Modelling aerosol concentration distributions from transient (puff) sourcesA paper submitted to the Journal of Environmental Engineering and Science.

2009 ◽  
Vol 36 (5) ◽  
pp. 911-922 ◽  
Author(s):  
Timothy J. DeVito ◽  
Xiaoying Cao ◽  
Gilles Roy ◽  
Johnathan R. Costa ◽  
William S. Andrews
Keyword(s):  
Gaussian Model ◽  
Aerosol Concentration ◽  
Atmospheric Conditions ◽  
Dispersion Coefficients ◽  
Scanning Lidar ◽  
Ann Models ◽  
Artificial Neural Network Ann ◽  
Gaussian Puff ◽  
Gaussian Puff Model ◽  
System Operating

A field trial involving 50 separate releases of inert aerosol (kaolin) was conducted to determine the concentration distribution within aerosol puffs resulting from near-instantaneous releases. Atmospheric conditions during the trial fell within Pasquill stability classes A and B (very and moderately unstable, respectively). Aerosol concentration measurements were made using a scanning lidar system operating at 1.06 μm. Artificial neural network (ANN) models were developed using the data to predict concentration distributions, given a number of meteorological parameters. The ANN predictions were compared to those from traditional Gaussian puff models, and provided better predictions than the Gaussian model parameterizations examined. The ANN models were also used to develop Gaussian fitting parameters to replace traditional Pasquill and Slade dispersion coefficients. The ANN-derived dispersion coefficients provided better predictions of measured puff concentration distributions than either the Pasquill or Slade parameterizations, though the full multi-input ANN models provided even better predictions than the Gaussian puff model using ANN-derived dispersion coefficients.

scholarly journals Inverse Design of Fe-Based Bulk Metallic Glasses Using Machine Learning

Metals ◽  
2021 ◽  
Vol 11 (5) ◽  
pp. 729
Author(s):  
Junhyub Jeon ◽  
Namhyuk Seo ◽  
Hwi-Jun Kim ◽  
Min-Ha Lee ◽  
Hyun-Kyu Lim ◽  
...  
Keyword(s):  
Thermal Properties ◽  
Metallic Glasses ◽  
Alloy Composition ◽  
Relevant Literature ◽  
Bulk Metallic Glasses ◽  
X Ray Diffraction ◽  
X Ray ◽  
Ann Models ◽  
Wide Scale ◽  
Artificial Neural Network Ann

Fe-based bulk metallic glasses (BMGs) are a unique class of materials that are attracting attention in a wide variety of applications owing to their physical properties. Several studies have investigated and designed the relationships between alloy composition and thermal properties of BMGs using an artificial neural network (ANN). The limitation of the wide-scale use of these models is that the required composition is yet to be found despite numerous case studies. To address this issue, we trained an ANN to design Fe-based BMGs that predict the thermal properties. Models were trained using only the composition of the alloy as input and were created from a database of more than 150 experimental data of Fe-based BMGs from relevant literature. We adopted these ANN models to design BMGs with thermal properties to satisfy the intended purpose using particle swarm optimization. A melt spinner was employed to fabricate the designed alloys. X-ray diffraction and differential thermal analysis tests were used to evaluate the specimens.

Evaluation of the performance of different short-range atmospheric dispersion models for the monitoring of CH4 emissions from industrial facilities

2021 ◽  
Author(s):  
Bonaventure Fontanier ◽  
Pramod Kumar ◽  
Grégoire Broquet ◽  
Christopher Caldow ◽  
Olivier Laurent ◽  
...  
Keyword(s):  
High Frequency ◽  
Meteorological Data ◽  
Atmospheric Dispersion ◽  
Local Scale ◽  
Lagrangian Model ◽  
Dispersion Models ◽  
Industrial Facilities ◽  
Gaussian Plume ◽  
Gaussian Puff ◽  
Gaussian Puff Model

<p>Methane (CH<sub>4</sub>) is a powerful greenhouse gas which plays a major role in climate change. The accurate monitoring of emissions from industrial facilities is needed to ensure efficient emission mitigation strategies. Local-scale atmospheric inversions are increasingly being used to provide estimates of the rates and/or locations of CH<sub>4</sub> sources from industrial sites. They rely on local-scale atmospheric dispersion models, CH<sub>4</sub> measurements and inversion approaches. Gaussian plume models have often been used for local-scale atmospheric dispersion modelling and inversions of emissions, because of their simplicity and good performance when used in a flat terrain and relatively constant mean wind conditions. However, even in such conditions, failure to account for wind and mole fraction variability can limit the ability to exploit the full potential of these measurements at high frequency.</p><p>We study whether the accuracy of inversions can be increased by the use of more complex dispersion models. Our assessments are based on the analysis of 25 to 75-min CH<sub>4 </sub>controlled releases during a one-week campaign in October 2019 at the TOTAL’s TADI operative platform in Lacq, France (in a flat area). During this campaign, for each controlled release, we conducted near-surface in situ measurements of CH<sub>4</sub> mole fraction from both a mobile vehicle and a circle of fixed points around the emission area. Our inversions based on a Gaussian model and either the mobile or fixed-point measurements both provided estimates of the release rates with 20-30% precision.  </p><p>Here we focus on comparisons between modeling and inversion results when using this Gaussian plume model, a Lagrangian model “GRAL” and a Gaussian puff model. The parameters for the three models are based on high-frequency meteorological values from a single stationary 3D sonic anemometer. GRAL should have relatively good skills under low-wind speed conditions. The Gaussian puff is a light implementation of time-dependent modeling and can be driven by high-frequency meteorological data. The performance of these dispersion models is evaluated with various metrics from the observation field that are relevant for the inversion. These analyses lead to the exploration of new types of definitions of the observational constraint for the inversions with the Gaussian puff model, when using the timeseries from fixed measurement points. The definitions explore a range of metrics in the time domain as well as in the frequency domain.</p><p>Eventually, the Lagrangian model does not outperform the Gaussian plume model in these experiments, its application being notably limited by the short scales of the transport characteristics. On the other hand, the Gaussian puff model provides promising results for the inversion, in particular, in terms of comparison between the simulated and observed timeseries for fixed stations. Its performance when driven by a spatially uniform wind field is an incentive to explore the use of meteorological data from several sonic stations to parameterize its configuration. The fixed-point measurements are shown to allow for more robust inversions of the source location than the mobile measurements, with an average source localization error of the order of 10 m.</p>

scholarly journals Invasion performance-similarity found among multiple cell systems

2020 ◽  
Author(s):  
Om Prakash
Keyword(s):  
Cell Lines ◽  
Drug Repositioning ◽  
Biological Research ◽  
Multiple Systems ◽  
Biologically Relevant ◽  
Selective Targeting ◽  
Ann Models ◽  
Multiple Cell ◽  
Artificial Neural Network Ann ◽  
Network Mapping

ABSTRACTUnderstanding of inter-system behavior develops biologically relevant intuition for drug repositioning as well as other biological research. But combining all the possible genes interactions into a system, and furthermore comparisons of multiple systems are a challenge on time ground with feasible experiments. In present study, 64 cell lines from 11 different organs were compared for their invasion performance. RNA expressions of 23 genes were used to create systems artificial neural network (ANN) models. ANN models were prepared for all 64 cell lines and observed for their invasion performance through network mapping. The resulted cell line clusters bear feasible capacity to perform experiments for biologically relevant research motivations as drug repositioning and selective targeting etc.; and can be used for analysis of invasion related aspects.

ARTIFICIAL NEURAL NETWORK (ANN)-BASED PREDICTIVE TOOL FOR ESTIMATING LIGHTNING DAMAGE IN COMPOSITES

2021 ◽  
Author(s):  
DEVIN NIELSEN ◽  
TYLER LOTT ◽  
SOM DUTTA ◽  
JUHYEONG LEE
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Network Architecture ◽  
Back Propagation ◽  
Training Dataset ◽  
Predictive Tool ◽  
Ann Models ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Hidden Layer

In this study, three artificial neural network (ANN) models are developed with back propagation (BP) optimization algorithms to predict various lightning damage modes in carbon/epoxy laminates. The proposed ANN models use three input variables associated with lightning waveform parameters (i.e., the peak current amplitude, rising time, and decaying time) to predict fiber damage, matrix damage, and through-thickness damage in the composites. The data used for training and testing the networks was actual lightning damage data collected from peer-reviewed published literature. Various BP training algorithms and network architecture configurations (i.e., data splitting, the number of neurons in a hidden layer, and the number of hidden layers) have been tested to improve the performance of the neural networks. Among the various BP algorithms considered, the Bayesian regularization back propagation (BRBP) showed the overall best performance in lightning damage prediction. When using the BRBP algorithm, as expected, the greater the fraction of the collected data that is allocated to the training dataset, the better the network is trained. In addition, the optimal ANN architecture was found to have a single hidden layer with 20 neurons. The ANN models proposed in this work may prove useful in preliminary assessments of lightning damage and reduce the number of expensive experimental lightning tests.

Neural Network Models

2016 ◽  
pp. 368-395
Author(s):  
Eliano Pessa
Keyword(s):  
Neural Network ◽  
Statistical Physics ◽  
Optimization Theory ◽  
Network Models ◽  
Cooperative Interaction ◽  
New Paradigm ◽  
Neural Network Models ◽  
Ann Models ◽  
Artificial Neural Network Ann ◽  
The One

The Artificial Neural Network (ANN) models gained a wide popularity owing to a number of claimed advantages such as biological plausibility, tolerance with respect to errors or noise in the input data, learning ability allowing an adaptability to environmental constraints. Notwithstanding the fact that most of these advantages are not typical only of ANNs, engineers, psychologists and neuroscientists made an extended use of ANN models in a large number of scientific investigations. In most cases, however, these models have been introduced in order to provide optimization tools more useful than the ones commonly used by traditional Optimization Theory. Unfortunately, just the successful performance of ANN models in optimization tasks produced a widespread neglect of the true – and important – objectives pursued by the first promoters of these models. These objectives can be shortly summarized by the manifesto of connectionist psychology, stating that mental processes are nothing but macroscopic phenomena, emergent from the cooperative interaction of a large number of microscopic knowledge units. This statement – wholly in line with the goal of statistical mechanics – can be readily extended to other processes, beyond the mental ones, including social, economic, and, in general, organizational ones. Therefore this chapter has been designed in order to answer a number of related questions, such as: are the ANN models able to grant for the occurrence of this sort of emergence? How can the occurrence of this emergence be empirically detected? How can the emergence produced by ANN models be controlled? In which sense the ANN emergence could offer a new paradigm for the explanation of macroscopic phenomena? Answering these questions induces to focus the chapter on less popular ANNs, such as the recurrent ones, while neglecting more popular models, such as perceptrons, and on less used units, such as spiking neurons, rather than on McCulloch-Pitts neurons. Moreover, the chapter must mention a number of strategies of emergence detection, useful for researchers performing computer simulations of ANN behaviours. Among these strategies it is possible to quote the reduction of ANN models to continuous models, such as the neural field models or the neural mass models, the recourse to the methods of Network Theory and the employment of techniques borrowed by Statistical Physics, like the one based on the Renormalization Group. Of course, owing to space (and mathematical expertise) requirements, most mathematical details of the proposed arguments are neglected, and, to gain more information, the reader is deferred to the quoted literature.

scholarly journals Newly Developed Correlations to Predict the Rheological Parameters of High-Bentonite Drilling Fluid Using Neural Networks

Sensors ◽  
2020 ◽  
Vol 20 (10) ◽  
pp. 2787
Author(s):  
Ahmed Gowida ◽  
Salaheldin Elkatatny ◽  
Khaled Abdelgawad ◽  
Rahul Gajbhiye
Keyword(s):  
Rheological Properties ◽  
Drilling Fluid ◽  
Rheological Parameters ◽  
Percentage Error ◽  
Cleaning Efficiency ◽  
Periodic Monitoring ◽  
Data Points ◽  
Ann Models ◽  
Artificial Neural Network Ann ◽  
Marsh Funnel

High-bentonite mud (HBM) is a water-based drilling fluid characterized by its remarkable improvement in cutting removal and hole cleaning efficiency. Periodic monitoring of the rheological properties of HBM is mandatory for optimizing the drilling operation. The objective of this study is to develop new sets of correlations using artificial neural network (ANN) to predict the rheological parameters of HBM while drilling using the frequent measurements, every 15 to 20 min, of mud density (MD) and Marsh funnel viscosity (FV). The ANN models were developed using 200 field data points. The dataset was divided into 70:30 ratios for training and testing the ANN models respectively. The optimized ANN models showed a significant match between the predicted and the measured rheological properties with a high correlation coefficient (R) higher than 0.90 and a maximum average absolute percentage error (AAPE) of 6%. New empirical correlations were extracted from the ANN models to estimate plastic viscosity (PV), yield point (YP), and apparent viscosity (AV) directly without running the models for easier and practical application. The results obtained from AV empirical correlation outperformed the previously published correlations in terms of R and AAPE.

