No abstract

This work developed an Advanced Boiling Water Reactor (ABWR) feedwater pump and controller model, which was incorporated into Personal Computer Transient Analyzer (PCTran)-ABWR, a nuclear power plant simulation code. The feedwater pump model includes three turbine-driven feedwater pumps and one motor-driven feedwater pump. The feedwater controller includes a one-element / three-element water level controller and a specific feedwater speed controller for each feedwater pump. The performance tests, including inadvertent closure of all turbine control valves and one feedwater pump trip at 100% power, demonstrate the feasibility of dynamic response of incorporated model. Furthermore, a diversity and defense-in-depth analysis is performed to demonstrate the feasibility for motor-driven feedwater pump as an Emergency Core Cooling System (ECCS) automatic diverse back-up. In Lungmen Nuclear Power Plant (NPP), a diverse manual initiation means for the High Pressure Core Flooder (HPCF) loop C is designed as the back-up of digitalized Engineered Safety Features Actuation System (ESFAS). If the Motor-Driven Feedwater Pump (MDFWP) can be an automatic digital diverse back-up for ESFAS, Lungmen NPP would be more robust to defend against software common cause failure (CCF).

This work developed a stand-alone ABWR (Advanced Boiled Water Reactor) feedwater pump and controller model which was incorporated with a simplified reactor vessel and steam line model. The purpose of this work is to improve the existing model in PCTran-ABWR, a nuclear power plant (NPP) simulation code. INER has been using this computer code as an NPP simulation model for Software Safety Analysis (SSA) and software Fault Injection (FI) of digital instrumentation and control (I&C) research for years. The feedwater pump model includes three turbine-driven feed water pumps and one motor-driven feed water pump. The feedwater controller includes a one-element / three-element water level controller and a specific feedwater speed controller for each feedwater pump. The feed water turbines are driven by the steam from main steam line. As a result, the reactor dome pressure can affect the driving force of the three turbine-driven feed water pumps. It means if the dome pressure becomes low enough, the turbine-driven feed water pumps cannot function normally. The reactor dome pressure transient also affects the pressure difference of feedwater pump discharge pressure and the reactor dome pressure, which can actually affect the feedwater flow rate and reactor water level. The time-lag of feedwater control valve is also considered in this model. Hence, the slower response of turbine-driven feed water pump than that of motor-driven feed water pump can be observed. A number of test cases namely step change of dome pressure, load rejection, and four tests of feedwater pumps transfer were performed in this work to demonstrate the feasibility of dynamic response of this model. Therefore, this model will be implemented into the existing PCTran-ABWR plant simulation code to improve the response of feedwater pump and controller model. This stand-alone model can also be a feedwater control strategy tool to observe the possible responses of various feedwater control architectures.

The digitalized Instrumentation and Control (I&C) system of Nuclear power plants can provide more powerful overall operation capability, and user friendly man-machine interface. The operator can obtain more information through digital I&C system. However, while I&C system being digitalized, three issues are encountered: 1) software common-cause failure, 2) the interaction failure between operator and digital instrumentation and control system interface, and 3) the non-detectability of software failure. These failures might defeat defense echelons, and make the Diversity and Defense-in-Depth (D3) analysis be more difficult. This work developed an integrated methodology to evaluate nuclear power plant safety effect by interactions between operator and digital I&C system, and then propose improvement recommendations. This integrated methodology includes component-level software fault tree, system-level sequence-tree method and nuclear power plant computer simulation analysis. Software fault tree can clarify the software failure structure in digital I&C systems. Sequence-tree method can identify the interaction process and relationship among operator and I&C systems in each D3 echelon in a design basis event. Nuclear power plant computer simulation analysis method can further analyze the available backup facilities and allowable manual action duration for the operator when the digital I&C fail to function. Applying this methodology to evaluate the performance of digital nuclear power plant D3 design, could promote the nuclear power plant operation safety. The operator can then trust the nuclear power plant than before, when operating the highly automatic digital I&C facilities.

No abstract