Abstract
The thesis aims to apply a model-based assessment for the different maintenance policies possible for a system and thereby aid the engineers in choosing the optimal maintenance policy. For this purpose, a system of separators and compressors used in Oil and Gas production is considered. AltaRica 3.0 is used as the simulation tool to carry out the performance assessment of the system. The work consists of two parts; the first one is the modeling and simulation of the system in AltaRica. The second part is a python script used to execute the AltaRica model, calculate the necessary output, and implement machine learning algorithms to find the solution.
The first stage consists of building, debugging, and compiling the model in AltaRica. Afterward, the necessary performance indicators are identified and saved as a description file. Then a mission description file is generated, which contains the information regarding the mission duration. Then the simulation is executed, and the results are stored in a .csv file.
The second stage uses a python script to carry out this process repeatedly for a series of candidate maintenance policies. The results of each run are used to identify the optimal maintenance solution. Machine learning algorithms can be used in this stage to improve the algorithm's efficiency, which enables us to skip unnecessary candidates, thereby making the process faster.
Using the combination of AltaRica tools and machine learning algorithm, the process of optimization becomes efficient. This work facilitates the connection of these two realms and achieves the required solution.