Real time reactor noise diagnostics for the Borssele (PWR) nuclear power plant

  • Published on

  • View

  • Download


<ul><li><p>Pergamon </p><p> </p><p>Progress in Nuclear Energy, Vol. 43, No. 1-4, pp. 137-143, 2003 Available online at 2003 Elsevier Science Ltd. All rights reserved </p><p>Printed in Great Britain s~tN l~D~mlevo 0149-1970/03/$ - see front matter </p><p>doi: 10.1016/S0149-1970(03)00020-9 </p><p>REAL TIME REACTOR NOISE DIAGNOSTICS FOR THE </p><p>BORSSELE (PWR) NUCLEAR POWER PLANT </p><p>B. BARUT~U a, S. SEKER b, E. AYAZ b, E. TURKCAN b </p><p>alnstitute for Nuclear Energy, "Istanbul Technical University, ITU", bElectrical Engineering Department, "Istanbul Technical University, ITU", </p><p>80626, Maslak, Istanbul, Turkey </p><p>ABSTRACT After the upgrade of Borssele NPP in 1997, core cycle 24, the power plant operated three years more with 91% availability. The authority of the power plant decided to enhance and upgrade the reactor trend monitoring and plant information recording system with higher frequencies than the plant data processing system (PPS) as well as installing a flexible and multiple-purpose reactor noise analysis system which may support the reactor maintenance group with on-line and off-line capabilities for several different signal processing applications. Two measuring and monitoring systems were built in 2001 and fully taken in implementation during the start-up of the new core 28. In this sense, the new system was used in power operation during the 29 th of September 2001. This paper will introduce the measuring system, the operational tasks, and the results obtained so far on the real-time core-bah'el motions (CBM) and the two-primary coolant pump vibrations measured through the reactor noise analysis. 2003 Elsevier Science Ltd. All rights reserved. </p><p>KEYWORDS Borssele nuclear power plant (PWR), NPP diagnostic system, reactor noise analysis, core barrel motions, coolant pumps vibrations </p><p>1. INTRODUCTION </p><p>The Borssele Nuclear Power Plant has, since the first fuel cycle of the power plant in 1974, served plant operational data for the reactor signal processing community. Off-line signal analysis was started in 1974 and continued until 1981. After this time a dedicated system was changed by Energy Research Centrum Netherlands (ECN) using the local measuring system with secure data transfer from Borssele NPP to ECN (200 km distance) and using array processors for real-time calculations. Well-defined algorithms resulted in an on-line real-time system, which was used successfully until 1997. After the plant modification, a new </p><p>137 </p></li><li><p>138 B. l~arm~, ct ~d. </p><p>measuring system was built with collaboration between EPZ and lnterfaculty Reactor Institute (IRI) for thc continuation of plant data analysis in the sense of transient monitoring as well as a general purpose system. used for reactor noise analysis (see e.g. Tarkcan et al., 2001). This system was built during 2001 and used with the start-up of the new core cycle 28 in September 200l. In this paper, we will introduce the system implementation of the core-barrel motions in context of reactor noise analysis and the measured vibrations of the primary coolant pumps. The paper will present the outcome of the experimental results of the last few months. </p><p>2. THE POWER PLANT, DATA MEASURING AND ANALYSING SYSTEM </p><p>The 480 MWe Borssele PWR (KCB) is owned and operated by NV Electriciteits-Productiemaatschappij Zuid-Nederland (EPZ), and located near the Westerschelde estuary. The single unit plant was built over the years 1968 to 1973 by Siemens/KWU and achieved a lifetime load factor above the 80% over the first 24 years. In the first half of 1997 the world's most ambitious nuclear backfitting project successfully finished and the power plant operation in the core cycles in '98 and '99 achieved to average load factor above 91% (see e.g. Gruhl and Kalverboer, 1997). The new plant data collection and processing system was described by E. Ttirkcan et