Detection of small leaks in gas pipelines is an important and persistent problem in the oil and gas industry. However, the industry is beginning to investigate how tools of Machine Learning, Artificial Intelligence, Big Data, etc. can be used to improve current industry processes. This work aims to study the ability of intelligent models to detect small leaks in a natural gas pipeline using operational parameters such as pressure, temperature and flowrate through existing industry performance metrics (sensitivity, reliability, robustness and accuracy). Observer design technique was applied to detect leaks in a gas pipeline using a regresso-classification hierarchical model where an intelligent model acts as a regressor and a leak detection algorithm acts as a classifier. Five intelligent models (Gradient Boosting, Decision Trees, Random Forest, Support Vector Machine and Artificial Neural Network) were used in this present work. Results showed that the Random Forest and Decision Tree models are the most sensitive as they can detect a leak of 0.1% of nominal flow in about 2 hours. All the intelligent models had high reliability with zero false alarm rate in testing phase. However, due to this level of reliability, the models had low accuracy with the Artificial Neural Network and Support Vector Machine performing best and better regressors than the others. All the intelligent models are robust. The average time to leak detection for different leak sizes for all the intelligent models were compared to a real time transient model in literature. The intelligent models had a time savings of 25% to 48%. Results in this present work further suggest that intelligent models could be used alongside a real time transient model to improve leak detection. Also, that the tools of big data, data analytics, artificial intelligence can be harnessed to improving leak detection results.

Accurate pressure losses prediction for flow in tubing installations is of great importance in the petroleum industry. Historically, the Bottom-Hole Pressure (BHP) determination was obtained using down-hole pressure gauges, because of the economic disadvantage and redundancy, this procedure seems to be less effective, this led to the adoption of the BHP prediction process in estimation. The wide acceptance of data-driven analytics makes the estimation procedure a valid approach in the industry today. Recently, the Artificial Neural Network (ANN), a technique which has been widely accepted because the model proved to predict better than the conventional correlations. This present work aims to develop a prediction model for BHP based on input and output data obtained from the Volve production field in Norway. Machine learning algorithm based on ANN was used to predict and further improve the accuracy of the prediction while considering a large production dataset from different wells of the field. In developing the model, the initial dataset was processed to about 2,500 data points; the model was trained, tested and cross-validated based on the parameters from the data. Results affirmed that ANN has the ability to handle large dataset, results also revealed that ANN outperformed other models, with a Coefficient of Determination of 0.99997, Root Mean Squared Error of 0.07405 and Mean Absolute Error of 0.02657, which shows high predictability of the model. These results indicated that the ANN model gives a better prediction of BHP when compared to other mechanistic models. Finally, this work supports the claim, that Production engineers can accurately predict the pressure at the sand-face of a producing well without the use of expensive BHP gauge.

Ever increasing and growing awareness of oil-spillage to water environment has led to the search for cost-effective unconventional remediating techniques. This study was carried out using agro-wastes (Rice Husks, Banana Peels and Groundnut Husks) adsorbents. They were activated with H₃PO₄ for crude oil spill removal from aqueous solution; results were analysed using atomic absorption spectroscopy. The raw agro-wastes and their activated forms were characterized by SEM. The sorption study for maximum adsorption capacity were carried out at different adsorbent concentrations, adsorbent dosage, contact time, pH and rotational speed. The experimental results were analysed using Langmuir, Freundlich, Temkin, Dubinin-Radushkevich isotherm models. The maximum average adsorption capacity (q_e) for raw and activated carbon adsorbents were observed in rice husks with 0.2750 and 0.3698 mg/g respectively. The Langmuir isotherm was found to well represent the measured sorption data for the raw and activated banana peel while the remaining raw and activated adsorbents followed the Temkin isotherm. The batch adsorption data on the effect of contact time were fitted into the pseudo-first-order and pseudo-second order models. The banana peel data and its activated form were best described by the pseudo-second-order model indicating chemisorption process while the remaining adsorbents followed the pseudo-first-order model indicating physisorption process. The average removal efficiency of oil by the various adsorbents used increase in the order: (Raw: Banana Peels (50.4 %) < Groundnut Husks (56.8 %) < Rice Husks (74.4 %); Activated Carbon: Banana Peels (61.4 %) < Groundnut Husks (65.6 %) < Rice Husks (82.8 %)). Results of this study (high values of R² and least values AARE and RMSE) revealed and confirmed that activated carbon adsorbents have better adsorption capacity than the raw forms to clean-up oil spills in aqueous solution.

