description
- Structura l Health Monitoring Systems (SHMSs) are increasingly used to assess the condition of the assets to which they are a pplied. However, many asset operators face chaJ/enges in the processing of vast amount of data that is avalla ble to them through SH MS, in order to get meaningful information and Intelligence a bout the underlying asset. The cha llenge is often d ue to a number of factors in cluding: - Com puting capa bility required to process real-time data - Lack of or limitations of physics-based degradation/ damage assessment models - lack of hlstorica! data to ena ble machine learning techniques to be used to make diagnostic or prognostic For a SHMS to provide val ue to operators, they need to provide inputs to decision support tools that enable operators to manage their assets. This PhD will focus on raw SHMS data and its processing such that a ppropriate inputs are provided to risk based decision support tools for managing the integrity of assets. Algorith ms will be developed to process data so that diagnostic or predictive information is extracted in order to provide inputs to decision making. The topic requires a multi-disciplinary a pproach with an appreciation of Non-destructive testing, Structural Health Monitoring systems, and risk based decision-making tools. Specialist knowledge and experience in the a pplication statistical techniq ues particula rly machine !earning / artiflcial !ntelllgence will be required for developing a ppropriate data processi ng methods.