scholarly journals A Neural Network-Inspired Matrix Formulation of Chemical Kinetics for Acceleration on GPUs

Energies ◽  
2021 ◽  
Vol 14 (9) ◽  
pp. 2710
Author(s):  
Shivam Barwey ◽  
Venkat Raman
Keyword(s):  
Neural Network ◽  
Chemical Kinetics ◽  
Graphics Processing Units ◽  
Source Term ◽  
Turbulent Flames ◽  
Peak Performance ◽  
Computational Time ◽  
Detailed Chemical Kinetics ◽  
The Matrix ◽  
Artificial Neural

High-fidelity simulations of turbulent flames are computationally expensive when using detailed chemical kinetics. For practical fuels and flow configurations, chemical kinetics can account for the vast majority of the computational time due to the highly non-linear nature of multi-step chemistry mechanisms and the inherent stiffness of combustion chemistry. While reducing this cost has been a key focus area in combustion modeling, the recent growth in graphics processing units (GPUs) that offer very fast arithmetic processing, combined with the development of highly optimized libraries for artificial neural networks used in machine learning, provides a unique pathway for acceleration. The goal of this paper is to recast Arrhenius kinetics as a neural network using matrix-based formulations. Unlike ANNs that rely on data, this formulation does not require training and exactly represents the chemistry mechanism. More specifically, connections between the exact matrix equations for kinetics and traditional artificial neural network layers are used to enable the usage of GPU-optimized linear algebra libraries without the need for modeling. Regarding GPU performance, speedup and saturation behaviors are assessed for several chemical mechanisms of varying complexity. The performance analysis is based on trends for absolute compute times and throughput for the various arithmetic operations encountered during the source term computation. The goals are ultimately to provide insights into how the source term calculations scale with the reaction mechanism complexity, which types of reactions benefit the GPU formulations most, and how to exploit the matrix-based formulations to provide optimal speedup for large mechanisms by using sparsity properties. Overall, the GPU performance for the species source term evaluations reveals many informative trends with regards to the effect of cell number on device saturation and speedup. Most importantly, it is shown that the matrix-based method enables highly efficient GPU performance across the board, achieving near-peak performance in saturated regimes.

scholarly journals BUCKLING LOAD ESTIMATION OF CRACKED COLUMNS USING ARTIFICIAL NEURAL NETWORK MODELING TECHNIQUE

2012 ◽  
Vol 18 (4) ◽  
pp. 568-579 ◽  
Author(s):  
Mahmut Bilgehan ◽  
Muhammet Arif Gürel ◽  
Recep Kadir Pekgökgöz ◽  
Murat Kısa
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Matrix Method ◽  
Buckling Load ◽  
Computational Time ◽  
Neural Network Modeling ◽  
Critical Buckling Load ◽  
The Matrix ◽  
Artificial Neural

In this paper, buckling analysis of slender prismatic columns with a single non-propagating open edge crack subjected to axial loads has been presented utilizing the transfer matrix method and the artificial neural networks. A multi-layer feedforward neural network learning by backpropagation algorithm has been employed in the study. The main focus of this work is the investigation of feasibility of using an artificial neural network to assess the critical buckling load of axially loaded compression rods. This is explored by comparing the performance of neural network models with the results of the matrix method for all considered support conditions. It can be seen from the results that the critical buckling load values obtained from the neural networks closely follow the values obtained from the matrix method for the whole data sets. The final results show that the proposed methodology may constitute an efficient tool for the estimation of elastic buckling loads of edge-cracked columns. Also, it can be seen from the results that the computational time reduces if the proposed method is used.

Novel Application to Recognize a Breakdown Pressure Event on Time Series Frac Data Vs. an Artificial Intelligence Approach

2021 ◽  
Author(s):  
Alberto Jose Ramirez ◽  
Jessica Graciela Iriarte
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Service Providers ◽  
Complex Model ◽  
Rate Of Change ◽  
Computational Time ◽  
Breakdown Pressure ◽  
Constant Multiple ◽  
Artificial Neural ◽  
Rule Based Approach

