Search Results (31)

To address the impact of the dynamic evolution of flood disaster chains and decision-makers’ (DMs’) risk preference heterogeneity on group decision-making, this study proposes a social network group decision-making method that integrates the evolutionary trend of the flood disaster chain with DMs’ risk preferences. First, a Bayesian network is constructed to quantify the disaster chain’s evolution, dynamically adjusting DMs’ evaluation values. Second, DMs’ risk preference types are identified based on the evaluation values, and a bounded confidence (BC) model, incorporating risk preferences, self-confidence and trust networks, is developed to promote consensus formation. Then, the optimal alternative is selected through weighted aggregation and used to update the Bayesian network dynamically during implementation. Finally, the effectiveness and superiority of the proposed method are verified using the flood disaster chain from the “7∙20” extreme rainfall disaster in Zhengzhou, Henan Province, China. The results show that risk-seeking DMs reduce BC values and resist consensus, whereas risk-averse DMs enlarge BC values and accelerate convergence. Moreover, worsening flood disaster chain trends drive DMs to update the optimal alternative. These findings show that the method captures both dynamic disaster evolution and behavioral heterogeneity, providing realistic and adaptive decision support in flood emergency scenarios. Full article

In regions characterized by complex terrain and diverse pollution sources, high-precision air pollution prediction remains challenging due to nonlinear spatiotemporal coupling and the difficulty of modeling local pollutant agglomeration. To address these issues, this study proposes a CNN–LSTM–Transformer multimodal prediction framework integrated with Bayesian Optimization. First, the Local Moran’s Index (LMI) is introduced as a spatial perception feature and concatenated with pollutant concentration sequences before being input into the CNN module. This design enhances the model’s ability to identify local pollutant clustering and spatial heterogeneity. Second, the LSTM architecture adopts a dual-channel structure: the main channel employs bidirectional LSTM to extract temporal dependencies, while the auxiliary channel uses unidirectional LSTM to capture evolutionary trends. A Transformer with a multi-head attention mechanism is then introduced to perform global modeling. Bayesian Optimization is employed to automatically adjust key hyperparameters, thereby improving the model’s stability and convergence efficiency. Empirical results based on atmospheric pollution monitoring data from Sichuan Province during 2021–2024 demonstrate that the proposed model outperforms various mainstream methods in predicting six pollutants in Chengdu. For instance, the MAE for PM_2.5 decreased by 14.9–22.1%, while the coefficient of determination (R²) remained stable between 87% and 89%. The accuracy decay rate across four-day forecasts was controlled within 12.4%. Furthermore, in PM_2.5 generalization prediction tasks across four other cities—Yibin, Zigong, Nanchong, and Mianyang—the model exhibited superior stability and robustness, achieving an average R² of 87.4%. These findings highlight the model’s long-term stability and regional generalization capability, offering reliable technical support for air pollution prediction and control strategies in Sichuan Province and potentially beyond. Full article

African ground squirrels (Xerus spp.), the inhabitants of African arid zones, face extreme heat and water scarcity driving selection for metabolic optimization. We assembled and annotated the first mitogenomes of Xerus inauris and Xerus rutilus (16,525–16,517 bp), revealing conserved vertebrate architecture with genus-specific traits. Key features include Xerus rutilus’s elongated ATP6 (680 vs. 605 bp), truncated ATP8–ATP6 spacers (4 vs. 43 bp), and tRNA-Pro control regions with 78.1–78.3% AT content. Their nucleotide composition diverged from that of related sciurids, marked by reduced T (25.78–26.9%) and extreme GC skew (−0.361 to −0.376). Codon usage showed strong Arg-CGA bias (RSCU = 3.78–3.88) and species-specific elevations in Xerus rutilus’s UGC-Cys (RSCU = 1.83 vs. 1.17). Phylogenetics positioned Xerus as sister to Ratufa bicolor (Bayesian PP = 0.928; ML = 1.0), aligning with African biogeographic isolation. Critically, we identified significant signatures of positive selection in key mitochondrial genes linked to arid adaptation. Positive selection signals in ND4 (ω = 1.8 × background), ND1, and ATP6 (p < 0.0033) correspond to enhanced proton gradient efficiency and ATP synthesis–molecular adaptations likely crucial for optimizing energy metabolism under chronic water scarcity and thermoregulatory stress in desert environments. Distinct evolutionary rates were observed across mitochondrial genes and complexes: Genes encoding Complex I subunits (ND2, ND6) and Complex III (Cytb) exhibited accelerated evolution in arid-adapted lineages, while genes encoding Complex IV subunits (COXI) and Complex V (ATP8) remained highly conserved. These findings resolve the Xerus mitogenomic diversity, demonstrating adaptive plasticity balancing arid-energy optimization and historical diversification while filling critical genomic gaps for this xeric-adapted lineage. Full article

