Monte Carlo Simulation using Excel for Predicting Reliability of a Geothermal Plant .pdf

Click the link above to download the pdf file






The transforming of the physical and functional block model of the system into a functional RBD is needed to make a reliability prediction “Kececioglu (1991)”of a complex system. Units or modules are placed in series if failure of one results in system failure. If redundant units are available, they are placed in parallel and system failure occurs only when the number of module failures exceeds a given number, say ‘m’ of the ‘n’ units present in the system. Very often, each of the blocks present in the RBD may be comprised of units placed in series, parallel or a combination of both. Moreover, the system may be so complex that it may not be possible to render its RBD as a conventional network of series / parallel modules. Under such circumstances, a formula for the system reliability may be difficult, if not impossible to derive. Use of Bayes decomposition technique may become extremely cumbersome and time consuming requiring several hours, or even days, of derivations and/or calculations. The possibility of introducing human error during derivations further aggravates the issue. There may also be a desire to perform studies of reliability performance; such as calculation of reliability over time (instead of at a point in time) or determining failure distributions. Under such circumstances, Monte Carlo simulation “Law and Kelton (1991)”, “Sobol (1994)” is the only time effective way of estimating reliability.