Traditional Physical Model Research Articles

Model Construction of College Engineering Drawing Teaching System Based on Artificial Intelligence Network Technology

The integration of artificial intelligence (AI) in education, particularly in engineering drawing courses, enhances content richness and engagement, stimulating student enthusiasm and improving teaching quality. This paper examines the design of a multimedia teaching system incorporating virtual reality (VR) for engineering drawing. Key technologies such as graphics processing, 3D modeling, and dynamic databases are researched. Tools like Dreamweaver, Fireworks, SolidWorks, and 3DS MAX are used to create and integrate 3D models into web pages, forming a 3D model library. This library provides a flexible, content-rich, online learning environment that promotes self-study, enhances spatial imagination, and improves the quality of engineering drawing education. Establishing a virtual model library supports multimedia teaching and replaces traditional physical model rooms.

Deep Learning Ensemble for Flood Probability Analysis

Predicting flood events is complex due to uncertainties from limited gauge data, high data and computational demands of traditional physical models, and challenges in spatial and temporal scaling. This research innovatively uses only three remotely sensed and computed factors: rainfall, runoff and temperature. We also employ three deep learning models—Feedforward Neural Network (FNN), Convolutional Neural Network (CNN), and Long Short-Term Memory (LSTM)—along with a deep neural network ensemble (DNNE) using synthetic data to predict future flood probabilities, utilizing the Savitzky–Golay filter for smoothing. Using a hydrometeorological dataset from 1993–2022 for the Nile River basin, six flood predictors were derived. The FNN and LSTM models exhibited high accuracy and stable loss, indicating minimal overfitting, while the CNN showed slight overfitting. Performance metrics revealed that FNN achieved 99.63% accuracy and 0.999886 ROC AUC, CNN had 95.42% accuracy and 0.893218 ROC AUC, and LSTM excelled with 99.82% accuracy and 0.999967 ROC AUC. The DNNE outperformed individual models in reliability and consistency. Runoff and rainfall were the most influential predictors, while temperature had minimal impact.

A comparative analysis of machine learning methods for display characterization

This paper explores the application of various machine-learning methods for characterizing displays of technologies LCD, OLED, and QLED to achieve accurate color reproduction. These models are formed from input (device-dependent RGB data) and output (device-independent XYZ coordinates) data obtained from three different displays. Training and test datasets are built using RGB data measured directly from the displays and corresponding XYZ coordinates measured with a high-precision colorimeter. A key aspect of this research is the application fuzzy inference systems for building interpretable models. These models offer the advantage of not only achieving good performance in color reproduction, but also providing physical insights into the relationships between the RGB inputs and the resulting XYZ outputs. This interpretability allows for a deeper understanding of the display’s behavior. Furthermore, we compare the performance of fuzzy models with other popular machine-learning approaches, including those based on neural networks and decision trees. By evaluating the strengths and weaknesses of each method, we aim to identify the most effective approach for display characterization. The effectiveness of each method is assessed by its ability to accurately reproduce and display colors, as measured by the ΔE00 visual error metric. Our findings indicate that the Fuzzy Modeling and Identification (FMID) method is particularly effective, allowing for an optimal balance between high accuracy and interpretability. Its competitive performance across all display types, combined with its valuable interpretability, provides insights for potential future calibration and optimization strategies. The results will shed light on whether machine learning methods offer an advantage over traditional physical models, particularly in scenarios with limited data. Additionally, the study will contribute to the understanding of the interpretability benefits offered by fuzzy inference systems in the context of LCD display characterization.

