Search

Automatic determination of manufacturing process sequences for the physical production of given part designs is key to facilitate on-demand cyber manufacturing. In this work, we propose an integrated framework that (i) identifies manufacturing features from 3D part designs using a Graph Neural Network (GNN), (ii) identifies the manufacturing processes necessary to produce all features in the part using a Convolutional Neural Network (CNN) that considers shape, material properties, and quality information, and (iii) outputs an ordered manufacturing sequence that can produce the designed part with the help of sequence mining. Using these methods, the knowledge required to enable automated manufacturing process selection is easily scalable and updatable without requiring manual population of ad-hoc or rule-based descriptions. We present exemplar implementations of the proposed framework by suggesting manufacturing sequences for discrete parts with multiple features. The suggested manufacturing sequences demonstrate the potential of the proposed framework for use in future on-demand cyber manufacturing applications. [ABSTRACT FROM AUTHOR]

Scheduling is a complex task due to the need to optimise multiple competing objectives and react to unpredictable events that may occur during production execution. The strategy of predictive-reactive scheduling can be used to reconcile the conflict between the original schedule and the current shop floor situation. This study seeks to present an inventory data-driven predictive-reactive production scheduling model that supports the evolving concepts of the Industry 4.0. Periodically, a machine learning technique provides predictive scheduling considering a best-case scenario according to an established Key Performance Indicator (KPI). Then, material non-availability causes disruptions in production, which triggers the Simulation-Based Optimization (SBO) method to handle these events. Thus, SBO provides a reactive schedule with the best set of priority rules to sequence jobs on each machine according to the data on the shop floor. This model was validated with a real case study using data collected from a metal-mechanical company. Considering the service level KPI, the results showed that the model is able to find a better solution in the compared scenarios. Therefore, even in a dynamic and stochastic scenario, with machine breakdowns, quality problems, raw material delays, and accuracy issues, the model proved efficient in mitigating these variations' effects. [ABSTRACT FROM AUTHOR]

A real-time capable core Ion Cyclotron Range of Frequencies (ICRF) heating model on NSTX and WEST is developed. The model is based on two nonlinear regression algorithms, the random forest ensemble of decision trees and the multilayer perceptron neural network. The algorithms are trained on TORIC ICRF spectrum solver simulations of the expected flat-top operation scenarios in NSTX and WEST assuming Maxwellian plasmas. The surrogate models are shown to successfully capture the multi-species core ICRF power absorption predicted by the original model for the high harmonic fast wave and the ion cyclotron minority heating schemes while reducing the computational time by six orders of magnitude. Although these models can be expanded, the achieved regression scoring, computational efficiency and increased model robustness suggest these strategies can be implemented into integrated modeling frameworks for real-time control applications.

Automated high-density storage systems (AHDSS) have attracted widespread attention in recent years owing to their advantages of high throughput and space utilisation. However, owing to the characteristics of large-scale, multi-disturbance, and short-period task scenarios, a system is required to make instant and efficient decisions. To this end, this paper proposes a data-driven real-time decision-making method to solve the real-time equipment scheduling and dynamic location assignment problem in AHDSS. The proposed method comprises two phases: decision scheme learning and real-time decision-making. The operation state attribute features of the AHDSS were constructed to generate training data for equipment scheduling and location assignment scheme learning. Thereafter, a hierarchical learning and decision-making mechanism based on the deep belief network (DBN) is proposed. The integrated learning of better scheduling solutions was realised by establishing three-stage models of lift selection, shuttle selection, and location priority. Additionally, the Taguchi method was adopted to determine the best performance parameters for DBNs at different learning stages. Compared with other well-known machine learning algorithms, DBNs have a higher learning accuracy. Finally, a real-world AHDSS problem is studied, and the results demonstrate that the proposed approach outperforms existing dispatching rules. [ABSTRACT FROM AUTHOR]

This study leveraged milling steel fibers to enhance the flexural capacity of ultra-high-performance concrete (UHPC). We assessed UHPC's flexural performance, measuring compressive strength and flowability, considering three fiber types: milling, straight, and a hybrid. UHPC with milling fibers displayed superior compressive strength and flowability compared to straight fiber. Milling fibers consistently enhanced UHPC's flexural capacity by over 120%, with a flowability reduction below 15%. Hybrid fiber demonstrated a balanced and commendable performance. Data-driven and theoretical flexural models were developed for UHPC with both fibers. the Gaussian process regression data-driven model outperformed the other 18 models, with an error below 3.5%. [ABSTRACT FROM AUTHOR]

Matrix cracking of composite laminates leads to stiffness degradation and dominates the medium leakage of liner-less composite vessels. A stiffness degradation model can be established through the machine learning method, which provides only discrete form equivalent moduli data without explicit analytical results. This paper further explored the use of the discrete form data set of stiffness degradation in the crack density estimation for composite laminates by using an iterative process. Furthermore, different mixed-mode fracture criteria were discussed. The results showed that Benzeggagh-Kenane (B-K) criterion and the power law criterion are more appropriate in mixed-mode crack density analyses. [ABSTRACT FROM AUTHOR]

The micro- and nanopore throats in shale oil reservoirs are finer than those in conventional oil reservoirs and have a larger specific surface area, potentially resulting in a more pronounced crude oil boundary effect. The prediction of recoverable reserves in shale oil reservoirs is influenced by factors such as geological complexity, fracture characteristics, and multiphase flow characteristics. The application of conventional reservoir seepage theories and engineering methods is challenging because of the unique characteristics of shale formations. A novel computational framework is proposed for the prediction of recoverable reserves and optimization of fracturing parameters by combining machine learning algorithms with causal discovery. Based on the theory of causal inference, the framework discovers the underlying causal relationships of the data, mines the internal laws of the data, and evaluates the causal effects, aiming to build an interpretable machine learning model to better understand the properties of shale oil reservoirs. Compared to traditional methods, the interpretable machine learning model has an outstanding prediction ability, with R2 of 0.94 and average error as low as 8.57%, which is 5.22% lower than that of traditional methods. Moreover, the maximum prediction error is only 21.84%, which is 25.2% smaller than the maximum error of traditional methods. The prediction robustness is good. An accurate prediction of recoverable reserves can be achieved. Furthermore, by integrating particle swarm optimization and TabNet, a fracturing parameter optimization model for shale oil reservoirs is developed. According to an on-site validation, this optimization results in an average increase of 13.45% in recoverable reserves. This study provides an accurate reference for reserve assessment and production design in the exploration and development of shale oil reservoirs. [ABSTRACT FROM AUTHOR]