al. (2001), at the Annual Meeting on Nuclear Technology, May 2001. The system consists of two sub-systems: </p><p>a) Monitoring of DC signals of the plant (max 96 signals) with fixed sampling rate of 10 samples/sec. used for operational history recording with the aid of plant transient analysis. </p><p>b) The reactor noise diagnostic system with measuring 32 AC/DC signals with selectable gain and sampling rate (MS measuring system). </p><p>Both systems are built on National Instruments hardware and software, using LABVIEW and are connected through the LAN for the continuous observation of the plant behaviour. In this paper, we used the signals of the continuous operation of the plant through the AC/DC measuring system (MS) with 32 process signals. The measuring system and the Borssele NPP are shown in the Fig. 1. </p><p>KCB MONITORING &amp; PROCESSING "i </p><p>(96Ch) MR System Iv 16 Bits Monitoring </p><p>I'o'c 'ane AC OC S' nalCon # oni AO , . o,seO,o nostics (32Ch) MS system HP-Gain-LP 16 Bits </p><p>I </p><p>I </p><p>I I </p><p>~][ [ LAN </p><p>b4 I ! </p><p>FIGURE 1 - Measuring system of EPZ Borssele Nuclear Power Plant </p></li><li><p>Real time reactor norse diagnostics [ 39 </p><p>Ch 0 1 2 3 4 5 6 7 8 9 tO 11 12 13 14 15 </p><p>TABLE 1 - Signal list of the MS System for Core Integrity </p><p>Signal Ident YQ032T006 YA001T095 SPOIOE001 YX003X082 YX003x062 Excore Neutron Channel (140 cw) YX003X072 Excore Neutron Channel (230 cw) YXO03X052 Excore Neutron Channel (320 cw) </p><p>lncore Neutron Channel YQo23xol 1 YQO23X015 YQ023X016 YA001P001 YA002P001 YDOOIV016 YD001V018 YD001V005 YDO01V017 Vibration Axial Seals </p><p>Signal Name Unit Ch Signal Ident Core Exit Temperature C 16 YD001V010 Core Inlet Temperature C 17 </p><p>Reactor Power MW 18 Excore Neutron Channel (50 cw) V 19 </p><p>V 20 V 21 V 22 V 23 YD002V018 </p><p>Incore Neutron Channel V 24 YD002V005 Incore Neutron Channel V 25 </p><p>Primary Pump Pressure (L-l) Bar 26 Primary Pump Pressure (L-2) Bar 27 Vibration Vert. Elect. Motor Mm/s 28 Vibration Axial Fly-wheel Mm/s 29 </p><p>Vibration Radial Fly-wheel (90 cw) Mm/s 30 Mm/s 31 </p><p>Signal Name Vibration Radial (90 cw) Seals </p><p>YD001V015 Vibration Hori. Elect. Motor YD001V013 Vibration Radial Elect. Motor YD001V006 Vibration Radial Fly-wheel YD001V009 Vibration Radial Seals YD00IV011 Vibration Radial Recirc. Pump YD002V016 Vibration Vert. Elect. Motor </p><p>Vibration Axial Fly-wheel Vibration Radial Fly-wheel (90 cw) </p><p>YD002V017 Vibration Axial Seals YD002V010 Vibration Radial (90 cw) Seals YD002V015 Vibration Hori. Elect. Motor YDO02V013 Vibration Radial Elect. Motor YD002V006 Vibration Radial Fly-wheel YD002V009 Vibration Radial Seals YD002V011 Vibration Radial Recirc. Pump </p><p>Unit mm/s mm/s mm/s mm/s mmIs mm/s mm]s mm/s mrn/s mm/s mrn/s mnVs rflm/s </p><p>mm/s mm/s mm/s </p><p>3. CORE BARREL MOTION ANALYS IS AND THE RESULTS </p><p>One of the items that interested the nuclear power plant operators since the 1970's was the core barrel motion (CBM) detected and evaluated from the fluctuations measured by ex-core neutron detectors around the reactor vessel of the NPP. The possibility of the detecting CBM was first indicated by the J. Thie 1975. The CBM was first evaluated by cross-spectral decomposition by Dragt and Ttirkcan (1977). The method later was extended in on line applications by Tiarkcan (1982, 1985, and 1986). Subsequently, this method was widely used over the world. Several PWR users in the world investigated research and extended the method in extend time and frequency domain analysis, a very extensive literature study is given by Pazsit et al. (1998) including time-domain presentation. Lipcsei and Kiss (2000) also implemented and extended the method for VVR 440/213 and VVER 1000. Extending error analysis depending on number of ex-core neutron detectors and their positions are given by Berta and P6r, (1999). In the earlier measurements of the Borssele NPP, four levels of (each level with 4 ex-core detectors) positions were used. Later, this was reduced to two levels (upper and lower levels) by summing up of the signals. These signals were used in the new measurements. An example of the recent measurement from the new core is given in Fig. 2. </p><p>10 4 </p><p>z </p><p>lO .3 . . . . . . . . . . . </p><p>14~, </p><p>10 "r </p><p>10 -a </p><p>10 "e . . . . </p><p>Ex -core Neut ron Detectors </p><p>82 N1 50 </p><p>- - 62 N2 140 - 72 N3 230 </p><p>52 N4 320 5 10 15 </p><p>Frequency (Hz) </p><p>~0 L ..... 