PROGNOSTICS AND HEALTH MANAGEMENT VIA QUANTUM MACHINE LEARNING IN THE OIL & GAS INDUSTRY
Quantum Machine Learning; Prognostic and Health Management; Fault Diagnosis; Oil and Gas Industry; Research and Development.
In the Oil and Gas (O&G) industry the concept of Technology Readiness Levels (TRLs) has been used to deal with these new technologies. In one of these levels is the so-called Prognostic and Health Monitoring (PHM) that has as one of its objectives to diagnose the failure modes of the equipment. In this sense, a new technique that has already been applied in different scenarios is quantum machine learning, which aims to bring improvements to conventional methods in terms of performance and results. This naster thesis dissertation aims to apply quantum machine learning models for the PHM of equipments which can be used in the O&G and energy industry. Concerning the methodological aspects, the QML models will be based on feature extraction from the original signal data. Then, the data will be encoded using the angle encoding technique. Next, the coded information will be passed through parameterized quantum circuits (PQC), whose angles will be trained and optimized by a classical neural network. A contribution of this dissertation is the use of different PQCs with different layers. Data from the literature will be used to demonstrate the applicability of the model. The results obtained in this study indicate the effectiveness of this model. Thus, showing the possibility of their application in the O&G context.