Non-stationary Data Research Articles

Hydrological Data Projection Using Empirical Mode Decomposition: Applications in a Changing Climate

This paper demonstrates the effectiveness and superiority of Empirical Mode Decomposition (EMD) in projecting non-stationary hydrological data. The study focuses on daily Sea Surface Temperature (SST) sequences in the Niño 3.4 region and uses EMD to forecast the probability of El Niño events. Applying the Mann–Kendall test at the 5% significance level reveals a significant increasing trend in SST changes in this region, particularly noticeable after 1980. This trend is associated with the occurrence of El Niño and La Niña events, which have a recurrence interval of approximately 8.4 years and persist for over a year. The modified Oceanic Niño Index (ONI) proposed in this study demonstrates high forecast accuracy, with 97.56% accuracy for El Niño and 89.80% for La Niña events. Additionally, the EMD of SST data results in 13 Intrinsic Mode Functions (IMFs) and a residual component. The oscillation period increases with each IMF level, with IMF7 exhibiting the largest amplitude, fluctuating between ±1 °C. The residual component shows a significant upward trend, with an average annual increase of 0.0107 °C. These findings reveal that the EMD-based data generation method overcomes the limitations of traditional hydrological models in managing non-stationary sequences, representing a notable advancement in data-driven hydrological time series modeling. Practically, the EMD-based 5-year moving process can generate daily sea temperature sequences for the coming year in this region, offering valuable insights for assessing El Niño probabilities and facilitating annual updates.

On metafeatures’ ability of implicit concept identification

Concept drift in data stream processing remains an intriguing challenge and states a popular research topic. Methods that actively process data streams usually employ drift detectors, whose performance is often based on monitoring the variability of different stream properties. This publication provides an overview and analysis of metafeatures variability describing data streams with concept drifts. Five experiments conducted on synthetic, semi-synthetic, and real-world data streams examine the ability of over 160 metafeatures from 9 categories to recognize concepts in non-stationary data streams. The work reveals the distinctions in the considered sources of streams and specifies 17 metafeatures with a high ability of concept identification.

Driving low-carbon mechanisms through smart investments in renewable resources and green finance initiatives among G20 nations

In light of the escalating concerns regarding climate change and environmental decline, major nations are actively exploring strategies to mitigate environmental harm and achieve future sustainability. The surge in economic expansion in developed economies is linked to an increase in CO2 emissions. Consequently, in their pursuit of carbon-neutral policies, these countries are increasingly turning towards renewable energy as a means to enhance resource conservation and efficiency. This research investigates the varied impacts of renewable energy investment, green finance, geopolitical risk, GDP growth, foreign direct investment, and gross fixed capital formation on the carbon emissions of G20 countries. The study uses the CUP-FM (Continuously Updated Fully Modified) and CUP-BC (Continuously Updated Bias-Corrected) estimators, which are sophisticated econometric approaches designed to handle non-stationary panel data and cross-sectional dependency, to produce robust long-term parameter estimates. The CUP-FM estimator adjusts for potential endogeneity and serial correlation, improving the accuracy of long-run relationships in panel data. The CUP-BC estimator provides bias-corrected estimates to further enhance the precision of these long-run connections.The long-term parameter estimates reveal a negative correlation between renewable energy investment, green finance, and carbon emissions. In contrast, foreign direct investment, gross fixed capital formation, GDP growth, and geopolitical risk are positively associated with CO2 emissions. This suggests that financial stability often leads to investments in carbon-heavy economic ventures, thereby implicating economic growth as a contributing factor to environmental degradation in G20 countries.

Machine Learning-Based Univariate Time Series Imputation Method for Estimating Missing Values in Non-Stationary Data

Handling missing values in time series data is crucial because they can disrupt data analysis and interpretation. Sequentially missing values in time series often pose a more complex challenge compared to randomly missing values. One of the promising recent methods is Machine Learning-Based Univariate Time Series Imputation (MLBUI), although it is still not widely used and its accessibility is limited. MLBUI employs Random Forest Regression (RFR) and Support Vector Regression (SVR) algorithms. This study evaluates the performance of MLBUI in addressing missing data scenarios in non-stationary univariate time series data. The data used in this research is the average temperature data from Bogor Regency. The missing data scenarios considered include rates of 6%, 10%, and 14%. Besides MLBUI, five other comparison methods are used: Kalman StructTS, Kalman Auto-ARIMA, Spline Interpolation, Stine Interpolation, and Moving Average. The results show that MLBUI performs poorly for non-stationary data, although the obtained Mean Absolute Percentage Error (MAPE) is below 10%.

