ML20050A936

From kanterella
Jump to navigation Jump to search
Nonproprietary Cpc/Ceac Software Mods for Sys 80.
ML20050A936
Person / Time
Site: 05000470
Issue date: 03/31/1982
From:
ABB COMBUSTION ENGINEERING NUCLEAR FUEL (FORMERLY
To:
Shared Package
ML19268D122 List:
References
NUDOCS 8204020417
Download: ML20050A936 (32)
v • d • e


Text

--

SYSTEM 80 l

DOCKETS ST'N-50-470F ENCLOSURE 1-NP to LD-82-039 i CPC/CEAC SOFTWARE MODIFICATIONS FOR l

SYSTEM 80 s

' MARCH, 1982 ,

COMBUSTION ENGINEERING, INC.

NUCLEAR POWER SYSTEf1S '

POWER SYSTEMS GROUP WINDSOR, CONNECTICUT F

=

o 6

l l

8204020417 820330 ,

l PDR ADOCK 05000470 A PDR l

____j

[

LEGAL NOTICE This response was prepared as an account of work sponsored by Combustion Engineering, Inc. Neither Combustion '

Engineering nor any person acting on its behalf:

a. Makes any warranty or representation, express or l implied including the warranties of fitness fo' a l

particular purpose or merchantability, with respect to I the accuracy, completeness, or usefulness of the information contained in this response, or that the use of any information, apparatus, method, or process j

disclosed in this response may not infringe privately owned rights; or

b. Assumes any liabilities with respect to the use of, or for damages resulting from the use of, any information, apparatus, method or process disclosed in this response.

, I

! I it h

< l i

f

)

s -k f TABLE OF CONTENTS Section Title Page No.

1.0 INTRODUCTION

1.1 Report Scope 1-1 1.2 Report Summary 1-1 1.3 Reference for Section 1.0 1-3 2.0 SOFTWARE MODIFICATION 2.1 Control Element Assembly Calculator (CEAC) Algorithm 2-1 Changes 2.2 Core Protection Calculator (CPC) Algorithm Changes 2 '- 10 2.3 CPC Addressable Constant Changes 2 - 23 2.4 Data Base Constants for the RPC Algorithm 2 - 25 2.5 References for Section 2.0 2 - 26 o

4 l

iii i _ - - _

i

1.0 INTRODUCTION

)

1.1 REPORT SCOPE The Core Protection Calculator (CPC) System developed by Combustion l '

Engineering is a digital mini-computer system which calculates the minimum Departure f rom Nucleate Boiling Ratio (DNBR) and the peak l

! local power density (LPD) on-line and generates a reactor trip if either the minimum DNBR or the peak LPD ar7 caches the Specified Acceptable Fuel Design Limit. The CPC System has been reviewed by NRC and approved for operation in Arkansas Nuclear One (ANO) Unit 2 and San Onofre Nuclear Generating Station (SONGS) Unit 2. The CPC software for the System 80 Plants contains additional functional design changes. This report presents the changes to the latest NRC approved CPC sof tware (ANO-2 Cycle 2). These changes were made to the CPC System in accordance with the NRC-approved CPC sof tware change procedure (References 1 and 2).

1.2 REPORT

SUMMARY

The bases for modifications made to the CPC/CEAC software are:

1) Upgrade the CPC capabilities to be compatible with the Reactor Power Cutback System (RPCS). The RPCS is designed to rapidly reduce the reactor power by dropping pre-selected Control Element y Assemblies (CEAs) in response to either a large load rejection or loss of one feedwater punp without tripping the plant. The CPC/CEAC modifications consist of an algorithm to detect the actuation of RPC event, and a more accurate calculation of power during the down-power transient associated with a RPC..
2) Modify the FLOW calculation to account for forward flow through a reactor coolant pump at or near zero revolution per minute

( RPM).

1-1

3) Upgrade the non-uniform heating correction factors (Fk) in the UPDATE program.
4) Correct the linear heat rate distributions in the STATIC program to be consistent with the input values from the POWER program.
5) Provide flexibility for eliminating inadvertent reactor trip due to single CEA drops by shifting the range limit of the addressable constants for the DNBR and LPD penalty factor multipliers.

