RU2503070C1 - Method for experimental research of coolant mixing in operating nuclear reactor - Google Patents

Abstract

FIELD: power engineering.

SUBSTANCE: method for experimental research of coolant mixing in an operating nuclear reactor consists in the fact that a system of emergency boron injection is used at any level of power in one or more loops to create weak unevenness in distribution of an indicator, which plays a role of a temperature indicator before entrance to the reactor, and in the reactor core - a role of a neutron absorbing indicator. At the same time the complex of standard systems of neutron and temperature monitoring in sections of a circulating circuit is used: before entrance to the reactor and the reactor core, in the reactor core and before exit from the reactor. To eliminate distortions and fluctuation of recorded signals, they apply processing of low disturbances of signals in monitoring systems.

EFFECT: possibility to perform research directly in process of normal operation of a reactor or when it is commissioned.

4 cl, 20 dwg

Description

Technical field

The invention relates to nuclear energy, and more specifically to a reactor installation of water-cooled nuclear reactors using thermal neutrons.

State of the art

Knowing the mixing characteristics of the coolant in a nuclear reactor is important for safety. It is investigated experimentally at stands and operating reactors. Measurements on the stands allow you to explore various aspects and scenarios, however, they are costly and leave the applicability problem for full-scale power units. Computational and theoretical studies using modern codes such as CFD have great potential in the future, but they are also costly and require verification by experimental data. The experimental studies carried out on existing nuclear power reactors are, first of all, the nature of tests and confirmation of the design characteristics of systems and equipment. At the same time, in order to solve the urgent tasks of improving the safety and competitiveness of nuclear energy, it is necessary to modify and expand these studies aimed at the future. There are natural limitations for such experimental studies on operating commercial reactors: (1) they should be carried out only in relatively narrow ranges of parameters allowed by the technological regulations for safe operation; (2) if possible, only standard systems should be used to create uneven distribution of the indicator and its registration; (3) the losses of the installed capacity utilization factor of nuclear power plants should be minimized or completely eliminated due to the experiment itself or because of its potentially negative impact on operation. Subject to these restrictions, field research is the least expensive and most representative and preferred.

There is a method [1, 2] of an experimental study of inter-loop mixing of a coolant in an existing nuclear reactor by creating a temperature difference (DT indicator) between the loops and, accordingly, a “temperature spot” in the core when operating at low power or in a subcritical state. The formation of a “temperature spot” was provided by changing the operating conditions of one steam generator, for example, by shutting off the steam shut-off device and turning off the feed water at a power of about 10% Nnom [1] or by briefly cooling the steam generator in a subcritical state [2]. The registration of the “temperature spot” was carried out by a system of temperature sensors at the outlet of the fuel assemblies in the core. The disadvantage of this method is the dissipation of the ΔТ indicator during heat exchange with structural elements and other steam generators, which significantly distorts the parameters under study — mixing coefficients and angular displacement of the coolant flow in front of the core. There is an inertia in the process of heating-cooling the steam generator, which is a disadvantage of the method, because mixing parameters are also important for fast processes. The research process can take up to several tens of hours of time, which is also a disadvantage of the known method.

Disclosure of invention

The objective of the invention is to increase the accuracy and reliability of mixing studies carried out using standard technical equipment, applicable to all power units of VVER reactors during their operation at any power level from the minimum controlled level (MCU) to 100% Nnom, at any time of the reactor campaign, directly in during normal operation (or during commissioning), and without reducing the installed capacity utilization factor.

The technical result of the invention is to obtain data to evaluate the characteristics of mixing the coolant in the complex - in all important parts of the circulation circuit (inside the reactor - before entering the active zone, in the active zone and before leaving the reactor, as well as in the main circulation pipe (MTC) before entrance to the reactor).

The technical result of the invention is also to obtain data for verification and tuning of the associated neutron-physical and thermohydraulic calculation programs, allowing to simulate the spatial kinetics of neutrons and real mixing of the coolant.

