WO1997012375A1

WO1997012375A1 - Method and device for supervision with respect to the stability of a boiling water reactor

Info

Publication number: WO1997012375A1
Application number: PCT/SE1996/001209
Authority: WO
Inventors: Jacek Bujak; Juan Jaliff; Camilla Rotander; Marek Stepniewski
Original assignee: Abb Atom Ab
Priority date: 1995-09-27
Filing date: 1996-09-27
Publication date: 1997-04-03
Also published as: SE504945C2; SE9503333L; SE9503333D0

Abstract

Supervision with respect to the stability of a reactor core in a boiling water reactor plant. In a first calculating means (17), the internal state of the reactor core is determined on the basis of known operating data from the reactor plant. In dependence on sensed operating data and the calculated internal state, the behaviour of the reactor core for a number of preselected events affecting the stability of the core is continuously simulated in a simulation means (19). In a second calculating means (20), the stability of the reactor core is determined, based on the simulated behaviour of the reactor plant, and an alarm signal is generated if a predetermined stability criterion is not fulfilled.

Description

Method and device for supervision with respect to the stability of a boiling water reactor

TECHNICAL FIELD

The present invention relates to a method and a device for supervision during operation of the stability in a boiling water reactor.

BACKGROUND ART

A core in a boiling water nuclear reactor comprises a plura¬ lity of parallel fuel assemblies. These are arranged verti¬ cally in the core at such a distance from each other that a self-sustaining nuclear reaction can be obtained. The core is enclosed in a reactor vessel and immersed into water which serves both as coolant and as neutron moderator. Each one of the fuel assemblies comprises a cooling channel through which water flows and is partially evaporated as it absorbs heat from the reactor fuel. The cooling channels will thus contain a steam water mixture with a steam content which increases towards the outlet. A plurality of control rods, which com¬ prise a neutron-absorbing substance, may be inserted between the fuel assemblies and control the reactivity of the core and hence the output power of the reactor. For measuring the power, a plurality of detectors measuring the neutron flux density, so-called PRM (Local Power Range Monitor) detectors, are usually arranged both radially and axially in the core. The output signals from these neutron flux detectors are utilized for supervision of the core and initiate actions, in the worst case reactor scram, in the event that something abnormal is detected.

A well-known problem, which in certain operating positions and in certain operation situations may arise in a boiling water reactor, is that the core becomes unstable, so-called thermo- hydraulic instability. When cooling water is pumped upwards through the cooling channels, bubbles with steam are formed. The steam bubbles are all the time in motion, which results in variations in the relationship between water and steam in the cooling channels. If these variations in the water flow are not damped in a natural way, for example by friction in the cooling channels, the channel inlets or external circuits, they may grow into continuous oscillations. Since water is better moderator than steam, these flux oscillations will also result in power oscillations.

A reactor core is unstable when its power oscillates in an undamped manner. The power oscillation may either be global, that is, the power in the entire core oscillates, or local, that is, the power in one or a few cooling channels oscillates. Oscillations in power and coolant flow may result in the fixed margins for the fuel being exceeded, which in turn may lead to fuel damage. An instability in the reactor core must therefore be prevented.

The stability properties for each individual boiling water reactor depend partly on the design of the reactor and partly on the operating point of the core with respect to power and coolant flow, the current burn-up states and the current power distribution. In addition, the stability properties are influenced by wear in the control system of the reactor, associated pumps, actuators, valves, etc. The margins with respect to thermohydraulic instability decrease with increasing power and decreasing coolant flow and, in addition thereto, depend on the distribution of the power in the cooling channels. From this follows that the risk of instability is greatest when the reactor is run in an operating range which is characterized by high power and low coolant flow. The operating range where it is possible that instability may occur will hereinafter be referred to as the potential non-stable region of the reactor. In certain cases, deteriorated stability may arise as a result of the occurrence of a fault in the reactor, for example due to malfunctioning of some component.

In practice, of course, it is not desirable to experience stability problems during operation and attempts are therefore made to avoid running within the potential non-stable region. To avoid arriving in the potential non-stable region, it is investigated beforehand which operating states and operating situations may be expected to have acceptable stability margins and which other states and situations should not be allowed for normal operation. These investigations take place by means of calculations in advance utilizing dynamic models of the whole reactor system. However, there is a considerable risk that the allowed operating ranges become unnecessarily narrow, which leads to the reactor not being utilized to its optimu .