Enhancing Photovoltaic Power Predictions with Deep Physical Chain Model

Predicting solar power generation is a complex challenge with multiple issues, such as data quality and choice of methods, which are crucial to effectively integrate solar power into power grids and manage photovoltaic plants. This study creates a hybrid methodology to improve the accuracy of short-term power prediction forecasts using a model called Transformer Bi-LSTM (Bidirectional Long Short-Term Memory). This model, which combines elements from the transformer architecture and bidirectional LSTM (Long–Short-Term Memory), is evaluated using two strategies: the first strategy makes a direct prediction using meteorological data, while the second employs a chain of deep learning models based on transfer learning, thus simulating the traditional physical chain model. The proposed approach improves performance and allows you to incorporate physical models to refine forecasts. The results outperform existing methods on metrics such as mean absolute error, specifically by around 24%, which could positively impact power grid operation and solar adoption.

Machine learning ellipsometry as a sensitive diagnostic tool to study reproductive biology in Zika virus infected murine models

Machine-learning spectroscopy is a recent scientific and fast-growing area where specific machine-learning solutions are used to process spectroscopic data beyond traditional physical modeling. Here, several machine learning models are optimized and trained for processing highly sensitive broadband variable angle spectroscopic ellipsometry measurements as a new tool in reproductive biology. This new method is applied to classify semen samples obtained from male offspring from ZIKV-infected and uninfected murine models. The method is label-free, cost-effective, simple, and reaches excellent performance. The combination of spectroscopic depolarization degree data with a well-optimized and trained SVM model resulted in the excellent performance of AUROC = 0.99, accuracy = 92%, specificity = 87% and sensibility = 97%. Optical modeling of ellipsometry spectra with Kramers–Kronigconsistent B-splines that takes care of incoherent reflections in the samples allowed for the extraction of the average semen sample refraction index for each class, revealing small differences that were attributed to the observed semen head defects, protamination failure, and DNA fragmentation.

Comparative analysis of HEC-HMS and machine learning models for rainfall-runoff prediction in the upper Baro watershed, Ethiopia

ABSTRACT Accurate streamflow simulation is crucial for effective hydrological management, especially in regions like the upper Baro watershed, Ethiopia, where data scarcity challenges conventional modeling approaches. This study evaluates the efficacy of three hydrological models: the Hydrologic Engineering Center's Hydrologic Modeling System (HEC-HMS), artificial neural network (ANN), and support vector regression (SVR) in predicting runoff. Using data from 2000 to 2016, the analysis focused on various performance metrics such as the Nash–Sutcliffe efficiency (NSE), root mean square error (RMSE), and coefficient of determination (R2). The results indicated that the ANN model significantly outperformed the others, achieving an NSE of 0.98, RMSE of 24 m3/s, and R2 of 0.99. In comparison, the HEC-HMS model yielded an NSE of 0.85, RMSE of 113.4 m3/s, and R2 of 0.89, while the SVR model displayed an NSE of 0.97, RMSE of 27 m3/s, and R2 of 0.99. These findings highlight the superior performance of ANN in regions with limited hydrological data, suggesting its potential as a reliable alternative to traditional physical models. By demonstrating the efficacy of machine learning models, this research facilitates the way for innovative approaches to water resource management, offering valuable insights for policymakers and practitioners.

Physics-informed geostatistical approach (PIGA) for real-time fluvial flood modeling

This study presents the Physics-Informed Geostatistical Approach (PIGA), a novel methodology aimed at improving flood prediction accuracy in riverine systems during hurricane events. PIGA integrates reduced spatial correlations derived from fluvial hydrodynamic models with a reduced geostatistical approach (RGA) based on monitoring data to provide real-time estimations of water surface elevations (WSE) across river channels and reaches. Focusing on the Savannah River basin, an area prone to riverine flooding during hurricane seasons, we developed a 1D-2D HEC-RAS model as the physical model and compared its performance with PIGA for various hurricane events, including Hurricane Matthew, Irma, Dorian, and Idalia. Our findings reveal that PIGA consistently outperforms the HEC-RAS model in predicting WSE, particularly in areas with limited bathymetry data. Statistical analyses demonstrate significant improvements in prediction accuracy achieved by PIGA, emphasizing its potential as a reliable tool for flood prediction in data-deficient regions. Additionally, we explore the transferability of reduced spatial correlations across different hurricane events. Our results indicate that spatial correlations derived from one hurricane event can effectively predict WSE for other events, with Hurricane Irma’s spatial correlations proving to be the most versatile for our study site. However, we also observe variability in spatial correlations among different hurricane events, highlighting the importance of selecting the optimal alternative spatial reference for prediction. Furthermore, we highlight PIGA’s computational efficiency, with simulations completed in significantly less time compared to traditional physical models, rendering its potential to provide real-time prediction of WSE during flooding events.