The total technical result of the invention is the identification of additional reserves according to design safety criteria in VVER nuclear reactors, which can be converted into economic gain by reducing neutron leakage and / or increasing the power of reactors without compromising their safety.

The problem is solved by the method of studying the mixing of the coolant, which is characterized in that the emergency input system of boron create unevenness in the distribution of the indicator in the reactor core. The role of the indicator in operating VVER-type reactors is played by a cold (~ 20-40 ° C) solution of high concentration boric acid (~ 40 g / kg), which is fed into the cold thread of one (or more) of the reactor loop. Prior to entering the reactor, the indicator plays the role of a temperature indicator, and its registration near the injection point (~ 1-3 m) is carried out by standard temperature sensors (TS) located along the perimeter of the MTC in cold strings of loops, which makes it possible to evaluate the features of the nature of the flow in the MTC ( rotation, stratification). Further, in the core, the indicator plays the role of a neutron-absorbing indicator. Such an indicator is devoid of a lack of dissipation during circulation in the MTC and passage through steam generators inherent in the known method [1, 2]. Unevenness in the distribution of the neutron-absorbing indicator causes unevenness in the distribution of energy release (and, correspondingly, heating of the coolant) in the reactor core, for the registration of which a complex of four independent standard neutron and temperature monitoring systems is used. These are neutron detectors - direct charge sensors (DPS) in the fuel assembly channels and ionization chambers (IR) of the ACNP system around the reactor, as well as thermocouples (TP) in the fuel assembly channels and thermal sensors (TS) in hot strings of loops.

Based on the qualitative and quantitative coincidence of the signals of these four monitoring systems, the integral coefficients of inter-loop mixing are estimated with a high degree of certainty — the proportion of the flow rate of each loop falling into this and the other loops after each revolution of the coolant. The angular displacement of the coolant flow at the entrance to the active zone. To a first approximation, local mixing coefficients are also estimated — the fractions of the coolant flow rate of each loop falling into each fuel assembly of the active zone. It is precisely these integral and local fractions that are proportional to the perturbations in the distribution of the neutron-absorbing indicator, recorded through perturbations in the distribution of energy release. That is, from the distribution of the perturbation of energy release or heating, the integral and local mixing coefficients of the coolant are directly estimated. The error of such estimates depends on the number of neutron and temperature sensors available in various VVER projects, on the performance of pumps that create unevenness, on the magnitude of distortions (which can be minimized) and some other reasons.

Analysis of the signals of direct charge neutron sensors (DPS) by the height of the active zone is useful for evaluating the mixing of the coolant during movement in the active zone. The erosion of the “boron” and “bezborny” spots due to the transverse inter-cassette turbulent exchange can cause a decrease in the unevenness of energy release when moving from the lower layer to the upper. A quantitative assessment of this transverse exchange can be made by subsequent numerical reconstruction of the concentration distribution of the neutron-absorbing indicator in the system code.

A feature of the invention is the creation of weak unevenness in the distribution of the indicator, which is essential for conducting research directly during normal operation, without any time costs and without loss of installed capacity utilization factor.

At low capacities, this method is limitedly applicable - only the AKNP (IR) system works. At higher powers, signals from temperature sensors are added, and above 35% Nnom, signals are added to the SCR (height layers in the core). The greatest capabilities of the method are possible when testing from 35 to 100% of rated power.

The invention is characterized by the fact that a standard safety system is an emergency input system for high-pressure boron [3], containing several independent channels, with pumps of small constant capacity ~ 6-15 t / h each (depending on the specific reactor design), and having the purpose of shutting down the reactor in the event of over-design accidents, it is used as a new tool for studying the mixing of the coolant.

The invention is also characterized by the fact that a special computer processing (normalization) of weak disturbances in the distribution of energy is applied, which are created by a low-performance emergency input system for boron, which allows you to filter out distorting noises and fluctuations and identify clear responses from all standard neutron and temperature monitoring systems.