An additional disadvantage is that, in spite of the fact that acceptable stability margins have been considered to exist during the preliminary calculations, there is a risk of an unexpected instability due to an event which has not been possible to envisage. It is difficult to predict in advance which are the most unfavourable operating conditions during a following operating period. Unforeseen events may occur which influence the stability and which have not been included in the calculations. An example of such a relatively common event which is difficult to foresee is that a control rod during operation adheres to the core (a so-called stuck rod) . Such an event initiates a change of the other control rod positions, which in turn influences the power distribution in the core and may hence change the margin of the core with respect to stability. The stability may be influenced both in a positive and in a negative way, depending on which of the control rods becomes stuck and in which position it becomes stuck. To overcome the above-mentioned disadvantages, methods have been developed for detecting as early as possible whether the core is unstable such that measures may be taken to cause it to become stable again before any damage occurs. US 5,174,946 describes equipment which senses whether the power in any part of the reactor core oscillates by detecting early oscillations in the output signals from the neutron flux detectors . hen such an oscillation has been detected, a reactor operator is alerted who may take appropriate measures to make the core stable again. Examples of measures the operator may take is to change the coolant flow or the control rod positions.

One problem is that the power oscillations may grow very rapidly. In the worst case, it may take only a few minutes from the point where the output signal from a neutron flux detector starts oscillating until the entire core is oscilla¬ ting. The risk is then great that the only way to stop the oscillation before it causes serious damage is to release a reactor scram. A reactor scram causes considerable inconven- ience because of power loss and always entails heavy costs. A scram released due to an unstable core also gives bad publi¬ city to the reactor and is, in itself, a risk factor when the safety equipment is put to the test.

The problem with a supervision of the stability based on the principle of detecting oscillations in the output signals from the neutron flux detectors is that the core is then already unstable and that it may then be too late to initiate measures to make the core stable again. A reactor scram may then be the only remaining solution.

SUMMARY OF THE INVENTION

The object of the invention is to suggest a method and a device which, as early as possible during operation, give the reactor operator an indication that the core runs the risk of becoming unstable, such that the operator may take appropriate measures to prevent the occurrence of instability.

hat characterizes a method and a device according to the invention will become clear from the appended claims.

When the core is already unstable, that is, when the power in the core has started oscillating, it may be too late to take measures to make the core stable again. The invention is based on the idea not only to detect, continuously during operation, whether the reactor is unstable but also to predict if there is a risk of the reactor becoming unstable. If there is a risk of the reactor becoming unstable, an alarm signal is generated. The prediction takes place on the basis of current data about the power distribution of the core, the burnup of the fuel, the coolant flow, the control rod positions, etc.

When the reactor is run within the potential non-stable region, several different factors come into play whether the reactor becomes unstable or not. It has been observed, for example, that the stability of the core is dependent on the axial and the radial power distribution in the core. A bottom- peaked power distribution, that is, the power is highest at the bottom of the core, has a negative influence on the stability of the reactor. From the point of view of stability, the reactor should instead have a top-peaked power distri¬ bution.

The power distribution is influenced by the coolant flow and is more bottom-peaked at a low coolant flow. When the coolant flow increases, the steam contents decrease and the power increases successively towards the top. Also the control rods influence the power distribution. Since the control rods contain neutron-absorbing material, the power generation decreases in that part of the core where the control rods are inserted. In a boiling water reactor the control rods are adapted to be inserted into the core from below. If the con¬ trol rods are inserted a certain distance into the reactor core, a top displacement of the power distribution is achieved and the core thus becomes more stable.

During operation, the positions of the control rods may need to be changed for various reasons. This change may displace the power distribution in the core towards a more bottom- peaked power distribution, which entails a deterioration of the stability. If the reactor is run within the potential non-stable region, a minor change of the control rod position may cause instability. In the same way, a minor change of the coolant flow may cause instability.

Certain changes do not influence the stability immediately but there is a certain delay before the result of the change is shown. For example, a change of the xenon content may give a delay of a number of hours before its influence on the stabi¬ lity manifests itself in full. Such an event is well suited for predicting.