The Millard’s rat (Dacnomys millardi), a threatened murid endemic to Southeast Asian montane rainforests and the sole member of its monotypic genus, faces escalating endangered risks as a Near Threatened species in China’s Biodiversity Red List. This ecologically specialized rodent exhibits diagnostic morphological adaptations—hypertrophied upper molars and cryptic pelage—that underpin niche differentiation in undisturbed tropical/subtropical forests. Despite its evolutionary distinctiveness, the conservation prioritization given to Dacnomys is hindered due to a deficiency of data and unresolved phylogenetic relationships. Here, we integrated morphological analyses with the first complete mitogenome (16,289 bp in size; no structural rearrangements) of D. millardi to validate its phylogenetic placement within the subfamily Murinae and provide novel insights into genetic diversity erosion. Bayesian and maximum likelihood phylogenies robustly supported Dacnomys as sister to Leopoldamys (PP = 1.0; BS = 100%), with an early Pliocene divergence (~4.8 Mya, 95% HPD: 3.65–5.47 Mya). Additionally, based on its basal phylogenetic position within Murinae, we propose reclassifying Micromys from Rattini to the tribe Micromyini. Codon usage bias analyses revealed pervasive purifying selection (Ka/Ks < 1), constraining mitogenome evolution. Genetic diversity analyses showed low genetic variation (CYTB: π = 0.0135 ± 0.0023; COX1: π = 0.0101 ± 0.0025) in fragmented populations. We propose three new insights into this genetic diversity erosion. (1) Evolutionary constraints: genome-wide evolutionary conservation and shallow evolutionary history (~4.8 Mya) limited mutation accumulation. (2) Anthropogenic pressures: deforestation-driven fragmentation of habitats (>20,000 km²/year loss since 2000) has reduced effective population size, exacerbating genetic drift. (3) Ecological specialization: long-term adaptation to stable niches favored genomic optimization over adaptive flexibility. These findings necessitate suitable conservation action by enforcing protection of core habitats to prevent deforestation-driven population collapses and advocating IUCN reclassification of D. millardi from Data Deficient to Near Threatened. Full article

Physics-informed neural networks (PINNs) have emerged as a transformative methodology integrating deep learning with scientific computing. This review establishes a three-dimensional analytical framework to systematically decode PINNs’ development through methodological innovation, theoretical breakthroughs, and cross-disciplinary convergence. The contributions include threefold: First, identifying the co-evolutionary path of algorithmic architectures from adaptive optimization (neural tangent kernel-guided weighting achieving 230% convergence acceleration in Navier-Stokes solutions) to hybrid numerical-deep learning integration (5× speedup via domain decomposition) and second, constructing bidirectional theory-application mappings where convergence analysis (operator approximation theory) and generalization guarantees (Bayesian-physical hybrid frameworks) directly inform engineering implementations, as validated by $72\%$ cost reduction compared to FEM in high-dimensional spaces ($p<0.01,n=15$ benchmarks). Third, pioneering cross-domain knowledge transfer through application-specific architectures: TFE-PINN for turbulent flows ($5.12\pm 0.87\%$ error in NASA hypersonic tests), ReconPINN for medical imaging ($▵\mathrm{SSIM}=+0.18\pm 0.04$ on multi-institutional MRI), and SeisPINN for seismic systems ($0.52\pm 0.18$ km localization accuracy). We further present a technological roadmap highlighting three critical directions for PINN 2.0: neuro-symbolic, federated physics learning, and quantum-accelerated optimization. This work provides methodological guidelines and theoretical foundations for next-generation scientific machine learning systems. Full article

