Neural Model Research Articles

Advancing Sustainable Building Practices: Intelligent Methods for Enhancing Heating and Cooling Energy Efficiency

Our work is dedicated to enhancing sustainability through improved energy efficiency in buildings, with a specific focus on heating and cooling control and the optimization of thermal comfort of occupants. With an energy consumption of more than 60% in buildings, HVAC systems are the biggest energy users. By integrating advanced technology, data algorithms, and digital twins, our study aims to optimize energy performance effectively. We have developed a Neural Network-based Model Predictive Control (NNMPC) to achieve this goal. Leveraging technologies such as MQTT communication, Wi-Fi modules, and field-programmable gate arrays will enhance scalability and flexibility. Our findings demonstrate the efficacy of the NNMPC system deployed on the PYNQ board for reducing sensible thermal energy usage for both cooling and heating purposes. Compared to traditional On/Off control systems, the NNMPC achieved an impressive 40.8% reduction in heating energy consumption and a 37.8% decrease in cooling energy consumption in 2006. In comparison to the On/Off technique, the NNMPC demonstrated a 25.6% reduction in annual heating energy consumption and a 28.8% drop in annual cooling energy consumption in the simulated year of 2017. We observed that, across all strategies and platforms, there were no instances where the Predicted Mean Vote (PMV) fell below −0.5. However, a significant proportion of PMV values (ranging from 65% to 83%) were observed between −0.5 and 0.5, signifying a high level of occupant comfort. Additionally, for PMV values between 0.5 and 1.0, percentages ranged from 16% to 33% for both years. Importantly, the NNMPC exhibited notable efficiency in maintaining occupants’ comfort within this range, requiring less energy while ensuring highly satisfactory environments.

Performance Evaluation of Graph Neural Network-Based RouteNet Model with Attention Mechanism

Graph representation is recognized as an efficient method for modeling networks, precisely illustrating intricate, dynamic interactions within various entities of networks by representing entities as nodes and their relationships as edges. Leveraging the advantage of the network graph data along with deep learning technologies specialized for analyzing graph data, Graph Neural Networks (GNNs) have revolutionized the field of computer networking by effectively handling structured graph data and enabling precise predictions for various use cases such as performance modeling, routing optimization, and resource allocation. The RouteNet model, utilizing a GNN, has been effectively applied in determining Quality of Service (QoS) parameters for each source-to-destination pair in computer networks. However, a prevalent issue in the current GNN model is their struggle with generalization and capturing the complex relationships and patterns within network data. This research aims to enhance the predictive power of GNN-based models by enhancing the original RouteNet model by incorporating an attention layer into its architecture. A comparative analysis is conducted to evaluate the performance of the Modified RouteNet model against the Original RouteNet model. The effectiveness of the added attention layer has been examined to determine its impact on the overall model performance. The outcomes of this research contribute to advancing GNN-based network performance prediction, addressing the limitations of existing models, and providing reliable frameworks for predicting network delay.

Machine learning-enhanced model-based scenario optimization for DIII-D

Scenario development in tokamaks is an open area of investigation that can be approached in a variety of different ways. Experimental trial and error has been the traditional method, but this required a massive amount of experimental time and resources. As high fidelity predictive models have become available, offline development and testing of proposed scenarios has become an option to reduce the required experimental resources. The use of predictive models also offers the possibility of using a numerical optimization process to find the controllable inputs that most closely achieve the desired plasma state. However, this type of optimization can require as many as hundreds or thousands of predictive simulation cases to converge to a solution; many of the commonly used high fidelity models have high computational burdens, so it is only reasonable to run a handful of predictive simulations. In order to make use of numerical optimization approaches, a compromise needs to be found between model fidelity and computational burden. This compromise can be achieved using neural networks surrogates of high fidelity models that retain nearly the same level of accuracy as the models they are trained to replicate while reducing the computation time by orders of magnitude. In this work, a model-based numerical optimization tool for scenario development is described. The predictive model used by the optimizer includes neural network surrogate models integrated into the fast Control-Oriented Transport simulation framework. This optimization scheme is able to converge to the optimal values of the controllable inputs that produce the target plasma scenario by running thousands of predictive simulations in under an hour without sacrificing too much prediction accuracy.

