BNL-NUREG63264

8'1 5- f /

A LEVEL CONTROL MODEL FOR BWR EMERGENCY PROCEDURE GUIDELINES *

*I 9 H. S. Cheng and U. S. Rohatgi

Brookhaven National Laboratory 12 Upton Rd. Bldg. 475B Upton,New York 11973

(516) 344-2611

ABSTRACT

The level control during an Anticipated Transient Without Scram (ATWS) event in a BWR as prescribed in the Emergency Procedure Guidelines (EPG) is a

water level at the top of active fuel (TAF) without uncovering the reactor core. Also the computer simulation of EPG level control will require many trial and error Calculatiorur with the Emergency Core Cooling System (ECCS). A level control system model has been developed and implemented in the RAh4ONA-4B code in order to simulate the EF'G level control without iterations. The model has been extensively tested aad the results demonstrate that the model can simulate the EPG level control strategy. The calculations also show that the level control system will speed up the boron circulation to shut down the reactor sooner than the manual control. Furthermore, the suppression pool temperature is predicted to remain within the Technical Specification limit during a MSIV closure ATWS with the proposed level control strategy.

I difficult task for the operator for he has to keep the

L INTRODUCTION

One of the major concerns for an ATWS in a BWR is the Pressure Suppression Pool (PSP) heatup due to the long period of cycling of SafetyRelief Valves (SRV). The Technical Specification for some BWR containments is that the pool temperature be below 79.4"C (175°F) during an ATWS event. In order to meet this requirement, the BWR owners' group developed Emergency Procedure Guidelines (EPG) [NEDO- 3 133 13, one of which is to keep the reactor power as low as practicable in the early phase of the ATWS by lowering the downcomer water level to the Top of Active

Fuel (TAF) until the Standby Liquid Control System (SLCS) is activated to inject boron into the reactor core. For the current BWR, the EPG level control is done via the High Pressure Coolant Injection (HPCI) system.

The EPG level control as proposed is to be carried out manually by an operator. This is an important task since it may lead to partial uncovery of the me, if it is not done very carehlly. The diBiculty of the manual level control has been demonstrated in AT WS simulations with an interactive BWR plant analyzer [WuH 19841. Simulation of the EPG level control can be done by means of a prescribed level setpoint history. This is an iterative process as the level controlling parameter is the HPCI flow rate and the relationdup between the level setpoint and HPCI flow is not known a priori. A general level control system model has been developed and incorporatedinRAMONA-4B [Rohatgi, et. al., 19951, a BWR system code being developed at Brookhaven National Laboratory. This level control model along with a timedependent level setpoint can keep the water level at TAF in the early phase of an ATWS event, then raise the level by increasing the flow, after a signiticant amount of boron SoIutiOn has been injected, to speed up the boron circulation in the Reactor Pressure Vessel (RPV). This will help shut down the reactor. The analysis demonstrated that the level controller satisfies the EPG level control objective, and that the suppression pool temperature was within the Tech Spec limit.

IL THE EPG LEVEL CONTROL MODEL

The level control per EPG objectives is modeled by means of a general-purpose leveVflow controller as illustrated in the block diagram in Figure 1. It consists of

* This work was performed under the auspices of the U.S. Nuclear Regulatory

~t~~~~~ OF VIIS M IS

DISCLAIMER

Portions of this document may be illegible in electronic image products. Images are produced from the best available original document.

This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or uw- fulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any spe- cific commercial product, process, or service by trade name, trademark, manufac- turer, or otherwise does not necessarily constitute or imply its endorsement, recom- mendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

two Proportional-Integral-Derivative (PID) control elements: one for a level error signal S, and the other for a combined level and flow error signal T,. The final combined error signal then determines the appropriate flow rate (in this case the HPCI flow) to maintain the level around a level setpoint L.

The level error signal is given by

where L,,, is the sensed water level and & is the requested water level setpoint (e.g., TAF).

