Yuh-Ming Ferng

A Physical Model to Predict Wear Sites Engendered by Flow-Assisted Corrosion

Flow-assisted corrosion (FAC), an aspect of erosion/ corrosion (E/C), is a mechanism of piping degradation that causes a loss of material from the inside of the piping and then thinning of the wall. FAC damage is believed to be accelerated by a single- or two-phase mixture flowing within the piping. A physical model is presented that attempts to predict the distributions of sites of FAC wear within the fitting; this model includes the E/C and the three-dimensional single- or two-phase-flow models. Based on the calculated results, the impact of centrifugal and gravitational forces on liquid droplet behavior can be reasonably simulated. Appropriate indicators derived from the E/C model are used to predict the FAC locations. Compared with the plant measured results, the proposed model can precisely predict the distribution of wear sites. The FAC pattern dominated by the upstream fittings can also be determined. The satisfactory agreement reveals that the indicators provided by the current models can be used to reasonably predict the FAC locations and explain the complicated phenomenon of FAC wear occurring within the fittings.

Thermal-hydraulic simulation of localized flow characteristics in a steam generator

Abstract Multidimensional thermal-hydraulic characteristics in the secondary side of a steam generator (SG) are simulated by way of flow-boiling models. These models essentially belong to the so-called first-principle models that are derived from the conservation laws. The calculated results can provide the whole picture of thermal-hydraulic phenomena and the localized distributions of velocity, pressure, enthalpy, and void fraction, etc. in the secondary side of the SG. In addition, with the help of these localized flow characteristics, the forcing sources can be estimated for predicting flow-induced vibration (FIV) damage suspected in the tube bundles around the U-bend region. These calculated results can provide important information to help the FIV prediction for SG U-tubes and to find where the most possible FIV damage is located.

Numerically investigating the influence of local flow behaviors on flow-accelerated corrosion using two-fluid equations

Flow-accelerated corrosion (FAC) is a piping degradation mechanism resulting in the loss of material from the inside of the piping that subsequently thins the wall. The FAC that causes costly plant repairs and personal injuries is generally accelerated by the single-phase fluid or two-phase mixture, which seems to be a very serious problem found in most of the power plants these days. Based on the measured data of pipe thickness, FAC phenomenon strongly depends on the piping layout and local flow conditions. A three-dimensional two-phase model is proposed with the aim of simulating two-phase behaviors found in the pipe and investigating the impact of these local parameters on FAC damage. Through three-dimensional calculation, liquid droplet impingement was found to dominate the FAC damage occurring in high-steam-quality flow. A simplified parameter is proposed to express an indicator of this normal impingement force. The magnitude of this parameter can represent the severity location of the FAC damage. Compared with plant-measured data of the wear rate, the predicted locations of serious FAC are in good agreement qualitatively. In addition, the phenomenon that different piping layouts will induce different FAC locations can be accurately captured in the current mode.

Investigation of system responses of the Maanshan nuclear power plant to the loss of residual heat removal during midloop operation using a RELAP5/MOD3 simulation

Because several abnormal incidents involving the loss of the residual heat removal (RHR) system during refueling and maintenance outages have occurred in pressurized water reactors, the importance of investigating the physical phenomena with respect to these events has been recognized. RHR cooling is a major means to remove core decay heat after a reactor power plant shutdown. If the RHR system is lost and an alternate means for heat removal cannot be established in time, the core will boil off, and the primary system will be pressurized, which potentially results in uncovering of the fuel rods and failure of the temporary boundaries. The objective of this paper is to simulate the Maanshan nuclear power plant (MNPP) responses to the loss of the RHR system during midloop operation under variable outage conditions. Without gravity feed, the current investigation concentrates on the effects of different liquid levels, the existence of vents, and the number of active steam generators. Based on the simulated results, the total heat removal capability of one active steam generator and the pressurizer venting process is sufficient to remove core decay heat of 11.1 MW, which corresponds to the power level 3 days after plant shutdown, in the event that RHR cooling fails during midloop operation. The primary system will be stabilized, and the pressure throughout the transient will not exceed the design pressure of the nozzle dams or the temporary seals. The heat removal capability of the pressurizer vent plays a crucial role in system pressurization during loss of RHR and in the severity of this event, as shown by the calculated results of the open and closed MNPP, respectively. If only one or two active steam generators serve as an alternate cooling means, the increased pressure will exceed the design criteria of the temporary low-pressure boundaries. Then, for the closed conditions of MNPP, the loss-of-RHR event during midloop operation has the potential to induce another loss-of-coolant accident and to cause more serious consequences.

A comparison of the RELAP5/MOD3 code with the IIST natural circulation experiments

A series of experiments dealing with variable secondary-side cooling conditions have been conducted at the IIST facility, including the natural circulation experiments under the secondary-side conditions of normal feedwater, loss of feedwater, and full of air. Different cooling conditions at the secondary side directly affect the primary-to-secondary heat transfer and then may influence the heat removal capability of natural circulation in the primary system. The corresponding analytical work is performed using the RELAP5/MOD3 code. Good agreement is reached both qualitatively and quantitatively between the experimental data and calculated results, demonstrating the satisfactory assessment of RELAP5/MOD3 code compared with the IIST natural circulation experiments. The cooling conditions at the secondary side have no significant effect on the heat removal capability of natural circulation as long as sufficient coolant exists on the steam generator secondary side, based on current IIST data and analytical results. Continuous increase of the core temperature and system pressure is also demonstrated experimentally and analytically in the test with the secondary side dry for the sake of deficient heat transfer capability at the steam generator secondary system.