Prediction of pressure evolution in liquid hydrogen (LH 2) tanks utilizing data-driven technology has gradually garnered significant attention. However, training an accurate data-driven model for pressure prediction in LH 2 tanks is a challenge due to inadequate experimental data. To end this, this paper presents a Backpropagation Neural Network model for pressure prediction in LH 2 tanks based on transfer learning on pre-trained models, using experimental data and results from thermodynamics models. Pre-trained models are firstly optimized by Bayesian optimization algorithms with large sample datasets from the thermal multi-zone model program and then performed the fine-tuning method with small sample datasets from experiments. The effectiveness of the proposed approach is validated by experimental data and numerical results. Compared with baseline models trained only with experimental data, the results demonstrate better model performance, with a maximum of 10.51% accuracy improvement. The proposed approach demonstrates considerable potential in facilitating the application of advanced machine learning algorithms for LH 2 tank pressure prediction, significantly reducing the dependence on experimental data collection. [Display omitted] • Data-driven pressure prediction model in liquid hydrogen tanks. • Data augmentation by integrating both experimental and simulation data. • Finding the optimal model by Bayesian optimization algorithms. • A maximum of 10.51% improvement in accuracy after transfer learning. [ABSTRACT FROM AUTHOR]

The accuracy of the control model is essential for the effectiveness of model-based control methods. However, factors such as model simplification, parameter variations, and environmental noise can introduce inaccuracies in vehicle state descriptions, thereby compromising the precision of path tracking. This study introduces data-driven enhancements for an MPC-based path tracking controller in autonomous vehicles (DD-PTC). The approach consists of two parts: firstly, Kolmogorov–Arnold Networks (KANs) are utilized to estimate tire lateral forces and correct tire cornering stiffness, thereby establishing a dynamic predictive model. Secondly, Gaussian Process Regression (GPR) is deployed to accurately capture the unmodeled dynamics of the vehicle to form a comprehensive control model. This enhanced model allows for precise path tracking through steering control. The superiority of DD-PTC is confirmed through extensive testing on the Simulink-CarSim simulation platform, where it consistently surpasses normal MPC and Linear Quadratic Regulator (LQR) strategies, especially in minimizing lateral distance errors under challenging driving conditions. [ABSTRACT FROM AUTHOR]

This work focusses on identifying the optimal stacking sequence for composite tubes in mountain bike frames using a data-driven model combined with evolutionary algorithms. The objective is to find a frame that is sufficiently stiff while meeting the requirements of weight, strength, and minimum tube wall thickness. The decision variables are the ply winding angles and the ply thicknesses of each tube. The study performs designs for two load cases – Starting and Uphill – and explores two types of winding: the gradual winding of individual layers (1ply) and the winding of layers between predefined inner and outer layers with variable thicknesses (TW). Additionally, the design process is applied to frames made of isotropic materials, such as steel, aluminium, and titanium, using the same methodology to allow for comparison of results. The article demonstrates the successful application of this methodology to common sports equipment, suggesting its potential for beneficial use in other common composite frame structures. [ABSTRACT FROM AUTHOR]

Data‐driven artificial intelligence (DDAI) prediction has gained much attention, especially in recent years, because of its power and flexibility compared to traditional approaches. In hydrology, streamflow forecasting is one of the areas that took advantage of utilizing DDAI‐based forecasting, given the weakness of the old approaches (e.g., physical‐based approaches). Since many different techniques and tools have been used for streamflow forecasting, there is a new way to explore them. This manuscript reviews the recent (2011–2023) applications of DDAI in streamflow prediction. It provides a background of DDAI‐based techniques, including machine learning algorithms and methods for pre‐processing the data and optimizing or enhancing the machine learning approaches. We also explore the applications of DDAI techniques in streamflow forecasting. Finally, the most common tools for utilizing DDAI techniques in streamflow forecasting are presented. [ABSTRACT FROM AUTHOR]

Purpose: Respiratory motion (RM) significantly impacts image quality in thoracoabdominal PET/CT imaging. This study introduces a unified data-driven respiratory motion correction (uRMC) method, utilizing deep learning neural networks, to solve all the major issues caused by RM, i.e., PET resolution loss, attenuation correction artifacts, and PET-CT misalignment. Methods: In a retrospective study, 737 patients underwent [18F]FDG PET/CT scans using the uMI Panorama PET/CT scanner. Ninety-nine patients, who also had respiration monitoring device (VSM), formed the validation set. The remaining data of the 638 patients were used to train neural networks used in the uRMC. The uRMC primarily consists of three key components: (1) data-driven respiratory signal extraction, (2) attenuation map generation, and (3) PET-CT alignment. SUV metrics were calculated within 906 lesions for three approaches, i.e., data-driven uRMC (proposed), VSM-based uRMC, and OSEM without motion correction (NMC). RM magnitude of major organs were estimated. Results: uRMC enhanced diagnostic capabilities by revealing previously undetected lesions, sharpening lesion contours, increasing SUV values, and improving PET-CT alignment. Compared to NMC, uRMC showed increases of 10% and 17% in SUVmax and SUVmean across 906 lesions. Sub-group analysis showed significant SUV increases in small and medium-sized lesions with uRMC. Minor differences were found between VSM-based and data-driven uRMC methods, with the SUVmax was found statistically marginal significant or insignificant between the two methods. The study observed varied motion amplitudes in major organs, typically ranging from 10 to 20 mm. Conclusion: A data-driven solution for respiratory motion in PET/CT has been developed, validated and evaluated. To the best of our knowledge, this is the first unified solution that compensates for the motion blur within PET, the attenuation mismatch artifacts caused by PET-CT misalignment, and the misalignment between PET and CT. [ABSTRACT FROM AUTHOR]