Abstract Breakdown pressure is the peak pressure attained when fluid is injected into a borehole until fracturing occurs. Hydraulic fracturing operations are conducted above the breakdown pressure, at which the rock formation fractures and allows fluids to flow inside. This value is essential to obtain formation stress measurements. The objective of this study is to automate the selection of breakdown pressure flags on time series fracture data using a novel algorithm in lieu of an artificial neural network. This study is based on high-frequency treatment data collected from a cloud-based software. The comma separated (.csv) files include treating pressure (TP), slurry rate (SR), and bottomhole proppant concentration (BHPC) with defined start and end time flags. Using feature engineering, the model calculates the rate of change of treating pressure (dtp_1st) slurry rate (dsr_1st), and bottomhole proppant concentration (dbhpc_1st). An algorithm isolates the initial area of the treatment plot before proppant reaches the perforations, the slurry rate is constant, and the pressure increases. The first approach uses a neural network trained with 872 stages to isolate the breakdown pressure area. The expert rule-based approach finds the highest pressure spikes where SR is constant. Then, a refining function finds the maximum treating pressure value and returns its job time as the predicted breakdown pressure flag. Due to the complexity of unconventional reservoirs, the treatment plots may show pressure changes while the slurry rate is constant multiple times during the same stage. The diverse behavior of the breakdown pressure inhibits an artificial neural network's ability to find one "consistent pattern" across the stage. The multiple patterns found through the stage makes it difficult to select an area to find the breakdown pressure value. Testing this complex model worked moderately well, but it made the computational time too high for deployment. On the other hand, the automation algorithm uses rules to find the breakdown pressure value with its location within the stage. The breakdown flag model was validated with 102 stages and tested with 775 stages, returning the location and values corresponding to the highest pressure point. Results show that 86% of the predicted breakdown pressures are within 65 psi of manually picked values. Breakdown pressure recognition automation is important because it saves time and allows engineers to focus on analytical tasks instead of repetitive data-structuring tasks. Automating this process brings consistency to the data across service providers and basins. In some cases, due to its ability to zoom-in, the algorithm recognized breakdown pressures with higher accuracy than subject matter experts. Comparing the results from two different approaches allowed us to conclude that similar or better results with lower running times can be achieved without using complex algorithms.

Daily Surface Solar Radiation Prediction Mapping Using Artificial Neural Network: The Case Study of Reunion Island

2019 ◽  
Vol 142 (2) ◽  
Author(s):  
Peng Li ◽  
Miloud Bessafi ◽  
Beatrice Morel ◽  
Jean-Pierre Chabriat ◽  
Mathieu Delsaut ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Wavelet Transform ◽  
Solar Radiation ◽  
Computational Time ◽  
Reunion Island ◽  
Surface Solar Radiation ◽  
Clear Sky ◽  
Réunion Island ◽  
Artificial Neural

Abstract This paper focuses on the prediction of daily surface solar radiation maps for Reunion Island by a hybrid approach that combines principal component analysis (PCA), wavelet transform analysis, and artificial neural network (ANN). The daily surface solar radiation over 18 years (1999–2016) from CM SAF (SARAH-E with 0.05 deg × 0.05 deg spatial resolution) is first detrended using the clear sky index. Dimensionality reduction of the detrended dataset is secondly performed through PCA, which results in saving computational time by a factor of eight in comparison to not using PCA. A wavelet transform is thirdly applied onto each of the first 28 principal components (PCs) explaining 95% of the variance. The decomposed nine-wavelet components for each PC are fourthly used as input to an ANN model to perform the prediction of day-ahead surface solar radiation. The predicted decomposed components are finally returned to PCs and clear sky indices, irradiation in the end for re-mapping the surface solar radiation's distribution. It is found that the prediction accuracy is quite satisfying: root mean square error (RMSE) is 30.98 W/m2 and the (1 − RMSE_prediction/RMSE_persistence) is 0.409.

scholarly journals Application of artificial neural network for predicting the performance of CO2 enhanced oil recovery and storage in residual oil zones

2020 ◽  
Vol 10 (1) ◽  
Author(s):  
Hung Vo Thanh ◽  
Yuichi Sugai ◽  
Kyuro Sasaki
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Oil Recovery ◽  
Early Stage ◽  
Computational Time ◽  
Residual Oil ◽  
Ann Model ◽  
Ann Models ◽  
Artificial Neural ◽  
And Storage

