Available online at www.sciencedirect.

com

Nuclear Energy and Technology 3 (2017) 297–301

www.elsevier.com/locate/nucet

Testing of the ICMS input data diagnostic system at unit 1 of

Novovoronezh NPP II
A.V. Semenikhin a, Yu.V. Saunin a, M.M. Zhuk b,∗
a JSC “Atomtechenergo”, Novovoronezh filial “Novovoronezhatomtechenergo” 1 Yuzhnoye shosse, Novovoronezh, Voronezh Reg. 396072, Russia
b “Novovoronezh Nuclear Power Plant”, Branch of JSC “Concern Rosenergoatom” 1 Promyshlennaya zona Yuzhnaya, Novovoronezh, Voronezh Reg. 396071,
Russia
Available online 16 November 2017

Abstract
A real-time in-core monitoring system (ICMS) input data diagnostic system was tested for the first time during the commissioning of
Novovoronezh NPP II’s unit 1. The system was developed by specialists of “Novovoronezhatomtechenergo”, the Novovoronezh filial of JSC
“Atomtechenergo”, with participation of the Kurchatov Institute National Research Center’s experts. The purpose of the diagnostic system
development is to provide for the continuous monitoring of the input data validity in connection with the fact that new and upgraded ICMS
designs include the performance of protective functions based on local in-core parameters.
Issues are formulated to be addressed in the development of the ICMS input data diagnostic system, the structure of the developed system
is shown and basic data on its operation is given. The paper presents the key system test results which have confirmed that the system is
capable to detect in real time the measuring monitoring channels with invalid readings and display respective diagnostic information.
The considered ICMS system may serve a prototype for the development of similar systems, to be completed with novel functions, in
measuring channels of the NPP monitoring and control systems to enable an integrated real-time analysis of data for identification of the
causes for, early detection and prediction of developing defects.
Copyright © 2017, National Research Nuclear University MEPhI (Moscow Engineering Physics Institute). Production and hosting by
Elsevier B.V. This is an open access article under the CC BY-NC-ND license. (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Keywords: In-core monitoring system; Measuring monitoring channels; Diagnostic system; Invalid readings.