1D -FNO - Residual convolutional neural network autoregressive prediction model for deep water wave train evolution based on high-order spectral (HOS) data

Accurate prediction of deep-water wave dynamics is essential for ocean engineering, meteorology, and maritime safety. Traditional physical models, though effective, face significant computational and time challenges with complex nonlinear dynamics. This study introduces an autoregressive prediction model using a one-dimensional adaptive Fourier Neural Operator-Residual Convolutional Neural Network (1D-FNO-ResNet) to enhance prediction accuracy. We developed a high-precision single-step prediction model capable of capturing wave characteristics such as height and fluid particle velocity. A fine-tuning training strategy based on multi-step prediction errors was proposed for autoregressive evolution. Performance comparisons between the ResNet and FNO-ResNet models were made under various fine-tuning steps. A correction model was also introduced to control error accumulation, improving autoregressive prediction steps. The model's generalization performance was validated with modulated and random wave trains under different conditions, optimizing predictive stability and accuracy. This approach demonstrates efficient 1D deep-water wave evolution prediction, offering a novel deep learning solution for wave dynamics prediction in ocean engineering, supporting wave energy conversion device design and optimization.

Spectral prediction method based on the transformer neural network for high-fidelity color reproduction

Color distortion often occurs during transmission and reproduction processes, and existing spectral prediction methods have the disadvantage of low prediction accuracy in halftone reproduction. Addressing this issue, this paper establishes a halftone dataset composed of four-color inks (CMYK) mixtures. Based on this, the transformer network is introduced to model and characterize the spectral features of mixed inks, and a forward color formulation prediction model and a reverse spectral prediction model combining halftone reproduction with spectral sequences are proposed, namely the spectrum-color transformer (SC-Former). Color reproduction quality assessment experiments are conducted using the dataset established in this paper and the international standard Ugra/Fogra Media Wedge V3.0 test set. The experimental results show that the SC-Former model outperforms traditional physical models and data-driven prediction models in terms of color reproduction effects and spectral prediction accuracy. This research contributes to the development of high-fidelity color reproduction techniques.

Advances in remote sensing based soil moisture retrieval: applications, techniques, scales and challenges for combining machine learning and physical models

Soil Moisture (SM) monitoring is crucial for various applications in agriculture, hydrology, and climate science. Remote Sensing (RS) offers a powerful tool for large-scale SM retrieval. This paper explores the advancements in RS techniques for SM estimation. We discuss the applications of these techniques, along with the advantages and limitations of traditional physical models and data-driven Machine Learning (ML) based approaches. The paper emphasizes the potential of combining ML and physical models to leverage the strengths of both approaches. We explore the challenges associated with this integration and future research directions to improve the accuracy, scalability, and robustness of RS-based SM retrieval. Finally, the paper also discusses a few issues such as input data selection, data availability, ML complexity, the need for public datasets for benchmarking, and analysis.

Investigating the drag impact on pendulum through numerical models and the stability analysis

As a traditional physical model, the pendulum plays a crucial role in the dynamics. However, most research focuses on the ideal pendulum structure which deviates from reality. In order to study the pendulum in practical situations, this paper not only investigates the ideal model but also the simple pendulum motion with drag. Firstly, this work derived analytical and numerical solutions of the simple pendulum motion equations, which come from physical phenomena, and compared these two solutions. Moreover, this study investigated the error of this numerical model and accomplished further research on demonstrating the conditions to meet the stability of the model.

