JPS607392A - Method of monitoring output from nuclear reactor - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

Detailed Description of the Invention [Field of Application of the Invention] The present invention relates to a method for monitoring the output of a boiling water nuclear reactor, and in particular, when an unstable phenomenon in which the output fluctuates over time occurs, A reactor output monitoring method that not only detects the unstable phenomenon but also determines whether the unstable phenomenon is occurring throughout the reactor core or locally, and if it occurs locally, its local location. It is related to.

[Background of the invention]

Traditionally, nuclear reactors have been designed and operated in such a way that unstable phenomena do not occur, but in the event that an unstable phenomenon occurs, it must be detected immediately and the type of instability and its occurrence must be investigated. There is a need to suppress instability phenomena based on location. This is because if the unstable phenomenon is left unattended, control becomes difficult.

Nuclear reactor output has traditionally been monitored in order to detect unstable phenomena, but with conventional methods it is not possible to detect unstable phenomena even though the species is in good condition. .

That is, the reactor power monitoring method according to the prior art is LPRM.
(Local power management
) detection signal and AP(Av) as the average of these signals.
er2gedpower Range Mo” for
%3) to detect the output and temporal fluctuations in the output. Figure 1 shows the plane of the reactor core of a 1.1 million kW class BW reactor (the size is expressed in inches). According to this figure, the reactor core 1 has 185 control rods 3 between 764 fuel assemblies 2, Forty-three LPRM4s are arranged as shown in the figure, and each LPRM4 also has four neutron detectors υ. As shown in Figure 2 <LPR
The four neutron detectors A-D that make up M are located at the bottom of the reactor.
The reactor core power is measured by detecting neutrons at that position.

If an unstable phenomenon occurs in which the amplitude of output fluctuation increases over time, we will promptly detect this fact, its type and location, and insert control rods to suppress the output fluctuation. Measures such as these are being taken. However, in the conventional method, if the amplitude of the output fluctuation is the same as or smaller than the amplitude of the noise contained in the LPRM signal and APRM signal, the fluctuation cannot be directly detected from those No. 111; In order to determine in what region within the core the source of power fluctuations lies, there is always a problem if the monitor has to compare the amplitudes of the individual LP and M signals.

For this reason, apart from the above method, it has recently been considered to apply spectrum analysis to the processing of LP-M signals (Japanese Patent Laid-Open No. 57-50780). Prepare the spectrum in advance when no instability phenomenon occurs, compare it with the spectrum of the LP-M signal measured at any time, and if the difference is larger than the set value, it is considered a failure. It is determined that a stability phenomenon is occurring. However, when using this method, even if the spectrum changes due to noise etc., it may be mistakenly determined that an unstable phenomenon has occurred. This makes it impossible to detect the location where the phenomenon occurs.

[Purpose of the invention]

Therefore, the purpose of the present invention is to If an unstable phenomenon is detected, it is possible to determine whether it is related to the entire core or localized, and if it is localized, the location of its occurrence can be detected. It is our goal to provide a method for monitoring reactor output as quickly as possible.

[Summary of the invention]

For this purpose, the present invention provides that the frequency of output oscillation when an unstable phenomenon occurs is approximately 0.3 to 0.5 Hz, although there are some differences depending on the furnace type and operating conditions. Focusing on the fact that the waveform of the output oscillation is different depending on whether the output oscillation is related to the entire core or local, the following method is proposed.

That is, by obtaining the power spectral density function (PSD) by spectral analysis of the LPR and M signals, and then calculating the coherence and transfer functions for all combinations of any two signals, it is possible to determine whether an unstable phenomenon is occurring. In addition, if an unstable phenomenon occurs, it is necessary to determine whether it is related to the entire core or localized, and if it is localized, the location of the occurrence is determined.

[Embodiments of the invention,]

The present invention will be explained below with reference to FIGS. 3 to 20.

First, before specifically explaining the present invention, its theoretical background will be explained.

In general, a reactor core tends to become unstable as its power increases and as its axial power distribution reaches a lower peak (a state in which the output peak UlfM is closer to the lower end of the core).

Here, as an analysis condition, if we assume that the core is divided into four regions 1 to 2 as shown in Fig. 3 and a power distribution is given to each region I to 2 as shown in Fig. 4, the area I has a higher output than other areas ■～■, and it is unstable because it has a lower peak, and areas ■～■
The analysis results show that it is stable. Local instability phenomena are occurring in the core under these conditions.

