Search Results (4,538)

Accurate field-scale surface soil moisture (SSM) measures are important for agricultural water management. Conventional satellite SSM products remain too coarse for within-field applications. Here, we developed sequence-aware deep learning models for growing-season SSM estimation by fusing data from Sentinel-1 C-band SAR, harmonized Landsat-8/Sentinel-2 (HLS), and local ancillary datasets. We assembled a multi-source dataset on Sentinel-1 overpass time for 2016–2024 (May–September), yielding 1469 samples and 65 features per sample, including SAR and optical features, meteorological data, soil texture and bulk density, topography, crop labels, irrigation-likelihood flag, and irregular-time-step encoding. We compared long short-term memory (LSTM) and temporal convolutional neural network (TCN) architectures together with attention-augmented variants, including feature attention (FA), temporal attention (TA), and the combined feature–temporal attention (FTA). Models were trained and tested on seven years of data and were validated based on a temporal generalization using combined data of a wet year (2016) and a dry year (2023). The best model, FTA-TCN, achieved R² = 0.851, RMSE = 0.024 m³.m⁻³, and MAE = 0.020 m³.m⁻³ on the withheld validation years, outperforming the base LSTM (R² = 0.422; RMSE = 0.053 m³.m⁻³; MAE = 0.043 m³.m⁻³) and the base TCN (R² = 0.746; RMSE = 0.034 m³.m⁻³; MAE = 0.022 m³.m⁻³). Shapley additive explanations (SHAP) analysis indicated that antecedent precipitation and short-term rainfall accumulations were dominant forcings, while soil texture, elevation, incidence angle, and vegetation indices modulated SSM variability. Satellite-derived features accounted for ~28.5% of aggregated SHAP importance. Overall, the results show that dual-attention temporal convolution can capture field-scale SSM dynamics across wet and dry seasons when satellite signals are coupled with local soil-meteorological-management context. Full article

Accurate rice yield estimation is essential for food security. Two key factors affecting estimation accuracy are the long-term upward trend in yield over time and regional heterogeneity across space. Current studies predominantly employ statistical detrending methods (e.g., moving averages, linear regression) to isolate temporal trends. However, such methods rely on prior assumptions about the time–yield relationship and may introduce systematic bias when these assumptions break down. Meanwhile, the individual contributions of temporal and spatial information, and their interactive effects, have not been systematically evaluated within a unified framework. We selected 112 rice-growing counties across six U.S. states (2000–2021), using vegetation index (Normalized Difference Vegetation Index), meteorological indicators (growing degree days, killing degree days, and cumulative precipitation), and spatiotemporal variables (year, longitude, and latitude). We designed six input configurations to compare conventional detrending against direct temporal variable inclusion, testing across four model architectures (Long Short-Term Memory, Random Forest, XGBoost, and Transformer). Results showed that: (1) directly inputting year significantly outperformed detrending across all models, with the combined spatiotemporal configuration achieving the best performance (LSTM R² = 0.61 vs. 0.54 for detrending); (2) year was the most important predictor in SHAP analysis, with spatiotemporal variables ranking higher than most meteorological and remote sensing variables; (3) spatial information consistently improved accuracy and mitigated systematic bias for extreme yield regions; (4) the combined configuration performed best across different states, years (including extreme climate events), and yield levels, achieving near-end-of-season accuracy at the grain-filling stage (1.5–2 months before harvest). This study demonstrates that integrating raw spatiotemporal data directly into deep learning models is more effective than statistical detrending, offering a simpler and more robust approach for large-scale crop yield estimation. Full article