Random forest with feature selection and K-fold cross validation for predicting the electrical and thermal efficiencies of air based photovoltaic-thermal systems

The recent global energy transition policies emphasize the utilization of renewable sources as a primary energy solution. Solar energy systems, in particular, offer significant advantages in terms of cost-effectiveness and environmental friendliness. This work focuses on the modeling of a photovoltaic-thermal (PV/T) air-based system using machine learning (ML) techniques. The primary contribution lies in estimating the electrical and thermal efficiencies of the PV/T system through the application of the random forest (RF) technique and K-fold cross-validation. The results obtained are notably impressive when compared to the most commonly used techniques reported in the literature: Artificial Neural Networks (ANN), Support Vector Machine (SVM), and linear regression (LR). The assessment demonstrates that Random Forest outperforms these techniques in predicting both Electrical efficiency and Thermal efficiency with R² values of 99.9996 % and 81.2 % respectively, and MAE of 0.0034 and 2.64 respectively. Furthermore, selecting the most important features for electrical efficiency not only reduced the processing time but also improved the performance. The Random Forest model exhibited robust stability and demonstrated strong generalization capabilities, as evidenced by its consistent and promising electrical efficiency performance across a 10-fold cross-validation assessment. The proposed method could be a key solution to accurately model the electrical and thermal performances and serve as an alternative to traditional physical modeling approaches.

An artificial intelligence approach for the estimation of conduction heat transfer using deep neural networks

PurposeThis study aims to use deep neural networks (DNNs) to learn the conduction heat transfer physics and estimate temperature distribution images in a physical domain without using any physical model or mathematical governing equation.Design/methodology/approachTwo novel DNNs capable of learning the conduction heat transfer physics were defined. The first DNN (U-Net autoencoder residual network [UARN]) was designed to extract local and global features simultaneously. In the second DNN, a conditional generative adversarial network (CGAN) was used to enhance the accuracy of UARN, which is referred to as CGUARN. Then, novel loss functions, introduced based on outlier errors, were used to train the DNNs.FindingsA UARN neural network could learn the physics of heat transfer. Within a few epochs, it reached mean and outlier errors that other DNNs could never reach after many epochs. The composite outlier-mean error as a loss function showed excellent performance in training DNNs for physical images. A UARN could excellently capture local and global features of conduction heat transfer, whereas the composite error could accurately guide DNN to extract high-level information by estimating temperature distribution images.Originality/valueThis study offers a unique approach to estimating physical information, moving from traditional mathematical and physical models to machine learning approaches. Developing novel DNNs and loss functions has shown promising results, opening up new avenues in heat transfer physics and potentially other fields.

Solar Radiation Forecasting: A Systematic Meta-Review of Current Methods and Emerging Trends

Effective solar forecasting has become a critical topic in the scholarly literature in recent years due to the rapid growth of photovoltaic energy production worldwide and the inherent variability of this source of energy. The need to optimise energy systems, ensure power continuity, and balance energy supply and demand is driving the continuous development of forecasting methods and approaches based on meteorological data or photovoltaic plant characteristics. This article presents the results of a meta-review of the solar forecasting literature, including the current state of knowledge and methodological discussion. It presents a comprehensive set of forecasting methods, evaluates current classifications, and proposes a new synthetic typology. The article emphasises the increasing role of artificial intelligence (AI) and machine learning (ML) techniques in improving forecast accuracy, alongside traditional statistical and physical models. It explores the challenges of hybrid and ensemble models, which combine multiple forecasting approaches to enhance performance. The paper addresses emerging trends in solar forecasting research, such as the integration of big data and advanced computational tools. Additionally, from a methodological perspective, the article outlines a rigorous approach to the meta-review research procedure, addresses the scientific challenges associated with conducting bibliometric research, and highlights best practices and principles. The article’s relevance consists of providing up-to-date knowledge on solar forecasting, along with insights on emerging trends, future research directions, and anticipating implications for theory and practice.