Brief Description of the Drawings

The invention is illustrated by drawings, which represent:

figure 1 is an example of a layout of the core, circulation loops (Loop 1-4) and six ionization chambers of the working power range (IK 1, 4, 6, 8, 12, 15) around the core, relative to the axes of the VVER-1000 reactor. The axis of the reactor divides the core into four quadrants adjacent to the circulation loops with the corresponding numbers;

figure 2 - layout of 54 channels of neutron measurements and temperature (KNIT) in the fuel assembly of the active zone of the Bushehr NPP. The partitioning of the core with fuel assembly numbering into six sectors of 60 ° symmetry is indicated. KNIT channels have thermocouples at the input and output of the fuel assemblies, as well as 7 direct charge neutron detectors (DPS) located along the height of the fuel assemblies;

figure 3 - layout of 64 channels of neutron measurements with a DPZ (7 pieces each in height) and 95 thermocouples at the outlet of the fuel assemblies of the active zone of most operating VVER-1000 reactors;

figure 4 - an example of the layout of the nozzles on the main circulation pipelines (MTC), into which the indicator is entered by pumps emergency input of high pressure boron (B2 on the cold threads of the circulation loops), as well as six located along the perimeter of the MTC thermal sensors (TC15-20) in cold threads and six thermal sensors located along the perimeter of the MTC in hot threads of the circulation loops;

figure 5 - a typical change in reactor power in six ionization chambers (IK 1, 4, 6, 8, 12, 15 (figure 1) with a recording period of dt = 1c) in the process of mixing studies, for example, the alternate operation of the pumps emergency input of boron high pressure (TW1 and TW3) on loops 1 and 3. It is seen that the operation of the emergency bur input pump introduces a relatively weak disturbance, which is also distorted by fluctuations and unsteady poisoning;

6 is a typical change in the deviation of the signals of the monitoring systems from its values averaged over all sensors (Dev, rel.%), using the example of measuring reactor power in six ionization chambers (Fig. 5). The largest deviations with a minus sign in the energy release or heating of the coolant characterize the presence of a “boron” spot, i.e. core fragment, where the maximum share of the coolant from the loop directly gets, with the TW pump running. On the contrary, the largest deviations with a plus sign characterize the presence of a “spotless” spot, i.e. core fragment, where the coolant does not directly get from the loop, with the TW pump running. It is seen that the operation of the emergency bur input pump introduces a relatively weak disturbance into the signals distorted by fluctuations, a change in power, and the associated non-stationary poisoning. Such information requires processing to avoid distortion;

Fig.7 - typical experimental information after special processing (normalization) of small perturbations with the elimination of distortions in the distribution of the neutron flux (energy release) and heating of the coolant in the active zone is presented by the example of alternate operation of the TW1 and TW3 pumps on loops 1 and 3. A comparison of the sensor signals according to three independent systems operating on different principles of operation - KV (based on the signals of the SCR - figure 2, 3), DT (based on the signals of thermocouples - figure 2, 3) and ionization chambers (figure 1). In this case, the KV and DT signals were summed over the spots (indicated in the figures as Spot_IKi, i = 1, 6, 12, 4, 8, 15), each containing 14 fuel assemblies located in the peripheral rows of the active zone near the corresponding ith ionization chamber ( figure 1). Fig. 7 consists of three figures composing it, in which changes in the normalized deviations (Dev_N, rel.%) Of the signals of the monitoring systems from their average values for the three monitoring systems are presented. A comparison of the three figures reveals a good qualitative and quantitative correspondence in spatial distributions and time dependence. Differences in behavior from time to time are characterized in particular by the inertia of a specific monitoring system;

Fig.8 is experimental information similar to Fig.7, but related to the alternate operation of the pumps TW2 and TW4 on loops 2 and 4 with similar conclusions;

Fig.9 - typical experimental normalized information is presented on the example of the alternate operation of the TW1-3 pumps. The comparison of sensor signals by two independent systems operating on different operating principles is presented - KV (the upper figure for entering the active zone is layer 1 (L1) and the middle figure for exiting the active zone is layer 16 (L16)) and temperature sensors after exiting reactor (bottom figure) in four hot threads (Hot Legj, j = 1, ..., 4), averaged over six azimuthal temperature sensors in each hot thread (figure 4). In this case, the KV signals were summed over the quadrants (Quadj, j = 1, ..., 4) of the active zone adjacent to their loops (Fig. 1). A comparison of the patterns reveals a good qualitative and quantitative correspondence in terms of spatial distributions and time dependence.