Research on the Application of CEEMD and CEEMDAN in Seismic Random Noise Suppression

Empirical Mode Decomposition (EMD) is an adaptive method for processing nonlinear and non-stationary data. Complementary Ensemble Empirical Mode Decomposition (CEEMD) and Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) are noise-assisted signal processing methods developed from EMD. During seismic data acquisition, various types of noise, including random noise, are inevitably captured, and current methods for dealing with such noise are limited. In this study, these two signal processing methods were applied in simulated experiments and real seismic data to suppress random noise. By analyzing the processing effects and runtime, it was found that CEEMDAN offers better processing performance and speed, making it a valuable tool for practical applications.

The condition monitoring scheme for industrial IoT scenario: A distributed modeling for high-dimensional nonstationary data

Based on large-scale data collection and high-speed transmission, Industrial Internet of Things (IIoT) promotes the rapid development of intelligent manufacturing. IIoT systems are usually disturbed by complex external factors, which lead to high-dimensional nonstationary operating data. Besides, unexpected data transmission interruptions, sensor failures, and network delays lead to data loss. This paper proposes a distribution & communication strategy for monitoring high-dimensional nonstationary processes with missing values in IIoT scenarios. First, a deep learning-based imputation network is proposed to impute the missing values. Then a decomposition strategy based on degree of cointegration is proposed, which decomposes a high-dimensional nonstationary process into multiple blocks. And a communication strategy is proposed to mine the internal relationship between different blocks. Finally, faulty information is detected by a distributed framework. Two real cases from IIoT are applied to illustrate the monitoring performance of the proposed method. The results show that the proposed method outperforms existing benchmarks in data imputation and monitoring performance.

Prediction of oil production based on multivariate analysis and self-attention mechanism integrated with long short-term memory

The accurate oil prediction of wells is essential for making informed decisions regarding the extension of the well lifespan and the enhancement of oil recovery rates. However, the prediction of oil well production is highly challenging due to the complex, nonlinear, and non-stationary data influenced by reservoir geological characteristics and operational adjustments. To address this, a novel prediction method for oil well production is proposed in this study. Firstly, the data-cleaning approach designed in this study is utilized to eliminate outliers from the raw dataset and impute missing values. Subsequently, relevant features are identified through analysis to form a usable dataset. The proposed deep learning neural network, namely the self-attention mechanism integrated with long short-term memory, is trained and learned on this dataset. Finally, the predictive performance of the model is validated using a set of actual oil production data. Through data experiments, the proposed model effectively predicts oil well production with superior accuracy compared to baseline models, achieving an R-squared of 0.872. This method provides reliable decision support for optimizing oil field development and management.

Fuzzy lattices assisted EJAYA Q-learning for automated pulmonary diseases classification

This work proposes a novel technique called Enhanced JAYA (EJAYA) assisted Q-Learning for the classification of pulmonary diseases, such as pneumonia and tuberculosis (TB) sub-classes using chest x-ray images. The work introduces Fuzzy lattices formation to handle real time (non-linear and non-stationary) data based feature extraction using Schrödinger equation. Features based adaptive classification is made possible through the Q-learning algorithm wherein optimal Q-values selection is done via EJAYA optimization algorithm. Fuzzy lattice is formed using x-ray image pixels and lattice Kinetic Energy (K.E.) is calculated using the Schrödinger equation. Feature vector lattices having highest K.E. have been used as an input features for the classifier. The classifier has been employed for pneumonia classification (normal, mild and severe) and Tuberculosis detection (presence or absence). A total of 3000 images have been used for pneumonia classification yielding an accuracy, sensitivity, specificity, precision and F-scores of 97.90%, 98.43%, 97.25%, 97.78% and 98.10%, respectively. For Tuberculosis 600 samples have been used. The achived accuracy, sensitivity, specificity, precision and F-score are 95.50%, 96.39%, 94.40% 95.52% and 95.95%, respectively. Computational time are 40.96 and 39.98 s for pneumonia and TB classification. Classifier learning rate (training accuracy) for pneumonia classes (normal, mild and severe) are 97.907%, 95.375% and 96.391%, respectively and for tuberculosis (present and absent) are 96.928% and 95.905%, respectively. The results have been compared with contemporary classification techniques which shows superiority of the proposed approach in terms of accuracy and speed of classification. The technique could serve as a fast and accurate tool for automated pneumonia and tuberculosis classification.

