Nonlinear Dependence Research Articles

Broad‐Temperature Optical Thermometry via Dual Sensitivity of Self‐Trapped Excitons Lifetime and Higher‐Order Phonon Anharmonicity in Lead‐Free Perovskites

AbstractBroad‐temperature optical thermometry necessitates materials with exceptional sensitivity and stability across varied thermal conditions, presenting challenges for conventional systems. Here, we report a lead‐free, vacancy‐ordered perovskite Cs2TeCl6, that achieves precise temperature sensing through a novel combination of self‐trapped excitons (STEs) photoluminescence (PL) lifetime modulation and unprecedented fifth‐order phonon anharmonicity. The STEs PL lifetime demonstrates a highly temperature‐sensitive response from 200 to 300 K, ideal for low‐to‐intermediate thermal sensing. In contrast, the Eg phonon mode undergoes significant linewidth broadening due to five‐phonon scattering processes, with a distinct nonlinear temperature dependence up to 500 K. This fifth‐order anharmonic effect enhances Raman‐based temperature sensitivity, yielding a specific sensitivity (Sr) of 0.577 % K–1 at 330 K and remaining above 0.5 % K–1 at elevated temperatures. This study presents the first evidence of fifth‐order anharmonic effects enhancing Raman‐based temperature sensitivity, establishing Cs2TeCl6 as a versatile candidate for broad‐temperature optical thermometry and opening new avenues for precise non‐contact temperature sensing in advanced technological applications.

Mechanical Properties and Fracture Mechanisms of Nanocomposites of Metal and Graphene with Overlapping Edges.

The fracture mechanism and mechanical response of Ni/graphene nanocomposites under nanoindentation are investigated by molecular dynamics simulations. We analyze the effect of the overlapping area on the microstructural transition, HCP atomic fraction, dislocation density, load-displacement relationship, and stress distribution. It was found that the maximum indentation depth of embedded graphene has a nonlinear dependence relation with the overlapping area, and it becomes smaller at the overlapping width comparable to the indenter diameter. The graphene layer is able to hinder the expansion of the dislocation into the interior of the Ni matrix in the initial stage. The densities of HCP atoms and dislocations in the composite gradually increase with increasing indentation depth. The stress concentration tends to cause nucleation of dislocations below the indentation surface. When the graphene is ruptured, the elastic recovery to some extent occurs in the deformed substrate. This work sheds light on modifying the mechanical properties of metal/graphene composites by tuning the overlapping boundary of graphene.

The End of Mean-Variance? Tsallis Entropy Revolutionises Portfolio Optimisation in Cryptocurrencies

Has the mean-variance framework become obsolete? In this paper, we replace traditional variance–covariance methods of portfolio optimisation with relative Tsallis entropy and mutual information measures. Its goal is to enhance risk management and diversification in complicated finance ecosystems. We utilize the S&P 500 and Bitwise 10 cryptocurrency indices’ daily returns (2019–2024 data) and conduct our analysis to the year 2020 under extreme shocks. Many models were trained with different configurations, like mean-variance (MV), mean-entropy (ME), and mean-mutual information (MI) traders and their corresponding variants, using Sharpe’s ratio, Jensen’s alpha, and entropy value of risk (EVAR). The findings indicate that entropic models outperform conventional models in terms of diversification and, especially, extreme risk management. Because the appropriate normalization conditions often fail to be satisfied, we can informally see that after a recalibration of the effective frontier, we obtain from EVAR an accumulated resilience aspect to these rare events while also observing the great potential of entropy-based models to replicate non-linear dependencies between assets. The results show that models combining entropy and mutual information optimise the gain–loss ratio (GLR), providing stable diversification and improved risk management, while maximising returns in complex and volatile market environments.

New strategy for predicting liquid–liquid equilibrium near critical point using global renormalization group theory

AbstractClassical liquid activity coefficient models, such as the nonrandom two‐liquid (NRTL) model, fail near the critical point of the liquid–liquid equilibrium (LLE), unless a highly nonlinear temperature dependency is introduced for the molecular interaction parameters. In this work, we propose an approach to predict the LLE data near the critical point using data away from the critical region based on the global renormalization group theory (GRGT). Specifically, we propose a non‐empirical approach to determine the GRGT parameters, which does not rely on experimental data. The performance of our method is examined using the NRTL model on 21 binary mixtures. Our results show that the predictive approach proposed in this work reduces the error in the critical solution temperatures by about 48% when compared to the classical NRTL model with linear temperature‐dependent interaction parameters.