Here, the analysis results of the changes in core power in each of regions I to ① using the three-dimensional nuclear thermal-hydraulic stability analysis code are shown in FIG. 5(a) as relative values to the initial values. Also, Figure 5 (1))
is an enlarged view of a part of the diagram, and as can be seen, the amplitude of the vibration in tracking area I is larger than that in other areas ■～■, and the phase is also different from that in other areas ■～■. After that, the waveform becomes different. Area■
The output also oscillates in ~■, but this is due to neutron diffusion from region I. The phase is still delayed because
In region ■, the output is larger than in other regions ■~■, so void steam tf is generated! l) The amount of voids is large, and the time that these voids exist in the flow path is longer than in other regions (1) to (2). If there are many voids, the effect of moderating neutrons will be reduced and the output will not increase.

Figure 6 shows the output response analyzed using the same three-dimensional nuclear thermal-hydraulic stability JVR analysis code when an instability phenomenon occurs in the entire core, as opposed to the initial capacity. In this case, the instability phenomenon is considered to be distributed throughout the core, so both the amplitude and phase are approximately equal.

The present invention has been made against the above background, and the present invention will be explained in detail below.

Figure 7 shows how the reactor power monitoring device according to the present invention is installed in the reactor core and LP.
This is shown together with Rin M. According to this, multiple L
The measurement signal (LPRM signal) from the neutron detector composing each P-M4 is sent to the amplifier 5 (
51 to 5N), multiplexer 6, A/D converter 7
The signal is converted into a digital signal and is temporarily stored in the buffer memory 8. The measurement signal stored in the buffer memory 8 is then selectively given to the spectrum analyzer 10 by the selector 9, and the spectrum is analyzed by the spectrum analyzer 10 to determine the PSD, transfer function, and coherence. The PSD, transfer function, etc. from the spectrum analyzer 10 are then processed and calculated in a predetermined manner by the calculation circuit 11. In this example, the number of neutron detectors is set to N, but all of the 11 signals between N ship types from the neutron detectors are not necessarily required, and only those between the two can be selected as necessary. Both are possible. If a measurement signal is appropriately selected and subjected to processing, the t'1'' determination process can be carried out very quickly. In addition, the spectrum analyzer 10 uses PSD, etc., and their meanings are as follows. That is,
PSD represents the contribution rate of each frequency component to the average power of random fluctuations, and therefore shows a large value in the frequency band of unstable vibrations. Next is coherence, which is an index that expresses the degree of association between two measurements.If the relationship is perfect, the value is ``1'', and if there is no relationship at all, the value is ``0''. It is defined in . The lightest transfer function also represents the ratio of absolute unity between the two measurement signals and the γ phase difference.

Now, FIG. 8 and FIG. 9 show the normalized PSD of the measurement signal expected to be measured by arbitrary two neutron detections ff1f5t and 5J when an unstable phenomenon occurs. These figures show the peak of the cis vector ito, 3 to 0.
．． 51-(Appears near Z, but does this fact alone indicate that an unstable phenomenon has occurred?-1,1'lJ cannot be determined.Even when it is stable, it has not been measured in an actual reactor. L
According to the noise analysis of the P-fired M signal, depending on the furnace type, the first
As shown in Figure 1, results have been obtained in which a spectrum peak appears at 0.351 (z).
Application of No I Re Ana
Iys I S Me thod t OA40n
1 torStabi eyes of Boiling
Water Reactor HI3゜1 also U9adh
yaya and 3 others, progress in Nucl
earEnergY 1982. Vo19. pp6
57-664). Therefore, even though the presence of a peak in a spectrum can serve as a rough guide, it is a problem in terms of uniaxiality to immediately determine that an unstable phenomenon has occurred based on the presence of a peak. Therefore, in order to improve the reliability, the coherence C0Hr of all combinations of two signals in seven rows is calculated.
It takes z. If an unstable phenomenon occurs, the coherence COHrs will show a value close to "1" in the frequency band of 0.3 to 0.5 Hz, as shown in FIG.

At this time, it is necessary to set a threshold value for determining whether a β-indeterminate phenomenon occurs and C is present, and a value of about [0.5 J is appropriate as the threshold value. Figure 12 shows the noise coherence measured between two representative points in the furnace.
tigations in J30i1ing-Wa
ter and pressurized-Water
R,actors;G,Kosaly,Prog
Ress in Nucle3r Energy Vol.
, 5. pp145-199). Of course, this is in a stable state, but the coherence is l-0 as the maximum value.
, 4J, it is considered appropriate to install it as described above.

