Use this URL to cite or link to this record in EThOS:
Title: Monitoring concept to detect engine oil condition degradations to support a reliable drive operation
Author: Rigol, Sascha
ISNI:       0000 0004 2718 4399
Awarding Body: University of East London
Current Institution: University of East London
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
Access from Institution:
The theoretical part of this research work summarised all the known potential lubricant degradation effects during engine operation and in particular with regard to the use of the current generation of biofuels. A qualitative risk assessment was conducted which outlined biodiesel as potentially the most critical fuel. A ‘black box’ model was used to outline the challenge of oil condition monitoring based on summative sensor measurement methods. The theoretical considerations were supported by a statistical analysis which investigated how the presence of multiple contaminants in the oil affects the most common sensor data of permittivity, conductivity and viscosity. ‘Design of Experiments’ (DoE) models were developed for the permittivity, conductivity and viscosity data and expressing mathematically the relationship between the contaminants having a significant influence on each of the sensor data. The findings of the multivariate analysis identified that the effects on permittivity and viscosity provided reliable information about oil condition changes. The concept of an oil condition algorithm in this research was aimed at addressing accuracy and efficiency in predicting the outcome. The core aspect of the algorithm was the use of characteristic maps based on bi-linear regression to predict the fuel, soot and oxidation levels in the oil using permittivity and viscosity as the input data. Based on the predictions using three contaminant components a method to assess the overall condition status is derived. The derived condition status provided the input for an oil drain forecast method which monitors the status within the predefined maximum mileage. Another achievement from this research was the vehicle simulation for different driving profiles and the corresponding simulation of oil degradations. The selected profiles were for a ‘Taxi’, ‘Normal’ driving and ‘Long Distance’ driving. The resulting simulation of fuel, soot and oxidation levels in the oil showed a high correlation compared with the use of real oil analysis based on engineering judgement. The quantitative assessment of the simulated contaminant levels compared with the predicted levels obtained from each characteristic map showed excellent prediction performance. The derived overall condition rating and mileage forecast prediction also showed very good results. The results from this research have shown that this new oil condition algorithm concept using bi-linear characteristic maps has enabled the compromise between predictive accuracy and an efficient and transparent algorithm structure. Validation of the results confirms that the algorithm has the potential to minimise and prevent oil condition related engine failures regardless of the actual fuel used.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral