Multilayer Model Research Articles

Analytical model for concrete crack width of steel lined reinforced concrete penstock

Cracking in steel lined reinforced concrete penstock (SLRCP) poses substantial challenges to both serviceability and safety, particularly when crack width exceeds specified limits. Despite its critical implications, accurately calculating crack width remains a persistent challenge. To address this issue, a multi-layer steel-concrete ring analytical model is derived through the concept of orthotropic and thick-walled cylinder theories. In this model, the cracked concrete is assumed to be an orthotropic material incorporating both elastic modulus reduction and tensile softening behavior based on its cracking characteristic of radial penetrating cracks uniformly distributed on the circumference. The steel liner and reinforcing steel bar are modeled as continuous steel rings using equivalent stress theory, while thick-walled cylinder theory is applied due to geometric features and bearing characteristics. Crack width is determined through equilibrium of force and compatibility of deformations at steel-concrete interfaces, utilizing these assumptions and the developed calculation model. To validate the proposed model, data from prototype observations and large-scale model tests of SLRCPs were systematically collected. The crack width results obtained from the proposed model exhibit strong agreement with the collected data, confirming its accuracy and reliability. In conclusion, the proposed analytical model provides an effective solution for calculating the crack width of SLRCP and offers valuable insights for durability evaluation.

Multilayer Perceptron-Based Wearable Exercise-Related Heart Rate Variability Predicts Anxiety and Depression in College Students.

(1) Background: This study aims to investigate the correlation between heart rate variability (HRV) during exercise and recovery periods and the levels of anxiety and depression among college students. Additionally, the study assesses the accuracy of a multilayer perceptron-based HRV analysis in predicting these emotional states. (2) Methods: A total of 845 healthy college students, aged between 18 and 22, participated in the study. Participants completed self-assessment scales for anxiety and depression (SAS and PHQ-9). HRV data were collected during exercise and for a 5-min period post-exercise. The multilayer perceptron neural network model, which included several branches with identical configurations, was employed for data processing. (3) Results: Through a 5-fold cross-validation approach, the average accuracy of HRV in predicting anxiety levels was 89.3% for no anxiety, 83.6% for mild anxiety, and 74.9% for moderate to severe anxiety. For depression levels, the average accuracy was 90.1% for no depression, 84.2% for mild depression, and 82.1% for moderate to severe depression. The predictive R-squared values for anxiety and depression scores were 0.62 and 0.41, respectively. (4) Conclusions: The study demonstrated that HRV during exercise and recovery in college students can effectively predict levels of anxiety and depression. However, the accuracy of score prediction requires further improvement. HRV related to exercise can serve as a non-invasive biomarker for assessing psychological health.

Motor imagery electroencephalography channel selection based on deep learning: A shallow convolutional neural network

Electroencephalography (EEG) motor imagery (MI) signals have recently attracted much attention because of their potential to communicate with the surrounding environment in a specific way without the need for muscular and physical movement. Despite these advantages, these signals are difficult to detect due to their low signal to noise (SNR) rate and non-stationary and dynamic nature. Convolutional neural networks (CNN) can extract appropriate spatial and temporal features without the need for separate feature extraction and classification steps. Clean input data for CNN significantly improves its performance, but sometimes, common interference occurs between multi-channel signals. This challenge, along with the noisy EEG signals, degrades the performance of CNN networks. This paper presents a channel selection method based on convolutional neural networks to obviate these challenges. The proposed shallow convolutional neural network (SCNN) is designed with temporal convolution and pointwise convolution (using 1×1 convolution) to select the best channel with minimal computational load. Following the channel selection step, the multi-layer fusion CNN model was used to classify the signals. Choosing suitable and relevant EEG channels for motor imagery tasks makes it possible to create appropriate inputs for the classification model so that the model can be improved in the end. For High Gamma Dataset and Brian-Computer Interface (BCI) Competition IV-2a, the following accuracies were obtained: 81.15% and 72.01%, which are higher than when the other channel selections such as Mutual Information, Sequential Feature Forward Selection (SFFS) and wrapper methods are used.