The linear canonical W-transform

The time-frequency spectrum of seismic data is commonly used as an attribute in the analysis and interpretation of nonstationary seismic signals. To improve the capability of the W-transform to analyze nonstationary data, we introduce the linear canonical W-transform (LCWT) by implementing an affine matrix along with scaling parameters to generate high resolution and energy-concentrated time-frequency spectra. More features of nonstationary signals can be explored in the time-linear canonical frequency domain by transforming the time-frequency plane, which makes the algorithm more applicable to seismic signals. For the completeness of the LCWT, we also develop the inverse LCWT to extend its applications. Examples of synthetic and field seismic data show that a highly resolved and energy-concentrated time-frequency spectrum can be estimated with the LCWT with negligible additional computational burden. In addition, our algorithm is promising for several seismic data processing and interpretation techniques, such as reservoir characterization and thin layer identification.

Temporal online self-learning stochastic configuration networks: A study on strip deviation prediction

Stochastic configuration networks (SCNs), with the rapid learning abilities and universal approximation properties, have garnered growing interest in the industrial domain in recent years. In the hot rolling process, strip deviation affects product quality and production safety. In this paper, a novel model named temporal online self-learning stochastic configuration networks (TOSL-SCNs) designed for early prediction of strip deviation in the hot rolling process is introduced. To address the challenges of time series prediction, conventional SCNs are extended to a two-dimensional format, enabling effective processing of time series data. Given the dynamic nature of industrial environments, the model incorporates an online self-learning capability to adapt to nonstationary data. Additionally, to ensure the stability of the model’s predictive performance, a data buffer strategy is introduced to dynamically refresh the training dataset. This paper demonstrates the capabilities of TOSL-SCNs in time series modeling, online self-learning, and data buffer strategy from multiple perspectives. The case study on strip deviation prediction highlights the necessity of these capabilities in industrial applications.

Scalable interpolation of satellite altimetry data with probabilistic machine learning

We present GPSat; an open-source Python programming library for performing efficient interpolation of non-stationary satellite altimetry data, using scalable Gaussian process techniques. We use GPSat to generate complete maps of daily 50 km-gridded Arctic sea ice radar freeboard, and find that, relative to a previous interpolation scheme, GPSat offers a 504 × computational speedup, with less than 4 mm difference on the derived freeboards on average. We then demonstrate the scalability of GPSat through freeboard interpolation at 5 km resolution, and Sea-Level Anomalies (SLA) at the resolution of the altimeter footprint. Interpolated 5 km radar freeboards show strong agreement with airborne data (linear correlation of 0.66). Footprint-level SLA interpolation also shows improvements in predictive skill over linear regression. In this work, we suggest that GPSat could overcome the computational bottlenecks faced in many altimetry-based interpolation routines, and hence advance critical understanding of ocean and sea ice variability over short spatio-temporal scales.

Transfer learning-based nonstationary traffic flow prediction using AdaRNN and DCORAL

Traffic flow prediction is an integral part of an intelligent transportation system (ITS) for proactive transportation planning and management in public transit network systems. However, due to the influence of passengers’ travel modes and the difficulty of data collection, some traffic flow datasets are not only nonstationary in temporal distribution but also insufficient in data size, leading to inaccurate prediction results. Therefore, a transfer learning-based method is proposed to predict nonstationary traffic flow with inadequate data in the public transit system. In the proposed model, a novel approach named AdaRNN-DCORAL is presented to achieve accurate traffic flow prediction in the temporal dimension. Specifically, AdaRNN is an adaptive prediction network for nonstationary traffic flow data, while DCORAL performs domain adaptation between the source and target domain in the public transit system. First, two domains of traffic flow datasets in two public transit modes are established, which are the target domain with insufficient data and the source domain with sufficient data. Subsequently, the traffic flow datasets are divided into weekday and holiday time series based on their distribution characteristics. On this basis, AdaRNN-DCORAL is trained on the constructed datasets, thus achieving the precise nonstationary prediction of data scarcity in the target domain. Finally, the experiments demonstrate the improved performance of the proposed method over that of state-of-the-art models in terms of prediction accuracy and computation time.