The general format used in describing each software modification contained in this report is a statement of the change, the reason for the change, and a detailed description of the change including algorithm descriptions in symbolic algebra. This is the same format used in CEN-143(A)-NPwhich described software changes for Arkansas Nuclear One - Unit 2 relative to the implementation of an improved DNBR calculation.

9 1-2 t

4

1.3 REFERENCES

FOR SECTION 1.0

1. CEN-39( A)?JP, Revision 02, The CPC Protection Algorithm Software Change Procedure, December 21, 1978.
2. CEN-39(A)/,'P, Supplement 1-P, Revision 01, January 5,1979.
3. CEN-143(A)pf, Revision 1-P, CPC/CEAC Software Modifications for Arkansas Nuclear One - Unit 2, September,1981.

t l

l i

I.

5 1-3

2.0 SOFTWARE MODIFICATIONS

. 2.1 CONTROL ELEMENT ASSEMBLY CALCULATOR (CEAC) ALGORITHM CHANGES

1. Change:

An algorithm has been added for detecting the actuation of a Reactor Power Cutback (RPC) event.

Reason: ,

The CPC/CEAC is part of the plant protection system and operates independently and physically separated from the RPC control system.

It is necessary for the CPC/CEAC System to have the capability of distinguishing between a RPC event and a CEA deviation event and to improve the DNBR calculation in order to prevent an inadvertent reactor trip.

Description:

During the occurrence of a reactor power cutback event, certain pre-selected CEA group (s) will be dropped to reduce the reactor power rapidly. The rate-of-change of the processed CEA positions are used to determine whether the CEAs are dropping. If and only if all of the CEAs in one or more of the pre-selected RPC groups are dropping, then the reactor power cutback flag will be set (IRPC=1) and remain set for

- a pre-determined time period. The flag will be reset automatically after a time interval equal to a preset value.

Insert the following algorithm after the raw CEA positions are converted to inches withdrawn and before the determination of deviations (between Sections 4.1.1 and 4.1.P of Reference ?.5.3).

2-1

('

The variables used in this section of the program are:

CB1(Ji ),CB2(J )

2 CR3(J3 ) = array of CEA number in cutback groups 1, 2 and 3. j (J y= 1, NCBG1; J 2= 1; NCBG2, J =

3 1, NCBG3)

CBSP = distance that determines if a CEA is dropping (in i inches)

CEA LIM

= total number of CEAs CONTAB = array of indices for defining subgroups and groups for each CEA CPOS(i .j) = CEA #1 position from j-1 CEAC execution cycles ago, (in inches withdrawn; i = 1, CEALIMI d " l' ICYCLE+1) ,

DROP (ii) = array of CEA number which are dropping into the core (11=1, CEALIM)

ICYCLE = number of CEA execution cycles used in determining CEA drop IRPC = RPC flag JR = a counter indicating the total number of CEAs

. dropping

. NCBG1,NCBG2 NCBG3 = Number of CEAs in cutback groups 1, ? and 3 NDi = a flag indicating CEA in cuthack group i is dropping (i = 1,3) ,

2-2

1 l

PBOT = bottom position for dropping CEA (in inches i withdrawn)

PPOS(i) = array of present CEA positions, inches withdrawn (i = 1, CEALIM)

TCBP = maximum time period that the RPC flag can remain set (in seconds)

TGROUP = delay time in comparing the CEA motions in,the RPC group (in se'conds) t RPC

= time elapsed since the RPC flag was first set (in seconds)

D 2-3 1

-. a u ,wWe,ae-- ---, . -4y.- -a -- -4 2 -. _g_uw.- -~,wm 4a-- . _ _ _-2 + h=A - a AT A-- - -- o.-- * .s i,2 & m e w._L-w- s-,

, e en ,

  1. 3 l

l t

b a

i L

T O

i e

i 4

.l t

l t

I I

i f

f f.

h l

i L

t 5

I I

i.

