Power converters are becoming an integral part of future smart grids. These provide a necessary interface between the source of power available and the load/network by conditioning the energy flow to meet pre-defined system specs and/or grid code requirements. Hence, their seamless operation is crucial to the healthy operation of future smart grids. For this reason, condition monitoring of power converters is gaining more attention to provide means of predicting component failures and scheduling preventive maintenance schemes. Typically, semiconductor device stresses, capacitor ageing, and inductor/transformer insulation need to be monitored and assessed.
This project will study the condition assessment of a typical power electronic converter interface used in smart grids with the scope of providing an intelligent framework for early warning of potential failures or end-of-life. The project will perform advanced data analytics on measurements obtained from a lab prototype to assess the state-of-health of the power electronics under different operation scenarios for prolonged operational durations. Scenarios will also include timed overload and simulated fault conditions. The aim is to monitor the performance of the converter semiconductor devices and passive components using various sensors and data acquisition means and processing the big sets of data generated through a dedicated Big Data server for machine learning.
The data will be analysed and used to provide system intelligence for: * Early warnings of component failures * Establishing scheduled preventive maintenance schemes * Predicting behaviour of power converter operation for periods of time beyond the typical experimental test time.
Funding Notes
The successful candidate should be able to demonstrate a solid educational background in at least one of the following areas: power electronics, machine learning and/or data science. It is also desirable to have strong simulation skills in Matlab/Simulink and relevant programming skills using relevant software such as Python. Proven record of academic research would be considered favourably.
You haven’t included a message. Providing a specific message means universities will take your enquiry more seriously and helps them provide the information you need. Why not add a message here