Introduction tio (DNBR). This leads to greatly increased requirements to

Apart from information functions, the in-core monitoring
system (ICMS) for new and upgraded NPP unit designs with
VVER reactors also performs safety functions [1,2]. These
functions consist in generation and transmission to the ini-
tiating part of the control and protection system (CPS) of
preventive and emergency protection signals based on in-core
local parameters such as the maximum linear power density
per fuel element and the departure from nucleate boiling ra-
∗ Corresponding author.
E-mail addresses: SemenikhinAV@nvate.ru (A.V. Semenikhin),
yuvsaunin@nvate.ru (Yu.V. Saunin), ZhukMM@nvnpp1.rosenergoatom.rus
(M.M. Zhuk).
Peer-review under responsibility of National Research Nuclear University
MEPhI (Moscow Engineering Physics Institute).
Russian text published: Izvestiya vuzov. Yadernaya Energetika (ISSN
0204-3327), 2017, n.3, pp. 88–97.
https://doi.org/10.1016/j.nucet.2017.11.004
2452-3038/Copyright © 2017, National Research Nuclear University MEPhI (Moscow Engineering Physics Institute). Production and hosting by Elsevier
B.V. This is an open access article under the CC BY-NC-ND license. (http://creativecommons.org/licenses/by-nc-nd/4.0/)
the ICMS maintenance activities, including periodic opera-
tional checks and tests, specifically for the validity of the
input measurement data coming to the ICMS input [3,4].
The input data validity check is one of the most important
primary processing functions in monitoring and control sys-
tems. The ICMS designs include functions for checking the
input data validity but they are limited because they do not
use to the full extent the structural features of data measuring
systems and monitored processes. Meanwhile, data validity
checking algorithms based on these features are applied ex-
tensively, in particular, in the ICMS physical tests during the
NPP unit commissioning and operation [5]. Physical ICMS
tests are full-scale tests conducted, subject to regulatory re-
quirements, to confirm the functionality of the system and its
compliance with the design characteristics.
Presently, the requirements to the use of online monitoring
(OLM) technologies for the NPP instrumentation and con-
298 A.V. Semenikhin et al. / Nuclear Energy and Technology 3 (2017) 297–301
trol (I&C) systems maintenance have already been actually
defined [6,7]. Therefore, activities are under way and differ-
ent methods and techniques are proposed to implement OLM
technologies at NPPs [8–11], including for the VVER NPP
ICMS [12]. With regard for these requirements, a real-time
ICMS input data diagnostic system has been developed by
specialists of “Novovoronezhatomtechenergo” (NVATE), the
Novovoronezh filial of JSC “Atomtechenergo”, with partici-
pation of the Kurchatov Institute National Research Center’s
experts. The development of the system was based on a large
scope of experimental data obtained by the authors as part of
the ICMS physical testing at VVER NPP units of different
designs.
The first ever full-scale test of the developed diagnos-
tic system was conducted during the commissioning of
Novovoronezh NPP II’s unit 1. The paper presents the basic
details of the system, the results obtained and its potential
uses.
Issues addressed in the development of the input data
diagnostic system
The key objective of the input data diagnostic system de-
velopment is the establishment of an automatic real-time diag-
nostic system. This requires the development and implemen-
tation of algorithms that make it possible to automate some
of the procedures performed by attending personnel in the
process of the ICMS periodic maintenance and status check-
ing in accordance with operating regulations. The attending
personnel actions during the performance of required proce-
dures for their formalization and algorithmization involved the
following steps:
– determination of the operating mode in which the reactor
plant is;
– determination of the state that meets the input data validity
diagnostics and checking requirements;
– checking and assessment of the parameter monitoring
channels for the validity of readings which meet the di-
agnostic conditions in a particular reactor plant operating
mode;
– output of required information messages to attending per-
sonnel on checking and diagnostic results;
– archiving of current diagnostic results.
The system shall start to perform diagnostic procedures
which consist in assessing the validity of the measuring chan-
nel readings with the “hot state” or “reactor at low power
level” mode when conditions are fulfilled for the primary
coolant temperature monitoring channel checking with cal-
culation of additive corrections. Channels for monitoring of
general technological parameters can be also checked in this
mode of the reactor plant operation, e.g., the pressure drop
in the reactor coolant pump set (RCPS), the RCPS supply
frequency, the RCPS power and others. During operation at
over 10% of the rated power, conditions are already fulfilled
for the channels to monitor the self-power neutron detectors
(SPND) currents and the parameters involved in the calcula-
tion of the reactor plant weighted average power.
All checks to assess the validity of the measuring chan-
nel readings need to be performed in stabilized states of the
monitored facility proceeding from the statistical processing
method used. One can find out if the stabilized state condi-
tions are fulfilled using the selected stability criteria of typical
parameters for each reactor plant operating mode.
The methods to assess the measuring monitoring channel
readings for validity used in the development of the diagnostic
system were tested repeatedly as part of the ICMS physical
tests during the commissioning and operation of VVER-1000
units of various designs [5]. Dedicated software is used at the
present time to process the test results [13,14]. For validity
assessments, the presented system used operating algorithms
of this software modified, however, for online applications.