This study aims to enhance the accuracy and interpretability of flood susceptibility mapping (FSM) in Seoul, South Korea, by integrating automated machine learning (AutoML) with explainable artificial intelligence (XAI) techniques. Ten topographic and environmental conditioning factors were selected as model inputs. We first employed the Tree-based Pipeline Optimization Tool (TPOT), an evolutionary AutoML algorithm, to construct baseline ensemble models using Gradient Boosting (GB), Random Forest (RF), and XGBoost (XGB). These models were further fine-tuned using Bayesian optimization via Optuna. To interpret the model outcomes, SHAP (SHapley Additive exPlanations) was applied to analyze both the global and local contributions of each factor. The SHAP analysis revealed that lower elevation, slope, and stream distance, as well as higher stream density and built-up areas, were the most influential factors contributing to flood susceptibility. Moreover, interactions between these factors, such as built-up areas located on gentle slopes near streams, further intensified flood risk. The susceptibility maps were reclassified into five categories (very low to very high), and the GB model identified that approximately 15.047% of the study area falls under very-high-flood-risk zones. Among the models, the GB classifier achieved the highest performance, followed by XGB and RF. The proposed framework, which integrates TPOT, Optuna, and SHAP within an XAI pipeline, not only improves predictive capability but also offers transparent insights into feature behavior and model logic. These findings support more robust and interpretable flood risk assessments for effective disaster management in urban areas. Full article

Hybrid electric vertical take-off and landing (HEVTOL) flying vehicles serve as effective platforms for efficient transportation, forming a cornerstone of the emerging low-altitude economy. However, the current lack of co-optimization methods for powertrain component sizing and energy controller design often leads to suboptimal HEVTOL performance. To address this, this paper proposes a hierarchical manifold-enhanced Bayesian evolutionary optimization (HM-BEO) approach for HEVTOL systems. This framework employs lightweight manifold dimensionality reduction to compress the decision space, enabling Bayesian optimization (BO) on low-dimensional manifolds for a global coarse search. Subsequently, the approximate Pareto solutions generated by BO are utilized as initial populations for a non-dominated sorting genetic algorithm III (NSGA-III), which performs fine-grained refinement in the original high-dimensional design space. The co-optimization aims to minimize fuel consumption, battery state-of-health (SOH) degradation, and manufacturing costs while satisfying dynamic and energy management constraints. Evaluated using representative HEVTOL duty cycles, the HM-BEO demonstrates significant improvements in optimization efficiency and solution quality compared to conventional methods. Specifically, it achieves a 5.3% improvement in fuel economy, a 7.4% mitigation in battery SOH degradation, and a 1.7% reduction in system manufacturing cost compared to standard NSGA-III-based optimization. Full article