Feasibility study and optimization of limestone calcined clay composites for compressive strength using multi-layered explainable artificial intelligence models

Interpretable white box machine learning models to predict the compressive strength of Limestone Calcined Clay (LC3) cementitious composite have been developed and analyzed. For this, a large dataset consisting of more than 4000 numerical instances has been collected from the literature, which has been tuned and normalized. Mainly based on the proportion of the chemical composition of its main ingredients – Limestone (0–31 by %mass of LC3), Calcined Clay (0–70 by %mass of LC3), and Cement (30–100 by %mass of LC3) with a water-to-binder ratio ranging from 0.40 to 0.60, predictive models have been developed for the 28 days compressive strength (fc) of LC3. The performance of various predictive models is then compared across the dataset, and suggestions are made for their reliability and explainability. The models are further interpretated using multi-layered explanations invoking model-agnostic explainable artificial intelligence (XAI) technique. Optimization and interaction plots have also been generated and shown as three-dimensional (3D) visualization. Percent of calcined clay, cement and the water to binder ratio were found to be the most influencing parameters for the compressive strength of LC3. Results of this study indicate that the optimized blends of LC3 can be expressed with up to 50 % replacement of OPC without any significant compromise in fc, contributing to the development of sustainable cementitious composites. The major advantage of using this framework lies in the model agnostic approach, so different predictive models can be compared in a same manner and thus, this approach is applicable to interpretable forecasting models for binary, ternary, quaternary sustainable cementitious composites.

Predicting Network Traffic Flow with a Multi-Layer Short-Term Memory Model

Predicting Network Traffic Flow with a Multi-Layer Short-Term Memory Model

Unveiling significant risk factors for intensive care unit-acquired weakness: Advancing preventive care

In this editorial, we discuss an article titled, “Significant risk factors for intensive care unit-acquired weakness: A processing strategy based on repeated machine learning,” published in a recent issue of the World Journal of Clinical Cases . Intensive care unit-acquired weakness (ICU-AW) is a debilitating condition that affects critically ill patients, with significant implications for patient outcomes and their quality of life. This study explored the use of artificial intelligence and machine learning techniques to predict ICU-AW occurrence and identify key risk factors. Data from a cohort of 1063 adult intensive care unit (ICU) patients were analyzed, with a particular emphasis on variables such as duration of ICU stay, duration of mechanical ventilation, doses of sedatives and vasopressors, and underlying comorbidities. A multilayer perceptron neural network model was developed, which exhibited a remarkable impressive prediction accuracy of 86.2% on the training set and 85.5% on the test set. The study highlights the importance of early prediction and intervention in mitigating ICU-AW risk and improving patient outcomes.

Thermomechanical Collaborative Design and Experimental Verification of a Sandwich Structure with a Polyimide Foam Core

This paper introduces the technology and performance of a new sandwich structure consisting of carbon fiber composite face sheets, a polyimide foam core, and an aluminum honeycomb panel with both heat insulation and load-bearing functions. The heat-resistant performance of the sandwich structure was evaluated. The vacuum thermal conductivity of the sandwich structure in the vertical direction ranges from 0.54 to 0.92 (W/[m⋅K]) with increasing temperature from −170 to 135°C. A multilayer finite element model of carbon fiber cloth, polyimide foam, and aluminum honeycomb was established. Finally, a comprehensive evaluation of the mechanical and thermal insulation performance of the sandwich structure was conducted through random vibration testing and thermal vacuum testing. The results showed that the sandwich structure had good stiffness and thermal insulation performance, successfully verifying the effectiveness of the collaborative design.