Track–bridge interaction effects in the acceleration and displacement response of high-speed railway bridges: Simplified vs refined modelling

The prediction of structural vibrations induced by high-speed trains is crucial for designing bridge structures, as well as for the assessment of the dynamic compatibility between bridges and rolling stock in large railway networks. The computational effort involved in such dynamic compatibility checks is also large. However, obtaining an accurate prediction of the dynamic response is challenging due to numerous uncertain factors still under research. Interaction phenomena like vehicle–bridge (VBI), soil–structure (SSI), and track–bridge (TBI) interactions are crucial but complex to model, given the multitude of input parameters and the computational cost required. Therefore, the choice of a modelling approach for simulating the dynamic response of a train crossing a bridge will depend on the focus of analysis. For the purpose of sensitivity analyses – typical in early design stages –, compatibility checks (screenings), and also in non-deterministic analyses, computational cost becomes crucial. Consequently, for practical purposes interaction mechanisms are often simulated by means of simplified and conservative approaches, usually aligned with some design code recommendations. Moreover, traditional physical models, where the problem is idealised as a beam traversed by constant moving loads (travelling load model, TLM), have been commonly employed for such purposes. However, they neglect the load distributive effect of the track on the vehicle axle loads, which has proven to be relevant and beneficial, especially for short bridge spans. The main purpose of this work is thus to investigate the influence of the load spreading effect exerted by the ballasted track on the displacement and acceleration response of single-track, simply-supported (S-S) railway bridges of short-to-medium span lengths under operating conditions, since these structures are prone to experience inadmissible vibration levels under more demanding traffic conditions. A comprehensive analysis of several simplified approaches proposed by regulations and researchers is performed, plus a subsequent comparison vs. a finite element (FE) strategy to consider TBI in a refined manner. Conclusions about the adequacy of the simplified approaches are provided, along with a new data-driven formula for the prediction of the vertical displacement response in resonant conditions, that can be exploited to reduce significantly the computational effort.

Power prediction using high-resolution SCADA data with a farm-wide deep neural network approach

Accurate loss estimation methods with a high level of temporal granularity are necessary to enable the implementation of efficient and adaptable control strategies for wind farms. Predictive models for the power of wind turbines within a wind farm are investigated using high-resolution SCADA data and deep learning methodologies. Traditional physical models offer detailed insights but are computationally expensive. Statistical models face limitations in handling wind energy variability. In this study, deep learning models are explored to capture spatial and temporal dynamics, with four models developed: Multi-Layer Perceptron (MLP), Convolutional Neural Network (CNN), Long Short-Term Memory network (LSTM), and a hybrid CNN-LSTM model. SCADA data from an offshore wind farm is preprocessed, anomalies removed, and annotated based on operational regions. The models are trained, validated, and tested, demonstrating improved accuracy over baseline methods. The hybrid model, incorporating spatial and temporal information, yields the highest predictive performance, showcasing the significance of these dimensions in wind power prediction.

Physics-guided neural network for grinding temperature prediction

Creep-feed grinding is a high-efficiency, high-precision grinding process widely used in the manufacturing of aviation engines. However, the workpiece burn and other quality issues caused by high processing temperature limit the yield of grinding. Therefore, the accurate model of grinding temperature has become the key to improving processing efficiency and quality. Different from the traditional physical or data-driven models, this paper attempts to combine both perspectives based on Physics-Guided Neural Networks (PGNN) to accurately predict grinding temperature with a small number of experiments. At the level of data acquisition, real grinding experiment data was obtained and a data augmentation method had been proposed. At the level of neural network structure, optimisation processes were implemented to enhance prediction performance, and a physics-guided loss function was inserted to guide network training. The experiment results shows that PGNN had better prediction accuracy than the physical model, while also mitigating the limitations of data-driven models on small sample sets. PGNN also performed better with noisy data and predictions out of the training data range, this reveals the benefits of PGNN for small sample problems in processing scenarios.