PilotCareTrans Net: an EEG data-driven transformer for pilot health monitoring

IntroductionIn high-stakes environments such as aviation, monitoring cognitive, and mental health is crucial, with electroencephalogram (EEG) data emerging as a keytool for this purpose. However traditional methods like linear models Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) architectures often struggle to capture the complex, non-linear temporal dependencies in EEG signals. These approaches typically fail to integrate multi-scale features effectively, resulting in suboptimal health intervention decisions, especially in dynamic, high-pressure environments like pilot training.MethodsTo overcome these challenges, this study introduces PilotCareTrans Net, a novel Transformer-based model designed for health intervention decision-making in aviation students. The model incorporates dynamic attention mechanisms, temporal convolutional layers, and multi-scale feature integration, enabling it to capture intricate temporal dynamics in EEG data more effectively. PilotCareTrans Net was evaluated on multiple public EEG datasets, including MODA, STEW, SJTUEmotion EEG, and Sleep-EDF, where it outperformed state-of-the-art models in key metrics.Results and discussionThe experimental results demonstrate the model's ability to not only enhance prediction accuracy but also reduce computational complexity, making it suitable for real-time applications in resource-constrained settings. These findings indicate that PilotCareTrans Net holds significant potential for improving cognitive health monitoring and intervention strategies in aviation, thereby contributing to enhanced safety and performance in critical environments.

Advanced Hybrid Models for Air Pollution Forecasting: Combining SARIMA and BiLSTM Architectures

This study explores a hybrid forecasting framework for air pollutant concentrations (PM10, PM2.5, and NO2) that integrates Seasonal Autoregressive Integrated Moving Average (SARIMA) models with Bidirectional Long Short-Term Memory (BiLSTM) networks. By leveraging SARIMA’s strength in linear and seasonal trend modeling and addressing nonlinear dependencies using BiLSTM, the framework incorporates Box-Cox transformations and Fourier terms to enhance variance stabilization and seasonal representation. Additionally, attention mechanisms are employed to prioritize temporal features, refining forecast accuracy. Using five years of daily pollutant data from Romania’s National Air Quality Monitoring Network, the models were rigorously evaluated across short-term (1-day), medium-term (7-day), and long-term (30-day) horizons. Metrics such as RMSE, MAE, and MAPE revealed the hybrid models’ superior performance in capturing complex pollutant dynamics, particularly for PM2.5 and PM10. The SARIMA combined with BiLSTM, Fourier, and Attention configuration demonstrated consistent improvements in predictive accuracy and interpretability, with attention mechanisms proving effective for extreme values and long-term dependencies. This study highlights the benefits of combining statistical preprocessing with advanced neural architectures, offering a robust and scalable solution for air quality forecasting. The findings provide valuable insights for environmental policymakers and urban planners, emphasizing the potential of hybrid models for improving air quality management and decision-making in dynamic urban environments.

Understanding Factors Influencing Generative AI Use Intention: A Bayesian Network-Based Probabilistic Structural Equation Model Approach

This study investigates the factors influencing users’ intention to use generative AI by employing a Bayesian network-based probabilistic structural equation model approach. Recognizing the limitations of traditional models like the technology acceptance model and the unified theory of acceptance and use of technology, this research incorporates novel constructs such as perceived anthropomorphism and animacy to capture the unique human-like qualities of generative AI. Data were collected from 803 participants with prior experience of using generative AI applications. The analysis reveals that social influence (standardized total effect = 0.550) is the most significant predictor of use intention, followed by effort expectancy (0.480) and perceived usefulness (0.454). Perceived anthropomorphism (0.149) and animacy (0.145) also influence use intention, but with a lower relative impact. By utilizing a probabilistic structural equation model, this study overcomes the linear limitations of traditional acceptance models, allowing for the exploration of nonlinear relationships and conditional dependencies. These findings provide actionable insights for improving generative AI design, user engagement, and adoption strategies.