Probabilistic fatigue life prediction using multi-layer perceptron with maximum entropy algorithm

In the structural fatigue fields, such as aerospace, maritime, and civil engineering, the accurate construction of the Probability-Stress-Life (P-S-N) curve is crucial for preventing unexpected structural failures and is fundamental to the engineering design process. Conventional methods usually assume that the fatigue life distribution follows a normal or weibull distribution, which overlooks the differences of life distribution at different stress levels and introduces subjective errors. To address this issue, this paper proposes a new approach based a multi-layer perceptron model with maximum entropy algorithm (MPME) for probabilistic fatigue life prediction. During the derivation of life probability distribution function in this approach, the type not require any pre-assumption, thus effectively avoiding the introduction of subjective errors. Specifically, the dispersed information of fatigue life is quantified as various life distribution statistical features such as mean, variance, skewness, and kurtosis. Next, learn the continuous functional relationship between stress and these features using a multi-layer perceptron (MLP). Finally, the target probability distribution is derived using the maximum entropy algorithm, with the predicted distribution features obtained from the MLP serving as constraints for it. In addition, to ensure the consistency between the prediction of MLP and physical laws, we encode physical knowledge into the network loss to guide the updating of network parameters. Experimental results of two different material fatigue data demonstrate that, compared to traditional deterministic model-based methods and machine learning methods, the MPME in this paper can more accurately predict the probabilistic fatigue life.

Ozone Loss on Painted Surfaces: Dependence on Relative Humidity, Aging, and Exposure to Reactive SVOCs.

Ozone and its oxidation products result in negative health effects when inhaled. Despite painted surfaces being the most abundant surface in indoor spaces, surface loss remains one of the largest uncertainties in the indoor ozone budget. Here, ozone uptake coefficients (γO3) on painted surfaces were measured in a flow-through reactor where 79% of the inner surfaces were removable painted glass sheets. Flat white paint initially had a high uptake coefficient (8.3 × 10-6) at 20% RH which plateaued to 1.1 × 10-6 as the paint aged in an indoor office over weeks. Increasing the RH from 0 to 75% increased γO3 by a factor of 3.0, and exposure to 134 ppb of α-terpineol for 1 h increased γO3 by a factor of 1.6 at 20% RH. RH also increases α-terpineol partitioning to paint, further increasing ozone loss, but the type of paint (flat, eggshell, satin, semigloss) had no significant effect. A kinetic multilayer model captures the dependence of γO3 on RH and the presence of α-terpineol, indicating the reacto-diffusive depth for O3 is 1 to 2 μm. Given the similarity of the kinetics on aged surfaces across many paint types and the sustained reactivity during aging, these results suggest a mechanism for catalytic loss.

The impact of social interventions on COVID-19 spreading based on multilayer commuter networks

From March to June 2022, Shanghai was struck by a new coronavirus variant, Omicron, resulting in the infected cases of at least 600,000 people. Despite implementing a strict containment policy of city-wide silence (i.e., residents were not allowed to go out unless necessary), the outbreak cannot be effectively prevented within a short period of time. A significant academic and practical question is: how could we prevent and control outbreak of COVID-19 in large, densely populated cities like Shanghai? It is necessary to develop a rational epidemic spreading model for large cities, in order to accurately predict the trend of disease and quantitatively assess the impact of non-pharmaceutical interventions. In this paper, a multilayer commuter metapopulation network model is constructed to capture commuting flows and the size of epidemic outbreak during commuting between districts. The model accurately predicts epidemic spreading in each district of Shanghai. Assuming strict city-wide lockdowns, with each district locked down and limited inter-district commuting as social zones, simulations demonstrate significant suppression of outbreaks due to social-level interventions. For example, a 1-fold increase in PCR (Polymerase Chain Reaction) testing efficiency reduces the size of epidemic outbreak by approximately 70%. Larger districts require stricter controls to prevent exponential growth. Lockdowns effectively prevent epidemic outbreak at low disease rates but less so at high rates. Liberalized policies lead to varied outbreak trends, with economically developed regions peaking earlier due to higher population densities. This study provides a comprehensive framework for quantitatively evaluating the impact of social and regional controls on urban epidemics.