Deep learning framework with Bayesian data imputation for modelling and forecasting groundwater levels

Although traditional physical models have been used to analyse groundwater systems, the emergence of novel machine learning models can improve the accuracy of the predictions. Deep learning has been prominent in environmental and climate change problems. In this paper, we present a framework for utilising deep learning models to predict groundwater levels based on nearby streamflow and rainfall data. We address the missing data problem using a Bayesian linear regression model within the deep learning framework. Our deep learning framework utilises models such as long-short term memory (LSTM) networks and convolutional neural networks (CNN) for multi-step ahead time series prediction. We examine the fluctuations in groundwater levels at various boreholes located near Middle Creek in New South Wales, Australia. We use the National Collaborative Research Infrastructure Strategy (NCRIS) groundwater database and utilise Bayesian linear regression to impute missing data. We investigate the accuracy of the selected models for individual and regional basins and univariate and multivariate strategies. Our results show that the LSTM-based regional model with multivariate strategy using rainfall data provided the best accuracy.

Failure analysis of 3D concrete printing bolted laminates mimicking geological strata

3D concrete printing has the potential to simulate physical models for engineering investigations. However, it has not been fully exploited and validated. In this study, 3D concrete printing was used to construct laminated strata, including bolting, to mimic an underground excavation environment. Loading tests were conducted using various experimental techniques, such as digital image correlation (DIC) and acoustic emissions (AE). Monotonic and cyclic loading were applied to investigate the laminate failure in an engineering scenario. Attention was focused on the bonding effect between the 3D printing layers and the orientation between printing and loading directions. This study has three main contributions. Firstly, it exploits 3D printing to accurately mimic laminated strata in a geological scenario. Secondly, it integrates rock bolts with 3D printing models to replicate the interlocking mechanism between rock bolts and strata. Finally, it highlights the benefits of using 3D concrete printing to mimic geological strata, which can be adapted to accurately simulate various engineering failure patterns. The results suggest that 3D concrete printing has the potential to decrease manual labor, enhance precision, and more effectively tackle engineering challenges in comparison to traditional physical models.

Application of Machine Learning in Ecological Red Line Identification: A Case Study of Chengdu–Chongqing Urban Agglomeration

China’s Ecological Protection Red Lines (ERLs) policy has proven effective in constructing regional ecological security patterns and protecting ecological space. However, the existing methods for the identification of high conservation value areas (HCVAs) usually use physical models, whose parameters and processes are complex and only for a single service, affecting the ERL delineation. In this study, the data-driven machine learning (ML) models were innovatively applied to construct a framework for ERL identification. First, the One-Class Support Vector Machine (OC-SVM) was used to generate negative samples from natural reserves and ecological factors. Second, the supervised ML models were applied to predict the HCVAs by using samples. Third, by applying the same ecological factors, the traditional physical models were used to assess the ecological services of the study area for reference and comparison. Take Chengdu–Chongqing Urban Agglomeration (CY) as a case study, wherein data from 11 factors and 1822 nature reserve samples were prepared for feasibility verification of the proposed framework. The results showed that the area under the receiver operating characteristic curve (AUC) of all ML models was more than 97%, and random forest (RF) achieved the best performance at 99.57%. Furthermore, the land cover had great contributions to the HCVAs prediction, which is consistent with the land use pattern of CY. High-value areas are distributed in the surrounding mountains of CY, with lush vegetation. All of the above results indicated that the proposed framework can accurately identify HCVAs, and that it is more suitable and simpler than the traditional physical model. It can help improve the effectiveness of ERL delimitation and promote the implementation of ERL policies.