This review proposes a novel integration of game-theoretical methods—specifically Evolutionary Game Theory (EGT), Stackelberg games, and Bayesian games—with deep reinforcement learning (DRL) to optimize electricity markets. Our approach uniquely addresses the dynamic interactions among power purchasing and generation enterprises, highlighting both theoretical underpinnings and practical applications. We demonstrate how this integrated framework enhances market resilience, informs evidence-based policy-making, and supports renewable energy expansion. By explicitly connecting our findings to regulatory strategies and real-world market scenarios, we underscore the political implications and applicability of our results in diverse global electricity systems. By integrating EGT with advanced methodologies such as DRL, this study develops a comprehensive framework that addresses both the dynamic nature of electricity markets and the strategic adaptability of market participants. This hybrid approach allows for the simulation of complex market scenarios, capturing the nuanced decision-making processes of enterprises under varying conditions of uncertainty and competition. The review systematically evaluates the effectiveness and cost-efficiency of various control policies implemented within electricity markets, including pricing mechanisms, capacity incentives, renewable integration incentives, and regulatory measures aimed at enhancing market competition and transparency. Our analysis underscores the potential of EGT to significantly enhance market resilience, enabling electricity markets to better withstand shocks such as sudden demand fluctuations, supply disruptions, and regulatory changes. Moreover, the integration of EGT with DRL facilitates the promotion of sustainable energy integration by modeling the strategic adoption of renewable energy technologies and optimizing resource allocation. This leads to improved overall market performance, characterized by increased efficiency, reduced costs, and greater sustainability. The findings contribute to the development of robust regulatory frameworks that support competitive and efficient electricity markets in an evolving energy landscape. By leveraging the dynamic and adaptive capabilities of EGT and DRL, policymakers can design regulations that not only address current market challenges but also anticipate and adapt to future developments. This proactive approach is essential for fostering a resilient energy infrastructure capable of accommodating rapid advancements in renewable technologies and shifting consumer demands. Additionally, the review identifies key areas for future research, including the exploration of multi-agent reinforcement learning techniques and the need for empirical studies to validate the theoretical models and simulations discussed. This study provides a comprehensive roadmap for optimizing electricity markets through strategic and policy-driven interventions, bridging the gap between theoretical game-theoretic models and practical market applications. Full article

Accurate remaining capacity prediction (RCP) of lithium-ion batteries (LIBs) is crucial for ensuring their safety, reliability, and performance, particularly amidst the growing energy crisis and environmental concerns. However, the complex aging processes of LIBs significantly hinder accurate RCP, as traditional prediction methods struggle to effectively capture nonlinear degradation patterns and long-term dependencies. To tackle these challenges, we introduce an innovative framework that combines evolutionary learning with deep learning for RCP. This framework integrates Temporal Convolutional Networks (TCNs), Bidirectional Gated Recurrent Units (BiGRUs), and an attention mechanism to extract comprehensive time-series features and improve prediction accuracy. Additionally, we introduce a hybrid optimization algorithm that combines the Sparrow Search Algorithm (SSA) with Bayesian Optimization (BO) to enhance the performance of the model. The experimental results validate the superiority of our framework, demonstrating its capability to achieve significantly improved prediction accuracy compared to existing methods. This study provides researchers in battery management systems, electric vehicles, and renewable energy storage with a reliable tool for optimizing lithium-ion battery performance, enhancing system reliability, and addressing the challenges of the new energy industry. Full article

In nature, closely related species often exhibit diverse characteristics, challenging simplistic line interpretations of trait evolution. For these species, the evolutionary dynamics of one trait may differ markedly from another, with some traits evolving at a slower pace and others rapidly diversifying. In light of this complexity and concerning the phenomenon of trait relationships that escape line measurement, we introduce a novel general adaptive optimal regression model, grounded on polynomial relationships. This approach seeks to capture intricate patterns in trait evolution by considering them as continuous stochastic variables along a phylogenetic tree. Using polynomial functions, the model offers a holistic and comprehensive description of the traits of the studied species, accounting for both decreasing and increasing trends over evolutionary time. We propose two sets of optimal adaptive evolutionary polynomial regression models of $k^{th}$ order, named the Ornstein–Uhlenbeck Brownian Motion Polynomial (${\mathrm{OUBMP}}_k$) model and Ornstein–Uhlenbeck Ornstein–Uhlenbeck Polynomial (${\mathrm{OUOUP}}_k$) model, respectively. Assume that the main trait value $y_t$ is a random variable of the Ornstein–Uhlenbeck (OU) process and that its optimal adaptive value ${\theta }_t^y$ has a polynomial relationship with other traits $x_t$ for statistical modeling, where $x_t$ can be a random variable of Brownian motion (BM) or OU process. As analytical representations for the likelihood of the models are not feasible, we implement an approximate Bayesian computation (ABC) technique to assess the performance through simulation. We also plan to apply models to the empirical study using the two datasets: the longevity vs. fecundity in the Mediterranean nekton group, and the trophic niche breadth vs. body mass in carnivores in a European forest region. Full article