Unveiling the microstructural evolution and interaction mechanisms for twisted structures

Twisting plays an important role in fabricating twisted actuators and energy harvesters, which require excellent microstructural deformation and interaction properties between different layers. However, the cross-layer microstructural evolution and interaction mechanism of helical structures under twisting and stretching are still unclear. Herein, a multi-layer model is established to theoretically investigate the response of the filaments under twisting and stretching. The results quantitatively indicate that the filaments with a higher twist have more structural evolution and less elongation deformation in the tensile process, which contributes to the increase of ductility. The evolution of the helical angle is also verified by in-situ experiments and finite element simulations. Besides, a theoretical method is provided for investigating the mechanical behavior of anisotropic twisted fibers, and the contact pressure is also promoted to understand the twisting-induced densification. The interlayer extrusion reveals that appropriate stretching of twisted fiber is an effective way to densify the loose fibers, and it also provides the theoretical reason for twisting and stretching simultaneously when wringing out a wet towel. We believe that this work would shed light on the interaction mechanism of twisted filaments and provide mechanistic insights into the twisting design of multi-layer helical structures.

An Investigation into the Transdermal Behavior of Active Ingredients by Combination of Experiments and Multiscale Simulations.

Transdermal behavior is a critical aspect of studying delivery systems and evaluating the efficacy of cosmetics. However, existing methods face challenges such as lengthy experiments, high cost, and limited model accuracy. Therefore, developing accurate transdermal models is essential for formulation development and effectiveness assessment. In this study, we developed a multiscale model to describe the transdermal behavior of active ingredients in the stratum corneum. Molecular dynamics simulations were used to construct lipid bilayers and determine the diffusion coefficients of active ingredients in different regions of these bilayers. These diffusion coefficients were integrated into a multilayer lipid pathway model using finite element simulations. The simulation results were in close agreement with our experimental results for three active ingredients (mandelic acid (MAN), nicotinamide (NIC), and pyruvic acid (PYR)), demonstrating the effectiveness of our multiscale model. This research provides valuable insights for advancing transdermal delivery methods.

Effects of diffuse light on the gross ecosystem primary productivity of a winter wheat (Triticum aestivum L.) cropland in northern China

Diffuse light is produced by clouds and aerosols in the atmosphere. Exploring the effects of diffuse light on ecosystem productivity is important for understanding the terrestrial carbon (CO2) cycle. Here, 2 years of gross ecosystem primary productivity (GEP) from a (winter) wheat cropland in China was assessed using eddy covariance technology to explore the effects of diffuse photosynthetic active radiance (PAR) on wheat GEP. Wheat GEP increased significantly and positively along with diffuse PAR. In addition, wheat GEP was significantly affected by total PAR, air temperature, and vapor press deficit in different diffuse PAR fraction (fDIF) change stages. Because significant autocorrelations existed among the controlling factors, a path analysis was used to quantify the effects of diffuse light on GEP. Diffuse PAR was the primary and secondary importance factors affecting GEP with direct path coefficients of 0.54 and 0.48, respectively, in different fDIF change stages. A multilayer canopy model revealed that the middle and lower canopy levels intercepted more light when diffuse PAR increased. This resulted in the photosynthetic enhancement of middle and lower canopy levels, which contributed approximately 65% and 35%, respectively, to the increase in photosynthesis for the entire canopy (~ 30.5%). Overall, our study provided new evidence regarding the importance of diffuse light for CO2 uptake in agroecosystems, which is important for predicting the responses of ecosystem CO2 budgets to future climate-related light changes.

Non-Invasive Fish Biometrics for Enhancing Precision and Understanding of Aquaculture Farming through Statistical Morphology Analysis and Machine Learning.

Aquaculture requires precise non-invasive methods for biomass estimation. This research validates a novel computer vision methodology that uses a signature function-based feature extraction algorithm combining statistical morphological analysis of the size and shape of fish and machine learning to improve the accuracy of biomass estimation in fishponds and is specifically applied to tilapia (Oreochromis niloticus). These features that are automatically extracted from images are put to the test against previously manually extracted features by comparing the results when applied to three common machine learning methods under two different lighting conditions. The dataset for this analysis encompasses 129 tilapia samples. The results give promising outcomes since the multilayer perceptron model shows robust performance, consistently demonstrating superior accuracy across different features and lighting conditions. The interpretable nature of the model, rooted in the statistical features of the signature function, could provide insights into the morphological and allometric changes at different developmental stages. A comparative analysis against existing literature underscores the competitiveness of the proposed methodology, pointing to advancements in precision, interpretability, and species versatility. This research contributes significantly to the field, accelerating the quest for non-invasive fish biometrics that can be generalized across various aquaculture species in different stages of development. In combination with detection, tracking, and posture recognition, deep learning methodologies such as the one provided in the latest studies could generate a powerful method for real-time fish morphology development, biomass estimation, and welfare monitoring, which are crucial for the effective management of fish farms.