25 30 </p><p>FIGURE 2 - Showing the Normalized Auto Power Spectral Density (NAPSD) function of the ex-core lower neutron detector (N1, N2, N3 and N4) signals spectra for one measurement. Spectral change depending on detector position and the 9.2 Hz is a global effect caused by the delta pressure of two primary coolant pumps (MP1 and MP2). Other peaks are related to core barrel motions (10 - 17 Hz's), 25 Hz peak is created by the pump speed. </p></li><li><p>140 B. Barutu et al. </p><p>~E v </p><p>0 n: </p><p>10 </p><p>E 8 =L </p><p>"-" 6 </p><p>rn 4 O </p><p>2 </p><p>' -" 300 </p><p>o) 200 </p><p>"o v 100 </p><p>0 r - </p><p>13.. -100 </p><p>x 10 .4 </p><p>8 10 12 14 16 18 </p><p>1'0 - - 1 '2 1'4 1'6 1 '8 </p><p>/ i i i </p><p>10 12 1. 16 18 Frequency (Hz) </p><p>Frequency versus direction (Hz), 90 20 </p><p>120 J J~- - . . . . . . 60 / /E: " ....... </p><p>- . , ' ~ - - \ \ </p><p>2 4 ~ ~ 1 3 0 0 270 </p><p>Amplitude versus direction (~m). 90 20 </p><p>270 </p><p>FIGURE 3 - An example of decomposed CBM analysis (on the left) for four (a,b,c,d) succesive weeks of experiments, the right hand figures are presented in polar coordinates. </p><p>In these measurements, only the rms value of the reactivity changes gradually due to the change of boron concentration in the time. On the right figure decomposed CBM frequencies versus direction and amplitudes versus direction are shown in the absolute scale of the movement. </p><p>4. PRIMARY COOLANT PUMP VIBRATIONS AND THE RESULTS </p><p>The main interest is to measure the reactor noise at full reactor power by ex-core and in-core neutron detectors, thermocouples and primary pressures and the coolant pump vibrations. The main emphasis is to develop measuring techniques and methodology to understand core behaviour during the operation. In the last IAEA Istanbul-meeting (1998), periodical measurements in the sense of main coolant pump vibrations were strongly recommended. In this manner, the same request and interest came from the maintenance group of the Borssele NPP. In Table-l, a given number of signals were selected by the maintenance group and continuously measured and the data were analysed using the MS-system. </p></li><li><p>Real t ime reactor noise diagnost ics 14l </p><p>10 </p><p>~o+_:__ __ __ ,o' t ,o ~o ~o </p><p>:i t .~ '. v18 I </p><p>o + L:: I . . . . . . ,+ + 0 10 20 30 Z 13" r . . . . . . . . . . . . . +c ] </p><p>,o + !' . . . . . o. &lt; ~ 0 lO 20 30 'o r - </p><p>I </p><p>o ,~ .+ + ~ ++.-. !L/+%~ ~. , '~ju' v 17 </p><p>+ o . . . . </p><p>. 0 10 20 30 </p><p>I, </p><p>1, / ~ VIO i 0 . . . . . . </p><p>0 10 20 30 Frequency (Hz} </p><p>/ \ v15 L , ) ' </p><p>i} V16' !V19 Electr ic Motor </p><p>I </p><p>V 1 4 " ~ ~ i i n g </p><p>V8~.~.~ 7 Seals </p><p>r~ ~ / t Recirculat ion ' 'o </p><p>10 2 ,__ </p><p>10 ~ k- V15 i </p><p>i 10 . . . . . . . . </p><p>10 20 30 10 z </p><p>V13 i . . . . . . . . . J lO </p><p>. lO 20 30 lO [ . . . . . </p><p>V6 </p><p>.0 10 20 30 10" r . . . . . </p><p>'-" v9 lO ' - - , </p><p>0 10 20 30 10" + . . . . . . . . . . . . </p><p>I </p><p>10 ; [ M' ,~ . . . . &gt;-+-" 4 </p><p>itl,,t V 11 </p><p>10 0 10 2~0 %0 </p><p>Frequency (Hz) </p><p>FIGURE 4 - Primary Pumps Vibration Sensors. Solid lines: Primary Coolant Pump 1 (MP1), dashed lines: Primary Coolant Pump 2 (MP2) </p><p>(a) Excore Neut ron Detectors </p><p>o+t | ~ 82+2 i! ~,,=.~ ~ ~ ~,~. 