Benchmark of RELAP5 Check Valve Models against Experimental Data

Transcription

1 Benchmark of RELAP5 Check Valve Models against Experimental Data Damian D. Stefanczyk Manager, Thermal Hydraulics Services Fauske & Associates, LLC (FAI) Contributors: Jens Conzen, Basar Ozar, Kevin Ramsden (Fauske) Jens Gardell (Linköping University) ASME Verification and Validation Symposium May 7 9, 2014 Tropicana Las Vegas Las Vegas, Nevada

2 Outline Purpose of the Investigation Overview of Experimental Loop Overview of RELAP5 Test Matrix RELAP5 Model Measured Data RELAP5 Check Valve Model Benchmark Results Conclusions and Future Efforts 2

3 Fauske & Associates LLC. Linköping University Fauske & Associates, LLC. Linköping University - Burr Ridge (Chicago), IL. - Linköping i (2 hrs from Stockholm) 3

4 Purpose of Investigation Demonstrate modeling capabilities of RELAP5 check valve models against experimental data Check valve slams 4

5 Check Valve Slam 1. Nominal flow 4. Column separation upstream side Flow stagnation downstream side 2. Flow reversal 5. Flow returns to condense void 3. Valve closed 6. Steam condensation induced waterhammer 5

6 Waterhammer Demonstration (Video) 6

7 Overview of Experimental Loop Elevated supply tank Pressurized reservoir Flume Mini flow line Check valve and pressure sensors upstream and downstream Alternateflowpath to induce check valve slam Flow meter Pump 7

8 Overview of RELAP5 RELAP5/MOD3.3 (Patch 03) Reactor Excursion and Leakage Analysis Program light water reactor transient analysis 1D code developed for the US NRC for simulation of a wide variety of hydraulic and thermal transients in both nuclear and nonnuclear systems involving mixtures of steam, water, noncondensable gases and solutes under single phase and two phase conditions. Includes two check valve models Diode (default) on/off switch based on flow direction Inertial Swing models dynamic behavior of the valve disc based on hydrodynamic forces imposed on the disc 8

9 Test Matrix Test Pump Speed Tank Pressure Check Valve Hz 32 psig Swing Hz 20 psig Swing Hz 11 psig Swing 9

10 RELAP5 Model 10

11 Measured Data Test 1 Mass Flow Rate Motor valve opened to induce reverse flow through check valve 11

12 Measured Data Test 1 - Pressure Upstream of Check Valve, Downstream of Check Valve Take away from this plot: Maximum pressure spike Pressure oscillations 12

13 Measured Data Test 2 Mass Flow Rate Motor valve opened to induce reverse flow through check valve 13

14 Measured Data Test 2 - Pressure Upstream of Check Valve, Downstream of Check Valve Take away from this plot: Maximum pressure spike Pressure oscillations 14

15 RELAP5 Check Valve Model Benchmark Results 1 st Attempt Manufacturer Data (Test 1 Conditions) Upstream of Check Valve, Downstream of Check Valve Valve fully reopens multiple times 15

16 RELAP5 Check Valve Model Benchmark Results 2 nd Attempt Adjusted Check Valve Data (Test 1 Conditions) Lowered cracking gpressure Upstream of Check Valve, Downstream of Check Valve More realistic disc behavior 16

17 RELAP5 Check Valve Model Benchmark Results 1 st Attempt Manufacturer Data (Test 2 Conditions) Upstream of Check Valve, Downstream of Check Valve Valve fully reopens multiple times 17

18 RELAP5 Check Valve Model Benchmark Results 2 nd Attempt Adjusted Check Valve Data (Test 2 Conditions) Lowered cracking pressure Adjusted loss coefficients through valve Upstream of Check Valve, Downstream of Check Valve More realistic disc behavior 18

19 Conclusions and Future Efforts Default input data provides conservative results for this study based on manufacturer supplied data Results are sensitive to check valve inputs Sensitivity studies recommended d to determine inputs resulting in most conservative results Comparison of calculations to relevant experimental data is highly recommended for commercial applications to build confidence in the results Future Work Expand experimental database Test different check valve designs Improve the current swing check valve model Develop models for other check valve designs 19

20 Thank you for your attention Work published in: Gardell, J., (2013), Benchmark of RELAP5 Check Valve Models against Experimental Data, LIU-IEI-TEK-G--13/00556 SE, Department of Management and Engineering Applied Thermodynamics and Fluid Mechanics, Bachelor s Thesis, Linköping University, Sweden. Questions? 20