Published on: **Mar 3, 2016**

- 1. EXPLORATORY DYNAMIC MODELING OF A NUCLEAR POWERED BRAYTON CYCLE FOR NUCLEAR ELECTRIC PROPULSION August 6, 2004 Robert M. Edwards Professor of Nuclear Engineering The Pennsylvania State University 228 Reber Building University Park, PA 16802 (814) 865-0037 rmenuc@engr.psu.edu NASA Glenn Research Center Instrumentation and Controls Division ABSTRACT The Jupiter Icy Moons Orbiter (JIMO) is envisioned to be a nuclear powered electric ion- propulsion system. A dynamic model of such a system will be needed to study the dynamic interactions of components and to verify that control structures and algorithms are adequate to ensure accomplishment of mission objectives. However, the design of the specific JIMO system has not yet begun that can provide data to initiate legitimate system integration studies. This research describes a possible approach to simplified modeling a closed Brayton cycle power conversion unit and its coupling to a nuclear reactor model. Data for this exploratory modeling was gleaned from a series of reports that were prepared by, or at the direction of, NASA in the early 1990’s. SUMMARY System integration is often performed after components are far along in their design or even construction. Required component changes that are identified late in the design and construction process can result in costly modifications and inefficient or ineffective system performance. The effort of this research is thus meant to provide an early indication of the data needed and techniques employed in system integration. Figure 1 is used to describe the characteristics of the system integration or control engineering effort for a central station nuclear power plant. Around the periphery of the diagram are the development of detailed models of system components, such as the reactor core with its neutronics and thermal hydraulics, the steam generator and electrical, mechanical components like pumps, valves and turbines, and finally pressure vessel and piping. The analysis performed in the component areas invariably involves the development of large computer design codes with special emphasis on steady state full power conditions. Each component code invariably grows to use all the available computer horsepower and uses an analysis platform that is incompatible with other component codes. It is then the job of system integration to work with these large component “round pegs” and fit them into the smaller “square holes” of system integration. System integration therefore invariably works with simplified models of all components in a platform that lends itself to the use of specialized system integration
- 2. software tools, again invariably incompatible with the component design codes. The job of system integration is a challenging and diplomatic one because it interacts with different organizations and philosophies. Component specialists may be reluctant to reveal data, thinking that they are the only ones that can deal with problems that come up in their component. The following publications were used to conduct the exploratory modeling: 1. Scoping Calculations of Power Sources for Nuclear Electric Propulsion, ORNL CR- 191133, 1994 2. Brayton Power Conversion System Parametric Design Modeling for NEP, NASA contractor report CR-191135, 1993 3. Modular Modeling System (MMS): A Code for the Dynamic Simulation of Fossil and Nuclear Power Plants: Overview and General Theory, EPRI CS/NP-2989, 1983 4. Preliminary Results of a Dynamic System Model for a Closed-Loop Brayton Cycle Coupled to a Nuclear Reactor, Steven Wright, Sandia National Lab. 5. “Dynamic Analysis and Control System Design for an Advanced Nuclear Gas Turbine Power Plant”, a dissertation in Mechanical Engineering, MIT 1990. The first two references present steady state design data for a reactor component and Brayton power conversion unit, respectively. The codes are capable of designing the reactor and Brayton unit to specific power levels and compatible steady state operating conditions; however, executable versions of these old codes were not readily available. The reports do provide full power steady state example output data for a 50 MW reactor and 500 kWe Brayton PCU, respectively. A dynamic model of a representative fuel pin of the 50 MW reactor was created to interface with a model of the 500 kWe Brayton unit. Modeling of the Brayton unit was conducted using a lumped parameter equation set presented in the 1983 Modular Modeling System Report. The paper from Wright and the MIT dissertation suggested a way to model the compressor and turbine of the Brayton Unit. Figure 2 presents the top level diagram of the nuclear powered Brayton Unit in the Mathwork’s SIMULINK dynamic modeling system. The Brayton unit simulation model was organized around the turbine/intermediate heat exchanger (ihx) and the compressor/recuperator components. Within the turbine/ihx block, a model of the 50 MW fuel pin is interfaced to the tube side of the ihx to provide the heat source for the Brayton Unit. A simplified shaft dynamics model is represented at the bottom of the diagram and simulation studies of the open loop response were conducted for step reductions in power removed from the shaft (load). As expected, the shaft rotational speed increases. But a counter intuitive reactor response shows an increase in power when the load is decreased. These results are consistent with those presented in the paper by Wright (reference 5) for another nuclear powered Brayton unit. Wright concludes that “the reactor control system will have to be used to reduce the reactor power when load decreases”. Much work lies ahead for system integration of the components of JIMO, and an early initiation of system integration capabilities will make the project progress more smoothly.
- 3. Figure 1: System Integration turbine exhaust high pressure recuperator exhaust Pi Ti md0 N mdRx rho Po To mdi J nr turbine_ihx rho mdot Pi Ti mdo N Po To mdi J compressor-recuperatorScope9 Scope8 Scope7 Scope6Scope5Scope4 Scope3 Scope2 Scope10 Scope1 Scope Pout Output Point1 Output Point 1 s N^2 Input Point2 Input Point1 sqrt(u(1)) Fcn1 2*u(1)*(60/(2*3.14159))^2/3.195 2*(Pin-Pout)*(60/2pi)^2/J Figure 2: SIMULINK model of a Nuclear Powered Brayton Cycle Steam Generator & Electrical: Pressure Vessel & Piping Core Design: neutronics thermal hydraulics Mechanical Pumps, valves turbines System Integration (Control Engineering) DETAILED MODELSDETAILED MODELS DETAILED MODELS DETAILED MODELS