Forecasting fine particulate matter with a diameter of 2.5 μm (PM_2.5) is critically important due to its adverse effects on human health and environmental sustainability. Although ground-based monitoring stations provide accurate measurements, their limited spatial coverage restricts large-scale PM_2.5 assessment, especially in complex urban regions. Consequently, aerosol optical depth (AOD) derived from satellite imagery, combined with advanced deep learning (DL) techniques, has emerged as an effective alternative by offering wide spatial coverage and rich spatio-temporal information. This paper proposed an optimization-driven fuzzy transformer–deep belief network (ODFT-DBN) for accurate PM_2.5 air pollution prediction. The proposed framework integrates a fuzzy inference module to model uncertainty and nonlinear environmental relationships, a transformer encoder to capture long-range spatio-temporal dependencies, and a DBN to extract hierarchical features and improve prediction robustness. In addition, a novel multi-objective gray wolf optimizer (NMOGWO) is employed to jointly optimize the model hyper-parameters and fuzzy membership functions. The proposed approach is implemented for the city of Tehran, Iran, using meteorological variables, topographical features, ground-based PM_2.5 measurements, and satellite-derived AOD data. The ODFT-DBN model is compared with several benchmark methods, including bidirectional encoder representations from transformers (BERT), transformer, long short-term memory (LSTM), gated recurrent unit (GRU), convolutional neural network (CNN), DBN, and extreme gradient boosting (XGBoost). Experimental results demonstrate that the proposed framework achieves superior predictive performance, attaining an R² value of 0.94 and root mean square error (RMSE) of 0.8 μg/m³. Scatter plot analyses indicate a strong agreement between predicted and observed PM_2.5 values, while the proposed model exhibits low variance, stable convergence behavior, and acceptable computational time. Overall, the results confirm the effectiveness, robustness, and practical applicability of the proposed ODFT-DBN framework for spatio-temporal PM_2.5 forecasting. Full article

Accurate carbon price forecasting is essential for emission trading risk management and low-carbon investment decisions. In existing decomposition-prediction frameworks, secondary decomposition targets are typically selected based on statistical complexity rather than domain-informed causality, and standard Long Short-Term Memory (LSTM)-Transformer architectures discard the cell state that encodes long-term temporal memory. These limitations are particularly pronounced where energy-driven causal structures and regime-switching volatility coexist. This study proposes Causal State-driven Network (CaStNet), an intelligent forecasting framework with two core innovations. A Policy-Causality-guided Residual Secondary Decomposition (PCRSD) module replaces entropy-based criteria with Granger causality to select intrinsic mode functions (IMFs) exhibiting significant energy-carbon causal linkages for targeted variational mode decomposition (VMD). A Cell-State-Driven Dual-function Attention (CSDA) mechanism repurposes the LSTM cell state for simultaneously injecting long-term memory into the Transformer and employing the cell-state differential velocity as a volatility proxy to adaptively regulate Top-k attention sparsity. The Artificial Lemming Algorithm (ALA) globally co-optimizes decomposition dimensions and attention boundaries. A Shapley Additive exPlanations (SHAP)–Local Interpretable Model-agnostic Explanations (LIME) interpretability analysis reveals horizon-dependent driver transitions from short-term autoregressive momentum to long-term energy fundamentals, uncovering threshold nonlinearities in energy-carbon transmission channels. Validation on the Shanghai market (2013–2025) achieves point-forecast RMSE = 0.8326 and R² = 0.9777, outperforming all twelve benchmark models. Cross-market testing on the Hubei market yields R² = 0.9487, and expanding-window five-fold cross-validation on the Shanghai dataset yields mean R² = 0.9704, jointly confirming generalization robustness. Full article

South Africa faces significant challenges in monitoring air pollution from different provinces due to the sparse nature of the sensor network and heterogeneous pollutant sources. Notably, some provinces continue to record a limited amount of data on air pollution, thus making monitoring in those locations problematic. Fortunately, the capabilities of deep learning models to facilitate effective monitoring in data-scarce locations have been highlighted by researchers; however, these models within the context of transfer learning still lack transparency and uncertainty quantification. Using air pollutants and meteorological factors, this study proposes a transfer learning model for particulate matter (PM_2.5) prediction in a data-scarce region. This transfer learning (TL) model leverages an adaptive Bi-directional Gated Recurrent Unit (adaBiGRU) with explainable artificial intelligence (xAI) and uncertainty quantification (UQ) to provide a novel uncertainty-aware adaptation transfer learning (UATL_adaBiGRU) model for a data-scarce location. Variant models based on the adaBiGRU technique, such as the temporal convolution network adaBiGRU (TCN-adaBiGRU) and domain-adversarial neural network adaBiGRU (DANNadaBiGRU), are presented as comparative models. The performance evaluation metrics are root mean squared, R² score and mean squared error. The R² score of pre-trained models in source domain is adaBiGRU (0.888), DANN_adaBiGRU (0.7788) and TCN_adaBiGRU (0.876). Furthermore, other comparative TL models include GRU (0.898), MLP (0.802) and adaptive LSTM (0.886). Afterwards, the pre-trained baseline model (adaBiGRU) was fine-tuned in the target domain dataset and the unpromising result contributed to the proposition of the UATL_adaBiGRU model for a data-scarce location, with R² score of 0.9618. Uncertainty assessment metrics results were also presented for the proposed model. Ablation assessment demonstrates that each component of the UATL_adaBiGRU contributes to enhancing the predictive performance. Again, the Diebold–Mariano (DM) test statistic demonstrates a statistically significant difference between baseline model and UATL_adaBiGRU model. Finally, the local interpretable model-agnostic explanation highlights multi-scaled features as contributing towards the prediction of PM_2.5 in the target domain. In view of this result, model fine-tuning is strongly recommended to enhance the robustness of the proposed uncertainty-aware adaption model in data-limited regions in South Africa. Full article