Applying a multi-layer perceptron model for predicting gasification process outcomes

Applying a multi-layer perceptron model for predicting gasification process outcomes

[formula omitted][formula omitted]ave [formula omitted]earner: Predicting wave farms power output using effective meta-learner deep gradient boosting model: A case study from Australian coasts

Precise prediction of wave energy is indispensable and holds immense promise as ocean waves have a power capacity of 30–40 kW/m along the coast. Utilising this energy source does not generate harmful emissions, making it a superior substitute for fossil fuel-based energy. The computational expense associated with simulating and computing intricate hydrodynamic interactions in wave farms restricts optimisation methods to a few thousand evaluations and makes a challenging situation for training in deep neural prediction models. To address this issue, we propose a new solution: a Meta-learner gradient boosting method that employs four multi-layer convolutional dense neural network surrogate models combined with an optimised extreme gradient boosting. In order to train and validate the predictive model, we used four wave farm datasets, including the absorbed power outputs and 2D coordinates of wave energy converters (WECs) located along the southern coast of Australia, Adelaide, Sydney, Perth and Tasmania. Furthermore, the capability of the transfer learning strategy is evaluated. The WECs used in this study are of the fully submerged three-tether converter type, similar to the CETO prototype. The effectiveness of the proposed approach is assessed by comparing it with 15 well-established and effective machine learning (ML) methods. The experimental findings indicate that the proposed model is competitive with other ML and deep learning approaches, exhibiting considerable accuracy of 88.8%, 90.0%, 90.3%, and 84.4% in Adelaide, Perth, Sydney and Tasmania and improved robustness in predicting wave farm power output.

A universal tool for estimating monthly solar radiation on tilted surfaces from horizontal measurements: A machine learning approach

In this study, a universal method for estimating monthly solar radiation on tilted surfaces from horizontal measurements was developed and validated, targeting the optimal design and performance of solar energy systems. Using data from the Photovoltaic Geographical Information System (PVGIS) across 740 global locations, three machine learning models—linear regression, random forest, and k-nearest neighbors (KNN)—along with a multilayer perceptron deep learning model, were evaluated. The dataset spanned tilt angles from 0° to 90° at 1° increments and 11,511,773 rows with six input features for predicting irradiance on any tilt angle. The KNN model was selected for further testing analysis due to its superior accuracy, achieving the lowest root mean square error (RMSE) of 1.42 kWh/m2 and mean absolute error (MAE) of 0.7 kWh/m2 on the validation dataset. Testing was conducted with 16 years of data from five locations not included in the training dataset, obtained from PVGIS and Solcast. Results show that the KNN model consistently outperformed the traditional isotropic model in four out of five cities using PVGIS data and in two using Solcast data, especially excelling in regions with stable climates and consistent irradiance profiles. Moreover, better prediction performance was observed with PVGIS data compared to Solcast data across all five cities, with PVGIS data yielding an RMSE of 3.03 kWh/m2 and an MAE of 2.41 kWh/m2, while Solcast data exhibited a higher RMSE of 5.22 kWh/m2 and MAE of 4.14 kWh/m2. Finally, these results advocate for a shift towards data-driven methodologies highlighting their enhanced reliability.

