Scholarly works
Permanent URI for this collectionhttps://repository.ui.edu.ng/handle/123456789/422
Browse
3 results
Search Results
Item Prediction of oilfield scale formation using artificial neural network (ANN)(SCIENCEDOMAIN International, 2016-07) Falode, O. A.; Udomboso, C.; Ebere, F.Scale formation and deposition is a recurring problem in many oil producing fields leading to operational problems, problems in reservoirs, pumps, valves and topside facilities. Scale is described economically as a menace to an oil-field because its build-up clogs the flow lines and causes loss of millions of dollars yearly. The ability to predict the onset and amount of scale formation has been a major challenge in the oil industry. Previous models for predicting scale formation have focused mainly on thermodynamics and limited solubility data, and can predict only the potential or tendency to form scale. However, no studies have considered the influence of kinetic and transport factors. In this paper, a comprehensive and robust model incorporating other factors that have been ignored in past studies is developed using the technique of artificial neural network (ANN). Field data on two types of scale namely Barium and Calcium sulphate were obtained, processed, trained and tested with Artificial Neural Network. The model obtained was validated with actual data. Results show that at constant pressure, the neural network structure with optimum performance for BaSO(4) was ANN {1,2,1} with the lowest Mean Square Value (MSE) of 0.0025 and the highest correlation determination R(2) of 0.9966 while at constant temperature, it was ANN{1,1,1} with MSE of 0.0017 and R(2) of 0.9956. The neural network structure with optimum performance for CaSO4 precipitation kinetics with temperature and pressure was ANN{2,5,1} with MSE of 8.7745e-005 and R(2) of 0.8206 while at constant flow rate it was ANN{1,4,1} with MSE of 2.3007e-006 and R(2) of 0.9953. This gave a very close agreement with actual data in terms of prediction and performance. The results of this study therefore will greatly help to reduce the amount of risk incurred (such as NORM, etc.) due to the deposition and formation of scale in an oilfieldthe cost of stimulating an oil flow line and also improve the productivity of an oil well, hence, increase revenue to the oil industry.Item On the level of precision of the wavelet neural network in rainfall analysis(2014) Udomboso, C. G.; Amahia, G. N.; Dontwi, I. K.This research combines the efficiency of the artificial neural network and wavelet transform in modelling rainfall. The data used were decomposed into continuous wavelet signals on a scale of 48. Each of the decomposed series was subjected to correlation test with the original data. Instead of using all the series, ten series were selected on the basis of high correlation with die original data. These series included CWT 1, CWT 2, CWT 4, CWT 3, CWT 6, CWT 8, CWT 5, CWT 10, CWT 12, and CWT 7 (according to rank). The analysis showed that except in extremely rare cases, all the series performed optimally compared to the original data. The result of the study has been able to show' that using the continuous w'avelet transform in the ANN technique, a better performance of the network is observed.Item Statistical and neural network approach for estimating monthly evapotranspiration at the international institute of tropical agriculture, Ibadan, Nigeria - a comparative study(2011) Chukwu, A. U.; Udomboso, C. G.; Onafeso, O.Evapotranspiration (ET) is one of the main components of the hydrological cycle as it accounts for more than two-thirds of the precipitation losses at the global scale. Reliable estimates of actual Evapotranspiration are crucial for effective watershed modelling and water resource management, yet direct measurements of the Evapotranspiration losses are difficult and expensive. The major objective of this study was to investigate the potential of the classical linear regression and neural network (NN) technique to estimate evapotranspiration, and to examine if a trained neural network with limited input variables can estimate ET efficiently. The study utilized daily climatic data of temperature, relative humidity, sunshine hours, wind speed, and rainfall for ten years collected from the International Institute of Tropical Agriculture. (IITA) Ibadan, Nigeria. Linear regression models in terms of the climatic parameters influencing the regions and, optimal neural network architectures considering these climatic parameters as inputs were developed. The linear regression models showed a satisfactory performance in the monthly estimation in the region selected for the present study. The NN models, however, consistently showed a slightly improved performance over linear regression models. The results also indicated that even with limited climatic variables an ANN can estimate ET accurately.