Development of Neural Network Models for a Crude Oil Distillation Column

2012 ◽  
Author(s):  
Khairiyah Mohd. Yusof ◽  
Fakhri Karray ◽  
Peter L. Douglas
Keyword(s):  
Neural Network ◽  
Steady State ◽  
Crude Oil ◽  
Distillation Column ◽  
Network Models ◽  
Ann Model ◽  
Neural Network Models ◽  
Ann Models ◽  
Artificial Neural Network Ann ◽  
Crude Oil Distillation

This paper discusses the development of artificial neural network (ANN) models for a crude oil distillation column. Since the model is developed for real time optimisation (RTO) applications they are steady state, multivariable models. Training and testing data used to develop the models were generated from a reconciled steady-state model simulated in a process simulator. The radial basis function networks (RBFN), a type of feedforward ANN model, were able to model the crude tower very well, with the root mean square error for the prediction of each variable less than 1%. Grouping related output variables in a network model was found to give better predictions than lumping all the variables in a single model; this also allowed the overall complex, multivariable model to be simplified into smaller models that are more manageable. In addition, the RBFN models were also able to satisfactorily perform range and dimensional extrapolation, which is necessary for models that are used in RTO.

Computational Performance Analysis of Neural Network and Regression Models in Forecasting the Societal Demand for Agricultural Food Harvests

2022 ◽  
pp. 1287-1300
Author(s):  
Balaji Prabhu B. V. ◽  
M. Dakshayini
Keyword(s):  
Neural Network ◽  
Crop Production ◽  
Demand Forecasting ◽  
Forecasting Model ◽  
Ann Model ◽  
Demand Prediction ◽  
Computational Performance ◽  
Societal Needs ◽  
Ann Models ◽  
Artificial Neural Network Ann

Demand forecasting plays an important role in the field of agriculture, where a farmer can plan for the crop production according to the demand in future and make a profitable crop business. There exist a various statistical and machine learning methods for forecasting the demand, selecting the best forecasting model is desirable. In this work, a multiple linear regression (MLR) and an artificial neural network (ANN) model have been implemented for forecasting an optimum societal demand for various food crops that are commonly used in day to day life. The models are implemented using R toll, linear model and neuralnet packages for training and optimization of the MLR and ANN models. Then, the results obtained by the ANN were compared with the results obtained with MLR models. The results obtained indicated that the designed models are useful, reliable, and quite an effective tool for optimizing the effects of demand prediction in controlling the supply of food harvests to match the societal needs satisfactorily.

scholarly journals Using the artificial neural networks for prediction and validating solar radiation

2019 ◽  
Vol 27 (1) ◽  
Author(s):  
Zahraa E. Mohamed
Keyword(s):  
Solar Radiation ◽  
Global Solar Radiation ◽  
Horizontal Surface ◽  
The Other ◽  
Rate Coefficients ◽  
Ann Models ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Statistical Indicators ◽  
Better Than

AbstractThe main objective of this paper is to employ the artificial neural network (ANN) models for validating and predicting global solar radiation (GSR) on a horizontal surface of three Egyptian cities. The feedforward backpropagation ANNs are utilized based on two algorithms which are the basic backpropagation (Bp) and the Bp with momentum and learning rate coefficients respectively. The statistical indicators are used to investigate the performance of ANN models. According to these indicators, the results of the second algorithm are better than the other. Also, model (6) in this method has the lowest RMSE values for all cities in this study. The study indicated that the second method is the most suitable for predicting GSR on a horizontal surface of all cities in this work. Moreover, ANN-based model is an efficient method which has higher precision.

Sign in / Sign up

Export Citation Format

Share Document