This study examines the decision-making optimization of Power-Generation Enterprises (PGEs) in the green certificate market, with a focus on balancing bidding strategies and carbon-reduction targets. Given the increasing complexity of the green certificate market, the research employs Bayesian games, evolutionary games, and Stackelberg games to systematically analyze the strategic behavior of PGEs and their interactions within the market framework. The findings demonstrate that game theory facilitates cost structure optimization and enhances adaptability to market dynamics under policy-driven incentives and penalties. Additionally, the study explores the integration of stochastic modeling and machine learning techniques to address market uncertainties. These results provide theoretical support for policymakers in designing efficient green electricity market regulations and offer strategic insights for PGEs aligning with carbon neutrality objectives. This work bridges theoretical modeling and practical application, contributing to the advancement of sustainable energy policies and the development of green electricity markets. Full article

To successfully meet the various requirements of modern storage systems, NAND flash memory should be highly optimized by precisely tuning a huge number of internal operating parameters. Although 3D NAND flash memory succeeds in increasing the capacity of storage systems, its complex architecture and unique error behavior make such optimization a more difficult and time-consuming process during NAND manufacturing. In this paper, we introduce HAIPO, a novel methodology for post-fabrication optimization of NAND flash memory, which is an essential step in the manufacturing process of modern 3D NAND flash memory to simultaneously meet various requirements on reliability, performance, yield, etc. HAIPO is based on simple machine-learning approaches that consist of (i) a lightweight deep-learning (DL) model to generate initial device parameters and (ii) an evolutionary algorithm (EA) to explore device parameters automatically. To more effectively explore device parameters, we introduce three key guidelines for each generation in the EA: (1) domain-specific rules, (2) recent optimization results, and (3) online Bayesian simulation, respectively, to enable quick optimization for a huge number of device parameters within the limited product turnaround time (TAT). In addition, we integrate two optimization modules with HAIPO to improve optimization efficiency even in environments with severe process variation. We demonstrate the feasibility and effectiveness of HAIPO using real 320 3D TLC/QLC NAND flash chips, showing significant performance and reliability improvements by up to 8.8% and 12% on average, respectively, within a quite limited optimization TAT. Full article

