Simulation Based Testing Methodology for LOM Protection Performance Assessment

All protection relays and schemes associated with the transmission system have to undergo type testing procedure before they can be approved for installation. The purpose of these tests is to ensure adequate and predictable behaviour under a variety of system conditions. However, the protection devices installed within the distribution systems are often not required to go through such rigorous procedures due to the relatively low impact of potential faults and also to limit the overall development cost. The prevailing approach is to maximise safety which sometimes results in protection arrangements which are over sensitive to faults and thus cause unnecessary tripping such as the current prevailing practice of applying over sensitive settings to Loss-Of-Mains protection. Up until recently this approach was considered acceptable as the integrity of the whole system did not depend on the non-dispatchable generation connected at the distribution level which was simply regarded as a negative load. However, with increased penetration levels of Distributed Generation (DG) the system operators have started to realise that the performance and the behaviour of the aggregated large amounts of DG can have a major impact on the system frequency and local voltage. In particular, the DG behaviour during the system transients such as those resulting from major system faults needs to be more predictable. The undesirable disconnection of high proportions of DG can have grave consequences as has been experienced during events such as the UCTE blackout event in November 2006 and the frequency deviation event in UK in May 2008 where 546MW of demand had to be disconnected to counterbalance unexpected loss of DG. Therefore, this presentation proposes a methodology for developing sets of standard testing procedures for the accurate dynamic evaluation of protection relay performance. An example of the Loss-Of-Mains (LOM) protection is used to demonstrate the application of this methodology which led to the definition of the testing scenarios for this type of function. Taking into account the two fundamental aspects of protection performance: (a) sensitivity to system events within the protected zone and (b) stability under conditions which fall outside the protected area, the proposed methodology for the development of comprehensive testing scenarios includes the following components: (1) the definition of the primary system models; (2) the definition of the sensitivity and stability test scenarios; (3) scenario simulation and relay testing approach; and (4) interpretation of the results and relay performance evaluation. An exhaustive set of testing scenarios for the evaluation of the LOM relay performance is presented and the potential uses of the scenarios are discussed, including the evaluation of the performance of the existing relays, assessment of the performance of the new LOM protection methods and type testing of the protection relays. As it may be difficult (or sometimes even impossible) to achieve both sensitivity and stability under all testing conditions the questions related to the best sensitivity/stability compromise or acceptable performance need to be addressed taking into account the specifics of the protected network and type of protection under test.