Nonlinear field dependence of Hall effect and high-mobility multi-carrier transport in an altermagnet CrSb

As a promising candidate for altermagnet, CrSb possesses a distinctive compensated spin split band structure that could lead to groundbreaking concepts in the field of spintronics. In this work, we have grown high-quality CrSb single crystals and comprehensively investigated their electronic and magneto-transport properties. We have observed large, positive, and non-saturated magnetoresistance (MR) in CrSb, which obeys Kohler's rule, indicating its classic Lorentz scattering origins. The largest MR can reach 27% at 10 K and 9 T. Remarkably, a nonlinear magnetic field dependence of Hall effect resembling the spontaneous anomalous Hall is identified over a wide temperature range. It was found that the nonlinearity mainly stems from the incorporation of different carriers in the magnetoconductivity. According to the Fermi surface analyses of CrSb, we applied the three-carrier model to fit the conductivity data, yielding good agreement. The extracted carrier concentration ranges from 1018 to 1021 cm−3, with a high mobility up to ∼3 × 103 cm2/V s below 50 K. These indicate that CrSb behaves more like a semimetal. Furthermore, calculations using the semiclassical Boltzmann transport theory have reproduced the main features of the experimental MR and Hall effect in CrSb. These exceptional transport properties make CrSb unique for applications in spintronics as an altermagnet.

Risk Spillover Effect of Crude Oil Prices Against Stock Price Indices: Case Study of Oil-Producing and Consumer Countries for The Period of 2000-2022

This research investigates the risk spillover effects of crude oil prices on stock price indices in both oil-producing and consumer countries during the period 2000–2022. The research employs the GARCH Copula Quantile Regression (CQR) model, a method capable of capturing nonlinear dependencies and tail risks between crude oil prices and stock indices across different risk levels. The data includes daily crude oil prices (WTI) and stock indices of oil-producing countries (S&P500, TASI, MOEX, TSX) and consumer countries (SSE, BSE SENSEX, Nikkei 225). The findings reveal significant downside and upside risk spillovers from crude oil prices to stock indices, with oil-producing countries experiencing higher spillover risks compared to consumer countries. Notably, the TASI index exhibits the greatest risk spillover among producing countries, while the SSE index is most affected among consumer countries. These results highlight the asymmetrical nature of risk spillovers, with downside risks posing greater challenges than upside risks. This research contributes to the existing literature by providing a more detailed analysis of risk dependencies using the GARCH CQR model and offering practical insights for investors and policymakers to better manage risks in volatile markets. The study underscores the importance of understanding tail dependencies in financial markets affected by crude oil price fluctuations.

Comparison of plasma start-up with high Z and low Z first wall in WEST

Abstract The choice of first wall material is of paramount importance for the plasma start-up conditions in ITER and future fusion power plants. In this context, the present work focuses on the correlations between first wall impurity sources and total radiated power during plasma start-up in the tungsten (W) Environment in Steady-state Tokamak (WEST). The objective is to highlight experimental indications for a preferable combination of start-up plasma scenario and first wall materials. Until 2019, WEST featured a full high Z first wall with all limiters exposing only W surfaces to the plasma. To study the impact of a low Z first wall in WEST, boron nitride (BN) tiles were installed in the central part of the inner and outer limiters in 2020. Although visible spectroscopy and bolometry measurements show respectively a strong weakening of the WI line intensity on the limiters and a reduction of radiated power after the changeover, a degradation occurs with the accumulation of plasma exposure. In addition, the different plasma facing elements of the main chamber do not influence equally the radiated power during start-up. In both high Z and low Z environments, a clear non-linear dependence is found between the start-up radiated power and the outer limiter W impurity source. Since W seems to be the main cause for core radiation, correlation between outer limiter W sources and other impurity sources are investigated. Finally, analysis of the legacy of B powder drops on a number of start-up plasmas suggests that it is less effective at reducing radiated power when the first wall is covered with W.