F e

4 I

9 l

L e

1 I

f 2~4 l

1

' * - -, , -~_ , w+ _. ,,, -- . .

9 MM 2-5

4 O

e 5

2-6

1 l

l l

1 l

I i

I I

2. Change:

The multiple CEA deviation, Case = 2, is determined only when the RPC flag is not set, s

2-7

Reason:

During a RPC event, multiple CEA deviation may exist when all the CEAs in the RPC group (s) are dropping into the core. Therefore, the Case =

2 which indicates abnormal multiple CEA deviations should not be used for RPC. .

Description:

The CASE 2 deviation is determined only when the RPC flag is set and both the most withdrawn and the "least withdrawn CEAs are outside the deadband.

If IRPC = 0, perform the CASE 2 deviation calculation, otherwise .

bypass the CASE 2 deviation calculation. -

3. Change:

The RPC flag is included in the 16 bits CPC/CEAC data communication link.

Reason:

The algorithm described in 2.1.1 can detect a RPC event. The result is transmitted to the CPC over the 16 bit data communication link.

One of the 16 bits is designated for the RPC flag.

l

Description:

)

The 16 bits are assigned in the following manner:

0 1 7 8 9 10 14 15

  • ** *** LLINK ****

DLINK O

2-8

where:

  • IFAIL - A failure flag which is set (bit 0 = "1") when the CEAC unit contains an excessive number of failed sensors.

DLINK - The transmitted scaled maximum DNBR penalty factor, which is equal to the calculated maximum DNBR penalty factor minus one. The range of DLINK is from 0 to 127 which represents a floating point range of 0.0 to 1.0 scaled over seven bits (least significant bit value is 0.0078125)

    • ICASE2 - A flag which is set (bit 8 = "0") when there are multiple CEA deviations in a subgroup. '
  • *
  • I RP C - A flag which is set (bit 9 = "1") when a reactor power cutback event is determined.

LLINK - The transmitted scaled maximum LPD penalty factor, which is equal to the calculated maximum LPD penalty factor minus one. The range of LLINK is from 0 to 31 which represents a floating point range of 0.0 to 1.0 scaled over five bits (least significant bit value is 0.03125)

        • ISCALE - A flag which is set (bit 15 = "0") when the maximum DNBR and LPD penalty factor are both less than the small scale limits., This flag indicates the method to be used in scaling the penalty f actors for packing into the 16-bit output buffer. (The CPCs include options in the unpacking elements which will adjust the transmitted penalty factors by the appropriate i scaling factor indicated by this flag.)

i 2-9 )

2.2 CORE PROTECTION CALCULATOR (CPC) ALGORITHM CHANGES 1.

Change:

Determine from the 16-bit penalty factor word output from the CEAC whether a Reactor Power Cutback (RPC) event is detected by the CEAC.

Reason:

During a RPC event, CPC may use a more accurate logic in the calculation of power and avoid inadvertent trips.

Description:

In the CPC UPDATE program the CPC extracts from each penalty factor word output from eaac CEAC4 (i = 1 or 2) the following infornation:

I CAS21 (i = 1 or 2) =

the case 2 deviation flag from CEACj where, after unpacking, O indicates no case ? deviation and 1 indicates a case 2 deviation.

=

PF j the DNBR penalty factor from CEAC4 PF Li

=

the LPD penalty factor from CEAC4 I PFCBi = the reactor power cutback bit from CFAC, 1

2-10 __

I____-

O i

1 i

I l-I O

2 - 11

O e

O O

N l

r 2 - 12

1 .

l l

l

  • l l

m.

2. Change:

During the period that the CPC RPC flag is set, IRPC = 1, the following factors are used in the DNBR and LPD calculations a) the of f-line calculated CEA deviatior; penalty -f actors are used in lieu of the penalty factors transmitted by the CEAC.

\* the out-of-sequence penalty factor PFos = 1.0 is used.

b) c) the last ' calculated subgroup deviation penalty factor is used.

d) the last calculated radial peaking f actors and rod shadewing factor are used.

