Reservoir Fluid Sample Decontamination: Application Example

Altowilib, Ali R.; Sequeira, Daryl S.; AlOtaibi, Fawaz; Sahl, Ahmed M.

doi:10.2118/196394-ms

Cited by 2 publications

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Outlier detection and selection of representative fluid samples using machine learning: a case study of Iranian oil fields

Hosseini,

Zaheri,

Roosta

2024

J Petrol Explor Prod Technol

View full text Add to dashboard Cite

During the development of a field, many fluid samples are taken from wells. Selecting a robust fluid sample as the reservoir representative helps to have a better field characterization, reliable reservoir simulation, valid production forecast, efficient well placement and finally achieving optimized ultimate recovery. First, this paper aims to detect and separate the samples that have been collected under poor conditions or analyzed in a non-standard way. Moreover, it introduces a novel ranking method to score the samples based on the amount of coordination with other fluid samples in the region. The dataset includes 136 fluid samples from five reservoirs in Iranian fields, each of them consisting of 21 key parameters. Five acknowledged machine learning based anomaly detection techniques are implemented to compare fluid samples and detect those whose results deviate from others, indicating non-standard samples. To ensure the proper detection of outlier data, the results are compared with the traditional validation method of gas-oil ratio estimation. All five outlier detection methods demonstrate acceptable performance with average accuracy of 79% compared to traditional validation. Furthermore, the fluid samples with the highest scores in scoring-based algorithms are introduced as the best reservoir’s representative fluid. Finally, fuzzy logic is used to obtain a final score for each sample, taking the results of the six methods as input and ranking the samples based on their output score. The study confirms the robustness of the novel approach for fluid validation using outlier detection techniques and the value of machine learning and fuzzy logic for sample ranking, excelling in considering all critical fluid parameters simultaneously over traditional methods.

show abstract

Outlier detection and selection of representative fluid samples using machine learning: a case study of Iranian oil fields

Hosseini,

Zaheri,

Roosta

2024

J Petrol Explor Prod Technol

View full text Add to dashboard Cite

show abstract

Machine Learning to Detect Oil-Based Mud Contamination in Oil Samples

Almalki,

Alotaibi

2024

Day 3 Wed, February 14, 2024

View full text Add to dashboard Cite

Current methods to decontaminate bottom hole oil samples from Oil-Based Mud (OBM) poses many challenges as it requires analytical solutions. The existing analytical practices are highly subjective, as contamination results can differ depending on the approach determined by the user. The proposed method involves training a newly developed machine learning model to seek contamination patterns in an oil or gas condensate sample and highlight the contaminated samples. A python-based program that utilizes machine learning techniques to predict the concentration of OBM contamination in hydrocarbon fluids was constructed. The proposed procedure involved collecting 397 records of analyzed liquid C1 – C44 weight percent and whether they were reported as contaminated with OBM or not. The records were normalized and standardized on a column-basis, to preserve the confidentiality of the data. The model then, was trained on different algorithms, then was subjected to real cases to determine the model’s prediction accuracy. The train/test ratio was chosen to be 80/20 across different algorithms. The model was subjected to different algorithms. The algorithms were logistic regression, manual logistic regression, linear support vector machine (SVM), kernel SVM, Decision Tree, and Random Forest. The Manual Logistic Regression (MLR) algorithm resulted in an accuracy of 97%, and an F-1 Score of 94%. This shows that the MLR algorithm is the most suitable to be used in this prediction scheme. Moreover, this algorithm can be used as a quick method to get a sense of whether the fluid is contaminated or not. This illustrates the model’s significance in OBM contamination detection and making it the tool to be used in deciding the validity of a hydrocarbon sample. The current methods that are used to analytically decontaminate a fluid are based on analytical approaches, which depend on the person interpreting the results. Machine learning will resolve the issue of subjectivity in determining the contamination existence, increase the efficiency of the process, and is one step forward towards digital transformation.

show abstract

Reservoir Fluid Sample Decontamination: Application Example

Cited by 2 publications

References 12 publications

Outlier detection and selection of representative fluid samples using machine learning: a case study of Iranian oil fields

Outlier detection and selection of representative fluid samples using machine learning: a case study of Iranian oil fields

Machine Learning to Detect Oil-Based Mud Contamination in Oil Samples

Contact Info

Product

Resources

About