Abstract Residual Oil Zones (ROZs) become potential formations for Carbon Capture, Utilization, and Storage (CCUS). Although the growing attention in ROZs, there is a lack of studies to propose the fast tool for evaluating the performance of a CO2 injection process. In this paper, we introduce the application of artificial neural network (ANN) for predicting the oil recovery and CO2 storage capacity in ROZs. The uncertainties parameters, including the geological factors and well operations, were used for generating the training database. Then, a total of 351 numerical samples were simulated and created the Cumulative oil production, Cumulative CO2 storage, and Cumulative CO2 retained. The results indicated that the developed ANN model had an excellent prediction performance with a high correlation coefficient (R2) was over 0.98 on comparing with objective values, and the total root mean square error of less than 2%. Also, the accuracy and stability of ANN models were validated for five real ROZs in the Permian Basin. The predictive results were an excellent agreement between ANN predictions and field report data. These results indicated that the ANN model could predict the CO2 storage and oil recovery with high accuracy, and it can be applied as a robust tool to determine the feasibility in the early stage of CCUS in ROZs. Finally, the prospective application of the developed ANN model was assessed by optimization CO2-EOR and storage projects. The developed ANN models reduced the computational time for the optimization process in ROZs.

Computational Issues for Simulating Finite-Rate Kinetics in LES

2002 ◽  
Author(s):  
Rajat Kapoor ◽  
Suresh Menon
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Essential Feature ◽  
Turbulent Flames ◽  
Finite Rate ◽  
Parallel Solvers ◽  
Effective Manner ◽  
Rate Kinetics ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

At present, large-eddy simulations (LES) of turbulent flames with multi-species finite-rate kinetics is computationally infeasible due to the enormous cost associated with computation of reaction kinetics in 3D flows. In a recent study, In-Situ Adaptive Tabulation (ISAT) and Artificial Neural Network (ANN) methodologies were developed for computing finite-rate kinetics in a cost effective manner. Although ISAT reduces the cost of direct integration considerably, the ISAT tables require significant on-line storage in memory and can continue to grow over multiple flow-through times (an essential feature in LES). Hence, direct use of ISAT in LES may not be practical, especially in parallel solvers. In this study, a storage-efficient Artificial Neural Network (ANN) is investigated for LES application. Preliminary studies using ANN to predict freely propagating turbulent premixed flames over a range of operational parameters are described and issues regarding the implementation of such ANNs for engineering LES are discussed.

scholarly journals An Optimal Approach for Heart Sound Classification Using Artificial Neural Network

2021 ◽  
Author(s):  
Yunendah Nur Fu’adah ◽  
Ki Moo Lim
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Heart Sound ◽  
Learning Technologies ◽  
Computational Time ◽  
Cardiovascular Disorders ◽  
Sound Classification ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Optimal Approach

Abstract Heart sound auscultation is one of the most widely used approaches for detecting cardiovascular disorders. Diagnosing abnormalities of heart sound using a stethoscope depends on the physician’s skill and judgement. Several studies have shown promising results in the automatic detection of cardiovascular disorders based on heart sound signals. However, the accuracy performance needs to be improved as automated heart sound classification aids in the early detection and prevention of the dangerous effects of cardiovascular problems. In this study, an optimal heart sound classification method based on machine learning technologies for cardiovascular disease prediction is performed. It consists of three steps: pre-processing that sets the 5 s duration of the Physionet Challenge 2016 datasets, feature extraction using mel-frequency cepstrum coefficients (MFCC), and classification using an artificial neural network (ANN) with one hidden layer that provides low parameter consumption. Ten-fold cross-validation was used to evaluate the performance of the proposed method. The best model obtained 94% accuracy and 93% AUC score, which were assessed using 1626 test datasets. Taken together, the results show that the proposed method obtained excellent classification results and provided low parameter consumption, thereby reducing computational time to facilitate a real-time implementation.

Hybrid scheduling and maintenance problem using artificial neural network based meta-heuristics

2017 ◽  
Vol 12 (3) ◽  
pp. 525-550
Author(s):  
Mehdi Abedi ◽  
Hany Seidgar ◽  
Hamed Fazlollahtabar
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Imperialist Competitive Algorithm ◽  
Parameter Tuning ◽  
Parallel Machine Scheduling ◽  
Computational Time ◽  
Aging Effects ◽  
Content Type ◽  
Maintenance Activities ◽  
Artificial Neural

Purpose The purpose of this paper is to present a new mathematical model for the unrelated parallel machine scheduling problem with aging effects and multi-maintenance activities. Design/methodology/approach The authors assume that each machine may be subject to several maintenance activities over the scheduling horizon and a machine turn into its initial condition after maintenance activity and the aging effects start anew. The objective is to minimize the weighted sum of early/tardy times of jobs and maintenance costs. Findings As this problem is proven to be non-deterministic polynomial-time hard (NP-hard), the authors employed imperialist competitive algorithm (ICA) and genetic algorithm (GA) as solution approaches, and the parameters of the proposed algorithms are calibrated by a novel parameter tuning tool called Artificial Neural Network (ANN). The computational results clarify that GA performs better than ICA in quality of solutions and computational time. Originality/value Predictive maintenance (PM) activities carry out the operations on machines and tools before the breakdown takes place and it helps to prevent failures before they happen.