As an environmentally responsible alternative to conventional concrete, geopolymer concrete recycles previously used resources to prepare the cementitious component of the product. The challenging issue with employing geopolymer concrete in the building business is the absence of a standard mix design. According to the chemical composition of its components, this work proposes a thorough system or framework for estimating the compressive strength of fly ash-based geopolymer concrete (FAGC). It could be possible to construct a system for predicting the compressive strength of FAGC using soft computing methods, thereby avoiding the requirement for time-consuming and expensive experimental tests. A complete database of 162 compressive strength datasets was gathered from the research papers that were published between the years 2000 and 2020 and prepared to develop proposed models. To address the relationships between inputs and output variables, long short-term memory networks were deployed. Notably, the proposed model was examined using several soft computing methods. The modeling process incorporated 17 variables that affect the CSFAG, such as percentage of SiO₂ (SiO₂), percentage of Na₂O (Na₂O), percentage of CaO (CaO), percentage of Al₂O₃ (Al₂O₃), percentage of Fe₂O₃ (Fe₂O₃), fly ash (FA), coarse aggregate (CAgg), fine aggregate (FAgg), Sodium Hydroxide solution (SH), Sodium Silicate solution (SS), extra water (EW), superplasticizer (SP), SH concentration, percentage of SiO₂ in SS, percentage of Na₂O in SS, curing time, curing temperature that the proposed model was examined to several soft computing methods such as multi-layer perception neural network (MLPNN), Bayesian regularized neural network (BRNN), generalized feed-forward neural networks (GFNN), support vector regression (SVR), decision tree (DT), random forest (RF), and LSTM. Three main innovations of this study are using the LSTM model for predicting FAGC, optimizing the LSTM model by a new evolutionary algorithm called the marine predators algorithm (MPA), and considering the six new inputs in the modeling process, such as aggregate to total mass ratio, fine aggregate to total aggregate mass ratio, FASiO₂:Al₂O₃ molar ratio, FA SiO₂:Fe₂O₃ molar ratio, AA Na₂O:SiO₂ molar ratio, and the sum of SiO₂, Al₂O₃, and Fe₂O₃ percent in FA. The performance capacity of LSTM-MPA was evaluated with other artificial intelligence models. The results indicate that the R² and RMSE values for the proposed LSTM-MPA model were as follows: MLPNN (R² = 0.896, RMSE = 3.745), BRNN (R² = 0.931, RMSE = 2.785), GFFNN (R² = 0.926, RMSE = 2.926), SVR-L (R² = 0.921, RMSE = 3.017), SVR-P (R² = 0.920, RMSE = 3.291), SVR-S (R² = 0.934, RMSE = 2.823), SVR-RBF (R² = 0.916, RMSE = 3.114), DT (R² = 0.934, RMSE = 2.711), RF (R² = 0.938, RMSE = 2.892), LSTM (R² = 0.9725, RMSE = 1.7816), LSTM-MPA (R² = 0.9940, RMSE = 0.8332), and LSTM-PSO (R² = 0.9804, RMSE = 1.5221). Therefore, the proposed LSTM-MPA model can be employed as a reliable and accurate model for predicting CSFAG. Noteworthy, the results demonstrated the significance and influence of fly ash and sodium silicate solution chemical compositions on the compressive strength of FAGC. These variables could adequately present variations in the best mix designs discovered in earlier investigations. The suggested approach may also save time and money by accurately estimating the compressive strength of FAGC with low calcium content. Full article

In the wake of the terrorist attacks of 11 September 2001, extensive research efforts have been dedicated to the development of computational algorithms for identifying contamination sources in water distribution systems (WDSs). Previous studies have extensively relied on evolutionary optimization techniques, which require the simulation of numerous contamination scenarios in order to solve the inverse-modeling contamination source identification (CSI) problem. This study presents a novel framework for CSI in WDSs using Bayesian optimization (BO) techniques. By constructing an explicit acquisition function to balance exploration with exploitation, BO requires only a few evaluations of the objective function to converge to near-optimal solutions, enabling CSI in real-time. The presented framework couples BO with EPANET to reveal the most likely contaminant injection/intrusion scenarios by minimizing the error between simulated and measured concentrations at a given number of water quality monitoring locations. The framework was tested on two benchmark WDSs under different contamination injection scenarios, and the algorithm successfully revealed the characteristics of the contamination source(s), i.e., the location, pattern, and concentration, for all scenarios. A sensitivity analysis was conducted to evaluate the performance of the framework using various BO techniques, including two different surrogate models, Gaussian Processes (GPs) and Random Forest (RF), and three different acquisition functions, namely expected improvement (EI), probability of improvement (PI), and upper confident bound (UCB). The results revealed that BO with the RF surrogate model and UCB acquisition function produced the most efficient and reliable CSI performance. Full article

The reliability of urban transportation systems is crucial for ensuring smooth traffic flow and minimizing disruptions caused by external factors. This study focuses on improving the stability and efficiency of transportation systems through the calibration of a refined link performance function while building upon the U.S. Bureau of Public Roads (BPR) model. To achieve this, we propose three customized algorithms—Newton’s method, Bayesian optimization, and the differential evolutionary algorithm—to calibrate the key parameters. Additionally, we conducted a sensitivity analysis to assess the influences of the model parameters on link performance. Numerical experiments conducted in Yuyao City demonstrate the applicability and efficacy of the proposed model and solution algorithms. Our results reveal that the Newton approach is notably more efficient than the Bayesian optimization algorithm and the differential evolutionary algorithm. Full article