Messages on detected invalid readings of the analyzed
monitoring channels are displayed in a generalized format
with color indication, this enables the system operation re-
sults to be easily perceived. Besides, there is also a format
with detailed information on the analysis results for analyzing
the causes for the invalidity.
Diagnostic system structure
The ICMS input data diagnostic system consists of a com-
puter (laptop) connected to the ICMS via the local network.
Fig. 1 presents a block diagram of the ICMS input data diag-
nostic system which provides for a graphic representation of
the system and the interactions among its key components.
The computer is fit with the MS Windows 7 operational
system and special-purpose software, Diagnostika SVRK,
which is the key component of the system and consists of:
– the executable file Diagnose.exe;
– the BD_SVRK.accdb database with information on param-
eters and settings for processing;
– setup files of the ∗ .ini and ∗ .txt type for storing information
on the reactor plant operating mode determination and the
selection of the time interval for the calculation;
– diagram templates for the rapid plotting of dependences
needed to analyze and assess the behavior of the monitored
ICMS parameters;
– text files with stored data processing information;
– binary files – archives of the values obtained from the
ICMS.
Diagnostic system operation
Following the receipt of data from the ICMS, the diagnos-
tic system buffers it. The buffer operates based on the first-in-
first-out (FIFO) principle and may have any size. Normally,
600 data deliveries are fair enough which corresponds to the
time of the received data storage (10 min). This value is sub-
ject to change depending on the pre-commissioning results.
Data needs to be buffered for checks on a set of values and
not on point data.
 A.V. Semenikhin et al. / Nuclear Energy and Technology 3 (2017) 297–301
Fig. 1. Block diagram of the ICMS input data diagnostic system.
Based on the buffered data, the system determines the reac-
tor plant operating mode. This procedure is needed to start the
required monitored parameter validity assessment algorithms.
In accordance with the process peculiarities, the following
operating modes can be identified:
– not determined (ND);
– cold state;
– hot state or reactor at low power level;
– reactor plant power below 10% of rated value;
– reactor plant power over 10% of rated value.
If the reactor plant operating mode is “cold state” or “ND”,
the diagnostic system does not start any task to assess the
status of the monitoring channels checked. When the oper-
ating mode is “reactor power below 10% of rated value”,
tasks are started to determine the validity for the readings
of the primary coolant temperature monitoring channels, the
RCPS pressure drops, the RCPS supply frequencies and other
general technological parameters. When the mode is “reactor
power over 10% of the rated value”, checking of the SPND
current monitoring channels and the parameters for the reac-
tor plant weighted average power calculation is added. In the
“hot state” or “reactor at low power level” mode, the primary
coolant temperature monitoring channels are checked with
calculation of additive corrections. Checks similar to checks
for the “reactor power below 10% of rated value” are per-
formed in this mode. Readings for the coolant temperature
monitoring channels in the primary coolant’s cold and hot
legs, the reactor power readings in the neutron power moni-
toring equipment monitoring channels, and the reactor power
values calculated in the ICMS using different techniques are
used to determine the reactor plant operating mode.
The validity of the measuring channel readings is assessed
by the Diagnostika SVRK system only in the steady-state
mode of the reactor plant operation. The required condition
is determined by monitoring a set of specific parameters for
299
the stability of readings in the channels for monitoring such
parameters. There is a set of monitored parameters for each
mode. The stability of readings for monitored parameters is
determined based on the buffered data.
The stability condition for readings is different in the case
of discrete and analog parameters. If there was a change in
a discrete parameter reading within the given time interval,
the reading is considered stable. For an analog parameter, the
stability condition in the considered time interval is assessed
from the rate of its change and based on the mean-root-square
deviation value. These parameter characteristics shall not ex-
ceed certain values. The algorithm for the approximation of
the linear dependence parameter values are used to estimate
the change rate. Fig. 2 shows the respective system interface
window.
Where the stability conditions are fulfilled for the current
state of the reactor plant, calculations are started to assess
the validity of data in the monitoring channels being checked.
Where invalidity signs are revealed, the group, in which the
monitoring channel with invalid data is detected, is shown in
red in the generalized format (Fig. 3).
Real-time diagnostic system test results
During the commissioning of Novovoronezh II’s unit 1, the
developed ICMS input data diagnostic system was installed
for the first time at the facility it had been designed for and
was therefore tested in full-scale conditions. The tests have
shown that the diagnostic system is serviceable and fits for its
purpose while making it possible to detect on a timely basis
more than one defect. In particular, the readings for one of
the RCPS-1 pressure drop monitoring channels (monitoring
channel code: 11JEC10CP802BXQ3) were found to be in-
valid based on a symptom of noncompliance with the mode
value during tests at the stage of reaching 50% of the rated
power level. The reading for this channel with RCPS-1 in
operation was about 0.002 MPa. The reading of the remain-
300 A.V. Semenikhin et al. / Nuclear Energy and Technology 3 (2017) 297–301