Tracking highly maneuverable UAVs in complex maritime environments faces multiple challenges: dynamic sea surface interference and low-altitude occlusion make UAV motion trajectories difficult to predict; the strong maneuvering behavior of UAVs imposes high demands on tracking real-time performance and accuracy; and marine environmental noise and unstable shipborne sensor data lead to measurement incompleteness. These factors collectively limit the adaptability and robustness of existing maneuvering UAV tracking methods in complex maritime scenarios. In this context, accurate model recognition for UAVs becomes a key factor in improving tracking performance. Traditional interactive multiple model (IMM) methods rely on probabilistic weighting for model selection, suffering from response delays during UAV maneuvers, and fixed model sets cannot adapt to diverse maneuvering scenarios, resulting in degraded UAV velocity estimation accuracy. To address the above issues, this study proposes a dual long short-term memory (LSTM) cooperative network architecture, targeting the two key problems of incomplete measurements in shipborne radar measurements and inaccurate model probability estimation, and presents corresponding solutions. First, an online-trained LSTM-based incomplete-measurement compensation method is proposed, which achieves real-time fitting and restoration of historical measurement data, providing continuous and stable measurement inputs for shipborne platform UAV tracking in maritime environments. Second, building on this, an LSTM-based UAV model recognition method is developed to directly identify the UAV’s current motion model from multi-frame historical measurement information, effectively reducing maneuvering delays. Furthermore, GPS data is used to generate optimal model probabilities as training labels, thereby improving model reliability. Simulation results show that, under incomplete-measurement conditions, the proposed method can effectively reconstruct missing measurements and ensure measurement continuity. Under complete-measurement conditions, the proposed LSTM-based model recognition method significantly improves UAV model recognition accuracy and three-dimensional velocity estimation performance, demonstrating the effectiveness of deep learning for maneuvering UAV tracking from shipborne platforms in maritime environments. Full article

Construction contracts are vital for governing responsibilities in large-scale infrastructure projects, but their increasing complexity often leads to interpretation difficulties, disputes, and delays. Despite advances in natural language processing (NLP), automated analysis of construction contract clauses remains limited in project management. This study proposes a text classification framework integrating transformer-based contextual embeddings (BERT, ALBERT, RoBERTa, and DistilBERT) with machine learning and deep learning models (RNN, GRU, and LSTM) to analyze FIDIC and JCT contract provisions. Two multi-class classification tasks were defined: Dataset 1 (DS1) focusing on obligations, operational actions, optional provisions, general statements, and Dataset 2 (DS2) covering cost, quality, and time dimensions. Experimental results show that deep learning models consistently outperform traditional machine learning algorithms. Specifically, LSTM combined with RoBERTa and DistilBERT achieved the highest accuracy levels of 98.06% and 98.33% for DS1. The framework may support transparent contract governance by enabling faster and more consistent identification of contractual clauses. From a sustainability perspective, the findings suggest potential process-level contributions to economic efficiency, administrative workload reduction, and decision-making support throughout the project lifecycle. Full article