The flow error signal is the mass inventory error due to the HPCI flow:

where WHCa is the sensed HPCI flow rate, and qb is the sensed reference HPCI flow rate (assumed to be a fixed value in this work, but can be a prescribed time- dependent function).

The level and flow signals are sensed by a first-order LAG due to instrument inertia:

dL, - L,-L, - - ' dt =L

(3)

where L is the level signal or calculated water level, W,, is the reference HPCI flow rate, Nc is the HPCI flow signal or the calculated HPCI flow given by Eq. (1 2). and the T'S are the time constants of the sensors. The requested (or desired) level setpoint L, is treated as a timedependent function, a tubular input of LR vs. time.

The level error signal E is fed into a PID, which generates an intermediate signal:

(6) dt

The intermediate level signal is limited by a limiter to prevent an excessively strong response:

The trimmed level signal is then combined with the flow error signal to form a total error signak

where K, is an input constant (a zero means that the level control is to be acmmphshed by the level setpoint alone).

Finally, the combined signal is supplied to another PID controller:

E,dt + T ~ , ~ - d E T ) * (9) dt

The intermediate combined signal T, is also limited as follows:

Tr = Tr,- if Tr > Tr,- 3

with fRwc being an input constant (e.g., 3.5).

The regulated HPCI flow rate to meet the requested level setpoint is given by the correlation:

where W,, is the reference HPCI flow rate and I$ is given by

The control system parameters are treated as input data. The typical values used for this work are given in Table 1.

HI. CALCULATIONAL MODEL

The RAMONAAB calculational model used for the present simulation is shown in Figure 2. It includes the RPV with all important internal components, steam lines, control systems, balance of plant (BOP), and the SLCS for boron injection.

The transient hydraulics is calculated with a drift flux model for two-phase flow and the. neutron kinetics is calculated with a threedimensional l%-group diffusion theory.

The reactor core was modeled with 101 neutronic channels and 25 hydraulic channels in a 1/8core ;syn.m&y. Twenty four axial cells were used in each of the neutronic a d hydraulic channels. Seven evenly distniuted spacers were located along the axial height of each hydraulic channel (except for the bypass channel) in order to account for the local form losses due to the spacer grids.

The nuclear parameters for the 3D neutron kinetics were based on the Browns Ferry Unit 1 at the end of cycle 5. The two-group cross sections and their feedback coefficients were generated from the lattice physics data provided by TVA in early 1980.

IV. TRANSIENT DESCRIPTION

The transient simulated was an ATWS event initiated by inadvertent closure of all MSIVs at 2 sec (This was done to make the initial power spike visible from the vertical axis). The normal scram system and alternate rod insertion (ART) system were assumed to have failed, thus ccmtituting an ATWS. Boron injection was initiated manually at 30 sec with 60-sec delay. The boron was injected by SLCS into the middle of lower plenum. (lb code allows other injection points such as the downcomer, core and riser). The water level in the RPV was controlled by the EPG level controller using a timedependent level setpoint specified by a tabular function as shown in Figure 3.

The initial condition was the hot full power and full flow condition. The key parameters calculated by RAMONA4B are:

Core Thermal Power

System Pressure

Feedwater Flow Rate

Core Flow Rate

Core Inlet Subcooling

3293 Mwt

6.962 MPA

1685 kgls

13,040 kg/s

10.65 "C

The only boundary codtion imposed was a manual closure of the MSIV within 4.5 secomls at t=2 second.

V. SIMULATION RESULTS

The ATWS eveut was initiated from the full power codition at t=2 sec by a manual closure of all MSIVs in 4.5 seconds, which resulted in a sharp rise of the system pressure from 6.% MPA to a peak of 8.95 MPA in 11 seconds as shown in Figure 4. The pressure oscillations are due to cycling of SRV. A break in the oscillation at around 250 s is due to the change in the bank of SRV cycling (from 2nd bank to 1st bank). The sharp rise in pressure caused a core-wide collapse of voids, resulting in a power spike of 305% of rated power as shown in Figure 5. The effect of EPG level control and the SLCS boron injection on the reactor power is evident in the power response. The reactor power was at decay-heat level after 800 seconds.

