Computer Assignment

Document Preview:

CISE 301 Term: 122 Computer Assignment 2 Instructions: -This assignment contains two parts -Both parts are required -Submit soft and hard copies by Monday 6th May 2013, by the end of the lecture I. Predicting Gas Solubility using Multiple Regression Models This project is about predicting gas solubility using multiple regression models. Gas solubility ( s R ) is a measure of the capability of oil to dissolve gas. It is in nature the result of complicated molecular interactions between all the individual components existing in oil and associated gas. But it is usually experimentally determined in the laboratory. Regardless of the specific lab experiments that are usually requested for a reservoir, one can find well-known correlations for this purpose in the literature. It is suggested here to use Multiple regression models to predict this property. Multiple regression analysis attempts to explain the relationships between the independent variables and the dependent variable. When there are p the independent variables 1 2 , ,..., p X X X , the multiple exponential regression equation is the general form as follows: 1 2 0 1 2 ... p Yi ?b X b X b X pb Where Y is a dependent variable; 1 2 , ,..., p X X X are the independent variables; 0 b is the constant that the regression line intercepts the Y axis, representing the amount the dependent Y will be when all the explanatory variables are 0, and 1 i b ?i?p are the regression coefficient, representing the amount the response variable Y changes when the explanatory variable changes. Using the dataset provided below, write a program code to indentify the following models suggested for gas solubility and saturated viscosity Perform the following functions for predicting gas solubility Model G1 1.2048 1.4 10 18.2 b x s g P R g ??? ?? ?? ? ??? ? ? ?? ???? ?? ?? 1 2x?a API?a (T?460) Determine Coefficients a1 and a2 Model G2 The following is another suggested model for gas solubility is suggested: 1.73184 0.8439 12.340 10 112.727...