MODIFICATION OF HAZEN-WILLIAM EQUATION TO DETERMINE PRESSURE DROP AT INJECTION WELL BASED ON EMPIRICAL DATA

• Main Author: CIO MARIO (NIM : 22207311), CIO
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/21526
• Institution: Institut Teknologi Bandung

It is common in XYZ Field that Hazen-William Equation is used to estimate pressure drop in the tubing, especially for estimating bottom hole injection pressure and calculation of ESP pump design for very high water cut wells. Although Hazen-William Equation is frequently used but the validation of the pressure drop calculation result to actual field data has never been done. This is <br /> <br /> <br /> causing no information about the accuracy of pressure drop calculation result. This thesis is developed by doing research comparing pressure drop calculation <br /> <br /> <br /> results in the injection wells by using Hazen-William Equation with actual field data which comes from Spinner Survey. It is found that calculation results not always giving good estimation of pressure drop comparing with actual field data, even in the certain rate, the pressure drop estimation is deviating very far from the <br /> <br /> <br /> actual data. Because of the difference between calculation result and actual field data, research <br /> <br /> <br /> is continued by trying to modify Hazen-William Equation based on actual empirical data from Spinner Survey. To modify the equation, dimensional analysis is used by applying Buckingham Pi Theorem. From the dimensional <br /> <br /> <br /> analysis, a new formula which is simpler but has better accuracy comparing with the original Hazen-William Equation was found. <br /> <br /> <br /> <br /> (The new equation) <br /> <br /> <br /> <br /> Calculation result using this new equation is compared with other methods, such as: Hagedorn&Brown, Beggs&Brill, the original Hazen-William (with C = 100), <br /> <br /> <br /> and Jain&Darcy-Weisbach. In general, the calculation result using new equation gives better estimation and close to actual spinner data in the field if we compare <br /> <br /> <br /> with above methods.