Multi-objective optimization of CO2 ejector by combined significant variables recognition, ANN surrogate model and multi-objective genetic algorithm

The ejectors are crucial in enhancing the efficiency of the CO2-based refrigeration cycle. Although the ejector can be optimized by experimental and computational fluid dynamic simulation to improve its performance, these methods are time-consuming and complicated. This research aims to present a fast and systematic approach for optimizing CO2 ejector by combining variance analysis, surrogate model, and non-dominated sorting genetic algorithm. Firstly, the key geometric structures affecting the performance of ejector were determined by variance analysis. Secondly, the influence of key geometric structures and boundary conditions on ejector efficiency and entrainment ratio was analyzed by using computational fluid dynamic simulation, and the database was generated. Next, the artificial neural network surrogate model was established to forecast the ejector performance. Finally, the non-dominated sorting genetic algorithm was involved to optimize the ejector for maximizing ejector efficiency and entrainment ratio. The results show that the efficiencies of the optimized ejectors are above 42%, and the entrainment ratios exceed 0.7. Compared with the basic model, the average efficiencies have increased about 13.20%. When the pressure lift is 0.7 MPa, the entrainment ratios of the optimized ejectors are 0.696 (−20 °C ≤ Te < −15 °C), 0.82(-15 °C ≤ Te < −5 °C), 0.60 (−5 °C ≤ Te < 5 °C) and 0.90 (5 °C ≤ Te < 15 °C), respectively. The above research demonstrates that the success of this approach in addressing time-consuming multi-optimization problems. It offers a fast and systematic approach for the multi-objective optimization of CO2 ejector as a valuable reference for engineering applications.

Tensor-based physics-encoded neural networks for modeling constitutive behavior of soil

Data-driven constitutive models are increasingly addressing non-elastic and three-dimensional scenarios. However, their robustness can be significantly impacted by the inadequate integration of physical information. Accordingly, this study introduces a tensor-based physics-encoded neural network to characterize the constitutive behavior of soil, exemplified by isotropic hypoplasticity with dependency on pressure and porosity. The framework ensures strict adherence to fundamental physical laws for small strain, rate-independent isotropic constitutive behavior. The network utilizes stress tensor invariants and soil state parameters (porosity) as inputs, and outputs crucial coefficients for the tensorial constitutive relations. The model has been calibrated using only symmetric triaxial test data (both drained and undrained). The effectiveness of the neural network-based constitutive model has been validated through simulations of drained and undrained triaxial tests under various initial conditions, as well as boundary value problems with complex loading. The results demonstrate that the proposed model offers the following distinguishing advantages: 1) Applicability to both two- and three-dimensional non-elastic cases, even when trained with two-dimensional data; 2) Strict adherence to fundamental physical laws, avoiding soft constraints; 3) An incremental, tensor-based architecture, suitable for integration in numerical software for boundary value problems.

Determining the minimum data size for the development of artificial neural network-based prediction models for rice pests in Korea