Ternary PEO/PVDF-HFP-Based Polymer Electrolytes for Li-Ion Batteries

The impetus to study and develop polymer electrolytes for metal-ion batteries is due to their enhanced safety compared to flammable organic liquid electrolytes, promising ionic conductivity, and broad electrochemical stability window, making them to viable candidates for battery application. In the current work, we present a simple fabrication procedure and a comprehensive physico–chemical study of various PVDF-HFP-based electrolyte formulations with a sufficient addition of PEO polymer, LiTFSI conducting salt, and EMIMTFSI ionic liquid. The ionic conductivity, activation energy for ionic movement and thickness of the resulting polymer electrolyte show a non-linear dependency on the PVDF-HFP/PEO ratio. The electrolyte composition with a 0.35PEO-0.65PVDF-HFP/1LiTFSI/1EMIMTFSI mass fraction exhibits the highest ionic conductivity among the compositions, revealing 7.7×10−5 S cm−1 at 30 °C. Electrochemical tests in half full and full Li-ion batteries with a LiFePO4 cathode and Li4Ti5O12 anode also emphasized this composition as the most promising one, providing an initial capacity in full cells of 120 mAh g−1 and a capacity retention of about 75% after 50 charge/discharge cycles at a 0.1 C current rate. In the PEO/PVDF-HFP polymer blend with EMIMTFSI as a plasticizer, the amount of crystalline parts, which are detrimental to a fast ionic diffusion, is significantly reduced.

FRACTAL ANALYSIS IN ALTERNATIVE INVESTMENTS: UNVEILING DYNAMICS AND DIVERSIFICATION BENEFITS

The realm of alternative investments, comprising commodities, real estate, private equity, and other non-traditional asset classes, presents a rich landscape for investors seeking diversification and enhanced returns. Amidst the complexities of these markets, fractal analysis emerges as a powerful tool for unraveling the intricate dynamics and exploring diversification benefits. This paper delves into the application of fractal theory in alternative investments, assessing its efficacy in analyzing asset price movements, correlations, and portfolio diversification. Through a comprehensive review of the literature and empirical evidence, we elucidate how fractal analysis techniques can deepen our understanding of alternative investment markets and inform strategic decision-making. Key findings highlight the ability of fractal analysis to capture nonlinear dependencies, self-similarity, and multi-scale dynamics across various asset classes. The study demonstrates how these insights enhance portfolio resilience, optimize asset allocation, and improve risk-adjusted returns. Additionally, this research contributes novel perspectives on integrating fractal methods with portfolio management and risk diversification strategies.

Broad-Temperature Optical Thermometry via Dual Sensitivity of Self-Trapped Excitons Lifetime and Higher-Order Phonon Anharmonicity in Lead-Free Perovskites.

Broad-temperature optical thermometry necessitates materials with exceptional sensitivity and stability across varied thermal conditions, presenting challenges for conventional systems. Here, we report a lead-free, vacancy-ordered perovskite Cs2TeCl6, that achieves precise temperature sensing through a novel combination of self-trapped excitons (STEs) photoluminescence (PL) lifetime modulation and unprecedented fifth-order phonon anharmonicity. The STEs PL lifetime demonstrates a highly temperature-sensitive response from 200 to 300 K, ideal for low-to-intermediate thermal sensing. In contrast, the Eg phonon mode undergoes significant linewidth broadening due to five-phonon scattering processes, with a distinct nonlinear temperature dependence up to 500 K. This fifth-order anharmonic effect enhances Raman-based temperature sensitivity, yielding a specific sensitivity (Sr) of 0.577% K-1 at 330 K and remaining above 0.5% K-1 at elevated temperatures. This study presents the first evidence of fifth-order anharmonic effects enhancing Raman-based temperature sensitivity, establishing Cs2TeCl6 as a versatile candidate for broad-temperature optical thermometry and opening new avenues for precise non-contact temperature sensing in advanced technological applications.

PyDRex: Predicting crystallographic preferred orientation in peridotites under steady-state and time-dependent strain

Abstract Crystallographic preferred orientation (CPO) of peridotite minerals is frequently invoked to explain the widespread dependence of seismic wave speed on propagation direction in Earth’s mantle — a property known as seismic anisotropy. As established by rock mechanics experiments, CPO constitutes a direct signature of past and ongoing strain regimes experienced by rocks during mantle flow. Therefore, an improved understanding of CPO generation promises to yield valuable information on the rheology and corresponding deformation mechanisms activated through mantle dynamics. Simulating CPO in geodynamical models is computationally challenging and has often been restricted to steady-state mantle flows. However, within Earth’s vigorously convecting mantle the steady-state assumption is questionable, thus motivating the need to couple CPO simulations with time-evolving mantle flow models. Here, we present a new Python implementation of the D-Rex CPO model, called PyDRex, which predicts salient features of mineral grain size and orientation evolution whilst providing a well-documented, user-friendly interface that supports flexible coupling to geodynamical modelling frameworks. PyDRex also packages numerous post-processing routines for strain analysis and visualisation of grain orientation distributions. We provide a set of benchmark simulations based on previous D-Rex implementations that validate PyDRex and demonstrate sensitivities to model parameters for both steady-state and time-dependent flows. Analysis of benchmark results highlights the role of dynamic recrystallisation in controlling competing grain growth in both the softest and hardest crystallographic orientations. When employing a commonly used value for the grain boundary mobility parameter (M* = 125), we also find that transient CPO textures are generally not well resolved if crystals are represented by fewer than 5000 ‘grains’ (weighted orientation samples) — a configuration rarely employed in most previously published studies. Furthermore, kinematic corner-flow models suggest that CPO produced at mid-ocean ridges has a non-linear dependence on depth, which implies that even ostensibly simple mantle flows can result in complex distributions of seismic anisotropy. Our analyses motivate further experimental calibration of parameters controlling dynamic recrystallisation and potential improvements to the numerical treatment of subgrain nucleation.