Internet of Things (IoT) environments, similarly to traditional network infrastructures, are highly vulnerable to volumetric Distributed Denial of Service (DDoS) attacks. Detecting such attacks remains challenging due to their bursty and short-lived nature, particularly in User Datagram Protocol (UDP) flood traffic, which often blends into normal traffic fluctuations. Conventional deep learning (DL) approaches, particularly Long Short-Term Memory (LSTM) networks, assign uniform importance to all time steps, limiting their ability to capture temporally localized burst patterns critical for identifying UDP-based volumetric attacks. To address this limitation, this study proposes LSTM-IoT, an attention-enhanced intrusion detection framework that integrates a temporal attention mechanism into an LSTM architecture. The model selectively emphasizes informative time intervals while suppressing irrelevant temporal segments, improving discrimination between benign and attack traffic. Evaluated on UDP traffic flows from the CICDDoS2019 dataset, LSTM-IoT achieves a detection accuracy of 99.93%, outperforming a baseline LSTM model. The results confirm that the proposed DL-based model effectively detects UDP-based volumetric DDoS attacks in IoT environments. Full article

Neurodevelopmental disorders (NDDs) are associated with impairments in communication, behavior, and social interaction, making accurate diagnosis clinically challenging. Autism Spectrum Disorder (ASD), a major NDD, often exhibits atypical speech patterns characterized by altered prosody and reduced emotional expressiveness. The study proposes a hybrid dual-path framework for ASD detection from emotional speech using two strategies: PCA–GMM-based acoustic modeling and a CNN–BiLSTM–Attention architecture for spectral–temporal feature learning. The proposed framework captures probabilistic, spectral, and temporal speech characteristics for robust ASD classification. Acoustic analysis demonstrated clear separability between ASD and non-ASD speech, while the deep learning framework achieved stable and reliable performance across multiple emotional conditions. Experimental evaluation achieved 98.3% accuracy, AUC values ranging from 0.9699 to 0.9864, and F1-scores up to 0.9891. The findings highlight the potential of AI-driven speech analysis as a scalable and non-invasive tool for early ASD screening and predictive healthcare applications. Full article

Social media has emerged as a crucial environment for examining public sentiment during disasters, providing immediate insights into collective emotions and urgent expectations. This research examines the emotional reactions expressed on Turkish posts shared on the X platform (formerly Twitter) following the 6 February 2023 earthquake by employing an integrated method that combines topic modeling and topic-based sentiment analysis. Data were collected between 10 February 2023 and 28 February 2023. A large dataset consisting of 305,000 tweets was compiled, and 296,836 tweets remained for analysis after preprocessing and filtering procedures. Latent Dirichlet Allocation (LDA), enhanced with term frequency-inverse document frequency weighting and bigram extraction techniques, was applied to identify prominent themes, including rescue operations, appeals for assistance, communication about missing persons, and disaster management. The sentiment polarity within each topic was determined using a hybrid deep learning model incorporating Bidirectional Encoder Representations from Transformers (BERT) embeddings Convolutional Neural Networks (CNN), Bidirectional Long Short-Term Memory (BiLSTM) layers, and FastText representations. This model reached a classification accuracy of 94%, with F1-scores of 0.91 and 0.95, recall values of 0.90 and 0.96, and precision values of 0.92 and 0.95, achieving higher performance than the evaluated baseline models. The findings indicate that supportive, solidarity-oriented, and resilience-related communication patterns were among the most frequently observed positive sentiment expressions, whereas negative sentiments appeared more frequently in discussions regarding delays in aid delivery and perceived shortcomings in institutional response. This study presents a scalable and flexible framework for analyzing sentiment in Turkish-language crisis communication, providing insights that may support disaster response monitoring and decision-making processes as well as the development of systems for tracking public reactions in real time. Full article