Sudden outbreaks of crop pests (insect pests and diseases) are increasing in Korea due to climate change and globalization. To prevent such outbreaks, it is necessary to predict and control pest occurrences in advance. Crop pests have been predicted through process-based or statistical modeling; however, the limitations of these models, which rely solely on historically acquired domain knowledge and data, have become increasingly prominent owing to climate change and rapidly changing agricultural ecosystems. To overcome these limitations, artificial neural network (ANN)-based models that use continuous pest survey data in the field have been investigated over the last decade. However, because pest survey data are collected by humans through process-mediated methods, fundamental problems exist in terms of data quality and size that may hinder the performance of the resulting ANN-based models. In this study, to determine the minimum pest data size required to ensure the optimal performance of ANN-based models, we developed feed-forward neural network models for 19 rice pests using 23 pest datasets collected from 149 districts by the Rural Development Administration of Korea over 19 years (2002–2020). Using each ANN-based model, the minimum data size required for the highest model performance achieved in this study was determined for all 19 rice pests. Furthermore, we developed a decision-tree rule to estimate the minimum data size based on the selected characteristics of each pest. The final Decision tree rule, based on the distinction between diseases and insect pests and the balance of pest data (the relative ratio of pest occurrence data to non-occurrence data), showed a relatively good performance (70.24 %) in the 3-fold cross-validation test. Overall, these results indicate that the minimum data sizes required for ANN modeling vary among rice pests, depending on the pest data characteristics, as indicated by the Decision tree rule developed in this study.

Analytical Approach for Forecasting the Load Capacity of the EN AW-7075-T6 Aluminum Alloy Joints Created Using RFSSW Technology.

To ensure the high reliability of aircraft structures, the Refill Friction Stir Spot Welding (RFSSW) process must be characterized by a high load capacity of the welds and a small standard deviation of the load capacity spread. This allows us to obtain uniform functional properties in the connections, ensuring the high quality of the process. This work aims to select the most favorable technological parameters for the welding process of EN AW-7075-T6 Alclad aluminum alloy sheets, which are used for the production of aircraft structures. The best networks were calculated using the Statistica 13.3 program. The obtained results were compared with the results of previous investigations. It has been shown that a model using neural networks allows for the determination of connection parameters with much greater accuracy than the classical model. The maximum error in estimating the load capacity of the connection for the mathematical model was 6.13%, and the standard deviation was 14.51%. In the case of neural networks, the maximum error value did not exceed 1.55%, and the standard deviation was 3.74%. It was shown that, based on the neural model, it is possible to determine the process parameters that ensure the required quality capacity of the process, ensuring a probability of obtaining the required load capacity of the connections amounting to P = 0.999935 with a defect rate of 0.0065%. This possibility is not provided by the classical model due to its large error in estimating the process spread and the high sensitivity of the process input parameters to the output parameters.

A WOA-BP neural network microstructure evolution prediction model of TC11 titanium alloy and application in hollow shaft during cross wedge rolling with mandrel

A WOA-BP neural network microstructure evolution prediction model of TC11 titanium alloy and application in hollow shaft during cross wedge rolling with mandrel

Deep neural network-based modeling and optimization methodology of fuel cell electric vehicles considering power sources and electric motors

This study proposes a modeling and optimization methodology for fuel cell electric vehicles (FCEVs). Among FCEV components, the traction motor, lithium-ion battery, fuel cell stack, and air supply system are mainly investigated. The FCEV modeling is performed based on the vehicle specifications, electromagnetic finite element analysis, and experimental data. To conduct design optimization, deep neural networks (DNNs) are adopted and trained to predict vehicle performance considering the fluctuation of applied direct current voltage. At this stage, the adaptive layering and sampling algorithm was suggested, which enables efficient DNN construction. To confirm the feasibility of the suggested training algorithm, the number of hidden layers and sampling points of constructed DNNs are investigated. Finally, DNN-based fuel economy optimization is performed considering the driving performance. The effectiveness of the proposed optimization methodology is validated by additional optimization results.

Computational Synaptic Modeling of Pitch and Duration Mismatch Negativity in First-Episode Psychosis Reveals Selective Dysfunction of the N-Methyl-D-Aspartate Receptor.