Vertical gradient of needle ozone uptake within the canopy of Cryptomeria japonica

Leaf ozone uptake through the stomata is an important index for the ozone risk assessments on trees. Stomatal conductance (gs) and ozone concentration ([O3]), determinants of the leaf ozone uptake, are known to show vertical gradients within a tree canopy. However, less is known about the within-canopy vertical gradient of leaf ozone uptake. This study was aimed to elucidate how the vertical gradient of [O3] and gs affect needle ozone uptake within a canopy of mature Cryptomeria japonica trees in a suburban forest at Tokyo, Japan. For this purpose, a multilayer gas exchange model was applied to estimate the vertical gradient of needle gs and the accumulated ozone uptake during the study period (POD1, Phytotoxic Ozone Dose above a threshold of 1 nmol m−2 s−1). In addition, we also tested several scenarios of vertical gradient of [O3] within the canopy for sensitivity analysis. The POD1 was declined from the top to the bottom of the canopy. This tendency strongly depended on the vertical gradient of gs and was hardly affected by the changes in simulated vertical reductions of the [O3]. We further assessed the photosynthesis of sunlit needles (needles absorbing both direct and diffuse light) and shaded needles (needles only absorbing diffuse light). The photosynthesis of shaded needles in the upper half of the canopy made a great contribution to the entire canopy photosynthesis. In addition, given that their POD1 was lower than that of sunlit needles, ozone may affect sunlit and shaded needles differently. We concluded that these considerations should be incorporated into modeling of the calculation of ozone uptake for mature trees to make accurate predictions of the ozone effects on trees at the canopy scale.

Development of a multi-layer canopy model for E3SM Land Model with support for heterogeneous computing

The vertical structure of vegetation canopies creates micro-climates. However, the land components of most Earth System Models, including the Energy Exascale Earth System Model (E3SM), typically neglect vertical canopy structure by using a single layer big-leaf representation to simulate water, CO2, and energy exchanges between the land and the atmosphere. In this study, we developed a Multi-Layer Canopy Model for the E3SM Land Model to resolve the micro-climate created by vegetation canopies. The model developed in this study re-implements the CLM-ml_v1 to support heterogeneous computing architectures consisting of CPUs and GPUs and includes three additional optimization-based stomatal conductance models. The use of Portable, Extensible Toolkit for Scientific Computation provides a speedup of 25–50 times on a GPU relative to a CPU. The numerical implementation of the model was verified against CLM-ml_v1 for a month-long simulation using data from the Ameriflux US-University of Michigan Biological Station site. Model structural uncertainty was explored by performing control simulations for five stomatal conductance models that exclude and include the control of plant hydrodynamics (PHD) on photosynthesis. The bias in simulated sensible and latent heat fluxes was lower when PHD was accounted for in the model. Additionally, six idealized simulations were performed to study the impact of three environmental variables (i.e. air temperature, atmospheric CO2, and soil moisture) on canopy processes (i.e. net CO2 assimilation, leaf temperature, and leaf water potential). Increasing air temperature reduced net CO2 assimilation and increased air temperature. Net CO2 assimilation increased at higher atmospheric CO2, while decreasing soil moisture resulted in lower leaf water potential.

A Single Image High-Perception Super-Resolution Reconstruction Method Based on Multi-layer Feature Fusion Model with Adaptive Compression and Parameter Tuning

We propose a simple image high-perception super-resolution reconstruction method based on multi-layer feature fusion model with adaptive compression and parameter tuning. The aim is to further balance the high and low-frequency information of an image, enrich the detailed texture to improve perceptual quality, and improve the adaptive optimization and generalization of the model in the process of super-resolution reconstruction. First, an effective multi-layer fusion super-resolution (MFSR) basic model is constructed by the design of edge enhancement, refine layering, enhanced super-resolution generative adversarial network and other sub-models, and effective multi-layer fusion. This further enriches the image representation of features of different scales and depths and improves the feature representation of high and low-frequency information in a balanced way. Next, a total loss function of the generator is constructed with adaptive parameter tuning performance. The overall adaptability of the model is improved through adaptive weight distribution and fusion of content loss, perceptual loss, and adversarial loss, and improving the error while reducing the edge enhancement model. Finally, a fitness function with the evaluation perceptual function as the optimization strategy is constructed, and the model compression and adaptive tuning of MFSR are carried out based on the multi-mechanism fusion strategy. Consequently, the construction of the adaptive MFSR model is realized. Adaptive MFSR can maintain high peak signal to noise ratio and structural similarity on the test sets Set5, Set14, and BSD100, and achieve high-quality reconstructed images with low learned perceptual image patch similarity and perceptual index, while having good generalization capabilities.