From a comparison of the upper and middle figures of Fig. 9, a noticeable decrease in unevenness is seen when moving from the lower layer L1 to the upper L16, which, apparently, indicates the erosion of the “boron” and “bezborny” spots due to the transverse inter-cassette turbulent exchange;

figure 10 - experimental information similar to figure 9, but related to the alternate operation of the pumps TW2-4 with similar conclusions;

11 - distribution in the core of the local mixing coefficients - the share of the flow rate of the coolant of each loop falling into each fuel assembly is shown in three figures that make up 11. The information was obtained on the basis of neutron signals (KV) for the steady-state distribution of mixing coefficients during the operation of the TW1 pump immediately before its shutdown. The distribution of these coefficients is smooth, and their normalized values vary from 0 to 2.3 (a total of 163). The “boron spot” (painted in white) consists of fuel assemblies into which the maximum proportion of boron from loop 1 falls. Accordingly, the “stain-free spot” (painted in black) consists of fuel assemblies in which the minimum proportion of boron falls. The angular displacement (swirl) of the coolant flow in front of the active zone during operation of the TW1 pump is approximately estimated by the displacement of the axis of the “boron spot” with respect to the axis of loop 1 (Fig. 1). The integral coefficients of inter-loop mixing, i.e. the distribution of the relative fractions of the loop consumption over four loops correspond to the sums of the corresponding quadrants of the core (Fig. 1) of the neutron signals of the SCR (KV) and the sums of the local mixing coefficients. The integral coefficients of inter-loop mixing are also consistent with the signals of thermal sensors in four hot threads (Figs. 9 and 10);

Fig. 12 is experimental information that differs from Fig. 11 in that it is obtained on the basis of the temperature signals of thermocouples DT. Fig.12 is in good agreement with Fig.11 according to the location of the "boron" and "bezborochnyh" spots in the active zone;

Fig.13 - differs from Fig.11 in that it relates to the operation of the TW3 pump immediately before it is turned off;

Fig.14 - differs from Fig.11 in that it relates to the operation of the TW2 pump immediately before it is turned off;

Fig. 15 - differs from Fig. 11 in that it relates to the operation of the TW4 pump immediately before it is turned off;

Fig.16 - differs from Fig.15 in that it is obtained on the basis of the temperature signals of thermocouples DT;

Fig - normalized signals of six azimuthally located along the perimeter of the MTC thermal sensors (TC15-20, see Fig.4) in the cold thread of loop 1 (ColdLegl) during alternate operation of the pumps TW1 and TW3 on loops 1 and 3. It is seen that the main part a cold coolant stream from the point of its injection into the MTC (pipe B2 in FIG. 4) gets to the ТС15 sensor located 1.8 m from the injection point and with a rotation of 120 ° ПЧС (if you look at the reactor). A small part of the cold stream hits the TC20 sensor, located 1.45 m from the injection point and with a 30 ° turn of the frequency converter. Even smaller shares fall on the sensors TC17, 18 (figure 4). Analyzing such parameters as injection speed and temperature, coolant speed and its temperature in the MTC, as well as sensor signals and their placement, we can conclude that in this section there is a translational-rotational stratified coolant flow with a rotation speed of at least 1 r / s ( PES if you look at the reactor). This is in favor of the existing hypothesis that there is a translational-rotational flow in the MTC caused by rotation in the same direction (PES) of the MCP blades;

Fig. 18 is the same as in Fig. 17, but for loop 3 (ColdLeg3). In contrast to loop 1 (Fig. 17), there is a greater deviation (amplitude) of the TC15 sensor signal. This can be explained by a lesser degree of local turbulization of the cold core falling on the TC15, which is also confirmed by smaller fractions (than in Fig. 17) falling on the sensors adjacent to it;

Fig.19 is the same as in Fig.18, but for loop 2 (ColdLeg2). Here there is an even greater deviation (amplitude) of the TC15 sensor signal and, accordingly, even smaller fractions of the cold stream falling on the sensors adjacent to it;

Fig.20 is the same as in Fig.17, but for loop 4 (ColdLeg4), with a degree of turbulent mixing close to Fig.17.