Polarized topological invariance frequency conversion on high-fidelity information transfer.

A cross-band information converter, which links communication and storage modules operating at different wavelengths, is crucial for the development of optical vortex-based information networks. However, due to nonlinear efficiency dependence on orbital angular momentum (OAM), the traditional intensity or polarization distribution-based recognition distortion emerges. Here, by employing the optical vector vortex with a tailored polarization mode as the information carrier of communication network, and the polarization topology one-to-one mapping as the mode identification method, we demonstrate a simple yet flexible frequency converter based on sum-frequency generation (SFG) that preserves the topological invariance of polarization, enabling high-fidelity information transmission. In a proof-of-concept experiment, vector vortex-encoded image information is successfully transferred from a 1064 nm band to 630.9 nm without any quality degradation. This approach has the potential to enable cross-platform information transmission in optical networks, paving the way for broader applications.

Optimizing Hydrogen Production in the Co-Gasification Process: Comparison of Explainable Regression Models Using Shapley Additive Explanations.

The co-gasification of biomass and plastic waste offers a promising solution for producing hydrogen-rich syngas, addressing the rising demand for cleaner energy. However, optimizing this complex process to maximize hydrogen yield remains challenging, particularly when balancing diverse feedstocks and improving process efficiency. While machine learning (ML) has shown significant potential in simulating and optimizing such processes, there is no clear consensus on the most effective regression models for co-gasification, especially with limited experimental data. Additionally, the interpretability of these models is a key concern. This study aims to bridge these gaps through two primary objectives: (1) modeling the co-gasification process using seven different ML algorithms, and (2) developing a framework for evaluating model interpretability, ultimately identifying the most suitable model for process optimization. A comprehensive set of experiments was conducted across three key dimensions, generalization ability, predictive accuracy, and interpretability, to thoroughly assess the models. Support Vector Regression (SVR) exhibited superior performance, achieving the highest coefficient of determination (R2) of 0.86. SVR outperformed other models in capturing non-linear dependencies and demonstrated effective overfitting mitigation. This study further highlights the limitations of other ML models, emphasizing the importance of regularization and hyperparameter tuning in improving model stability. By integrating Shapley Additive Explanations (SHAP) into model evaluation, this work is the first to provide detailed insights into feature importance and demonstrate the operational feasibility of ML models for industrial-scale hydrogen production in the co-gasification process. The findings contribute to the development of a robust framework for optimizing co-gasification, supporting the advancement of sustainable energy technologies and the reduction of greenhouse gas (GHG) emissions.

Time series data analysis and association rule mining in financial recommendation systems using Hadoop and Spark

Increasing amounts of financial data demand sophisticated analytics to develop sound recommendation models. This article discusses combining time series analysis and association rule mining for big data in Hadoop and Spark to enrich financial product recommendation engines. The paper is an integrated analysis of two types of prediction algorithms: AutoRegressive Integrated Moving Average (ARIMA) and Long Short-Term Memory (LSTM) networks to forecast user behavior and demand for financial services in the future from transactional history. The ARIMA model is used as the default while the LSTM model is used to represent non-linear dependencies and give a more dynamic forecast. association rule mining in particular the Apriori algorithm is used to find latent patterns and relationships between user transactions and financial products. This article illustrates how time series forecasting and association rule mining can be merged to bring a more useful financial recommendation. The hybrid approach, which combines both approaches, proves to increase user interaction and recommendation accuracy by 20% compared to the previous systems, according to experiments. The paper emphasises the possibilities of using big data in the construction of scalable, individualized financial recommendation systems.

