Recent advancements in computer engineering have provided effective solutions for processing and analyzing complex systems and big data. Consequently, the adjustment and standardization of this data play a crucial role in addressing issues related to the monitoring of industrial systems. In this study, we propose a reliability approach for gas turbines to identify and characterize their degradation using operational data. We introduce a method for adjusting turbine reliability data, which resolves the challenges associated with the nature of these operating data. This enables us to determine a mathematical function that models the relationships between turbine reliability parameters and evaluate the impact of reliability practices in terms of availability. Additionally, we determine the survival function and employ it as a lifespan distribution model by estimating the parameters of the Johnson SB function. Furthermore, we calculate the failure rates and mean time between good operations for this rotating machine under different operating conditions. The obtained results allow us to estimate the parameters of the distribution that best fit the turbine reliability data, which are validated through statistical and graphical tests. We assess the goodness-of-fit using mean square error and reliability tests such as Kolmogorov-Smirnov.

Included in

Risk Analysis Commons