Shut-in Stabilization Time Optimization for Better Reservoir Pressure Monitoring Harnessing Neural Network Modeling

Abstract Stabilization time is an essential key for pressure measurement accuracy. Obtaining representative pressure points in build-up tests for pressure-sensitive reservoirs is driven by optimizing stabilization time. An artificial intelligence technique was used in the study for testing pressure-sensitive reservoirs using measuring gauges. The stabilization time function of reservoir characteristics is generally calculated using the diffusivity equation where rock and fluid properties are honored. The artificial neural network (ANN) technique will be used to predict the stabilization time and optimize it using readily available and known inputs or parameters. The values obtained from the formula known as the diffusion formula and the ANN technique are then compared against the actual values measured from pressure gauges in the reservoirs. The optimization of the number of datasets required to be fed to the network to allow for coverage over the whole range is essential as opposed to the clustering of the datasets. A total of about 3000 pressure derivative samples from the wells were used in the testing, training, and validation of the ANN. The datasets are optimized by dividing them into three fractional parts, and the number optimized through monitoring the ANN performance. The optimization of the stabilization time is essential and leads to the improvement of the ANN learning process. The sensitivity analysis proves that the use of the formula and ANN technique, compared to actual datasets, is better since, in the formula and ANN technique, the time was optimized with an average absolute relative error of 3.67%. The results are near the same, especially when the ANN technique undergoes testing using known and easily available parameters. Time optimization is essential since discreet points or datasets in the ANN technique and formula would not work, allowing ANN to work in situations of optimization. The study was expected to provide additional data and information, considering that stabilization time is essential in obtaining the pressure map representation. ANN is a superior technique and, through its superiority, allows for proper optimization of time as a parameter. Thus it can predict reservoir log data almost accurately. The method used in the study shows the importance of optimizing pressure stabilization time through reduction. The study results can, therefore, be applied in reservoir testing to achieve optimal results.

Download Full-text

Predicting Pressure Sensitivity to Luminophore Content and Paint Thickness of Pressure-Sensitive Paint Using Artificial Neural Network

Sensors ◽

10.3390/s21155188 ◽

2021 ◽

Vol 21 (15) ◽

pp. 5188

Author(s):

Mitsugu Hasegawa ◽

Daiki Kurihara ◽

Yasuhiro Egami ◽

Hirotaka Sakaue ◽

Aleksandar Jemcov

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Data Augmentation ◽

Percentage Error ◽

Pressure Sensitivity ◽

Pressure Sensitive Paint ◽

Pressure Sensitive ◽

Artificial Neural ◽

Paint Thickness ◽

Artificial Neural Network Ann

An artificial neural network (ANN) was constructed and trained for predicting pressure sensitivity using an experimental dataset consisting of luminophore content and paint thickness as chemical and physical inputs. A data augmentation technique was used to increase the number of data points based on the limited experimental observations. The prediction accuracy of the trained ANN was evaluated by using a metric, mean absolute percentage error. The ANN predicted pressure sensitivity to luminophore content and to paint thickness, within confidence intervals based on experimental errors. The present approach of applying ANN and the data augmentation has the potential to predict pressure-sensitive paint (PSP) characterizations that improve the performance of PSP for global surface pressure measurements.

Download Full-text

Predicting Fatigue Life of Pre-Corroded LC4 Aluminum Alloy by Artificial Neural Network

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.118-120.221 ◽

2010 ◽

Vol 118-120 ◽

pp. 221-225 ◽

Cited By ~ 1

Author(s):

Cheng Long Xu ◽

Sheng Li Lv ◽

Zhen Guo Wang ◽

Wei Zhang

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Aluminum Alloy ◽

Fatigue Life ◽

Network Architecture ◽

Fatigue Tests ◽

Data Sets ◽

Absolute Relative Error ◽

Artificial Neural ◽

Artificial Neural Network Ann

The purpose of this work was to predict the fatigue life of pre-corroded LC4 aluminum alloy by applying artificial neural network (ANN). Specimens were exposed to the same corrosive environment for 24h, 48h, and 72h. Fatigue tests were conducted under different stress levels. The existing experimental data sets were used for training and testing the construction of proposed network. A suitable network architecture (2-15-1) was proposed with good performance in this study. For evaluating the method efficiency, the experimental results have been compared to values predicted by ANN. The maximum absolute relative error for predicted values does not exceed 5%. Therefore it can be concluded that using neural networks to predict the fatigue life of LC4 is feasible and reliable.