FIG. 13 shows the process flow of spectrum analysis in the spectrum analyzer, and FIG. 14 shows the process flow of determining whether the system is stable or insecure, which is performed by the arithmetic circuit.

According to this, in spectrum analysis, first the PSD is determined for the measurement signals from N neutron detectors, and then the transfer function '1' lJ and coherence CQHIJ are determined for all two combinations of those measurement signals. There is. When the processing in the spectrum analyzer is completed, the operator n1 circuit is activated next, and the processing shown in FIG. 14 is executed. In this stable/unstable process, the presence or absence of a peak is determined within the range of 0.3 to 0.5 If Z for all PSDs corresponding to measurement signals, and if it is determined that there is a peak, the It is determined whether or not the coherence of all the combinations of those determined to exist are equal to or greater than a reference value (approximately 0.5), and the number of combinations that are equal to or greater than the reference value is counted. In this case, the criterion for determining the presence or absence of a peak is 0.
It is assumed that the spectrum in the 3 to 0,5 Hz (1 band) range is 1.1 times or more than that in other bands, but this value should be set appropriately depending on the furnace type and operating conditions. The same holds true for the reference values for reeds and coherence.

In this way, the number of combinations for which the coherence is greater than the reference value is counted, but once the processing to determine whether the coherence is greater than or equal to the reference value is completed, the total number of combinations of the b1-determined individuals is counted. number of pieces (NC2=N (N-1
) /2 ) It is determined whether an unstable phenomenon is occurring or not based on the ratio to ). In this example, whether or not an unstable phenomenon has occurred is determined based on whether the counted number is a dog or not compared to NC2/2, but this reference value is also set optimally depending on the furnace type, etc. It is a rubebe thing.

As described above, it is possible to know whether or not an unstable phenomenon is occurring.Next, when an unstable phenomenon occurs, the process of determining whether it is related to the entire core or local. explain.

FIG. 15 shows the flow of the process. As already explained in Fig. 6, when the instability phenomenon affects the entire core, the relative values of the amplitude of the output in each region within the reactor are almost equal, and there is almost no phase difference. I understand.

That is, assuming the transfer function TIJ of two arbitrary measurement signals, its absolute value l TrJl and phase difference arg(TIJ
) are shown in Figures 16 and 17, respectively, and 0
．． Frequency 1y from 3 to 0.5Hz, band: loquat vs. direct &, j:
t(odl), the phase difference is approximately equal to 0.

Therefore, if it is determined in the flow shown in FIG. 14 that an unstable phenomenon has occurred, the flow shown in FIG. ε<ITIJ+<1+ε, and −φ<a r
g ('I'rJ) < φ) is taken as an argument. If the final count result is worse than NC2/2, it is determined that the instability phenomenon is occurring throughout the core.

In this case, the values of ε and φ are generally set as 0.05 and π/20, but should be set appropriately depending on the furnace type and operating conditions. Also, 1 of NC, /2 used as a reference
The same goes for straightening.

If an unstable phenomenon occurs, but it is determined that it does not affect the entire core, it means that the unstable phenomenon is occurring locally. needs to identify its local location. FIG. 18 shows the flow for determining the local position.

The principle of judgment in this flow is that when an unstable phenomenon occurs locally as shown in Figures 5 (a) and (b) and Figure 6, the measurement signal related to the unstable region and the stable region The absolute value and phase difference of the transfer function between the
The result is as shown in FIG.

That is, if the measurement signals from the neutron detectors 51 and 5J are both related to the stable region, the absolute value of the transfer function TIJ is close to 1 (OdB), and the phase difference is close to 0; Neutron detector 5! If the measurement signal from I is related to the unstable region, the transfer function T from I to J
The absolute value of rJ becomes smaller by 4 times than l, and the phase difference a r g (TIJ) becomes larger. Therefore, by detecting the transfer function that has the smallest absolute value and the largest phase difference among the transfer functions related to the combination of all the two 11" measurement signals, the local position where the unstable phenomenon is occurring can be found. be.

Now, to explain the flow shown in FIG. 18, first, the transfer function TIJ determined in the flow shown in FIG.
I, a r g (TJx) = a r g (TI
J). For example, Den 7! A transfer function aT21 is determined from the function T12.