Hebbian spatial encoder with adaptive sparse connectivity

Biologically plausible neural networks have demonstrated efficiency in learning and recognizing patterns in data. This paper proposes a general online unsupervised algorithm for spatial data encoding using fast Hebbian learning. Inspired by the Hierarchical Temporal Memory (HTM) framework, we introduce the SpatialEncoder algorithm, which learns the spatial specialization of neurons’ receptive fields through Hebbian plasticity and k-WTA (k winners take all) inhibition. A key component of our model is a two-part synaptogenesis algorithm that enables the network to maintain a sparse connection matrix while adapting to non-stationary input data distributions. In the MNIST digit classification task, our model outperforms the HTM SpatialPooler in terms of classification accuracy and encoding stability. Compared to another baseline, a two-layer artificial neural network (ANN), our model achieves competitive classification accuracy with fewer iterations required for convergence. The proposed model offers a promising direction for future research on sparse neural networks with adaptive neural connectivity.

Forecasting mooring tension of offshore platforms based on complete ensemble empirical mode decomposition with adaptive noise and deep learning network

The tension sequence data of offshore platforms mooring cables exhibit randomness, strong nonlinearity, and nonsmoothness. Previous studies have shown that single neural network models, such as long short-term memory (LSTM) and recurrent neural network (RNN) models, can effectively predict nonlinear data. Dealing with nonstationary data presents several challenges, nevertheless, like a noticeable delay in forecast outcomes and significant prediction errors. Aiming to address the aforementioned challenges, this article proposes an innovative hybrid prediction model named CEEMDAN-CLL for predicting mooring tension in semi-submersible platforms based on complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN), frequency-domain analysis, convolutional neural network–long short-term memory (CNN-LSTM), and LSTM. The data used in this study are those monitored during the operation of “DeepSea One”, which is the first deepwater semi-submersible production platform in the South China Sea, is taken as the research object and cover a wide range of orientations and sea states.

Multichannel seismic attenuation compensation and interpolation with curvelet sparse constraint of frequency-wavenumber spectrum

Abstract High-resolution exploration is hampered by seismic attenuation, caused by the viscosity and heterogeneity of underground media. Conventional single-channel attenuation compensation methods can increase the corresponding vertical resolution partially. While the lateral continuity of compensated seismic data is always ignored and the noise resistance can be improved. Therefore, multichannel attenuation compensation methods were proposed, including the algorithm with sparse curvelet coefficient constraint of the time-space (t-x) data. However, nonstationary seismic data may be affected by irregular missing traces in field cases, which severely degrades its lateral continuity and negatively affects the performance of multichannel attenuation compensation. Therefore, we concentrate on simultaneous multichannel attenuation compensation and missing trace interpolation in a unified framework. Based on the inversion framework of sparsity promotion, we propose an approach for simultaneous multichannel compensation and interpolation, utilizing sparse curvelet coefficient constraint of the recovered principal frequency-wavenumber (f-k) spectrum. The size of the principal f-k spectrum is reduced by at least half compared with that of the corresponding t-x band-limited data. It significantly reduces the computational expense of curvelet transform-based processing. Synthetic and field data experiments validate the effectiveness of the proposed method in efficiency improvement with consistent performance when compared to the multichannel method with sparse curvelet coefficient constraint of the t-x data in improving the vertical resolution and lateral continuity. Furthermore, we have discussed a potential acceleration strategy based on random sampling in the normal compensation issue.

Crop Water Status Analysis from Complex Agricultural Data Using UMAP-Based Local Biplot

To optimize growth and management, precision agriculture relies on a deep understanding of agricultural dynamics, particularly crop water status analysis. Leveraging unmanned aerial vehicles, we can efficiently acquire high-resolution spatiotemporal samples by utilizing remote sensors. However, non-linear relationships among data features, localized within specific subgroups, frequently emerge in agricultural data. Interpreting these complex patterns requires sophisticated analysis due to the presence of noise, high variability, and non-stationarity behavior in the collected samples. Here, we introduce Local Biplot, a methodological framework tailored for discerning meaningful data patterns in non-stationary contexts for precision agriculture. Local Biplot relies on the well-known uniform manifold approximation and projection method, such as UMAP, and local affine transformations to codify non-stationary and non-linear data patterns while maintaining interpretability. This lets us find important clusters for transformation and projection within a single global axis pair. Hence, our framework encompasses variable and observational contributions within individual clusters. At the same time, we provide a relevance analysis strategy to help explain why those clusters exist, facilitating the understanding of data dynamics while favoring interpretability. We demonstrated our method’s capabilities through experiments on both synthetic and real-world datasets, covering scenarios involving grass and rice crops. Moreover, we use random forest and linear regression models to predict water status variables from our Local Biplot-based feature ranking and clusters. Our findings revealed enhanced clustering and prediction capability while emphasizing the importance of input features in precision agriculture. As a result, Local Biplot is a useful tool to visualize, analyze, and compare the intricate underlying patterns and internal structures of complex agricultural datasets.