Download Full-text

Prediction of Cutting Temperatures by Using Back Propagation Neural Network Modeling when Cutting Hardened H-13 Steel in CNC End Milling

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.576.91 ◽

2012 ◽

Vol 576 ◽

pp. 91-94 ◽

Cited By ~ 8

Author(s):

Erry Yulian Triblas Adesta ◽

Muataz H.F. Al Hazza ◽

M.Y. Suprianto ◽

Muhammad Riza

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

End Milling ◽

Back Propagation ◽

Cutting Temperature ◽

Neural Network Modeling ◽

Feed Forward Back Propagation ◽

Artificial Neural ◽

Artificial Neural Network Ann ◽

Cnc End Milling

Machining of hardened steel at high cutting speeds produces high temperatures in the cutting zone, which affects the surface quality and cutting tool life. Thus, predicting the temperature in early stage becomes utmost importance. This research presents a neural network model for predicting the cutting temperature in the CNC end milling process. The Artificial Neural Network (ANN) was applied as an effective tool for modeling and predicting the cutting temperature. A set of sparse experimental data for finish end milling on AISI H13 at hardness of 48 HRC have been conducted to measure the cutting temperature. The artificial neural network (ANN) was applied to predict the cutting temperature. Twenty hidden layer has been used with feed forward back propagation hierarchical neural networks were designed with Matlab2009b Neural Network Toolbox. The results show a high correlation between the predicted and the observed temperature which indicates the validity of the models.

Download Full-text

Perancangan Sistem Pendukung Keputusan Untuk Prediksi Penerima Beasiswa Menggunakan Metode Neural Network Backpropagation

Jurnal Teknologi Informasi dan Ilmu Komputer ◽

10.25126/jtiik.201852631 ◽

2018 ◽

Vol 5 (2) ◽

pp. 157 ◽

Cited By ~ 1

Author(s):

Ade Pujianto ◽

Kusrini Kusrini ◽

Andi Sunyoto

Keyword(s):

Neural Network ◽

Decision Making ◽

Artificial Neural Network ◽

Human Brain ◽

Learning Rate ◽

Neural Network Modeling ◽

Complex Problems ◽

Backpropagation Method ◽

Artificial Neural ◽

Artificial Neural Network Ann

Seleksi di Amikom masih mengalami kendala pada proses pengambilan keputusan, banyaknya data menyebabkan pengambil keputusan membutuhkan tools yang dapat membantu dalam menentukan penerima beasiswa, salah satu metode yang sering digunakan adalah artificial neural network (ANN). Metode ini meniru jaringan pemodelan saraf otak manusia berupa neuron-neuron untuk menyelesaikan suatu permasalahan. Salah satu penerapan neural network adalah untuk melakukan prediksi atau peramalan terhadap suatu peristiwa tertentu serta dianggap mampu menyelesaikan masalah yang komplek seperti penalaran otak manusia. Untuk menyelesaiakn masalah yang komplek neural network memerlukan banyak neuron atau yang biasa disebut layer (lapis). Salah satu metode neural network multi lapis adalah backpropagation yang mampu mengoptimalisasi bobot pada neuron dan menyelesaikan masalah yang komplek. Hasil dari penelitian ini adalah sebuah perancangan sistem prediksi dengan menggunakan metode neural network backpropagation untuk melakukan peramalan terhadap mahasiswa yang mendaftar beasiswa. hasil akhir penelitian ini adalah nilai akurasi sebesar 90% dan nilai error terkecil sebesar 0,000101 pada epoch ke 329 dengan jumlah 3000 data dengan pembagian data training 2.250 dan 750 data testing serta konfigurasi learning rate sebesar 0,2 dan momentum 0,2. Kata kunci: Artificial Neural netwok, Backpropagarion, Prediksi, beasiswa, Pengambilan Keputusan. AbstractSelection in Amikom is still constrained in the decision-making process, the number of data causing decision makers need tools that can assist in determining scholarship recipients, one of the most commonly used method is artificial neural network (ANN). This method mimics the neural network modeling of the human brain in the form of neurons to solve a problem. One application of neural network is to make predictions or forecasting of a particular event and is considered capable of solving complex problems such as human brain reasoning. To solve the problem the complex neural network requires many neurons or so-called layers. One method of multi layer neural network is backpropagation that is able to optimize the weight of neurons and solve complex problems. The result of this research is a prediction system design using neural network backpropagation method to forecast the students who apply for scholarship. the final result of this research is the accuracy value of 90% and the smallest error value of 0.000101 on epoch to 329 with the amount of 3000 data with sharing training 2,250 and 750 data testing and learning rate configuration of 0.2 and momentum 0.2.Keywords: Artificial Neural Netwok, Backpropagarion, Prediction, Scholarship, Decision Making.