The implementation of the invention

One of the most dangerous design reactive accidents is a break in the steam line of a steam generator (GH). In this GHG, the pressure sharply decreases, and in the corresponding loop, the temperature sharply decreases. This leads to an increase in energy release in the core, especially in the sector adjacent to the emergency loop, and with aggravating additional failures (for example, failure of emergency protection), it can lead to severe damage to the core. The presence of mixing of the coolant in the area in front of the active zone and in subsequent sections before leaving the reactor eliminates such consequences. In one pass through the reactor circulation path, each of the loop flows exchanges a certain mass of coolant with neighboring loop flows due to turbulent mixing at their boundaries, but mainly due to angular displacement (swirl). The amount of coolant from each loop falling into it and into neighboring loops is characterized by the values of the coefficients of inter-loop mixing. The experiments on determining the mixing coefficients were carried out both on the model of the reactor on a scale of 1: 5, and during commissioning tests at the stages of hot run-in and power development at a number of operating units with VVER-1000 reactors, starting from Unit 5 of the Novo-Voronezh NPP in 1979, up to 1 unit of Bushehr NPP in 2011. Measurements at the existing power units were carried out in a known manner [1, 2].

The proposed method is devoid of the disadvantages inherent in the known methods and has a number of advantages over them. 04/26/2012, the first test test of the proposed method was carried out at Unit 1 of Bushehr NPP, which demonstrated its real feasibility and effectiveness.

The main embodiments of the invention:

1. In the coolant of the operating nuclear reactor, the emergency input system of boron at any power level, in one or more loops create a weak unevenness in the distribution of the indicator, which plays the role of a temperature indicator before entering the reactor, and in the core - a neutron-absorbing indicator, while they use the whole complex of standard neutron and temperature monitoring systems in all important parts of the circulation circuit, and to eliminate distortions and fluctuations in the recorded signals, they use a special processing of small disturbances in the signals of monitoring systems. The greatest capabilities of the method are possible when testing from 35 to 100% of rated power.

The pumps of the emergency bur input system are switched on and off sequentially with possible time delays until the distribution of the indicator is equalized (after the pump is turned off) or until the established uneven distribution of the indicator (after the pump is turned on).

To eliminate distortions and fluctuations in the signals recorded by monitoring systems, special processing of small perturbations in the distribution of the neutron flux (energy release) and heating of the coolant in the core are used. This processing is broadly summarized as follows:

- at each moment of time, the deviations of the signals of individual sensors from the average value over the entire array of sensors are calculated;

- arrays of received signals are averaged over extended time intervals, for example, dt = 35 s;

- special corrections are introduced to eliminate bias effects associated with power distortions and transient poisoning.

This application of the method itself is accompanied by a decrease in power during testing, which gives distortions due to the redistribution of energy release, including due to unsteady xenon poisoning.

The possibility of implementing the proposed method is confirmed by tests at Bushehr NPP, the main results of which are presented in figure 1-figure 20.

Figure 1 shows that the difference in the azimuthal arrangement of loops 1 and 3 from loops 2 and 4 leads to differences in the angular displacement of the coolant flows and mixing coefficients.

At Bushehr NPP, measurements were made using both methods, which facilitates their comparison.

The known method [1, 2] gives the coefficients of inter-loop mixing, ie the distribution of the relative fractions of the loop consumption in four loops (in the direction of the PPS) for loops 1 and 3 (the shares in the loops) of 64, 25, 7 and 4% and the twist angle of about 20 ° PPS. For loops 2 and 4, the old method gives fractions of loops 64, 27, 5, and 4% and a twist angle of about 0 °.

