines in performance. Stamina predictions enabled individualized training adjustments that were in sync with athletes’ physiological thresholds. Bias correction mechanisms proved effective in minimizing systematic prediction errors. Moreover, real-time adaptations of the model led to enhanced training periodization strategies, reducing the risk of overtraining and improving overall athletic performance. Conclusions: In sports performance analytics, the AI-assisted model using IMU multivariate time series data is effective. Training can be tailored and constantly altered because the model accurately predicts fatigue and stamina. AI models can effectively forecast the beginning of weariness before any physical symptoms appear. This allows for timely interventions to prevent overtraining and potential accidents. The model shows an exceptional ability to customize training programs according to the physiological reactions of each athlete and enhance the overall training effectiveness. In addition, the study demonstrated the model’s efficacy in real-time monitoring performance, improving the decision-making abilities of both coaches and athletes. The approach enables ongoing and thorough data analysis, supporting strategic planning for training and competition, resulting in optimized performance outcomes. These findings highlight the revolutionary capability of AI in sports science, offering a future where data-driven methods greatly enhance athlete training and performance management. Full article

The stock index is actively used for the realization of profits using derivatives and via the hedging of assets; hence, the prediction of the index is important for market participants. As market uncertainty has increased during the COVID-19 pandemic and with the rapid development of data engineering, a situation has arisen wherein extensive amounts of information must be processed at finer time intervals. Addressing the prevalent issues of difficulty in handling multivariate high-frequency time-series data owing to multicollinearity, resource problems in computing hardware, and the gradient vanishing problem due to the layer stacking in recurrent neural network (RNN) series, a novel algorithm is developed in this study. For financial market index prediction with these highly complex data, the algorithm combines ResNet and a variable-wise attention mechanism. To verify the superior performance of the proposed model, RNN, long short-term memory, and ResNet18 models were designed and compared with and without the attention mechanism. As per the results, the proposed model demonstrated a suitable synergistic effect with the time-series data and excellent classification performance, in addition to overcoming the data structure constraints that the other models exhibit. Having successfully presented multivariate high-frequency time-series data analysis, this study enables effective investment decision making based on the market signals. Full article

Multi-variable time series (MTS) information is a typical type of data inference in the real world. Every instance of MTS is produced via a hybrid dynamical scheme, the dynamics of which are often unknown. The hybrid species of this dynamical service are the outcome of high-frequency and low-frequency external impacts, as well as global and local spatial impacts. These influences impact MTS’s future growth; hence, they must be incorporated into time series forecasts. Two types of normalization modules, temporal and spatial normalization, are recommended to accomplish this. Each boosts the original data’s local and high-frequency processes distinctly. In addition, all components are easily incorporated into well-known deep learning techniques, such as Wavenet and Transformer. However, existing methodologies have inherent limitations when it comes to isolating the variables produced by each sort of influence from the real data. Consequently, the study encompasses conventional neural networks, such as the multi-layer perceptron (MLP), complex deep learning methods such as LSTM, two recurrent neural networks, support vector machines (SVM), and their application for regression, XGBoost, and others. Extensive experimental work on three datasets shows that the effectiveness of canonical frameworks could be greatly improved by adding more normalization components to how the MTS is used. This would make it as effective as the best MTS designs are currently available. Recurrent models, such as LSTM and RNN, attempt to recognize the temporal variability in the data; however, as a result, their effectiveness might soon decline. Last but not least, it is claimed that training a temporal framework that utilizes recurrence-based methods such as RNN and LSTM approaches is challenging and expensive, while the MLP network structure outperformed other models in terms of time series predictive performance. Full article

The European Space Agency’s (ESA) “SAR for REDD” project aims to support complementing optical remote sensing capacities in Africa with synthetic aperture radar (SAR) for Reducing Emissions from Deforestation and Forest Degradation (REDD). The aim of this study is to assess and compare Sentinel-1 C-band, ALOS-2 PALSAR-2 L-band and combined C/L-band SAR-based land cover mapping over a large tropical area in the Democratic Republic of Congo (DRC). The overall approach is to benefit from multi-temporal observations acquired from 2015 to 2017 to extract statistical parameters and seasonality of backscatters to improve forest land cover (FLC) classification. We investigate whether and to what extent the denser time series of C- band SAR can compensate for the L-band’s deeper vegetation penetration depth and known better FLC mapping performance. The supervised classification differentiates into forest, inundated forest, woody savannah, dry and wet grassland, and river swamps. Several feature combinations of statistical parameters from both, single and multi-frequency observations in a multivariate maximum-likelihood classification are compared. The FLC maps are reclassified into forest, savannah, and grassland (FSG) and validated with a systematic sampling grid of manual interpretations of very-high-resolution optical satellite data. Using the temporal variability of the dual-polarized backscatters, in the form of either wet/dry seasonal averages or using the statistical variance, in addition to the average backscatter, increased the classification accuracies by 4–5 percent points and 1–2 percent points for C- and L-band, respectively. For the FSG validation overall accuracies of 84.4%, 89.1%, and 90.0% were achieved for single frequency C- and L-band, and C/L-band combined, respectively. The resulting forest/non-forest (FNF) maps with accuracies of 90.3%, 92.2%, and 93.3%, respectively, are then compared to the Landsat-based Global Forest Change program’s and JAXA’s ALOS-1/2 based global FNF maps. Full article