| Abstract |
Corrosion due to exposure to geothermal fluids is one of the technological problems faced by the geothermal industry. The flow of geothermal fluid causes deterioration of equipment due to corrosion and erosion; thus monitoring of facilities that come into contact with the fluid is vital to minimizing safety and environmental risks during geothermal operation. Since the geothermal well delivers two-phase fluid from the reservoir to the surface, its casing is the first equipment on the field that encounters the fluid. Corrosion and erosion reduce the casings’ integrity by causing pitting and thinning. Caliper surveys may be conducted to monitor the thickness of well casings and detect pits on the casing inner wall that may eventually become holes. Effective management of the effects of corrosion requires a good understanding of how different corrosion factors contribute to the overall casing degradation. In EDC, to date, selection of wells for caliper surveying is largely dependent on the experience of the field reservoir engineer and depends on the age, pH, and TSS levels of the well. In this paper, statistical methods were used to explore the contribution of factors such as temperature, flow velocity, pH, well geometry, and corrosion species to the overall thinning rate of a geothermal casing. Downhole fluid chemistry of the wells were projected from the measured wellhead chemistry using chemical speciation software. Physical flow parameters, on the other hand, were simulated from measured wellhead conditions using wellbore modeling software. Linear regression models were built to correlate data estimated thinning rates used of two-phase production wells of the Leyte Geothermal Production Field (LGPF) with different factors affecting corrosion rate. Regression results were analyzed statistically using ANOVA, p-value test, and multiple regression coefficients. Adjusted R2 of the regression analysis and mean absolute percentage error (MAPE) were used in selecting good regression models. The objectives are to determine which of the factors affecting casing thinning rate are most significant and eventually, predict the thinning rate of the casing based on surface measurements and known well data. The data can then be used to aid in the prioritization of wells lined up for caliper surveys. |