The advance of high-throughput biomedical technology makes it possible to access massive measurements of gene expression levels. An important statistical issue is identifying both under-expressed and over-expressed genes for a disease. Most existing multiple-testing procedures focus on selecting only the non-null or significant genes without further identifying their expression type. Only limited methods are designed for the directional problem, and yet they fail to separately control the numbers of falsely discovered over-expressed and under-expressed genes with only a unified index combining all the false discoveries. In this paper, based on a three-classification multiple testing framework, we propose a practical data-driven procedure to control separately the two directions of false discoveries. The proposed procedure is theoretically valid and optimal in the sense that it maximizes the expected number of true discoveries while controlling the false discovery rates for under-expressed and over-expressed genes simultaneously. The procedure allows different nominal levels for the two directions, exhibiting high flexibility in practice. Extensive numerical results and analysis of two large-scale genomic datasets show the effectiveness of our procedure. [ABSTRACT FROM AUTHOR]

Since 2002, with the SARS outbreak, infectious diseases, including the ongoing COVID-19 pandemic, have continued to be a major global public health threat. It is critical to develop effective data science methods to quickly detect disease outbreaks and contain their rapid globalized spread. However, in practice, limited testing availability, and thus insufficient testing data poses significant challenges in effective analysis and real-time monitoring of infectious diseases, especially during early stages of a novel disease outbreak. To tackle these challenges, this article proposes adaptive testing resource allocation strategies integrated with a physics-informed model to dynamically allocate limited testing resources across different communities. The physics-informed model accounts for transmission dynamics and health disparities, enabling effective health risk assessment despite limited data. By integrating nonstationary Multi-Armed Bandit techniques that strike superior balance between exploring the communities with high uncertain risks and exploiting those with high risk levels, the proposed methodology facilitates test allocation to collect high-quality testing data for early outbreak detection. Theoretical analysis is carried out to evaluate the performance of the proposed allocation strategies, ensuring either sublinear or linear dynamic pseudo-regret under regularity assumptions. A comprehensive simulation study is conducted under three transmission scenarios to thoroughly evaluate the proposed methodology. [ABSTRACT FROM AUTHOR]

Time-varying system is a widespread class of systems in reality. Model of such system is difficult to obtain because the parameters vary with time. Hence, a data-driven state estimation for unknown discrete time-varying system is investigated with zonotopic reachability analysis. Since the system model is unknown, a time-varying matrix zonotope which contains all possible models is computed by the available prior input-state trajectories. On this basis, an over-approximated reachable zonotope of system state is obtained through iteration. Then, the boundedness of over-approximated reachable zonotope is proved by introducing P -radius definition. This method guarantees that the actual system state falls within the estimation range and maintains the compactness of the estimation result. Finally, a numerical example and a circuit system simulation show the validity and applicability of the proposed method. • The prior input-state trajectories are utilized. • Zonotope is introduced in data-driven estimation. • The system model is represented by a matrix zonotope. • The matrix zonotope updates over time. • The data-driven estimation contains all possible syatem states. [ABSTRACT FROM AUTHOR]

Fairness in peer review is of vital importance in academic activities. Current peer review systems focus on matching suitable experts with proposals but often ignore the existence of outliers. Previous research has shown that outlier scores in reviews could decrease the fairness of these systems. Therefore, outlier detection in peer review systems is essential for maintaining fairness. In this paper, we introduce a novel method that employs data-crossing analysis to detect outlier scores, aiming to improve the reliability of peer review processes. We utilize a confidential dataset from a review organization. Due to the inability to access ground truth scores, we systematically devise data-driven deviations from an estimated ground truth through data-crossing analysis. These deviations reveal inconsistencies and abnormal scoring behaviors of different reviewers. Subsequently, the review process is strengthened by providing a structured mechanism to identify and mitigate biases. Extensive experiments demonstrate its effectiveness in improving the accuracy and fairness of academic assessments, contributing to the broader application of AI-driven methodologies to achieve more reliable and equitable outcomes. [ABSTRACT FROM AUTHOR]

This study addresses the coherence between performance and resource investment because prior studies have suggested diverse approaches to assessing performance in achieving the sustainable development goals (SDGs). However, a serious imbalance in resource investment still exists in current practices because there is an urgent need for countries to pursue the SDGs to achieve the 2030 Agenda. Paraguay is striving to issue sustainable bonds to address the SDGs. To fill the gaps in relevant research, this study proposes a data-driven hybrid method that (1) employs a normalization procedure to obtain unit-free values for comparison, (2) adopts factor analysis (FA) and reliability testing (RT) to ensure the validity and reliability of the structured goals, and (3) uses performance-resource analysis to identify the coherence between performance and resource investment. The results show that performance and resource investment for zero hunger (SDG 2) and decent work and economic growth (SDG 8) are below average and urgently need improvement. Responsible consumption and production (SDG 12) is the SDG for which Paraguay has reached a level of coherence. Most provinces in Paraguay are overconcentrating on a single goal in pursuing the SDGs. Alto Paraná, Itapúa and the Central Department are provinces in rich areas that are enabling the country to reach a level of coherence. However, Itapúa is the only province that is balancing at least four SDGs. [ABSTRACT FROM AUTHOR]

