Rawal, Aditi ORCID: 0000-0001-6546-8577 (2020) Diagnostic and prognostic methodology for monitoring silk fade in a museum environment. PhD thesis, University of Greenwich.

Preview

PDF Aditi Rawal 2020 Final.pdf - Published Version Available under License Creative Commons Attribution Non-commercial. Download (4MB) | Preview

Abstract

Amongst the natural fibres found in museums, silk is reported to be the most vulnerable to damage, especially because of photodegradation of silk dyes. Deterioration of silk, on open display in museums, cannot be prevented, and this degradation often results in complete or partial loss of the silk fabric, requiring expensive interventive restorations. Research conducted, so far, in this field, has utilised methods that expose sensitive and sometimes rare fabric to a destructive process of analysis.

The aim of this research is to develop a non-invasive and non-destructive methodology that can predict the remaining useful life (RUL) of silk within the controlled museum environment. The benefit arising from this research is to improve the decision-making process, relating specifically to the conservation of silk, through the application of this novel methodology and prognostic model.

The prognostic and health management methodology developed in this work is applied and demonstrated on three chairs, originally made between 1768 to 1770, and re-upholstered with historically authentic silk in 1956. These chairs form part of the Shrewsbury Set displayed in the Great Gallery, at The Wallace Collection in London.

This research develops upon the widely researched Prognostic and Health Management (PHM) approach of engineering systems, used in aerospace and for structural monitoring of built environments, including heritage structures. To achieve this, a new mathematical fade model is postulated, to predict the fading, (colour degradation) of the silk that is a function of the exposure to its environmental conditions, such as, temperature, light levels, and relative humidity. This is the first time such a model has been presented that seeks to combine these factors. A relatively inexpensive portable instrument was used to take in situ measurements of silk samples, from the chairs displayed in the museum environment. The data driven diagnostic technique adopted the international colour standard, such as CIE XYZ, with Euclidian distance analysis to identify the colour condition of the silk. Statistical data-driven prognostic techniques have been utilised to model cumulative degradation of the colorimetric condition data of the silk and to predict its remaining useful life. The diagnostic and prognostic tools presented in this research are validated though numerical optimisation and through demonstration examples.

Based on this research, the brand-new colour condition of the silk was determined to be 376.12 and it was found that the brand-new silk would have a life of 55 years, based on the typically maintained museum environmental conditions. This new prognostic methodology for the silk colour condition in situ can be used to predict the cumulative degradation to silk fabric over different time horizons and under different museum environmental conditions. Sensitivity analysis scenarios are proposed that enable the prediction of the Remaining Useful Life (RUL) of the silk samples that are currently on display at the Wallace Collection.

This novel non-invasive and non-destructive, data driven PHM methodology for the conservation of silk encourages timely and optimal actions based on the diagnostic and prognostic outputs for cumulative fade of silk in situ. This methodology can be further extended to other silk samples, as well as to other textiles.

Item Type:	Thesis (PhD)
Uncontrolled Keywords:	Silk colour fade; Colorimetry; PHM; RUL; Non-invasive and Non-destructive; Predictive modelling;
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Faculty / School / Research Centre / Research Group:	Faculty of Engineering & Science Faculty of Engineering & Science > School of Computing & Mathematical Sciences (CMS)
Last Modified:	04 Jun 2024 13:28
URI:	http://gala.gre.ac.uk/id/eprint/35059

Actions (login required)

View Item

Downloads