Proactive Scheduling for Steelmaking-Continuous Casting Plant with Uncertain Machine Breakdown Using Distribution-Based Robustness and Decomposed Artificial Neural Network

2015 ◽  
Vol 32 (02) ◽  
pp. 1550010 ◽  
Author(s):  
Kiatkajohn Worapradya ◽  
Purit Thanakijkasem
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Continuous Casting ◽  
System Performance ◽  
Traditional Approach ◽  
Model Complexity ◽  
Experimental Result ◽  
Computational Time ◽  
Proactive Scheduling ◽  
Artificial Neural

An unpredictable breakdown often occurs and tends to complicate production scheduling in a steelmaking-continuous casting (SCC) plant. Because of particular characteristics and technology constraints of the SCC plant, traditional robust scheduling often provides an excessively conservative solution. This paper proposes an effective proactive scheduling that utilizes robustness adopting a distribution curve of a system performance created as a mix-integer model. The proposed robustness is designed to work effectively with the existing factory operation and is based on uncertainty assessment. In this paper, artificial neural network (ANN) is adopted with a challenge of designing an accurate model due to the model complexity from the discrete and nonlinear properties of the system performance. The ANN model is achieved by applying k-mean clustering, which decomposes machines to smaller groups having similar effect to the uncertainty. A case study based on data from a real SCC plant is conducted to demonstrate the methodology. The experimental result shows that the proposed methodology is successful in designing a robust schedule that provides a lower production cost under an acceptable breakdown probability while also consuming less computational time compared with the traditional approach.

scholarly journals Artificial Neural Network to Estimate the Paddy Yield Prediction Using Climatic Data

2020 ◽  
Vol 2020 ◽  
pp. 1-11 ◽  
Author(s):  
Vinushi Amaratunga ◽  
Lasini Wickramasinghe ◽  
Anushka Perera ◽  
Jeevani Jayasinghe ◽  
Upaka Rathnayake
Keyword(s):  
Neural Network ◽  
Climate Change ◽  
Climatic Factors ◽  
Computational Time ◽  
Training Algorithm ◽  
Climate Data ◽  
Nonlinear Relationships ◽  
The Future ◽  
Artificial Neural ◽  
Paddy Yield

Paddy harvest is extremely vulnerable to climate change and climate variations. It is a well-known fact that climate change has been accelerated over the past decades due to various human induced activities. In addition, demand for the food is increasing day-by-day due to the rapid growth of population. Therefore, understanding the relationships between climatic factors and paddy production has become crucial for the sustainability of the agriculture sector. However, these relationships are usually complex nonlinear relationships. Artificial Neural Networks (ANNs) are extensively used in obtaining these complex, nonlinear relationships. However, these relationships are not yet obtained in the context of Sri Lanka; a country where its staple food is rice. Therefore, this research presents an attempt in obtaining the relationships between the paddy yield and climatic parameters for several paddy grown areas (Ampara, Batticaloa, Badulla, Bandarawela, Hambantota, Trincomalee, Kurunegala, and Puttalam) with available data. Three training algorithms (Levenberg–Marquardt (LM), Bayesian Regularization (BR), and Scaled Conjugated Gradient (SCG)) are used to train the developed neural network model, and they are compared against each other to find the better training algorithm. Correlation coefficient (R) and Mean Squared Error (MSE) were used as the performance indicators to evaluate the performance of the developed ANN models. The results obtained from this study reveal that LM training algorithm has outperformed the other two algorithms in determining the relationships between climatic factors and paddy yield with less computational time. In addition, in the absence of seasonal climate data, annual prediction process is understood as an efficient prediction process. However, the results reveal that there is an error threshold in the prediction. Nevertheless, the obtained results are stable and acceptable under the highly unpredicted climate scenarios. The ANN relationships developed can be used to predict the future paddy yields in corresponding areas with the future climate data from various climate models.

Sign in / Sign up

Export Citation Format

Share Document