Electric vehicles (EVs) powered by lithium-ion batteries have emerged as a global development trend. To ensure the safe and stable driving of EVs, it is imperative to address battery safety and thermal management issues, which rely heavily on the precise state-of-charge (SOC) estimation of the battery. However, estimating SOC under uncontrolled environmental temperatures remains an unresolved challenge. This study proposes a patch-level representation learning model based on domain knowledge to estimate the SOC over a wide temperature range. First, patches were adopted as inputs instead of traditional points, thereby mitigating error accumulation and capturing dynamic changes in the battery from these more informative representations. Second, the open-circuit voltage (OCV)-SOC-temperature relationship was incorporated to obtain the temperature-related SOC priors. Subsequently, the prior was updated recursively along the time dimension to obtain a more precise SOC estimate. The accuracy of the proposed model was confirmed experimentally for three driving cycles at six ambient temperatures, significantly reducing the root mean square error by 48.19% compared to popular existing models. Notably, the performance of the proposed method had an excellent improvement of 51.52% and 57.20% at −10°C and −20°C, respectively. Moreover, the parameter size of the proposed method was 39.748 KB, which significantly promoted the deployment and application of data-driven models in the real world. [ABSTRACT FROM AUTHOR]

Buffer allocation, which is an important research topic in manufacturing system design, typically focuses on system performance and cost. However, few previous studies have been performed to investigate energy-saving buffer allocation, which can decrease operational energy consumption in green manufacturing. Furthermore, the computational efficiency of solving the buffer allocation problem requires further investigation. This paper proposes a data-driven hybrid algorithm based on multi-evolutionary sampling strategies for solving energy-saving buffer allocation that can maximize the throughput rate and minimize energy consumption. Two evolutionary sampling strategies, that is, global search and surrogate-assisted local search, are integrated to balance exploitation and exploration. In addition, a database containing historical data pertaining to buffer allocation solutions is used to develop surrogate models that can rapidly predict the throughput and energy consumption and improve the evaluation efficiency of the local search strategy. Experimental results demonstrate the efficacy of the proposed algorithm. This study contributes to an efficient buffer allocation and presents a new perspective on energy-saving measures for green manufacturing designs. [ABSTRACT FROM AUTHOR]

Neural implicit representations, including Neural Distance Fields and Neural Radiance Fields, have demonstrated significant capabilities for reconstructing surfaces with complicated geometry and topology, and generating novel views of a scene. Nevertheless, it is challenging for users to directly deform or manipulate these implicit representations with large deformations in a real-time fashion. Gaussian Splatting (GS) has recently become a promising method with explicit geometry for representing static scenes and facilitating high-quality and real-time synthesis of novel views. However, it cannot be easily deformed due to the use of discrete Gaussians and the lack of explicit topology. To address this, we develop a novel GS-based method (GaussianMesh) that enables interactive deformation. Our key idea is to design an innovative mesh-based GS representation, which is integrated into Gaussian learning and manipulation. 3D Gaussians are defined over an explicit mesh, and they are bound with each other: the rendering of 3D Gaussians guides the mesh face split for adaptive refinement, and the mesh face split directs the splitting of 3D Gaussians. Moreover, the explicit mesh constraints help regularize the Gaussian distribution, suppressing poor-quality Gaussians (e.g., misaligned Gaussians, long-narrow shaped Gaussians), thus enhancing visual quality and reducing artifacts during deformation. Based on this representation, we further introduce a large-scale Gaussian deformation technique to enable deformable GS, which alters the parameters of 3D Gaussians according to the manipulation of the associated mesh. Our method benefits from existing mesh deformation datasets for more realistic data-driven Gaussian deformation. Extensive experiments show that our approach achieves high-quality reconstruction and effective deformation, while maintaining the promising rendering results at a high frame rate (65 FPS on average on a single commodity GPU). [ABSTRACT FROM AUTHOR]

Seismic inversion can be divided into time-domain inversion and frequency-domain inversion based on different transform domains. Time-domain inversion has stronger stability and noise resistance compared to frequencydomain inversion. Frequency domain inversion has stronger ability to identify small-scale bodies and higher inversion resolution. Therefore, the research on the joint inversion method in the time-frequency domain is of great significance for improving the inversion resolution, stability, and noise resistance. The introduction of prior information constraints can effectively reduce ambiguity in the inversion process. However, the existing modeldriven time-frequency joint inversion assumes a specific prior distribution of the reservoir. These methods do not consider the original features of the data and are difficult to describe the relationship between time-domain features and frequency-domain features. Therefore, this paper proposes a high-resolution seismic inversion method based on joint data-driven in the time-frequency domain. The method is based on the impedance and reflectivity samples from logging, using joint dictionary learning to obtain adaptive feature information of the reservoir, and using sparse coefficients to capture the intrinsic relationship between impedance and reflectivity. The optimization result of the inversion is achieved through the regularization term of the joint dictionary sparse representation. We have finally achieved an inversion method that combines constraints on time-domain features and frequency features. By testing the model data and field data, the method has higher resolution in the inversion results and good noise resistance. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Donghua University (Natural Science Edition) is the property of Journal of Donghua University (Natural Science) Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Deformation is a critical indicator for the safety control of high‐arch dams, yet traditional statistical regression methods often exhibit poor predictive performance when applied to long‐sequence time series data. In this study, we develop a robust predictive model for deformation behavior in high‐arch dams by integrating signal dimensionality reduction with deep learning (DL)‐based residual correction techniques. First, the fast Fourier transform is employed to decompose air and water temperature sequences, enabling the extraction of temperature cycle characteristics at the dam boundary. A data‐driven statistical monitoring model for dam deformation, based on actual temperature data, is then proposed. Subsequently, an improved Bayesian Ridge regression model is used to construct the dam deformation monitoring framework. The residuals that traditional statistical methods fail to capture are input into an enhanced Long Short‐Term Memory (LSTM) network to effectively learn the temporal characteristics of the sequence. A high‐arch dam with a history of long‐term service is used as a case study. Experimental results indicate that the data dimensionality reduction method effectively extracts relevant information from observed temperature data, reducing the number of input variables. Comparative evaluation experiments show that the proposed hybrid predictive model outperforms existing state‐of‐the‐art benchmark algorithms in terms of predictive efficiency and accuracy. Additionally, this approach combines the interpretability of statistical regression methods with the powerful nonlinear modeling capabilities of DL‐based models, achieving a synergistic effect. [ABSTRACT FROM AUTHOR]

