Use this URL to cite or link to this record in EThOS: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.789590
Title: Quantification of plausibility cross-checks in safety related control system architecture design for automotive applications
Author: Williams, Andrew Robert
Awarding Body: Birmingham City University
Current Institution: Birmingham City University
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Access from Institution:
Abstract:
When designing safety critical systems for automotive applications it is imperative that the chosen architecture can fulfil the designated safety goals. One significant aspect of this is proving architectural metrics are satisfied. The method developed in this thesis demonstrates, very early in the design process, that a system architecture can be systematically described and analysed to show that the final architectural metric targets for functional safety will be met. The system architecture model proposed can be used to explain a very complex system to other engineers / managers in an easily understood concept diagram, specifically tailored to examine the achievable diagnostic coverage of potential failures in the electrical /electronic system. Once the first architectural model is established, the method analyses architectural metrics in a quantified way, identifies potential weak areas and guides the designer towards additional Plausibility Cross-checks, or, in some cases, completely different architectures to improve the architectural metrics. The metrics can be calculated very quickly in comparison to the level of detail required for the final design. This permits quantified analysis of each candidate architecture allowing an informed decision to be made on which architecture to take through to the final design process. Often, multiple solutions will meet functional requirements, however, only a subset will meet functional safety requirements. The necessity to build safety into products has always been an important aspect of overall system design. This method allows decisions based on justifiable data, early in a project timeline to influence design decisions and ensure that concepts are correct. As demonstrated through examples this is achieved with a high level of confidence.
Supervisor: Srai, Manjit Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.789590  DOI: Not available
Keywords: H300 Mechanical Engineering
Share: