Andrea Vercellotti
Diagnostic and prognostic maintenance decision making. A case study regarding the main equipment of a cogeneration plant.
Rel. Andrea Carpignano, Paolo Tarasco, Raffaella Gerboni. Politecnico di Torino, Corso di laurea magistrale in Ingegneria Energetica E Nucleare, 2019
|
PDF (Tesi_di_laurea)
- Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives. Download (34MB) | Preview |
Abstract: |
The present work is framed in the field of maintenance engineering. Its primary aim is the application of two statistical techniques and the evaluation of their effectiveness on monitoring performance and health condition of complex equipment of a cogeneration plant. More specifically, the case study deals with one internal combustion diesel engine. The first part of the thesis is dedicated to data processing and to the creation of a reference model for the internal combustion engine, on the basis of the available operational data. The data to investigate are concerning signals whose acquisition is achieved through the continuous service of implemented systems of measurement. The sample for model delineation refers to historical data about ICE operation. The monograph contains the definition of computational codes and the verification of model robustness and accuracy, with the support of maintenance history for the components under examination. The second section of the study is devoted to the application of Principal Component Analysis and Hotelling T2 statistic to this set of measurements. PCA aims to reduce the magnitude of a data set consisting of a large number of interrelated variables, retaining as much as possible of the variation present in the data set. This is accomplished by transforming the examined variables into a new set of variables, which are uncorrelated and ordered so that the first few hold most of the information contained in the whole database. Hotelling T2 statistic is a multivariate quality control procedure, exploited to highlight the presence of outliers that might be helpful to predict an impending failure event. These calculations, that have a purely statistical nature, are then compared with recorded deviations to check their efficacy; in other words, on the basis of system operating history, it is important to ascertain whether or not some ICE failures and/or degraded working conditions could have been forecasted and, therefore, avoided through suitable maintenance actions. |
---|---|
Relators: | Andrea Carpignano, Paolo Tarasco, Raffaella Gerboni |
Academic year: | 2018/19 |
Publication type: | Electronic |
Number of Pages: | 82 |
Subjects: | |
Corso di laurea: | Corso di laurea magistrale in Ingegneria Energetica E Nucleare |
Classe di laurea: | New organization > Master science > LM-30 - ENERGY AND NUCLEAR ENGINEERING |
Aziende collaboratrici: | UNSPECIFIED |
URI: | http://webthesis.biblio.polito.it/id/eprint/11349 |
Modify record (reserved for operators) |