Aiming at the uncertainty of target motion and observation models in multi-maneuvering target tracking (MMTT), this study presents an innovative data-driven approach based on a Gaussian process (GP). Traditional multi-model (MM) methods rely on a predefined set of motion models to describe target maneuvering. However, these methods are limited by the finite number of available models, making them unsuitable for handling highly complex and dynamic real-world scenarios, which, in turn, restricts the adaptability and flexibility of the filter. In addition, traditional methods often assume that observation models follow ideal linear or simple nonlinear relationships. However, these assumptions may be biased in actual application and so lead to degradation in tracking performance. To overcome these limitations, this study presents a learning-based algorithm-leveraging GP. This non-parametric GP approach enables learning an unlimited range of target motion and observation models, effectively mitigating the problems of model overload and mismatch. This improves the algorithm's adaptability in complex environments. When the motion and observation models of multiple targets are unknown, the learned models are incorporated into the cubature Kalman probability hypothesis density (PHD) filter to achieve an accurate MMTT estimate. Our simulation results show that the presented approach delivers high-precision tracking of complex multi-maneuvering target scenarios, validating its effectiveness in addressing model uncertainty. [ABSTRACT FROM AUTHOR]

Existing dam displacement statistical methods simulate the thermal effects using simple harmonic functions ignoring the effects of ice periods, extreme heat, and seasonal weather. Moreover, existing data‐driven methods usually utilize a separate modeling strategy, inevitably ignoring the spatiotemporal correlation of multiple displacement points in dams, resulting in poor predictive performance. To overcome these shortcomings, this study proposes a novel machine learning (ML)—aided multiple‐point dam displacement predictive model considering the temperature hysteresis effect. Firstly, an improved hydraulic‐Air_temperture_Time (HTairT) statistical monitoring model is developed using the measured air temperature lagging monitoring data. On this basis, the multitask Gaussian process regression (multipoint GPR) algorithm with an improved kernel function to construct a multipoint deformation prediction model for ultrahigh arch dams. Then, the improved meta‐heuristic physics‐driven Frost algorithm is utilized to determine the optimal parameters of the multipoint GPR model. A high arch dam with a height of 305 m is used as the case study, and five displacement monitoring points are used for validation. Five advanced ML‐based algorithms are used to comparatively evaluate and verify the performance of the proposed method in terms of forecast accuracy and interpretability. The HTairT statistical model can better simulate the hysteresis effect of temperature on dam deformation. Moreover, the Frost‐optimized dam multipoint displacement prediction model with the RQ kernel functions outperforms the other comparison methods in terms of R2, mean absolute error (MAE), and root mean squared error (RMSE) evaluation indicators. This indicates the proposed method can mine the spatiotemporal correlation among multiple monitoring points of ultrahigh arch dams, further improving the overall deformation prediction and uncertainty estimation. [ABSTRACT FROM AUTHOR]

Functional coatings, including organic and inorganic coatings, play a vital role in various industries by providing a protective layer and introducing unique functionalities. However, its design often involves time‐consuming experimentation with multiple materials and processing parameters. To overcome these limitations, data‐driven approaches are gaining traction in materials science. In this paper, recent advances in data‐driven materials research and development (R&D) for functional coatings, highlighting the importance, data sources, working processes, and applications of this paradigm are summarized. It is begun by discussing the challenges of traditional methods, then introduce typical data‐driven processes. It is demonstrated how data‐driven approaches enable the identification of correlations between input parameters and coating performance, thus allowing for efficient prediction and design. Furthermore, carefully selected case studies are presented across diverse industries that exemplify the effectiveness of data‐driven methods in accelerating the discovery of new functional coatings with tailored properties. Finally, the emerging research directions, involving integrating advanced techniques and data from different sources, are addressed. Overall, this review provides an overview of data‐driven materials R&D for functional coatings, shedding light on its potential and future developments. [ABSTRACT FROM AUTHOR]

Bio-inspired robots are devices that mimic an animal's motions and structures in nature. Worm robots are robots that are inspired by the movements of the worm in nature. This robot has different applications such as medicine and rescue plans. However, control of the worm robot is a challenging task due to the high-nonlinearity dynamic model and external noises that are applied to that robot. This research uses an optimal data-driven controller to control the worm robot. First, data are obtained from the nonlinear model of the worm robot. Then, the Koopman theory is used to generate a linear dynamic model of the Worm robot. The dynamic mode decomposition (DMD) method is used to generate the Koopman operator. Finally, a linear quadratic regulator (LQR) control method is applied for the control of the worm robot. The simulation results verify the performance of the proposed control method. [ABSTRACT FROM AUTHOR]

To ensure the operational safety of oil transportation stations, it is crucial to predict the impact of pressure and temperature before crude oil enters the pipeline network. Accurate predictions enable the assessment of the pipeline's load-bearing capacity and the prevention of potential safety incidents. Most existing studies primarily focus on describing and modeling the mechanisms of the oil flow process. However, monitoring data can be skewed by factors such as instrument aging and pipeline friction, leading to inaccurate predictions when relying solely on mechanistic or data-driven approaches. To address these limitations, this paper proposes a Temporal-Spatial Three-stream Temporal Convolutional Network (TS-TTCN) model that integrates mechanistic knowledge with data-driven methods. Building upon Temporal Convolutional Networks (TCN), the TS-TTCN model synthesizes mechanistic insights into the oil transport process to establish a hybrid driving mechanism. In the temporal dimension, it incorporates real-time operating parameters and applies temporal convolution techniques to capture the time-series characteristics of the oil transportation pipeline network. In the spatial dimension, it constructs a directed topological map based on the pipeline network's node structure to characterize spatial features. Data analysis and experimental results show that the Three-stream Temporal Convolutional Network (TTCN) model, which uses a Tanh activation function, achieves an error rate below 5%. By analyzing and validating real-time data from the Dongying oil transportation station, the proposed hybrid model proves to be more stable, reliable, and accurate under varying operating conditions. [ABSTRACT FROM AUTHOR]