Mismatch negativity (MMN) to pitch (pMMN) and to duration (dMMN) deviant stimuli is significantly more attenuated in long-term psychotic illness compared to first-episode psychosis (FEP). It was recently shown that source-modeling of magnetically recorded MMN increases the detection of left auditory cortex MMN deficits in FEP, and that computational circuit modeling of electrically recorded MMN also reveals left-hemisphere auditory cortex abnormalities. Computational modeling using dynamic causal modeling (DCM) can also be used to infer synaptic activity from EEG-based scalp recordings. We measured pMMN and dMMN with EEG from 26 FEP and 26 matched healthy controls (HCs) and used a DCM conductance-based neural mass model including α-amino-3-hydroxy-5-methyl-4-isoxazole propionic acid, N-methyl-D-Aspartate (NMDA), and Gamma-aminobutyric acid receptors to identify any changes in effective connectivity and receptor rate constants in FEP. We modeled MMN sources in bilateral A1, superior temporal gyrus, and inferior frontal gyrus (IFG). No model parameters distinguished groups for pMMN. For dMMN, reduced NMDA receptor activity in right IFG in FEP was detected. This finding is in line with literature of prefrontal NMDA receptor hypofunction in chronic schizophrenia and suggests impaired NMDA-induced synaptic plasticity may be present at psychosis onset where scalp dMMN is only moderately reduced. To the best of our knowledge, this is the first report of impaired NMDA receptor activity in FEP found through computational modeling of dMMN and shows the potential of DCM to non-invasively reveal synaptic-level abnormalities that underly subtle functional auditory processing deficits in early psychosis.

Gaussian Aquila optimizer based dual convolutional neural networks for identification and grading of osteoarthritis using knee joint images

Degenerative musculoskeletal disease known as Osteoarthritis (OA) causes serious pain and abnormalities for humans and on detecting at an early stage, timely treatment shall be initiated to the patients at the earliest to overcome this pain. In this research study, X-ray images are captured from the humans and the proposed Gaussian Aquila Optimizer based Dual Convolutional Neural Networks is employed for detecting and classifying the osteoarthritis patients. The new Gaussian Aquila Optimizer (GAO) is devised to include Gaussian mutation at the exploitation stage of Aquila optimizer, which results in attaining the best global optimal value. Novel Dual Convolutional Neural Network (DCNN) is devised to balance the convolutional layers in each convolutional model and the weight and bias parameters of the new DCNN model are optimized using the developed GAO. The novelty of the proposed work lies in evolving a new optimizer, Gaussian Aquila Optimizer for parameter optimization of the devised DCNN model and the new DCNN model is structured to minimize the computational burden incurred in spite of it possessing dual layers but with minimal number of layers. The knee dataset comprises of total 2283 knee images, out of which 1267 are normal knee images and 1016 are the osteoarthritis images with an image of 512 × 512-pixel width and height respectively. The proposed novel GAO-DCNN system attains the classification results of 98.25% of sensitivity, 98.93% of specificity and 98.77% of classification accuracy for abnormal knee case–knee joint images. Experimental simulation results carried out confirms the superiority of the developed hybrid GAO-DCNN over the existing deep learning neural models form previous literature studies.

Periodic analysis of scenic spot passenger flow based on combination neural network prediction model

Abstract To prevent in a short time the rapid increase of tourists and corresponding traffic restriction measures’ lack in scenic areas, this study established a prediction model based on an improved convolutional neural network (CNN) and long- and short-term memory (LSTM) combined neural network. The study used this to predict the inflow and outflow of tourists in scenic areas. The model uses a residual unit, batch normalization, and principal component analysis to improve the CNN. The experimental results show that the model works best when batches’ quantity is 10, neurons’ quantity in the LSTM layer is 50, and the number of iterations is 50 on a workday; on non-working days, it is best to choose 10, 100, or 50. Using root mean square error (RMSE), normalized root mean square error (NRMSE), mean absolute percentage error (MAPE), and mean absolute error (MAE) as evaluation indicators, the inflow and outflow RMSEs of this study model are 82.51 and 89.80, MAEs are 26.92 and 30.91, NRMSEs are 3.99 and 3.94, and MAPEs are 1.55 and 1.53. Among the various models, this research model possesses the best prediction function. This provides a more accurate prediction method for the prediction of visitors’ flow rate in scenic spots. Meanwhile, the research model is also conducive to making corresponding flow-limiting measures to protect the ecology of the scenic area.