In the flow of this example, the transfer function To + T2z +...
・Although TNNs are also included, as will be explained later, these are deemed unnecessary and will be handled appropriately. Once transfer functions related to all required combinations have been found, the next transfer function TxJ with the minimum absolute value and maximum phase difference is detected as updatable. For all transfer functions, ptransfer function 1 that satisfies the above conditions.
If 'tz is detected, it can be determined that an unstable phenomenon is occurring in the region centered on the neutron detector 5. Therefore, by taking measures such as inserting control rods that are compatible with this robust area, unstable vibrations in the exit direction can be suppressed.

〔Effect of the invention〕

As explained above, the present invention obtains the PAD by spectrum analysis of the LP ILM signal including the total d11] signal, and then calculates the coherence and transfer function related to all combinations of any two L P ILM signals. , based on the PSD spectrum and coherence, it is possible to determine the presence or absence of an unstable phenomenon, and if an unstable phenomenon occurs, the transfer function and the type of cine stable phenomenon are determined, and furthermore, if the unstable phenomenon is local, the The local position is determined by Therefore, the present invention has the advantage of being able to detect the presence or absence of an unstable phenomenon, its gender, and further the local position related to the occurrence of the unstable phenomenon with good accuracy.

[Brief explanation of drawings]

Figure 1 is a diagram showing the BW reactor core plane, Figure 2 is
Figures 3 and 4, which are diagrams for explaining the configuration of the LPRM arranged in the reactor core, are diagrams showing the case where the core is divided into four regions as an analysis condition and a power distribution is given to each region. , Figures 5(a) and 5(b) are diagrams showing the output response when local instability occurs in the nuclear reactor, as determined by analysis, and an enlarged view of a part of it. FIG. 7 is a diagram showing the 4111 configuration of the reactor power monitoring device according to the present invention together with the reactor core, and FIG. 8 and FIG. 9 are diagrams showing the output response when an unstable phenomenon occurs. Figure 10 is a diagram showing the coherence between two arbitrary measurement signals when an unstable phenomenon occurs (PSL of two arbitrary measurement signals), and Figure 11 is a diagram showing the coherence between two arbitrary measurement signals when an unstable phenomenon occurs. LPR at time
Figure 12 is a diagram showing the noise analysis of No. M-=, Figure 12 is a diagram showing the coherence of noise measured between two representative points in the furnace during stable conditions, and Figure 13 is the spectrum measured by the spectrum analyzer. Diagram showing the flow of analysis processing, No. 14
15 and 18 are diagrams showing the entire flow of the process of determining the occurrence of an unstable phenomenon, the LIJ determination process of the type of unstable phenomenon, and the process of determining an unstable local position, respectively, which are executed by the arithmetic circuit. Figures 16 and 17 are diagrams for explaining the principle of determining the type of unstable phenomenon, and Figures 19 and 20 are diagrams for explaining the principle of determining unstable local positions. This is a diagram. 1... Core, 4... LPRM, 5 (5t~5N)
...Amplifier, 6...Multiple breech, 7...A/
D converter, 8... Buffer memory, 9... Selector, 1o... Spectrum analyzer, 11... Rhetoric circuit. Agent Patent Attorney Masamito Akimoto Kaya 2nd b Group'rrME (5εC) Hiroshi 7 Figure 1 Figure 10 Waist W: (H'L) Fu 11 Figure 1Z Kaya No 3 Figure 14 Gun 15 Moon spear 76 nomata I ring, f-m = I... old? #p < Hg3 I・7・Tsuruba□ I → '1 \\'1 Zu゛茅 lθ 目$lqKugawa imaro &<〃Zノ輼2θ Diagram number of cycles (〃l) 546-

Claims (1)

[Claims] 1. A power spectral density function is determined by spectrum analysis of measurement signals from a plurality of neutron detectors installed scattered within the reactor, and the power spectrum density function is calculated based on all combinations of any two measurement signals. A nuclear reactor power monitoring method characterized by determining at least the presence or absence of an unstable phenomenon based on coherence based on coherence and transfer function, using the presence of a spectral peak in a specific frequency band of a power spectral density function as a precursor. 2. The reactor power monitoring method according to claim 1, wherein when it is determined that an unstable phenomenon exists, the type of the unstable phenomenon is determined based on the absolute value and phase difference of the transfer function. 3. The reactor power monitoring method according to claim 2, wherein when it is determined that local instability is occurring, the unstable local position is determined based on the absolute value and phase difference of the transfer function.