The semiconductor industry faces increasing pressure to improve energy efficiency while maintaining competitiveness and sustainability. Apart from more conventional energy efficiency measures look at equipment modernization and process and design optimization, this paper explores the potential of data-driven approaches to address these challenges and optimize energy consumption across both the facility and manufacturing space of a semiconductor manufacture plant. By harnessing advanced analytics, machine learning algorithms, and IoT technologies, semiconductor manufacturers can gain real-time insights into energy usage patterns, and identify areas of opportunities that leads to the implementation of targeted interventions to optimize performance. The paper first looks into the challenges and measures of enabling and enhancing data visibility which is the foundation of the data-driven approach, then it examines case studies, best practices and various systematic approaches, demonstrating the transformative impact of data-driven energy efficiency measures which leads to operational efficiency, cost reduction, and environmental sustainability. Ultimately, this paper aims to provide a fresh angle into the energy efficiency study for peers in semiconductor industries to leverage in their journey towards a more sustainable and energy efficient future. [ABSTRACT FROM AUTHOR]

The synthesis, characterization, and application of carbon nanotubes (CNTs) have long posed significant challenges due to the inherent multiple complexity nature involved in their production, processing, and analysis. Recent advancements in machine learning (ML) have provided researchers with novel and powerful tools to address these challenges. This review explores the role of ML in the field of CNT research, focusing on how ML has enhanced CNT research by (1) revolutionizing CNT synthesis through the optimization of complex multivariable systems, enabling autonomous synthesis systems, and reducing reliance on conventional trial-and-error approaches; (2) improving the accuracy and efficiency of CNT characterizations; and (3) accelerating the development of CNT applications across several fields such as electronics, composites, and biomedical fields. This review concludes by offering perspectives on the future potential of integrating ML further into CNT research, highlighting its role in driving the field forward. [ABSTRACT FROM AUTHOR]

Accurate energy consumption prediction in the injection molding process is crucial for optimizing energy efficiency in polymer processing. Traditional parameter optimization methods face challenges in achieving optimal energy prediction due to complex energy transmission. In this study, a data-driven approach based on the Rolling Learning Informer model is proposed to enhance the accuracy and adaptability of energy consumption forecasting. The Informer model addresses the limitations of long-sequence prediction with sparse attention mechanisms, self-attention distillation, and generative decoder techniques. Rolling learning prediction is incorporated to enable continuous updating of the model to reflect new data trends. Experimental results demonstrate that the RL-Informer model achieves a normalized root mean square error of 0.1301, a root mean square error of 0.0758, a mean absolute error of 0.0562, and a coefficient of determination of 0.9831 in energy consumption forecasting, outperforming other counterpart models like Gated Recurrent Unit, Temporal Convolutional Networks, Long Short-Term Memory, and two variants of the pure Informer models without Rolling Learning. It is of great potential for practical engineering applications. [ABSTRACT FROM AUTHOR]

KDDSL open the black box of neural network to achieve interpretable prediction of SL (The two shields represent two SL genes). This can help us utilize SL to kill cancer cell. [Display omitted] • Synthetic lethality offers a promising means for expanding drug target space for cancer treatment. • We developed a knowledge and data dual-driven AI framework for synthetic lethality prediction (KDDSL). • KDDSL is a "visible" AI framework that integrates domain knowledge to explore synthetic lethality and its implications in cancer treatment. This "visible" framework enhances the trust between biologists and AI. • KDDSL can help in identifying SL interactions and guiding the design of synergistic anti-cancer combination therapies. • By applying KDDSL, we identified the combination of MDM2 and CDK9 inhibitors, which demonstrated significant anticancer effects in vitro and in vivo. Synthetic lethality (SL) provides an opportunity to leverage different genetic interactions when designing synergistic combination therapies. To further explore SL-based combination therapies for cancer treatment, it is important to identify and mechanistically characterize more SL interactions. Artificial intelligence (AI) methods have recently been proposed for SL prediction, but the results of these models are often not interpretable such that deriving the underlying mechanism can be challenging. This study aims to develop an interpretable AI framework for SL prediction and subsequently utilize it to design SL-based synergistic combination therapies. We propose a knowledge and data dual-driven AI framework for SL prediction (KDDSL). Specifically, we use gene knowledge related to the SL mechanism to guide the construction of the model and develop a method to identify the most relevant gene knowledge for the predicted results. Experimental and literature-based validation confirmed a good balance between predictive and interpretable ability when using KDDSL. Moreover, we demonstrated that KDDSL could help to discover promising drug combinations and clarify associated biological processes, such as the combination of MDM2 and CDK9 inhibitors, which exhibited significant anti-cancer effects in vitro and in vivo. These data underscore the potential of KDDSL to guide SL-based combination therapy design. There is a need for biomedicine-focused AI strategies to combine rational biological knowledge with developed models. [ABSTRACT FROM AUTHOR]

This paper addresses the adaptive wavelet estimations for density derivatives by using data-driven methods. Based on the classical linear wavelet estimator of density derivatives, we provide a point-wise estimation under the local Hölder condition firstly. Moreover, we introduce a data-driven wavelet estimator for adaptivity and prove a point-wise oracle inequality, which does not require any assumption on the underlying function. Finally, by using the point-wise oracle inequality, the point-wise estimation under the local Hölder condition and L p -risk ( 1 ≤ p < ∞ ) estimation on Besov spaces are investigated respectively. [ABSTRACT FROM AUTHOR]

