Drilling rate prediction from petrophysical logs and mud logging data using an optimized multilayer perceptron neural network

Anemangely, Mohammad; Ramezanzadeh, Ahmad; Tokhmechi, Behzad; Molaghab, Abdollah; Mohammadian, Aram

doi:10.1088/1742-2140/aaac5d

Cited by 101 publications

(31 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All of the data were shuffled randomly to avoid the possible dependency of different samples on each other. We also scaled each variable by the maximum number of that variable type (each variable divided by the maximum of that variable type) (Anemangely et al, 2018b). For the figures, we scaled them back to reach a realistic evaluation.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…In the field of artificial neural network application in the oil industry, Anemangely, Ramezanzadeh, Tokhmechi, Molaghab, and Mohammadian (2018b) and Anemangely, Ramezanzadeh, and Tokhmechi (2017), used artificial neural network capabilities to estimate the penetration rate in well drilling as well as estimate the travel time of shear wave in the rock. They used petrophysical and mud logging data for the training and estimation of artificial neural network and obtained acceptable results (Anemangely et al, 2017(Anemangely et al, , 2018b.…”

Section: Estimated Stress Field By the Neural Networkmentioning

confidence: 99%

See 1 more Smart Citation

Analysis of Deep Stress Field Using Well Log and Wellbore Breakout Data: A Case Study in Cretaceous Oil Reservoir, Southwest Iran

Abdideh

Alisamir

2018

Geodesy and Cartography

View full text Add to dashboard Cite

To identify the wellbore instability of Bangestan oil reservoir in the southwestern Iran, the direction and magnitude of stresses were determined using two different methods in this study. Results of injection test and analysis of wellbore breakouts were used to verify the accuracy of the stress profiles. In this study the Bartoon method, which using the breakout angle and strength of rock, was used. In addition, the ability of artificial neural network to estimate the elastic parameters of rock and stress field was used. The output of the neural network represents a high accuracy in the estimation of the desired parameters. In addition, the Mohr-Coulomb failure criterion was used to verify stress profiles. Estimated stresses show relative compliance with the results of injection test and Barton method. The required minimum mud pressure for preventing shear failures was calculated by using the Mohr-Coulomb failure criterion and the estimated stress profiles. The results showed a good compliance with failures which have been identified in the caliper and image logs. However, a number of noncompliance is observed in some depth. This is due to the concentration of fractures, collisions between the drill string and the wellbore wall, as well as swab and surge pressures. The stress mode is normal and strike-slip in some depth based on the estimated stress profiles. According to direction of breakouts which is clearly visible in the caliper and image logs, the minimum and maximum horizontal stresses directions were NW-SE and NE-SW, respectively. Thses directions were consistent with the direction of regional stresses in the Zagros belt.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Estimated Stress Field By the Neural Networkmentioning

confidence: 99%

Analysis of Deep Stress Field Using Well Log and Wellbore Breakout Data: A Case Study in Cretaceous Oil Reservoir, Southwest Iran

Abdideh

Alisamir

2018

Geodesy and Cartography

View full text Add to dashboard Cite

show abstract

“…It was also compared to wavelet analysis and proved superior to it. Furthermore, the successful use of Savitsky-Golay was also demonstrated in [39][40][41][42] to smooth noisy data and solve the problem when the amount of noise in the dataset becomes a hindrance to the prediction process.…”

Section: Preprocessing Smoothing Filtersmentioning

confidence: 99%

Ground-level Ozone Prediction Using Machine Learning Techniques: A Case Study in Amman, Jordan

Aljanabi

Shkoukani

Hijjawi

2020

Int. J. Autom. Comput.

View full text Add to dashboard Cite

Air pollution is one of the most serious hazards to humans′ health nowadays, it is an invisible killer that takes many human lives every year. There are many pollutants existing in the atmosphere today, ozone being one of the most threatening pollutants. It can cause serious health damage such as wheezing, asthma, inflammation, and early mortality rates. Although air pollution could be forecasted using chemical and physical models, machine learning techniques showed promising results in this area, especially artificial neural networks. Despite its importance, there has not been any research on predicting ground-level ozone in Jordan. In this paper, we build a model for predicting ozone concentration for the next day in Amman, Jordan using a mixture of meteorological and seasonal variables of the previous day. We compare a multi-layer perceptron neural network (MLP), support vector regression (SVR), decision tree regression (DTR), and extreme gradient boosting (XGBoost) algorithms. We also explore the effect of applying various smoothing filters on the time-series data such as moving average, Holt-Winters smoothing and Savitzky-Golay filters. We find that MLP outperformed the other algorithms and that using Savitzky-Golay improved the results by 50% for coefficient of determination (R 2) and 80% for root mean square error (RMSE) and mean absolute error (MAE). Another point we focus on is the variables required to predict ozone concentration. In order to reduce the time required for prediction, we perform feature selection which greatly reduces the time by 91% as well as shrinking the number of features required for prediction to the previous day values of ozone, humidity, and temperature. The final model scored 98.653% for R 2 , 1.016 ppb for RMSE and 0.800 ppb for MAE.

show abstract

“…A hybrid model composed of MLP with either a particle swarm optimization (PSO) algorithm or a cuckoo optimization algorithm (COA) was proposed in [15]. Their approach based on plus-l-take-r for feature selection as it increases the number of parameters used for the training model with the aim of minimizing the error rate of the model.…”

Section: Related Workmentioning

confidence: 99%

Enhancing Multi-Layer Perceptron for Breast Cancer Prediction

Al-Shargabi¹,

Al-Shami²,

Alkhawaldeh³

2019

IJAST

View full text Add to dashboard Cite

Breast cancer (BC) is a standout disease of the most well-known cancers among women around the world. The analysis and prediction of BC leads to early manage the disease and protect the patients from further medical complications. In the light of its noticeable focal points in basic highlights identification from complex BC datasets, Machine Learning (ML) is generally perceived as the technique of decision in BC design order and gauge displaying. Because of the high performance of the Multi-layer Perceptron (MLP) algorithm as one of the ML techniques, we conducted experiments in order to enhance the accuracy rate of MLP by tuning its hyper-parameters along with studying the effect of feature selection methods and feature reduction of MLP. As feature selection results indicated that an increase in the number of input parameters tends to reduce the error associated with the estimator model. The tuned MLP proposed in this paper, based MLP best fit hyper-parameters along with feature selection is applied for breast cancer classification using Wisconsin Diagnostic Breast Cancer (WDBC) dataset. As the tuned MLP experimental result shows an accuracy reached 97.70% as it outperforms the basic MLP.

show abstract

Drilling rate prediction from petrophysical logs and mud logging data using an optimized multilayer perceptron neural network

Cited by 101 publications

References 31 publications

Analysis of Deep Stress Field Using Well Log and Wellbore Breakout Data: A Case Study in Cretaceous Oil Reservoir, Southwest Iran

Analysis of Deep Stress Field Using Well Log and Wellbore Breakout Data: A Case Study in Cretaceous Oil Reservoir, Southwest Iran

Ground-level Ozone Prediction Using Machine Learning Techniques: A Case Study in Amman, Jordan

Enhancing Multi-Layer Perceptron for Breast Cancer Prediction

Contact Info

Product

Resources

About