Adaptive neural network surrogate model for solving the nonlinear elastic inverse problem via Bayesian inference

Abstract In this paper, we consider a Bayesian method for nonlinear elastic inverse problems. As a working model, we are interested in the inverse problem of restoring elastic properties from measured tissue displacement. In order to reduce the computational cost, we will use the following multi-fidelity model approach. First, we construct a surrogate low-fidelity DNNs-based model in the prior distribution, then use a certain number of simulations of high fidelity model associated with an adaptive strategy online to update the low-fidelity model locally. Numerical examples show that the proposed method can solve nonlinear elastic inverse problems efficiently and accurately.

Detection of Pneumonia Using A Hybrid Approach Consisting of MobileNetV2 and Squeeze-and-Excitation Network

Pneumonia is a global health concern, responsible for a significant number of deaths. Its diagnostic challenge arises from visual similarities it shares with various respiratory diseases, such as tuberculosis, complicating accurate identification. Furthermore, the variability in acquiring and processing chest X-ray (CXR) images can impact image quality, posing a hurdle for dependable algorithm development. To address this, resilient data-centric algorithms, trained on comprehensive datasets and validated through diverse imaging methods and radiology expertise, are imperative. This study presents a deep learning approach designed to distinguish between normal and pneumonia cases. The model, a hybrid of MobileNetV2 and the Squeeze-and-Excitation (SE) block, aims to reduce learnable parameters while enhancing feature extraction and classification. Integration of the SE block enhances classification performance, despite a slight parameter increase. The model was trained and tested on a dataset of 5856 CXR images from Kaggle's medical imaging challenge. Results demonstrated the model's exceptional performance, achieving an accuracy of 98.81%, precision of 98.79%, recall rate of 98.24%, and F1-score of 98.51%. Comparative analysis with various Convolutional neural network-based pre-trained models and recent literature studies confirmed its superiority, solidifying its potential as a robust tool for pneumonia detection, thus addressing a critical healthcare need.

Neural network-based Raman spectral recognition model for milk powder and wheat flour

The problem of counterfeit milk powder has occurred in recent years, and one of the standard techniques is to adulterate milk powder with wheat flour to reduce the cost. Since the morphology and colour of wheat flour are similar to that of milk powder, it is not easy to distinguish it from the naked eye, so a rapid method for detecting wheat flour adulteration in milk powder is urgently needed. This study used Raman spectral analysis as a novel and efficient screening technique. With the advantages of high sensitivity, non-destructive detection, low cost, no contamination, and online analysis, this method can play an essential role in rapid food detection and production control. By processing and analyzing the spectral data, a neural network model for identifying milk powder and wheat flour based on Raman spectroscopy has been developed in this study, and the model's performance has been evaluated. The results show that the model can quickly and accurately identify the proportion of wheat flour in milk powder to determine whether milk powder is adulterated or not. The results of this study are of great significance in ensuring the safety of infant milk powder. In the future, the method can be further optimized, for example, to improve detection speed and accuracy and to provide a more reliable quality control method for the milk powder industry.

Small-scale districts identification of Boletus bainiugan from Yunnan province of China based on residual convolutional neural network continuous classification models

Small-scale districts identification of Boletus bainiugan from Yunnan province of China based on residual convolutional neural network continuous classification models

Performance analysis of three recommendation algorithms on Amazon datasets

Recommendation systems play a crucial role in enhancing user satisfaction and driving sales for businesses. They are essential in todays marketplaces, as they are able to suggest products and services that may interest a particular individual based on their past purchases. In this paper, empirical research is conducted on three distinct subsets of the Amazon dataset, namely Sports &amp; Outdoors, Movies &amp; TV, and Video Games, to comprehensively evaluate the performance of three distinct recommendation methods based on deep learning algorithms. These methods include the dot product method, (an updated version of the singular value decomposition algorithm), the neural network method, and the neural collaborative filtering model with natural language processing method. The results of this study reveal that deep learning-based recommendation systems can achieve more accurate results compared to traditional recommendation systems for three types of products. The implementation of these methods on Amazons dataset can help improve sales by correctly identifying customers interests and suggesting relevant items.