Ship fuel consumption plays a crucial role not only in understanding ships' energy efficiency but also in gaining insights into their emissions. However, enhancing the accuracy of these predictions poses significant challenges due to data limitations and the methods employed. Due to these factors, such as data variability and equipment characteristics, ship fuel consumption exhibits certain fluctuations under specific conditions. Previous fuel consumption prediction methods primarily generate a single specific value, making it difficult to capture the volatility of and variability in fuel consumption. To overcome this limitation, this paper proposes a novel method that integrates Gaussian process prediction with quantile regression theory to perform interval predictions of ship fuel consumption, providing a range of possible outcomes. Through comparative analyses with traditional methods, the possibility of using the method is verified and its results are validated. The results indicate the following: (1) at a 95% confidence level, the proposed method achieves a prediction interval coverage probability of 0.98 and a prediction interval normalized average width of 0.123, which are significantly better than those of the existing backpropagation neural network (BPNN) and gradient boosting decision tree (GBDT) quantile regression models; (2) the prediction accuracy of the proposed method is 92% for point forecasts; and (3) the proposed method is applicable to main datasets, including both noon report and sensor datasets. These findings provide valuable insights into interval predictions of ship fuel consumption and highlight their potential applications in related fields, emphasizing the importance of accurate interval predictions in intelligent energy efficiency optimization. [ABSTRACT FROM AUTHOR]

By analyzing the results of compliance minimization of thermoelastic structures, we observed that microstructures play an important role in this optimization problem. Then, we propose to use a multiple variable cutting (M–VCUT) level set-based model of microstructures to solve the concurrent two–scale topology optimization of thermoelastic structures. A microstructure is obtained by combining multiple virtual microstructures that are derived respectively from multiple microstructure prototypes, thus giving more diversity of microstructure and more flexibility in design optimization. The effective mechanical properties of microstructures are computed in an off-line phase by using the homogenization method, and then a mapping relationship between the design variables and the effective properties is established, which gives a data-driven model of microstructure. In the online phase, the data-driven model is used in the finite element analysis to improve the computational efficiency. The compliance minimization problem is considered, and the results of numerical examples prove that the proposed method is effective. [ABSTRACT FROM AUTHOR]

In modern aerospace engineering, fault-tolerant control (FTC) has become increasingly vital for maintaining spacecraft functionalities, including attitude control. The attitude determination module incorporates data from different sensors. Star trackers and gyroscopes provide attitude and angular rate measurements affected by noise. To minimize the latter, the so-called gyro-stellar estimator exploits optimal state estimation such as Kalman filters. However, gyros are subject to drifts which beyond some limits could affect estimator convergence. In this paper, a hybrid supervision scheme is proposed for attitude control affected by gyroscope faults. The data-driven fault detection and identification (FDI) module localizes and coarsely identifies the emerging drift. Then, the supervision system reconfigures the gyro-stellar estimator and updates the controller gains to guarantee FTC capability. The final hybrid system ensures attitude control function until fine-tuning is performed by attitude determination and control engineers. Simulation results with in-orbit data validate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Thermal metamaterials are typically achieved by mixing different natural materials to realize effective thermal conductivities (ETCs) that conventional materials do not possess. However, the necessity for multifunctional design of metamaterials, encompassing both thermal and mechanical functionalities, is somewhat overlooked, resulting in the fixation of mechanical properties in thermal metamaterials designed within current research endeavors. Thus far, conventional methods have faced challenges in designing thermal metamaterials with configurable mechanical properties because of intricate inherent relationships among the structural configuration, thermal and mechanical properties in metamaterials. Here, a data‐driven approach is proposed to design a thermal metamaterial capable of seamlessly achieving thermal functionalities and harnessing the advantages of microstructural diversity to configure its mechanical properties. The designed metamaterial possesses thermal cloaking functionality while exhibiting exceptional mechanical properties, such as load‐bearing capacity, shearing strength, and tensile resistance, thereby affording mechanical protection for the thermal metadevice. The proposed approach can generate numerous distinct inverse design candidate topological functional cells (TFCs), designing thermal metamaterials with dramatic improvements in mechanical properties compared to traditional ones, which sets up a novel paradigm for discovering thermal metamaterials with extraordinary mechanical structures. Furthermore, this approach also paves the way for investigating thermal metamaterials with additional physical properties. [ABSTRACT FROM AUTHOR]

A central principle in neuroscience is that neurons within the brain act in concert to produce perception, cognition, and adaptive behavior. Neurons are organized into specialized brain areas, dedicated to different functions to varying extents, and their function relies on distributed circuits to continuously encode relevant environmental and body-state features, enabling other areas to decode (interpret) these representations for computing meaningful decisions and executing precise movements. Thus, the distributed brain can be thought of as a series of computations that act to encode and decode information. In this perspective, we detail important concepts of neural encoding and decoding and highlight the mathematical tools used to measure them, including deep learning methods. We provide case studies where decoding concepts enable foundational and translational science in motor, visual, and language processing. Understanding how the brain works has been a formidable challenge in neuroscience, but advances at the computational level are now driving us closer. This perspective discusses core concepts of neural encoding and decoding and how moving toward causal modeling can enable translational and foundational insights into the neural code. [ABSTRACT FROM AUTHOR]