Respiratory diseases are a major global health challenge. However, identification of respiratory diseases is often limited by subjectivity, environmental noise and inter-clinician variability. This study presents an explainable multimodal deep learning framework for recording-level multiclass classification of respiratory audio signals. The proposed system integrates two complementary representations—a spectro-temporal encoder based on a CNN–BiLSTM-attention architecture and a handcrafted acoustic-feature encoder capturing acoustic descriptors commonly used in respiratory-audio analysis, including MFCCs, zero-crossing rate, spectral centroid, spectral bandwidth, chroma, RMS energy, and spectral rolloff features. These branches are combined through late-stage fusion to leverage both data-driven representation learning and domain-informed acoustic cues. The proposed model was trained and internally evaluated on the Asthma Detection Dataset Version 2, comprising five respiratory categories: bronchial disease, asthma, COPD, healthy, and pneumonia. Mono conversion, resampling to 16 kHz, 100–2000 Hz band-pass filtering, amplitude normalisation, fixed 4 s trimming or zero-padding, training-only augmentation, handcrafted-feature extraction, mel-spectrogram generation, quality control auditing, and stratified recording-level partitioning have been applied in the pre-processing steps. Across five repeated experiments with different random seeds, the proposed hybrid model achieved a mean held-out recording-level test accuracy of $0.9099\pm 0.0163$, balanced accuracy of $0.8936\pm 0.0152$, macro F1-score of $0.8937\pm 0.0177$, macro ROC–AUC of $0.9867\pm 0.0010$, and macro PR–AUC of $0.9489\pm 0.0044$. Conventional machine learning baseline comparisons showed that the proposed model achieved stronger internal accuracy, balanced accuracy, macro recall, macro F1-score, and macro ROC–AUC than classical machine learning algorithms trained on handcrafted acoustic features, although Random Forest remained competitive in macro PR–AUC. Ablation analysis shows that the deep spectro-temporal branch was the primary contributor to predictive performance, while the handcrafted branch provided complementary interpretable acoustic information rather than consistently improving all classification metrics. Explainability was incorporated using Grad-CAM and Integrated Gradients for spectrogram-based interpretation and SHAP for handcrafted-feature attribution. Domain-shift evaluation on the ICBHI Respiratory Sound Database and a COPD-focused cohort revealed substantial dataset shift effects, including poor healthy-case recognition on ICBHI and seed-dependent COPD recognition in the COPD-focused cohort. Identifier-aware sensitivity analyses showed lower performance than the main recording-level split, suggesting that subject-like or source-level overlap may inflate internal performance estimates. The findings should be interpreted as promising internal held-out recording-level algorithmic performance with limited external transfer, rather than evidence of readiness for clinical use. Full article

Modern artificial intelligence methods are increasingly used in power systems for renewable energy generation and electricity load forecasting. However, the limited interpretability of complex machine learning and deep learning models constrains their adoption in critical energy applications where transparency and trust are essential. Explainable Artificial Intelligence (XAI) provides tools for interpreting model behavior, yet its application to multivariate time series remains associated with significant methodological challenges. This paper presents a systematic review of XAI applications in solar power, wind power, and electricity load forecasting based on 154 peer-reviewed journal articles published between 2019 and 2026, identified through searches in Scopus, IEEE Xplore, ScienceDirect, and MDPI, following the PRISMA 2020 methodology. The review covers widely used forecasting architectures, including LSTMs, Transformers, and tree-based ensembles, as well as XAI methods. The analysis identifies a fundamental limitation of conventional XAI approaches for multivariate time series, referred to as the curse of dimensionality in XAI-based interpretation of time series, in which each time step is treated as an independent feature, resulting in explanations that are difficult to interpret in practice. To address this challenge, eight categories of XAI adaptations for time series forecasting are systematized. A classification of knowledge extraction mechanisms is proposed, including feature-level, temporal, regime-based, causal, diagnostic, model-level, and decision-support knowledge. The results demonstrate a gradual transition from explainability toward knowledge extraction, where XAI serves not only to explain individual forecasts but also to generate actionable knowledge about data, models, and energy processes. The review is limited to peer-reviewed English-language journal articles published between 2019 and 2026. The findings suggest that Knowledge Extraction represents a key mechanism for building trust in intelligent energy forecasting systems. Full article