S 2-13

Reason:

When one or more of the RPC group CEAs are dropped into the core, the differences in CEA dropping speed may cause multiple CEA deviation and subgroup deviation. In previous design, large penalty factors which would trip the plant were used for these cases because they were not part of the design basis. 'Nevertheless, when the RPC group CEAs are dropped, the actual DNBR increases because of rapid core heat flux decrease. This software change allows the CPC calculation to more closely model core conditions (Reference 2.5.1) and preeents an inadvertent trip in the event of" a RPC.

Description:

a) In the CPC UPDATE program, if the reactor pow- cutback flag is set.

IRPC = 1, use the off-line calculated RPC penalty factors.

If IRPC

  • I then PF = PF DRPC and PF LPD = PF LPRC where:

PF = total CEA DNBR deviation penalty factor i

PF LPD

= total CEA LPD deviation penalty factor l

PF DRPC' PF LRPC

= corresponding DNBR and LPD penalty f actors for the reactor power cutback event 1

2 - 14

l b) In the CPC POWER program, if the reactor power cutback flag is set, IRPC = 1, set the out-of-sequence penalty factor PF05 = 1.0,

- othenvise calculate the out-of-sequence penalty factor as described in the previous CPC functional specification (reference 2.5.2).

c) In the CPC Power program, if the addressable "CEAC/RSPT Inoperable" flag, CIN0P, is less than three, then determine PF3g as follows:

If " I' IRPC then PF 3g = PF 3gp where:

PF3g = subgroup deviation penalty factor for this POWER execution cycle PF 3gp = subgroup deviation penalty factor calculated from last execution of the POWER program.

d) In the CPC POWER program if the reactor power cutback flag is set IRPC = 1, skip the planar radial peaking factors and CEA shadowing factors calculations. The planar radial peaking factors and CEA shadowing factors remain the same as those calculated in the last execution of the POWER program. If IRPC = 0, calculate the planar radial peaking factors and the CEA shadowing factors as described in the previous CPC functional specification.

3. Change:

The calculation of the pressure rise across a reactor coolant (RC) pump in the Fl.0W Program has been modified to account for fonvard flow through a RC pump with the pump rotor locked at or near zero RPM.

l 2-15

Reason:

- Up to now it has been assumed that affinity laws applied to all coastdown events, and algorithms modeling these laws have been used in the FLOW Program to determine the pressure rise across each RC pump at off-design speeds. During~ the part-loop flow analysis for System 80, it was found that special coastdown events, such as a Locked Rotor Event, did not follow these laws. For these special events, it was found that the flow coastdown was more rapid than was being modeled and predicted. As a result, for these special events an algorithm must be included in the calculation of the pressure rise to account for a very rapid occurrence of negative head across a RC pump with a locked rotor and to predict a more rapid coastdown. Therefore, to model special coastdown events such as a Locked Rotor. Event, an additional algorithm is being implemented in the FLOW Program in the calculation of the pressure rise across each RC pump to account for forward flow through the pump with the pump rotor locked at or near zero RPM.

Description:

The calculation of the pressure rise across each reactor coolant pump is modified as follows:

e l

t l

l l

l

! 2-16 l

l l  ;

I A new constant, a10, will be added to the curve fit coefficients for pump characteristics (the a array) for use in equations 4.1-24A.

4. Change:

4 The non-uniform heating correction factor (F )gin the UPDATE program is adjusted by two additional constants based on the direction of the change in quality margin.

Reason:

The new form of the non-uniform heating correction factor calculation in UPDATE results in a more accurate calculation of Fg for steady state operations and a conservative calculation of Fg during transient conditions.

Description:

. In the previous design, the non-uniform heating correction f actor Fg is calculated in the UPDATE program as:

3-2-17

where: Fg = current non-uniform heating correction f actor for hot-channel node K F

KST

= non-uniform heating correction f actor calculated in the static DNBR program X = update quality at the node of minimum DNBR XST

= static quality at the node of minimum DNBR calculated in static DNBR program In the new CPC design the F-correction factor is calculated as follows:

l where: Fg = current F-correction factor F

KST

= static F-correction factor calculated in the static DNBR program QLC0F1 OLC0F2 = coefficients for the F-correction factor calculation

5. Change:

The calculation of the four linear heat distributions in the STATIC

- program has been modified to account for the difference between the channel and hot pin relative powers.