Speech emotion recognition plays a crucial role in analyzing psychological disorders, behavioral decision‐making, and human‐machine interaction applications. However, the majority of current methods for speech emotion recognition heavily rely on data‐driven approaches, and the scarcity of emotion speech datasets limits the progress in research and development of emotion analysis and recognition. To address this issue, this study introduces a new English speech dataset specifically designed for emotion analysis and recognition. This dataset consists of 5503 voices from over 60 English speakers in different emotional states. Furthermore, to enhance emotion analysis and recognition, fast Fourier transform (FFT), short‐time Fourier transform (STFT), mel‐frequency cepstral coefficients (MFCCs), and continuous wavelet transform (CWT) are employed for feature extraction from the speech data. Utilizing these algorithms, the spectrum images of the speeches are obtained, forming four datasets consisting of different speech feature images. Furthermore, to evaluate the dataset, 16 classification models and 19 detection algorithms are selected. The experimental results demonstrate that the majority of classification and detection models achieve exceptionally high recognition accuracy on this dataset, confirming its effectiveness and utility. The dataset proves to be valuable in advancing research and development in the field of emotion recognition. [ABSTRACT FROM AUTHOR]

With the advancement of artificial intelligence technology, the field of space satellite tracking is also progressing towards intelligent systems. However, directly incorporating a purely data-driven learning approach into a satellite control system not only exhibits low efficiency but also necessitates a substantial number of interactive samples. To address these limitations, we propose an innovative data-knowledge joint-driven reinforcement learning algorithm based on guided policy and state-prediction (GP-SPRL) for continuous-thrust satellite tracking problems. By assuming linear quadratic tracking cost, GP-SPRL leverages orbital dynamics to facilitate an intelligent learning process and enables bidirectional guidance and update through knowledge-driven and data-driven approaches. Furthermore, to mitigate high interaction costs, GP-SPRL employs real samples to generate predicted states using Gaussian processes, thereby reducing the need for extensive interactions. Simulation results validate the efficacy of joint-driven in GP-SPRL by enhancing learning efficiency and minimizing reliance on real samples through predicted state utilization. [ABSTRACT FROM AUTHOR]

Supercritical water gasification (SCWG) has been utilized for producing hydrogen from organic wastes, which is associated with sustainable development. Nevertheless, identifying the optimal SCWG conditions and appropriate catalysts for diverse waste materials to yield hydrogen-rich syngas is consistently a laborious and costly endeavor. The current research presents a comprehensive framework that combines machine learning models (Decision Tree, Ensembled Learning Tree, Support Vector Machine, and Gradient Boost Regression) with Differential Evolution Optimization (DEO) to predict and optimize hydrogen production from Supercritical Water Gasification (SCWG) using 24 different biomass characteristics and process parameters.The findings indicate that ELA-DEO emerges as the preferred method for forecasting hydrogen yield (R2 test = 0.95, RMSE Test = 0.091), demonstrating its effectiveness in handling intricate variable-target relationships. Conversely, Support Vector Machine (SVM) displayed weaker performance with an R2 Test of 0.73 and RMSE Test of 0.116. In the SHAP feature importance analysis, temperature, catalyst amount, catalyst type, hydrogen and oxygen were highlighted as important parameters affecting the process. In addition, by fine-tuning the machine learning hyperparameters, the DEO optimization method was used to maximize the production of H 2. The optimized ELA-DEO model was used to identify the ideal features and conditions applicable to our experimental setup. Based on these findings, a recommended biomass table was finally formulated, which was validated by the ELA-DEO model. According to our laboratory results, Dunaliella salina produced 12 mmol of hydrogen with a 35% selectivity. Black liquor with 3% (wt) wood has the capacity to produce 17.21 mmol of hydrogen. Also, when crude glycerol was reacted with RuCl 3 , 18.7 mmol of hydrogen were generated. However, Dunaliella salina produced a 64% mole fraction of total gas output, which matched the maximum value predicted by our ELA-DEO models. Schematic diagram of the machine learning based optimization for H 2 -rich syngas production. [Display omitted] • An innovative ensemble forecasting method evolved to estimate SCWG (Supercritical water gasification) performance. • The DEO (Differential Evolution Optimization) methodology optimizes H2 yields using SHAP (SHapley Additive exPlanations) value and feature importance. • Offer desired biomass to maximize hydrogen yield, which adopts the temperature till 430C and pressure 22–24 MPa. • PDP (Partial Dependence Plot) facilitates comparative study to determine the best value for a feature. [ABSTRACT FROM AUTHOR]

The rise of digital twin technology has provided innovative methods for monitoring and optimizing ship cooling pumps. This paper proposes a digital twin-based framework for the status monitoring and visualization of ship cooling pumps. By establishing highly realistic physical and mathematical models and integrating actual operational data, a comprehensive virtual environment was created to simulate the operational status of ship cooling pumps. Using the random forest algorithm for data training and testing, the results showed that the root mean square error for the training set was 0.0037873, and for the test set, it was 0.008929, indicating high accuracy in predicting the status of cooling pumps. This system enables real-time monitoring, problem diagnosis, performance optimization, and decision support for cooling pumps. This study aims to leverage digital twin technology to design and apply a visualization monitoring system to enhance the intelligence of ship operation and maintenance. [ABSTRACT FROM AUTHOR]

In this paper, a novel data-driven distributed optimal control (DDOC) scheme via neighbourhood optimization is proposed for affine nonlinear interconnected systems with unknown dynamics. In this scheme, each subsystem merely broadcasts its local information to its neighbours, and receives information only from them. The proposed distributed method is established based on a hybrid structure mainly consisting of both algebraic and analytical parts, in which two neural networks (NN) are simultaneously tuned in an online manner by using neighbourhood information for each subsystem. To tackle the unknown dynamics, a new distributed data-driven model is established to reconstruct subsystem dynamics and estimate system states. Moreover, by using the Hamilton–Jacobi–Bellman (HJB) based formulation, a distributed NN-based optimal control is designed to minimize both the performance of the corresponding subsystem and its neighbours, in the presence of unknown dynamics. Through Lyapunov's direct method, the uniformly ultimately boundedness (UUB) of the overall closed-loop system is guaranteed, and the provided results of the stability are exploited to find updating rules for the NN-based observer and controller. Finally, two illustrative examples are investigated to certify the efficacy of the new distributed optimal control (DOC). [ABSTRACT FROM AUTHOR]