Experimental Investigation of Early Initiation of Primary Cooldown by Secondary-Side Depressurization in a PWR Inadequate Core-Cooling Accident

Abstract The thermal-hydraulic phenomena of inadequate core cooling caused by a cold-leg small-break loss-of-coolant accident (SBLOCA) were investigated experimentally at the Institute of Nuclear Energy Research Integral System Test facility. The experiments were performed under the conditions of different break sizes (0.5 and 2%) in the cold leg followed by failure of the high-pressure injection system. The primary system cooldown is implemented by the secondary-side depressurization. The effectiveness of early initiation of the recovery action on reactor safety and related thermal-hydraulic phenomena are examined. The initiation criterion for recovery action considered here is determined by core water levels instead of core exit temperature based on the current emergency operating procedures. The impact of emergency core-cooling flow bypass phenomenon may significantly deteriorate the effectiveness of the recovery operation for a cold-leg SBLOCA. The results showed that the early initiation of secondary-side depressurization can effectively minimize the risk of core damage by preventing fuel rods from heating up throughout the transient. In addition, the core suffers a rather moderate thermal stress during the cooldown process.

Using an IIST SBLOCA experiment to assess RELAP5/MOD3.2

The RELAP5/MOD3.2 code is used at the Institute of Nuclear Energy Research Integral System Test Facility to analyze a 2% cold-leg-break experiment that includes failure of the high-pressure injection system. The assessment code predictions include primary pressure, inventory distribution in the reactor coolant system (RCS), loop flow rate, break flow rate, and core thermal hydraulics. A comparison between the calculated results and the experimental data shows (a) a good match with the predictions of the RCS pressure and hot- and cold-leg fluid temperatures, (b) underprediction of the core and down-comer levels, (c) overprediction of the loop flow rates in single- and two-phase natural circulation, and (d) inadequate prediction of asymmetric coolant holdup in the three stean generators. Also presented are sensitivity studies of choked flow associated with the defaulted values of discharge coefficients in the simulation of the break flow, and of the core bypass area to evaluate the effect of core level depression.

RELAP5/MOD3 simulation of the station blackout experiment conducted at the IIST facility

Thermal-hydraulic responses in the station blackout experiment conducted at the IIST facility are simulated through the use of the advanced system code RELAP5/MOD3. Typical behaviors occurring in the IIST station blackout transient are characterized by secondary boiloff, primary saturation and pressurization, and subsequent core uncovery and heatup. As the coolant inventory within the steam generator secondary system boils dry, the primary system pressure increases as a result of degradation of the heat removal ability of the steam generator secondary side. This pressurization phenomenon causes the pressurizer safety valve to open and the primary coolant to deplete through the valve, causing the core to eventually become uncovered. The same response can be exactly simulated by the current model. The current calculated results show fairly good agreement with the experimental data, but the timing of the events occurring in the station blackout transient is calculated earlier than the measured value. The overall comparison of key parameters between the calculated results and IIST test data, however, reveals that the current RELAP5/MOD3 model can provide rear sonable station blackout modeling for simulating longterm system behavior.

Investigation of scale-up capability of RELAP5/MOD3 and scaling distortion in the IIST facility via natural circulation experiments

During the past years, a number of reduced-scale test facilities have been constructed to investigate the physical phenomena of transients or accidents occurring in nuclear power plants. Since the behavior of a nuclear power plant is complicated, it is quite impossible for a small-scaled facility to simulate all the physical phenomena during the transient process. But, by way of proper scaling, most of the important aspects of transient behavior can be simulated. Calculations using RELAP5/MOD3 investigate whether most of the key thermal-hydraulic phenomena observed in the Institute of Nuclear Energy Research Integral System Test (IIST) facility can be expected in a prototype plant. When compared with experimental data, the calculated results of two different scale models show reasonable agreement with the natural circulation transients. The scale-up capability of RELAP5/MOD3 is demonstrated by simulating the single-phase and two-phase natural circulation transients. Also, the scaling distortions in the heat transfer areas of the IIST facility do not strongly distort the thermal-hydraulic behavior of experimental data.

Investigation of two-phase phenomena occurring within moisture separator reheater high-level reactor trips at the Maanshan nuclear power plant

During the operating history of the Maanshan nuclear power plant (MNPP), five reactor trips have occurred as a result of the moisture separator reheater (MSR) high-level signal. These MSR high-level reactor trips have been a very serious concern, especially during the startup period of MNPP. Consequently, studying the physical phenomena of this particular event is worthwhile, and analytical work is performed using the RELAPS/MOD3 code to investigate the thermal-hydraulic phenomena of two-phase behaviors occurring within the MSR high-level reactor trips. The analytical model is first assessed against the experimental data obtained from several test loops. The same model can then be applied with confidence to the study of this topic. According to the present calculated results, the phenomena of liquid droplet accumulation and residual liquid blowing in the horizontal section of cross-under-lines can be modeled. In addition, the present model can also predict the different increasing rates of inlet steam flow rate affecting the liquid accumulation within the cross-under-lines. The calculated conclusion is confirmed by the revised startup procedure of MNPP.