Dynamic stability of the VVER-1200 power unit

The paper presents the results of critical experiments to study the dynamic stability of a power unit with the VVER-1200 reactor conducted as part of the pre-commissioning activities at the pilot operation stage of Novovoronezh NPP II's unit 1. The following dynamic tests were conducted: – trip of one main feedwater pump (MFP) with no standby MFP starting to operate at the power level of 100% Nnom, involving a detailed analysis of the variation in the process parameters of such mode and the process dynamics, and an assessment of the test results on a full-scale simulator; – trip of one out of four reactor coolant pump sets (RCPS) in operation at the power level of 100% Nnom and the reactor plant safety assessment in the context of the reactor core thermal reliability; – turbine generator (TG) load shedding to the auxiliary level with assessments for the behavior of the key reactor plant characteristics. The paper presents records for transients and safety-related process parameters, and describes the operation of the unit components and essential controls in the dynamic test process. A conclusion is made based on an analysis of the test results that the VVER-1200 unit has a high dynamic stability. The results of the dynamic stability studies for unit 1 of Novovoronezh II make it possible to provide a number of recommendations for further designs, including specifically the following: – accelerated warning protection (AWP) should be used instead of power reduction and limiting for modes with tripped main feedwater pumps; – generator-grid timing devices should be used for modes with the unit operating for auxiliary power supply; – Russian-developed software and hardware tools should be fully switched to in implementing both normal operation and safety control systems, since the adjustment of protection and interlocking algorithms used in the AREVA software and hardware package introduced at Novovoronezh II requires the developer's authorization which involves substantial time and financial expenditures.

Abstract licence: CC BY

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.