The new inventive method of estimating the mixing parameters based on the signals of the SCR (KV) for loops 1 and 3 gives the share of loops 40-41, 38-39, 13-14 and 7-8% and the swirl angle is 48-54 ° PS. For loops 2 and 4, the new method gives fractions in loops of 34-36, 41-42, 17-18 and 6% and a twist angle of 26-32 ° PSR. A new method for estimating mixing parameters based on thermocouple signals (heating) shows similar mixing parameters.

Thus, the new method gives approximately 30 ° greater angles of rotation of the PSD than the old method. The new method also gives 13-15 abs.% (Or 52-55 rel.%) Higher fractions of the coolant flow coming from each loop to the nearest (adjacent in the direction of the PPS) loop: 38-39 abs.% Against 25 abs .% (for loops 1 and 3) and 41-42 abs.% versus 27 abs.% (for loops 2 and 4) than the old method.

According to the new method, the local mixing coefficients are estimated — the fraction of the flow rate of each loop falling after each revolution of the coolant into each of the 163 fuel assemblies of the active zone. The distribution of these coefficients is smooth, and their normalized values vary from 0 to 2.2-2.4 (a total of 163).

2. The method according to paragraph 1, characterized in that simultaneously with the operation of the emergency input pumps of boron, pure condensate is introduced into all loops, while the flow of pure condensate is numerically selected and the reactor constant power is maintained.

This research scenario is applicable if pure condensate recharge is supplied through a common collector to all loops. By supplying pure condensate, it is only possible to approximately ensure a constant power, but this will reduce distortion and allow the indicator to be supplied longer by emergency input pumps (until the input branch pipes are completely filled with it). In this case, the stationarity of the initial state should be ensured, which will allow for less distortion in comparison with paragraph 1.

3. The method according to paragraph 2, characterized in that the pure condensate is introduced into one or more loops other than the loops (loops) into which the indicator is supplied.

This research scenario is applicable only in those projects where it is possible to feed pure condensate into separate loops. It is useful to maintain constant power (as in paragraph 2) and, at the same time, to enhance the uneven distribution of the indicator.

4. The method according to paragraphs 1, or 2, or 3, characterized in that, simultaneously with the operation of the pumps, an automatic power regulator is used and a constant reactor power is maintained.

This research scenario gives the least distortion, especially for the methods described in paragraphs 2 and 3.

The relatively low design capacity of the TW system creates weak perturbations in the distribution of energy release, which allows such a study to be maintained at a constant power, directly during normal operation (or during commissioning) at any stationary power level up to 100% Nnom.

Industrial applicability

It is most expedient to use the proposed solutions for use in nuclear power water-cooled thermal neutron reactors.

Literature

1. - Bezrukov Yu.A., Dragunov Yu.G., Logvinov S.A., Ulyanovsky V.N. A study of the mixing of coolant flows in the VVER reactor vessel. 13 AER Symposium on VVER Physics and Safety. September 22-26, 2003, Dresden, Germany.

2. - Yu.V. Saunin. Development of methods for complex testing of VVER internal reactor control systems. Abstract of dissertation for the degree of candidate of technical sciences. Mytishchi-Novovoronezh, 2010

3. - Engineering. Encyclopedia. Section IV Calculation and design of machines. Volume IV-25. Nuclear engineering. Book 2. Edited by E.O. Adamova, V.I. Solonina, K.S. Kolesnikova, Moscow, "Engineering", 2005.

Claims (1)

A method for the experimental study of mixing coolant in an existing nuclear reactor, which consists in creating an uneven distribution of the indicator in the reactor coolant, characterized in that the emergency input system of boron at any power level in one or more loops creates an uneven distribution of the indicator, which acts as a temperature indicator to the input into the reactor, and in the core — a neutron-absorbing indicator, using a set of standard neutron and tempo systems On-site monitoring: neutron detectors - direct-charge sensors, temperature sensors, and ionization chambers, or part of this complex, in sections of the circulation circuit: before entering the reactor and core, in the core and before leaving the reactor, and to eliminate distortions and fluctuations in the recorded signals apply the processing of small disturbances of the signals of monitoring systems.

Images