Fig. 2. “List of stability determination paremeters” window.

Fig. 3. Diagnostika SVRK generalized format showing groups with invalid readings.

ing monitoring channels for this parameter was 0.630 MPa
(see Fig. 3). It turned out that such an apparent defect had
not been identified by standard ICMS check and reject pro-
cedures. Concerning this monitoring channel, it was further
recognized by the ICMS developer that such readings in this
operating mode were really invalid and it was required to
withdraw them from further processing.
During tests at the stage of reaching 90% of the rated
power level, invalid readings were detected by the diagnostic
system in the RCPS-2 pressure drop monitoring channel in
the second rack of software and hardware complex for pro-
tection (SHC-P) set 1. The invalidity was identified from the
symptom of the readings deviation from the parallel RCMS-
2 pressure drop measurements in the other SHC-P racks in
excess of the permissible value.
Invalid RCMS-2 pressure drop readings led to DNBR cal-
culation for this SHC-P rack with an error 5% greater than for
the other racks. With no measures taken to remedy the defect
and to stop its further development, there would be a delay
in this SHC-P rack with the alarm generation in response to
 A.V. Semenikhin et al. / Nuclear Energy and Technology 3 (2017) 297–301
DNBR decrease to below the permissible value, which did
not happen only thanks to DNBR being sufficiently great.
This invalid RCPS-2 pressure drop reading not rejected was
used at the upper ICMS level in the power calculation for leg
2 based on the primary circuit parameters. So the power value
for leg 2 was erroneously 8% greater than the expected read-
ing in accordance with the current operating condition. This
error led further to an erroneously 0.7% overestimated value
of the reactor weighed average power. The presented chain of
algorithmic dependences shows to which consequences just a
single invalid value in the ICMS input data may ultimately
lead. After invalid readings were detected for the RCPS-2
pressure drop monitoring channel, a recommendation was is-
sued to withdraw these readings from the upper-level calcula-
tions, as well as to take technical measures consisting in the
blowdown of the pulse-forming lines and the adjustment of
the zero value.
One more typical example of the test results was that the
ICMS diagnostic system recognized as invalid the readings
in one of the SPND current monitoring channels at the stage
of reaching 50% of the rated power value. The invalidity of
the readings was identified from the “error in excess of per-
missible value” symptom meaning the presence of a noise
leading to an increase of the random error in this monitor-
ing channel in excess of the permissible value. Therefore, the
reactor plant power level was also calculated based on the
SPND readings with a greater error both at the upper ICMS
level and in one of the SHC-P rack sets where the readings
from this monitoring channel come to. The noisy SPND cur-
rent monitoring channel could lead to the calculation with an
increased error of the linear power density at the sensor lo-
cation. Accordingly, in the event of a particular operational
state, an excessive linear power density setting could be er-
roneously achieved and a false preventive and/or emergency
protection signal could be generated. After such defect was
detected, a recommendation was issued to withdraw the data
from this channel from the processing procedure both at the
lower and the upper ICMS levels, which increased the pre-
sentability and validity of the system operation in terms of
the in-core power density monitoring.
Conclusion
Tests of the ICMS input data diagnostic system during the
commissioning of Novovoronezh NPP II’s unit 1 has demon-
strated its functionality and the real-time performance of re-
quired functions in the basic operating modes. It has been
confirmed that the system detects on a timely basis the mon-