A Weighted Bayesian Kernel Machine Regression Approach for Predicting the Growth of Indoor-Cultured Abalone

The cultivation of abalone, a species with high economic value, faces significant challenges due to its slow growth rate and sensitivity to environmental conditions, resulting in prolonged cultivation periods and increased mortality risks. To address these challenges, we propose a novel probabilistic machine learning approach based on a Bayesian framework to predict abalone growth by modeling key environmental factors, including water temperature, pH, salinity, nutrient supply, and dissolved oxygen levels. The proposed method employs a weighted Bayesian kernel machine regression model, integrating Gaussian processes with a spike-and-slab prior to identify influential variables. This approach accommodates heteroscedasticity, capturing varying levels of variance across observations, and models complex, non-linear relationships between environmental factors and abalone growth. Our analysis reveals that time, dissolved oxygen, salinity, and nutrient supply are the most critical factors influencing growth, while water temperature and pH play relatively minor roles under controlled indoor farming conditions. Interaction analysis highlights the non-linear dependencies among factors, such as the combined effects of salinity and nutrient supply. The proposed model not only improves prediction accuracy compared to baseline methods, but also provides actionable insights into the environmental dynamics that optimize abalone growth. These findings underscore the potential of advanced machine learning techniques in enhancing aquaculture practices and offer a robust framework for managing complex, multi-variable systems in sustainable farming.

A device for measuring soil moisture in the field

RELEVANCE. When building roads, preparing foundations for high–rise buildings, information is needed about such an important parameter as the density of the soil skeleton – the ratio of the mass of soil particles to the volume of the sample of an undisturbed structure. The precise determination of this parameter is an important task, since this parameter has a nonlinear dependence on humidity. Consequently, in a narrow range of humidity changes, the density of the soil skeleton becomes maximum, which makes such a soil the most favorable for construction work. PURPOSE. To develop an autonomous device for measuring soil moisture in the field. METHODS. The dielkometric method was used to develop a device for measuring soil moisture. This is an indirect method for measuring the humidity of substances based on the dependence of the dielectric constant of these substances on their humidity. RESULTS. An autonomous device for determining soil moisture using the dielkometric method has been developed, designed for use in the field, which significantly speeds up the process of analyzing and processing measurement results, and, consequently, the process of processing information about the state of the soil. During the project, experiments were carried out to measure capacity depending on humidity and temperature at various frequencies from 100 Hz to 100 kHz in various types of soils. CONCLUSION. A device for determining soil moisture was developed and a method for determining the moisture and density of the soil skeleton by the dielkometric method was proposed, preliminary experiments were conducted. However, during the experiments, a temperature dependence was found for low frequencies, so additional temperature calibration is necessary.

Recovery Method of Continuous Missing Data in the Bridge Monitoring System Using SVMD‐Assisted TCN–MHA–BiGRU

Due to the influence of complex service environments, the bridge health monitoring system (BHMS) has to face issues such as sensor failures and power outages of data acquisition systems, leading to frequent occurrences of data missing events including continuous and discrete data missing. By comparison, the continuous data missing can cover up the time‐series characteristic and make the corresponding recovery present a greater difficulty, especially for the data with a large loss rate or complicated features. To this end, this paper develops a novel signal recovery method based on the combination of successive variational mode decomposition (SVMD) and TCN–MHA–BiGRU, which is the hybrid of temporal convolutional networks (TCNs), multihead attention (MHA), and bidirectional gated recurrent unit (BiGRU). In this method, SVMD with high reliability and strong robustness is initially employed to decompose the original signal into multiple stable and regular subseries. Then, TCN–MHA–BiGRU incorporating the concept of “extraction‐weighting‐description of crucial features” is designed for the independent recovery of each subseries, with the ultimate recovery result derived through the linear superposition of all individual recoveries. This method not only can effectively extract the data time‐frequency characteristics (e.g., nonstationarity) but also can accurately capture the data time‐series characteristics (e.g., linear and nonlinear dependences) within the data. The case study and the subsequent applicability analysis grounded in the monitoring data from BHMS are employed to comprehensively evaluate the effectiveness of the proposed method. The results indicate that this method outperforms compared methods for the recovery of continuous missing data with different missing rates.