Performance of power system protection is determined by the protection system objectives such as dependability, security, selectivity and speed. However, these objectives often conflict one another where an enhancement of performance with respect to one objective causes a deterioration of another. Therefore, the protection design and setting must provide the optimal level of performance based on carefully considered compromise among the objectives. This thesis proposes that the best compromise can be achieved when the minimum overall risk introduced by a given protection scheme is considered as a decision guiding principle. The risk is calculated as a product of the likelihood of the system failure (resulting from protection not operating as intended) and the anticipated failures cost and it can be used to rank alternative solutions or schemes. Such assessment can assist in a protection selection process, evaluation of existing schemes, or finding optimal settings. In order to facilitate the utilisation of risk assessment for power system protection, this thesis proposes a dedicated framework which can help the assessor to perform the risk assessment and report the result to decision makers in an efficient and systematic manner. The framework consists of risk assessment objectives, terminologies, metrics, knowledge of protection and the protected system, scenarios, data, assumptions, and assessment steps. Three case studies are presented to illustrate different intended uses and modelling levels of risk assessment implementation. The first case study evaluates of the existing distance protection on a transmission line after installation of a quadrature booster transformer. The risk result informs the decision making process relating to whether the protection can be maintained or needs to be changed or modified.In the second case study, setting of ROCOF protection is selected based on risk introduced by the settings, hence the optimum setting is the setting with the least risk. The third case study demonstrates application of protection selection i.e. adaptive versus conventional overcurrent protection for a distribution network with DG. The case studies show that risk assessment has been successful in quantifying the overall protection performance according to the intended use and the framework has provided a useful guidance for the assessment.
|Date of Award||13 Jun 2015|
- University Of Strathclyde
|Supervisor||Adam Dysko (Supervisor) & Graeme Burt (Supervisor)|