Vertically linked commodity markets—global futures, regional spot, and farm-gate prices—transmit information through directed cross-market channels whose strength varies with latent volatility regimes. Standard deep learning forecasters absorb both the directed cross-market dependence and the regime dependence of intrinsic-mode-aligned latent components into shared model weights, with no explicit architectural mechanism that exposes either as an inspectable structure. This paper proposes HVB-RA, a modular framework that combines two such mechanisms with a per-tier Variational Mode Decomposition and bidirectional LSTM backbone: (i) a directed cross-market attention layer in which the upstream-to-downstream topology is supplied from domain knowledge and the time-varying upstream-source attention intensities at the farm-gate tier (the regional-spot tier, with a single upstream key, reduces algebraically to a fixed residual upstream fusion) are extracted from data, and (ii) a regime-informed modal-weighting layer that mixes two trainable softmax weight profiles over IMF-aligned latent components through a filtered Markov-switching state probability fitted in a separate stage. An auxiliary post hoc projection enforces an exact linear constraint defined by long-run sample-mean ratios across tiers; the paper does not claim that these descriptive ratios are cointegrating relations or equilibrium coefficients. The framework is evaluated on three tiers of daily natural-rubber prices spanning 2038 trading days, against three external benchmarks (random walk, ARIMA(2,0,2), and an exogenous-only LSTM) and a contemporary neural hierarchical-interpolation forecaster (NHITS). Root mean squared error is reported per tier-horizon cell; a decision-aware income-smoothing metric quantifies the operational value of $h=5$ farm-gate forecasts under a 5-day selling rule; and a within-method comparison evaluates the marginal contribution of the auxiliary constraint projection. On the present single-regime test window, HVB-RA attains a lower point error than the contemporary NHITS baseline at every tier-horizon cell, while no method—including HVB-RA—improves on the random-walk floor at most cells; the regime-conditional components of the architecture are not identifiable because every calibration and test origin is classified as a high-volatility regime by the trained Markov-switching model. The paper contributes to machine learning and knowledge extraction by demonstrating how time-varying upstream-source attention intensities at the farm-gate tier and regime-dependent latent-component-weight profiles—two forms of latent structure typically absorbed into model weights—can be exposed as explicit, inspectable, and individually testable components of a multi-tier forecasting architecture, and by providing a reproducibility package documenting the conditions under which each component is expected to be identifiable. Full article

This study investigates whether shareholder value creation, defined as beta-adjusted outperformance relative to a market benchmark, can be effectively predicted in an emerging market using a sequential machine learning framework. While prior research has predominantly focused on profitability forecasting or stock return prediction, the prediction of risk-adjusted shareholder value creation remains relatively underexplored, particularly in emerging economies such as Morocco. To address this gap, the study develops a predictive framework that combines market-based indicators, macroeconomic variables, and accounting fundamentals using only information realistically available to investors at each decision date. These variables are organized into firm-level temporal sequences based on a monthly decision-date panel of non-financial firms listed on the Casablanca Stock Exchange over the period 2010–2024. To capture nonlinear relationships and temporal dependencies in financial data, the empirical analysis compares baseline models with deep learning architectures, including GRU, LSTM, and CNN1D. The results indicate that deep learning models consistently outperform naïve and linear benchmark models, suggesting that shareholder value creation exhibits a measurable degree of predictability. With an AUC of 0.700 and a PR-AUC of 0.727, CNN1D achieves the strongest performance in the final evaluation setting and ranks as the best-performing model according to the primary AUC criterion. The findings also reveal that macroeconomic variables generate the strongest standalone predictive signal, whereas market-based variables exhibit comparatively weaker predictive power when considered in isolation. By extending financial prediction toward a risk-adjusted, benchmark-based, and investor-oriented framework, and by providing new empirical evidence on the value of temporal modeling and multi-source financial information for forecasting shareholder value creation in an emerging market context, this study contributes to the growing literature at the intersection of financial forecasting and artificial intelligence. Full article

Accurate break size assessment is critical for the safety response of small modular reactors (SMRs) during loss-of-coolant accidents (LOCAs). Traditional methods struggle with the rapid transient features, strong spatiotemporal coupling, and complex uncertainty characteristics of SMR-LOCA, leading to low accuracy and poor stability. To address these issues, this study proposes an entropy-driven intelligent diagnosis approach based on a CNN-LSTM-Attention hybrid model. The framework adopts information entropy for data uncertainty quantification, adaptive weighting, and loss constraint, so as to realize high-precision break size assessment. A time-series dataset covering break sizes from 0.05 to 10 cm² was constructed using the PCTRAN/SMART platform. The CNN module extracts spatial coupling features of multi-sensor parameters, the LSTM module captures long-term temporal dependencies, and the attention mechanism dynamically weights key information to enhance feature representation under high uncertainty. Experimental results show that the model achieves a mean absolute error (MAE) of 0.096311, reducing errors by over 64.4% compared with baseline models; more than 90% of prediction errors are within ±5%, and the correlation coefficient reaches 0.994902. Based on the well-validated PCTRAN/SMART simulation platform, the proposed entropy-informed spatiotemporal learning framework provides a technical solution for intelligent LOCA diagnosis, uncertainty quantification, and safety assessment of SMRs. Full article