itoring channels with invalid readings and displays respective
data for the NPP personnel, being either operating personnel
or maintenance and/or engineering and physical support per-
sonnel within respective NPP departments. The correctness
of the diagnostic system operation was confirmed by com-
paring the detected monitoring channels with invalid readings
against the list of monitoring channels with invalid readings
prepared based on data obtained by the traditional “manual”
301
data handling technique as part of the ICMS commissioning
tests at the power levels of 50, 75 and 100% Nnom .
Provided it is further tested successfully during long-term
operation as part of Novovoronezh II’s unit 1 and in other
VVER units to be commissioned, the system may be rec-
ommended as a standard system for improving the operating
reliability of the ICMS and the unit as the whole thanks to
the automation of a range of routine ICMS maintenance and
support procedures.
The system may serve as the basis for the development of
similar systems, to be completed with novel functions, to en-
able an integrated real-time analysis of data for identification
of the causes for prediction and early detection of develop-
ing defects in measuring channels of the NPP I&C systems
[15,16].
References
[1] V.A. Bragin, I.V. Batenin, M.N. Golovanov, et al., Sistemy Vnutrireak-
tornogo Kontrolya AES s Reaktorami VVER (In-core Monitoring Sys-
tems of NPPs with WWER Reactors), Energoatomizdat Publ., Moscow,
1987, p. 128. (in Russian).
[2] A.Y. Kalinushkin, V.V. Kozlov, V.I. Mitin, Y.M. Semchenkov, Atomnaya
Energiya 106 (1) (2009) 3–8 (in Russian).
[3] Obshchiye polozheniya obespecheniya bezopasnosti atomnykh stantsiy
(NP-001-15) [General Provisions for Ensuring Safety of Nuclear Power
Plants]., Federal Service for Environmental, Technological and Nuclear
Supervision, Moscow, 2015, p. 78. (in Russian).
[4] Y.B. Saunin, S.I. Ryasny, Kompleksnye Ispytaniya Sistem Vnutrireak-
tornogo Kontrolya (Full-Scale Tests of In-Core Monitoring Systems),
VANT. Ser. Obespecheniye bezopasnosti AES, Vol. 29, 2011, pp. 12–18.
(in Russian).
[5] Y.V. Saunin, Razrabotka Metodik Kompleksnykh Ispytaniy Sistem Vnu-
trireaktornogo Kontrolya VVER: Avtoref. diss. kand. tekh. Nauk. (Devel-
opment of Techniques for Full-scale Tests of the VVER In-core Moni-
toring Systems), OKB “Gidropress”, Podolsk, 2010, p. 30. (in Russian).
[6] On-Line Monitoring for Improving Performance of Nuclear Power
Plants. Part 1: Instrument Channel Monitoring, IAEA, Vienna, 2008,
p. 122.
[7] IAEA Nuclear Energy Series. No. NP-T-1.2. On-Line Monitoring for
Improving Performance of Nuclear Power Plants. Part 2: Process and
Component Condition Monitoring and Diagnostics. IAEA, Vienna,
2008. 82 p.
[8] U.M. Hashemian, Maintenance of Process Instrumentation in Nuclear
Power Plants, Springler Verlag, 2006, p. 308.
[9] Y.M. Hashemian, Int. J. Nucl. Energy Sci. Technol. 4 (4) (2009)
350–354.
[10] J.W. Hines, E. Davis, Prog. Nucl. Energy 46 (3–4) (2005) 176–189.
[11] J. Ma, J. Jiang, Prog. Nucl. Energy 53 (3) (2011) 255–266.
[12] R.V. Alyev, Raspoznavaniye Sostoyaniya Aktivnoy Zony i Analiz Dos-
tovernosti Informatsii Sistemy Vnutrireaktonogo Kontrolya Pri Eksplu-
atatsyi Toplivnykh Zagruzok VVER-1000. Avtoref. diss. kand. tekh. nauk.
(Identification of the Core State and Analysis of the ICMS Data Validity
During Operation of VVER-1000 Reactor Fuel Loads), Obninsk, 2013,
p. 22. (in Russian).
[13] Y.V. Saunin, A.N. Dobrotvorskiy, A.V. Semenikhin, Tyazheloye
Mashinostroyeniye (11) (2008) 18–22 (in Russian).
[14] Y.V. Saunin, A.N. Dobrotvorskiy, A.V. Semenikhin, in: Sbornik tezisov
dokladov VI Mezhdunar. nauchn.-tekhnich. konf. “Obespecheniye be-
zopasnosti AES s VVER” [Abstracts of the 6th Int. Conf. “Safety As-
surance of NPPs with VVER Reactors”], Podolsk, OKB “Gidropress”,
2009, p. 105. (in Russian).
[15] H.M. Hashemian, Kerntechnik 75 (5) (2010) 231–242.
[16] H. Kim, M.G. Na, G. Heo, Nucl. Eng. Technol. 46 (6) (2014) 737–752.