-3-

Figure 6 shows the transient response of the collapsed liquiil level m the downcomer. The water level dropped rapidly as a result of the feedwater flow cuastdowm to zero within 20 seconds. Shortly after, the L2 level setpoint (-1.5 m below downcomer entrance) initiated the HPCURCIC flow, then followed by the EPG level control by regulatiig the HPCI flow rate as shown in Figure 7. The level controller kept the water level around TAF until 900 seconds, then raised the level up to about 3 m above TAF by increasing the HPCI flow rate in accordance with a timedependent inpt function for the level setpoint (see Figure 3). This was to help accelerate the boron circulation in the RFV so that the m t o r can be shut down sooner.

Figure 8 presents the transid response of the core- average boron concentration. Boron started to enter the reactor core at 100 seconds and its concentration increesed steadily to 300 ppm. The boron accumulation in the core was then accelerated at 600 seconds when the water level was raised to a higher level and the core flow rate increased. The core average boron concentration reached about 450 ppm at 800 seconds when the reactor appeared to be in hot shutdown. The boron circulation in the RPV is illustrated in Figure 9 W ~ t h C 3 b o ~ ~ ’011s in the lower plenum and riser are shown. The boron concentration is in the liquid phase and it increases as vaporization in the core decreases the water fraction.

Figure 10 shows the total core inlet flow response. The flow increase by the EPG level control in the later phase of the transient is evident. Figure 11 shows the transient response of the suppression pool temperature. The initial pool temperature was 32.2”C (90 OF). It i n c r e a s e d steadily to 71°C (160°F) at 500 seconds, then started to level off t h e d e r . The pool temperature remaineri below the Tech Spec limit of 79.4”C (175°F).

VI. CONCLUSIONS

The present simulation of the ATWS event using the RAMONA-4B code demonstrated the effectiveness of the level control strategy as prescn’bed in the EPG. The following conclusions are drawn from the present shuly:

Without this capability, laborious iterations will be required to accomplish the objective.

2. A time-dependent function for the level setpoint is necessary for the present model to accomplish the EPG requirements. This can be easily done via a tabular function of level setpoint vs. time as used in this study.

ACKNOWLEDGMENTS

This work was performed under the auspices of the U.S. Nuclear Regulatory Commission.

REFERENCES

1. BWR Owners’ Group (1987), “Emergency Procedure Guidelines, Revision 4,” NEDO-3 133 1, Emergency procedures Committee, General Electric Company, and Operations Engineering, Inc.

2. Wulff, W., et. al. (1984), “The BWR Plant Analyzer,” NUREG/CR-3943, BNL-NUREG- 5 18 12, Brookhaven National Laboratory.

3. Rohatgi, U. S., et. al. (1995), “R4MONAdB: A Computer Code with Three-Dimensional Neutron Kinetics for BWR and SBWR System Transients,” To Be Published, Brookhaven National Laboratory.

Table 1 EPG Level Control Parameters

Value 0.10 s 0.25 s 0.25 s 25.0 0.0 4.2 60.0 1.OE+8 0.0 0.0 -100.0 100.0 -100.0 3.5

1. The level control system model developed herein works well to meet the EPG level control requirement as demonstrated in this simulation.

-4-

Figure 1. Block Diagram of the EPG Level Control System.

F Figure 2. RAMONA-QB Calculational Model

-a -I Figure 3. Time-Dependent Level Setpoint for the EPG Level Control.

-5-

? :w :8 . ?

M e .- 2 n i 2 0 a v)

2

2 3 v) v)

a

i 2 1 M

L L ."

I 0 > 3

2

iz 1 an

-6-

3 0 9 :: P 2 2 0 5: h D N

(Udd) NOI1UUlN33N03

0 s x

H

H

9

?

3 8 h

9 H - v,

? - :? -

I-

s s,

B

w

8

9

9

9

? .a

0

?

E 0 .d

ul ul

E a a s rA

2

iz 1 M

vi 3 k 4

-7-