82-52 I! </p><p>0.6 ~l,~ 62-72 ' </p><p>04 ~ ~ -~ +2+52 </p><p>0.2 </p><p>8 </p><p>3 </p><p>0.8 </p><p>~, o6 O.4 </p><p>o </p><p>0.2 </p><p>0 </p><p>(c) P r imary Pump 1-2 P ressures </p><p>o+ , [U g2-YA2P01 </p><p>62-YA2P01 </p><p>(e) P r imary Pump Pressure 1 </p><p>4 8 12 16 20 </p><p>F requency (Hz) </p><p>- Vibrat ions </p><p>- - Y D lV03 1t YDIV15 Y D2V03 Y D2V15 </p><p>24 28 32 </p><p>(b) Excore Neut ron Detector 82 - Incore Neut ron Detectors </p><p>1 </p><p>08 </p><p>06 </p><p>04 </p><p>0 .2 </p><p>0 </p><p>1 </p><p>0 .8 </p><p>0 .6 </p><p>0 .4 </p><p>0 .2 </p><p>0 </p><p>1 </p><p>0 .8 </p><p>0 .6 </p><p>0 .4 </p><p>0 .2 </p><p>0 0 </p><p>i l 15 t6 </p><p>J 'i .I . </p><p>(d) P r imary Pump 1-2 P ressures </p><p>(f) P r imary Pump Pressure 2 - V ib ra t ions </p><p>t ' ' ~ YD1V15 L; ', : ~o2vo3 t~/|'~r/ "i i ",' Y D2V 15 l t tk , i ! L </p><p>4 8 12 16 20 24 28 32 </p><p>F requency (Hz) </p><p>FIGURE 5 - Different combinations of Coherence informations between the various sensors. </p></li><li><p>142 B. 14arut~u ct aL </p><p>Very interesting results were obtained from both primary pump vibration signals as follows. Fhc vibration characteristics of both coolant pumps are nearly the same and the absolute value of the vibration for a given Vibration Criterion Chart indicates the range of "allowable". There are several conclusions that can be tracked from the coherence information. One clear example can be seen from the coherence between ex-core and in-core detector signals, e.g. that the in-core neutron detector 11 is not functioning. For monitoring system integrity, it is important to follow the coherences between all ex-core and in-core neutron detector signals together with pressure and main vibration signals. Fig. 5 shows the main coherence functions between all these signals. Coherence and phase information were calculated in real- time. Different combinations of coherences are shown in Fig. 5 and the findings can be summarized as follows. (a) Coherence between the ex-core neutron detectors indicates strongly coherent 9.2 Hz, which related to </p><p>the reactivity effect (e.g. phase between them all zero). (b) Coherence between the in-core neutron detectors where in-core detector 11 is not very coherent it is </p><p>degradated sensor. (c) Coherence between the ex-core neutron detectors and primary system pressure sensors, reactivity effect </p><p>at 9.2 Hz caused by the delta pressure between the two pressures is dominant. (d) Coherence between the pressure signals of the two primary pumps (MPI, MP2) shows very high </p><p>coherence at 6.5 Hz. This 6.5 Hz indicated as a standing wave in the primary loop and proportional to the sound velocity, which is temperature dependent (E. Ttirkcan, 1984). </p><p>(e) , (f) in Fig. 5 indicates the coherence between pressure signals of each pump with the vibration sensor signals of the same pumps. These effects are summarised in the physical benchmark test of Borssele (PWR) (E. Ttirkcan, 1984). </p><p>O0 </p><p>C3 CO </p><p>&lt; Z </p><p>1 , -1P </p><p>I r </p><p>%80 </p><p>10 12 14 16 18 20 Frequency (Hz) </p><p>FIGURE 6 - NAPSD changing of one ex-core neutron detector between 80% - 100% of nominal power. </p></li><li><p>Real time reactor noise diagnostics 143 </p><p>Fig. 6 is an example for NAPSD of the ex-core neutron detector N1 signal given during the power-up operation. This figure indicates the development of the power spectra at hot condition between 80% - 100% power. </p><p>5. CONCLUSIONS </p><p>The new system for reactor diagnostics is functioning satisfactorily. Data collection and the presentations worked well. Some of the presentations with the LabView programs should be enhanced in the sense of direct use of results for the maintenance people. The monitoring of the vibration signals indicated that tins- velocity limits were in the range of allowable at full power operation that satisfies the maintenance needs. </p><p>ACKNOWLEDGEMENT </p><p>The authors gratefully acknowledge the interest of the NV-EPZ Borssele NPP authorities as well as the special support given by the EPZ maintenance personal Ing. P. Stok, J. Brevoord, and P. Tange. The first author B. Barutqu specially thank the EPZ management for inviti...</p></li></ul>


View more >