Invasion of mud filtrate into surrounding formation affects most data acquired by logging tools, which impedes accurate formation evaluation. The invasion process may result in formation damage and reduction in well producibility. The objectives of this work are to determine the effects of invasion on induction well logs, identify the invasion profile present, and to determine the factors influencing invasion. Therefore, a realistic model depicting the actual wellbore conditions is required in order to accurately analyze the invasion process, and its effects on well logs. In this work, a numerical model is developed. The physics of mud filtrate invasion in overbalanced drilling is numerically simulated using a 3D multiphase fluid flow simulator. Sensitivity analysis is done to quantify the influence of several petrophysical parameters on the spatial distribution of mud filtrate away from the wellbore, and the corresponding effects on induction logging tools. Result showed that the distribution of salt concentration due to the invasion of fresh water-base mud into a formation containing highly saline formation water affects resistivity measurements. Based on the plot of resistivity against radial distance of invasion from the wellbore, annulus invasion profile is found to be present. Also, sensitivity analysis showed that formation parameters such as porosity and permeability, in-situ fluid saturation and mudcake permeability are important factors to consider when determining the extent of filtrate invasion. A plot of formation resistivity versus formation porosity shows that a reduction of formation porosity from 0.25 to 0.1 increased the radial distance of invasion by about 70ft, and also shifted the annulus deeper into the formation. Result also showed that both medium and deep induction logs can be affected by invasion, and this depends on the total time of invasion, the formation rock and fluid properties and mudcake properties.

Filtrate losses and filter cake properties of drilling fluids are of concern in the oil industry because they alter near wellbore permeability and can reduce well productivity. Therefore, it is desirable to accurately characterize filtration process during oil well drilling. A mathematical model for analyzing mud filtration in deviated wells was developed in this study. The model determined solid pressure distribution within cake, cake thickness, cumulative volume of filtrate, and extent of invasion under different conditions. Results show assumptions of isotropy in previous studies greatly overestimate the magnitude of the damage. It was also confirmed that mud filtration tend to be higher in deviated than in vertical wells. The model was validated with experimental data.

Wax formation from paraffinic crude oil and deposition in reservoir pores; production and transportation lines are a foremost severe challenge to flow assurance. Chemical wax control has received huge acceptance and is mostly used in the petroleum industry. This work experimentally examined the Paraffin Inhibition tendencies of Jatropha oil (JTO), Jatropha oil-based Polyaminoamine (JPA), JTO dissolved in xylene (JTOX), and JPA dissolved in xylene (JPAX) in Niger Delta waxy crude-oils. Four performance indicators; Pour Point (PP), Weight of Wax Deposited (WWD), Wax Appearance Temperature (WAT), and Wax Crystal Morphology (WCM) were used to test the wax inhibition tendency of three crude oils. All experiments were performed on the blank and additives-treated crude-oils. The results obtained confirm that JTOX and JPAX act as PP depressants. Also, WWD decreases with increasing cold finger temperature correlating to the higher Wax Inhibition Efficiency (PIE) of the chemical additive. Below the WAT, the viscosity increased at a higher rate and the least wax amount was deposited, hence, the highest PIE (Crude-oil A:71%; Crude-oil B:72%; Crude-oil C:70%) was observed and close to the industrial control paraffin inhibitor (Crude-oil A:67%; Crude-oil B:65%; Crude-oil C:75%). JPAX has a significantly improved performance as a PP depressant than JTO alone.