Learning Network-Structured Dependence From Non-Stationary Multivariate Point Process Data

Learning Network-Structured Dependence From Non-Stationary Multivariate Point Process Data

Short-term load forecasting of mixed industrial users in distribution networks using DARTS-LSTM method

Abstract Load forecasting has perennially stood as a fundamental pillar within the power industry, with precision in prognostication constituting a paramount necessity for critical functions including grid management, optimization of energy storage, and response to demand fluctuations. In the present study, we introduce a comprehensive global modeling prediction framework aimed at forecasting large-scale power demand within time series data, alongside the development of a novel short-term power load forecasting model. By integrating all time series data pertaining to the same regression task and employing a unified univariate prediction function across the entire time series dataset, we efficaciously forecast large-scale heterogeneous loads aggregated at various levels within distribution networks. The proposed framework leverages the cutting-edge DARTS (Differentiable Architecture Search) deep learning framework for implementation. Through comprehensive experimentation on real-world data, our model framework demonstrates superior performance in terms of overall prediction accuracy, scalability, and robustness in handling missing and non-stationary time series data.

A Flexible Forecasting Stack

Forecasting extrapolates the values of a time series into the future, and is crucial to optimize core operations for many businesses and organizations. Building machine learning (ML)-based forecasting applications presents a challenge though, due to non-stationary data and large numbers of time series. As there is no single dominating approach to forecasting, forecasting systems have to support a wide variety of approaches, ranging from deep learning-based methods to classical methods built on probabilistic modelling. We revisit our earlier work on a monolithic platform for forecasting from VLDB 2017, and describe how we evolved it into a modern forecasting stack consisting of several layers that support a wide range of forecasting needs and automate common tasks like model selection. This stack leverages our open source forecasting libraries GluonTS and AutoGluon-TimeSeries , the scalable ML platform SageMaker , and forms the basis of the no-code forecasting solutions ( SageMaker Canvas and Amazon Forecast ), available in the Amazon Web Services cloud. We give insights into the predictive performance of our stack and discuss learnings from using it to provision resources for the cloud database services DynamoDB, Redshift and Athena.

IMPLEMENTATION OF VECTOR AUTOREGRESSIVE (VAR) AND VECTOR ERROR CORRECTION MODEL (VECM) METHOD IN PNEUMONIA PATIENTS WITH WEATHER ELEMENTS IN PANGKALPINANG CITY

The news about Covid-19 is no longer as scary as in previous years. As COVID-19 cases decrease, health protocols are becoming more relaxed, making it easier for the virus to spread. Besides COVID-19, ARI is one of Indonesia's leading causes of death for children under five. Around 20-40% of hospital admissions are children due to ARI, with around 1.6 million deaths due to pneumonia alone in children under five per year. Currently, ARI dominates the diseases most suffered by the people of Bangka Belitung. Based on this, using the VAR and VECM method to analyze pneumonia sufferers in toddlers regarding weather elements in the Pangkalpinang City. The VAR model has a simpler structure with a minimal number of variables where all the variables are endogenous, with the independent variable being the lag. Meanwhile, VECM can be used to model cointegrated and non-stationary time series data. The data used in this research is the number of monthly cases of toddlers suffering from pneumonia and data on climate conditions, namely rainfall, air temperature, air humidity and duration of sunlight during 2019-2021 in Pangkalpinang City. The results of the Granger Causality test show that the pneumonia variables regarding rainfall, temperature, duration of sunlight and humidity only have a one-way causality pattern. The VAR estimation results show that weather elements (rainfall, temperature and duration of sunlight) do not significantly affect pneumonia in the short term. Meanwhile, the VECM estimation results show that in the long-term pattern, humidity variables affect pneumonia. For this reason, it is recommended that the relevant agencies carry out outreach to the public, especially to pneumonia sufferers, to avoid damp weather. Because the lower the humidity value, the greater the potential for pneumonia in Pangkalpinang City, Bangka Belitung Islands Province.