Download Full-text

Artificial Neural Network Modeling of MoS2 Supercapacitor for Predicative Synthesis

International Journal of Innovative Technology and Exploring Engineering - Special Issue ◽

10.35940/ijitee.b6516.129219 ◽

2019 ◽

Vol 9 (2) ◽

pp. 554-560

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Energy Storage ◽

Neural Network Modeling ◽

Ann Model ◽

Energy Storage Devices ◽

Storage Devices ◽

Artificial Neural ◽

Forward Calculation ◽

Artificial Neural Network Ann

Energy storage systems are fundamental to the activity of intensity frameworks. They guarantee coherence of vitality supply and improve the dependability of the framework. The first area is centered on various energy storage frameworks, considering capacity limit, voltage and current proportions, and energy accessibility. Among the energy storage devices, supercapacitor is widely used because it is a high-limit capacitor with capacitance esteem a large amount than different capacitors. In the supercapacitor we have used MoS2 material synthesized with various Electrolytes. In perspective on the above mentioned, we report an Artificial Neural Network (ANN) strategy to achieve the predictable results. Levenberg- Marquardt feed-forward calculation prepares the neural network. We measure the exhibition of the ANN model with respect to mean square error (MSE) and the relationship coefficient between anticipated yield and yield given by the system. Results confirm the stability of supercapacitor over the other energy storage devices. To show such kind of conduct, we give Synthesis technique, Electrolyte, Cycle Life as an info esteems and Specific limit as yield esteem. For the amalgamation technique info esteem we have taken both compound and physical strategies by normalizing it. The practiced ANN demonstrating confirmations a higher number of concealed neuron design showing ideal execution as respects to expectation exactness

Download Full-text

Neural Network Modeling for Organizational Psychology

Deep Learning and Neural Networks ◽

10.4018/978-1-7998-0414-7.ch072 ◽

2020 ◽

pp. 1297-1309

Author(s):

Eliano Pessa

Keyword(s):

Neural Network ◽

Computational Models ◽

Network Modeling ◽

Organizational Psychology ◽

Neural Network Modeling ◽

Recurrent Networks ◽

Agent Based ◽

Ann Models ◽

Artificial Neural Network Ann ◽

The One

The nature itself of organizational psychology makes the study and modeling of emergence processes the key topic of this science. In this regard we can distinguish between two kinds of emergence: the one related to individual constructs and the other to collective constructs. In the former case the presence of suitable individual and contextual features gives rise to the emergence of suitable individual attitudes of holistic nature. In the latter case the features of single individuals belonging to a group, and reciprocally interacting, give rise to the occurrence of collective features and phenomena. In the last years both kinds of emergence have been studied through computational models. In this chapter we focus on the contribution of Artificial Neural Network (ANN) models to this modeling activity. As regards the emergence of individual constructs there is a consistent number of ANN-based models, most of which formulated in terms of recurrent networks. A review of their successes and failures constitutes a first part of the chapter. Instead, the emergence of collective constructs has been so far modelled by resorting to agent-based models. However, in recent times the ANN models have begun to be used with increasing frequency in this field. Namely, each agent can be modelled in an easier way by representing its structure through a suitable neural network. The final part of the chapter is, therefore, devoted to the problems underlying the use of ANNs as constituents of agent models.

Download Full-text

Fluoride contamination - Artificial neural network modeling and inverse distance weighting approach

Revue des sciences de l eau ◽

10.7202/1011606ar ◽

2012 ◽

Vol 25 (2) ◽

pp. 165-182 ◽

Cited By ~ 3

Author(s):

Imran Ahmad Dar ◽

K. Sankar ◽

Mithas Ahmad Dar ◽

Mrinmoy Majumder

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Neural Network Modeling ◽

Chemical Parameters ◽

Underground Waters ◽

Distance Weighting ◽

Distance Weighted ◽

Artificial Neural ◽

Artificial Neural Network Ann ◽

Inverse Distance

The underground waters in the Mamundiyar basin, India, present real chemical quality problems. Their fluoride content always exceeds the recommended levels. The Inverse Distance Weighted (IDW) method has been used for spatial interpolation of various key chemical parameters. Artificial Neural Network (ANN) modeling was applied to understand the correlation and sensitivity of all chemical parameters with respect to fluorides. The correlation of all the considered parameters is found to be poor where the highest correlation observed was only 0.37. This result showed that four of the parameters, namely pH, chlorides, sulphates and calcium, were found to have greater capacity of influencing fluorides than the other eight parameters. Chlorides were found to be the parameter that was the most sensitive and most correlated to fluorides.