2-18

Reason:

The STATIC algorithm (CETOP2) uses the axial power distribution and radial peaking factors, which are calculated in the POWER algorithm as hot pin values. At present, they are treated as hot channel values in the STATIC algorithm. This discrepancy was discussed in detail in reference 2.5.4 and was accounted for in the previous sof tware implementations in the calculation of the CETOP2 uncertainty factor l (E). This software modification correc.ts this discrepancy and modifies the channel linear heat rate distributions to be consistent l with the hot pin input values.

Description:

In the previous design the hot pin axial heat flux distributions were calculated as follows:

@c(I)"hCALC TR' PF PDj [~ ']

Fou r [ ] linear heat distributions were computed for the four modeling channels. The [ -] hot pin axial heat flux distribution was combined with the integrated one pin radial peak, and collapsed to [ ' ..- ] distributions, as follows:

~

e Q' AVG 0 AVG

  • ERR 1 *

~

q' HOT OHOT

  • BERR1
  • For j= .

q(j) = 0;ya x 0)(4.4-15) 3 P

( 3 D

. qh(j) =

) - Q' AVG H2

_ (4.4-16) am

l

' P l 93(J) = - ( /) - Q'3ys = 0 33 4

(4.4-17) 9(j) *

  • Q' AVG w

DH4 (4.4-18) l where

~

qi = [] element core-region linear heat distribution (Btu /ft/sec) q = []elementhot-assemblylinearheatdistribution (Btu /ft/sec) -

~

q3 =

[ ] element buffer channel linear heat distribution (Btu /ft/sec) qk =

[ ] element hot channel linear heat distribution (Btu /ft/sec)

= [ ] element hot pin relative axial power c

distribution (tc(i) = relative power in axial segment i of the hot pin).

PI ST = Integrated one pin radial peak.

0 AVG

= core average heat flux at full power (Btu /ft2/sec), constant.

QHOT

= hot pin heat flux at full power (Btu /ft2/sec),

constant.

- = 3.14159 ... constant.

P

  • 2 l P

3

)

P =

4 2 - 20

DH1-Dg4 = Heated diameter of respective channel (ft),

constant.

BERR1 = Addressable DNBR uncertainty f actor E = Region dependent Algorithm uncertainty factor O VG

= Adjusted core average heat flux at full power OHOT

= Adjusted hot pin heat flux at full power This modification corrects the calculation of the four linear heat rate distribution as follows: ,

9)(j) =

-(.01) - Qjyg x DH1(4.4-15) 4 (j) =

2 -(.01) P20VRPN Q' AVG w

DH2 (4.4-16) 9 (j) =

3 -(.01) P30VRPN -

Q'gyg x Dg3 (4.4-17) 9 (j)

  • 4 ,

i -( .01) . P40VRPN - Q' AVG w

DH4 (4.4-18)

I

(

2 - 21

where q = element core-region linear heat distribution, BTU /ft/sec q2 =

element hot-assembly linear heat distribution, BTU /ft/sec q3 =

element buffer channel linear heat distribution, BTU /f t/sec gh = element hot channel linear heat distribution, BTU /ft/sec P20VRPN = Ratio of channel 2 to hot ' pin relative power, constant.

P30VRPN = Ratio of channel 3 to hot' pin relative power, constant.

P40VRPN = Ratio of channel 4 to hot pin relative power, constant.

= element hot pin relative axial power c __ distribution (tc(1) = relative power in axial segment i of the hot pin).

PI ST = Integrated one pin radial peak.

0 AVG

= core average heat flux at full power, BTU /ft2 /sec, constant QiVG

= core average heat flux at full power ith penalties applied.