Neural Network Based Technical Analysis of Football Games

Football is the most famous sports in the world, and English Premier League is the number one league in the world for three consecutive years (Fédération internationale de football association, FIFA). It is always interesting to apply technical analysis to understand what makes the best football player and football team. In this article, we try to answer the question &amp;quot;what kind of tactical play is the most advanced&amp;quot; through statistical analysis on the game data of English Premier League in the 21-22 season. To be more specific, firstly, we applied descriptive statistics to analyze the technical and tactical play of each team, and then screen out the technical and tactical indicators that significantly affect the outcome of the game through one-way analysis of variance (ANOVA) and discriminate analysis, and the preliminary target conclusions were obtained. BP neural network was then carried out to predict the rankings of the Premier League teams by using the indicators selected by ANOVA and discriminant analysis, as input value. BP neural network prediction model is then established to predict the ranking of each team in the 22-23 season. A general conclusion and make suggestions on the planning of the technical and tactical playing methods of our country&amp;apos;s youth soccer sports.

Gibbs sampler for Bayesian prediction of triple seasonal autoregressive processes

Researchers have extended autoregressive (AR) time-series models to adequately fit and model time-series with triple seasonality. These AR extensions can be referred to as triple seasonal AR (TSAR) models. For these TSAR time-series models, only Bayesian estimation and identification have been introduced. Therefore, in this article, we aim to extend the existing work for presenting the Bayesian prediction for TSAR models using the Gibbs sampler algorithm. In this Bayesian prediction, we first assume the TSAR errors are identically normally distributed, and we employ normal-gamma and g priors for the TSAR parameters. Based on the normally distributed TSAR errors and the specified TSAR parameters’ priors, we are able to derive the conditional predictive and posterior densities. Particularly, we show the conditional posterior densities of TSAR coefficients and variance are the multivariate normal and inverse-gamma, respectively. In addition, for the future TSAR observations, we show the conditional predictive density is the multivariate normal. Using these derived full conditional distributions, we propose the Gibbs sampler to efficiently approximate the joint predictive and posterior densities and to easily carry out multiple-step ahead predictions. We validate the accuracy of forecasting for the proposed Gibbs sampler by conducting a simulation study. Moreover, we apply the proposed Gibbs sampler to hourly electricity-load time-series datasets in some European countries, along with a comparison with the well-known long-short term memory (LSTM) recurrent neural model.

Coupling Graph Neural Networks with Fractional Order Continuous Dynamics: A Robustness Study

In this work, we rigorously investigate the robustness of graph neural fractional-order differential equation (FDE) models. This framework extends beyond traditional graph neural (integer-order) ordinary differential equation (ODE) models by implementing the time-fractional Caputo derivative. Utilizing fractional calculus allows our model to consider long-term memory during the feature updating process, diverging from the memoryless Markovian updates seen in traditional graph neural ODE models. The superiority of graph neural FDE models over graph neural ODE models has been established in environments free from attacks or perturbations. While traditional graph neural ODE models have been verified to possess a degree of stability and resilience in the presence of adversarial attacks in existing literature, the robustness of graph neural FDE models, especially under adversarial conditions, remains largely unexplored. This paper undertakes a detailed assessment of the robustness of graph neural FDE models. We establish a theoretical foundation outlining the robustness characteristics of graph neural FDE models, highlighting that they maintain more stringent output perturbation bounds in the face of input and graph topology disturbances, compared to their integer-order counterparts. Our empirical evaluations further confirm the enhanced robustness of graph neural FDE models, highlighting their potential in adversarially robust applications.