Download Full-text

Kinetics, equilibrium isotherm and neural network modeling studies for the sorption of hexavalent chromium from aqueous solution by quartz/feldspar/wollastonite

RSC Advances ◽

10.1039/c5ra22851d ◽

2016 ◽

Vol 6 (7) ◽

pp. 5837-5847 ◽

Cited By ~ 13

Author(s):

B. Kavitha ◽

D. Sarala Thambavani

Keyword(s):

Neural Network ◽

Aqueous Solution ◽

Artificial Neural Network ◽

Adsorption Process ◽

Network Modeling ◽

Back Propagation ◽

Neural Network Modeling ◽

Training Algorithm ◽

Equilibrium Isotherm ◽

Artificial Neural Network Ann

A three layer feed forward artificial neural network (ANN) with back propagation training algorithm was developed to model the adsorption process of Cr(vi) in aqueous solution using riverbed sand containing quartz/feldspar/wollastonite (QFW) as adsorbent.

Download Full-text

Neural Network Modeling in Dithiothreitol Reduction and Ion Treatment of Recombinant Human Insulin Obtained from the Circular Dichroism (CD) Spectral Information

The Open Bioinformatics Journal ◽

10.2174/1875036200903010001 ◽

2009 ◽

Vol 3 (1) ◽

pp. 1-7

Author(s):

Sardari Soroush ◽

Soltani Saeed

Keyword(s):

Neural Network ◽

Human Insulin ◽

Model Building ◽

Three Dimensional ◽

Dimensional Structure ◽

Neural Network Modeling ◽

Recombinant Human Insulin ◽

Recombinant Peptide ◽

Peptide Molecule ◽

Artificial Neural Network Ann

In this study, induced changes in the secondary structure of the human insulin were carried out by addition of various reagents causing modification in the disulfide bond such as dithiothreitol (DTT) three dimensional structure of insulin. CD spectra were taken accordingly and the spectra recorded. There are methods to predict and estimate spectral changes of a peptide molecule, however there is no method to process CD spectral data and correlate them with that of inducing factor. Artificial intelligence backpropagation algorithm, as a strong model building tool was used here for prediction and data mining. Therefore, artificial neural network (ANN) methodology was used to build a model to study the effect of selected biochemical factors in the downstream process of a recombinant peptide.

Download Full-text

Photodegradation of roxarsone in the aquatic environment: influencing factors, mechanisms and artificial neural network modeling

10.21203/rs.3.rs-452155/v1 ◽

2021 ◽

Author(s):

Jizhong Meng ◽

Arong Arong ◽

Shoujun Yuan ◽

Wei Wang ◽

Juliang Jin ◽

...

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Influencing Factors ◽

Aquatic Environment ◽

Feed Additive ◽

Neural Network Modeling ◽

Factors Affecting ◽

Indirect Photolysis ◽

Artificial Neural ◽

Artificial Neural Network Ann

Abstract Roxarsone (ROX) is an organoarsenic feed additive, and can be discharged into aquatic environment. ROX can photodegrade into more toxic inorganic arsenics, causing arsenic pollution. However, the photodegradation behavior of ROX in aquatic environment is still unclear. To better understand ROX photodegradation behavior, this study investigated the ROX photodegradation mechanism and influencing factors, and modeled the photodegradation process. The results showed that ROX in the aquatic environment was degraded to inorganic As(III) and As(V) under light irradiation. The degradation efficiency was enhanced by 25 % with the increase of light intensity from 300 µW/cm2 to 800 µW/cm2 via indirect photolysis. The photodegradation was temperature dependence, but was only slightly affected by pH. Nitrate ion (NO3−) had an obvious influence, but sulfate, carbonate, and chlorate ions had a negligible effect on ROX degradation. Dissolved organic matter (DOM) in the solution inhibited the photodegradation. ROX photodegradation was mainly mediated by reactive oxygen species (in the form of single oxygen 1O2) generated through ROX self-sensitization under irradiation. Based on the data of factors affecting ROX photodegradation, ROX photodegradation model was built and trained by an artificial neural network (ANN), and the predicted degradation rate was in good agreement with the real values with a root mean square error of 1.008. This study improved the understanding of ROX photodegradation behavior and provided a basis for controlling the pollution from ROX photodegradation.

Download Full-text