QH0T

= hot pin heat flux at full power with penalties applied.  !

= 3.14159 ... constant.

B ERR 1

= Multiplicative power uncertainty factor for DNBR calculation.

E = Algorithm uncertainty f actor.

0 HOT = H t pin surface heat flux at full power.

DH1-Dg4

= Heated diameter of respective channel, ft, constant.

2 - 22  ;

i 2.3 CPC ADDRESSABLE CONSTANT CHANGES

- 1. Change:

Shift the positive range limit on the CEAC penalty factor multipliers from [ ].

Reason:

In the CPC design, the CEAC penalty factors are modified by the addressable constant multipliers PFRTD and PF gg. Based on the ANO-2 operating experience, inadvertent CPC trip due to single rod drop may be prevented with PFg g and PFK TL set within a range limit of [ ]. The new software provides the flexibility to set PFK TD and PFRTL to values that would eliminate inadvertent trips due to single rod drops, when, at some time in the future, analyses might be done to justify such an approach. This change in the range limits does not by itself change the actual PFKTD and PFgg values set into the CPC.

Description:

~

The ranges for the DNBR penalty factor multipler PFE TD are [

].

The ranges for the LPD penalty factor muiltiplier PFKTL are [

. 1

2. Change: >

Addressable constants have been added to the CPC and CEAC to define the duration that the RPC flags can remain set.

2-23

Reason:

r This change makes the CPC/CEAC functional design specification generic and applicable,to the plants with or without the RPC feature. For those plants without the RPC system, the RPC algorithm can be nullified by setting the RPC duration to 0.

Description:

Addressable constants are added to the CPC and CEAC Point ID tables to .

allow the operator to change the duration that the RPC flags can remain set.

I t

e i

p 2-24 i __

2.4 DATA BASE CONSTANTS FOR THE RPC ALGORITHM (typical for System 80)

RP FLAG TIMER SETPOIhT 1 SECOND TCBSP [_ (j.

CP 8 EXECUTION TIME, 1 SECOND DT8 [_ (J DELAY TIME IN COMPAPIAG 8OTH CEACS T SECONDS FOR RPC -

T80TH []

PENALTIES FOR DN8R ANC LPD DURING REACTOR 1 DIMENSIONLESS POWER CUTRACK

~

PFORPC PFLRPC . -

ARRAY OF CEA NUMBER Ih RPC GR04E. 1 DIMENSIONLESS CB1(1-4)

CBl(5-8)

CB2(I-4)

C82(5-8)

CB3(1-4)

C83(5-8) _

EUELER OF CEAS IN RPC GROUP 1 OIMENSIONLESS NCPG1 NC8G2 NCBG3 _ ,

DISTANCE THAT DETERMINES CEA IS DROPPING T INCH CBSP [ ]

NUMBER OF CEAC EXECUTION CYCLE USED IN 1 DIMENSIONLESS DETERMINING CEA DROP '

ICYCLE []

A0TTOW POSITION FOR OROPPING ROD 9 INCH P807 []

l MAXIMUM TIME THAT THE RPC FLAG CAN REMAIN 1 SEC0hD SET TCBA []

DELAY TIME IN COMPARING CEA MOTIONS IN THE 1 SECONDS RPC GROUP "

TGROUP [_ ]

2 - 25

j

2.5 REFERENCES

FOR SECTION 2.0

1. Enclosure 3-NPof this submittal LD-82-039, Safety Evaluation of the Reactor Power Cutback System, March,1982.

~

2. CEN-147(S)-NP, Functional Design Specification for a Core Protection Calculator, Janua ry 1981.
3. CEN-148(S)-NP, Functional Design Specification for a Control Element Assembly Calculator, January 1981.

4 CEN-157(A)-NP. Amendment 3-P, Response to Questions on Documents Supporting the ANO-2 Cycle 2 License Submittal, August 1981.

\

i e

1 4

l 1

l l

1 1

2-26 il

. - - ,--s , w n -

Retrieved from "https://kanterella.com/w/index.php?title=ML20050A936&oldid=2545817"