During normal operation, the reactor is outside the potential non-stable region. Certain unpredictable events may cause a reactor which is outside the potential non-stable region to arrive inside the potential non-stable region. Examples of such events are that some of the circulation pumps stops whereby the coolant flow decreases, a valve is stuck in a certain position, or a preheater to the feedwater stops functioning. When the reactor has arrived within the potential non-stable region, it may be sufficient with a small change of, for example, the control rod positions for the reactor to become unstable.

The prediction means that with knowledge of the current state of the core, such as the current power distribution, the burn¬ up of the fuel, the current control rod positions, and the current coolant flow, the behaviour of the reactor core is simulated for a number of preselected events which influence the stability of the core. Based on the simulated behaviour of the reactor plant, a stability determination of the reactor core is performed and an alarm signal is generated if a pre¬ determined stability criterion is not fulfilled.

The advantages of the invention is that it gives an early indication that the core is becoming unstable. The fact that all the time the current state of the core is the basis of the prediction gives a high accuracy in the prediction. Unforeseen events which have occurred earlier during the operating cycle, such as one or more control rods having become stuck, are included in the prediction. The invention can be implemented as part of the core supervision system which is already arranged in most boiling water reactors. The invention may use the same computer and the same presentation system as the core supervision system, which means that both costs for installa¬ tion of the invention and space are saved.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a schematic picture of a boiling water reactor comprising a core supervision system.

Figure 2 is a block diagram of a core supervision system comprising a stability supervision according to the invention.

Figure 3 shows in the form of a flow diagram the principle for carrying out the stability prediction.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

Figure 1 shows an example of a boiling water reactor compri¬ sing a core supervision system. The core 1 of the reactor contains fuel in the form of fuel rods between which cooung water is pumped. The core is surrounded by a pressure- maintaining reactor vessel 2. The heat development in the fuel rods cause the cooling water to boil and steam to be formed. The power of the reactor is controlled with control rods 3 and with the circulation pumps 4 which pass the cooling water upwards through the core. The produced steam is delivered via steam pipes 5 to the turbine 6 which drives the generator 7 where electrical energy is generated, whereupon the steam is condensed to water in the condenser 8. The water is returned to the reactor vessel 2 via a feedwater pipe 9 with the aid of the feedwater pump 10.

For supervision of the reactor, a number of operating data are sensed, utilizing special measuring equipment. Examples of such sensed operating data are the feedwater flow, the coolant flow, the feedwater temperature, the control rod positions, the pressure in the reactor vessel, and the power of the reactor. The physical input signals to the measuring equipment are converted therein into electrical output signals which are forwarded as input signals to a core supervision system 11. Based on these input signals, a number of quantities of importance for the supervision of the core are calculated in the core supervision system. These quantities are presented to a reactor operator 12 who judges whether measures must be taken.

Figure 2 is a block diagram of a core supervision system 11. The core supervision system 11 comprises a stability detector unit 13 which detects whether the core is unstable, a stabi¬ lity prediction unit 14 which continuously predicts whether the core will become unstable, and a supervision unit 15 which calculates the other quantities which are necessary for the supervision of the core. The output signals from the different units are compiled and presented to the operator in a common presentation unit 16. A stability parameter reflects the degree of instability of the core. The stability parameter most widely used today is the decay ratio for the power. The decay ratio is defined as the ratio between two consecutive amplitude maxima in the response to a disturbance. The decay ratio 1 characterizes undamped oscillations with a constant amplitude and indicates the limit to instability. For damped oscillations, the decay ratio is less than 1, and the difference between 1 and the calculated decay ratio is a good measure of the stability margin. The greater the difference, the more stable the core. In an unstable system, oscillations occur which grow exponentially with time. The decay ratio then exceeds the value 1.

To determine whether the core runs the risk of becoming un¬ stable, knowledge of the current power distribution in the core is needed. To this end, a three-dimensional core calculator 17 is used which comprises a mathematical model of the core and which, with the aid of measured values of para¬ meters included such as the total coolant flow, the control rod positions, the total power of the reactor, etc., is able to calculate the current power distribution in the core. For the core calculator 17 to be able to perform the necessary calculations, access to detailed information about the fuel and the core, for example material and geometry, is needed. To be able to predict how the core reacts to changes in the input signals, it is also necessary to know the history of the core, for example where the core is in the operating cycle. All of these necessary data have been collected in a data library 18. The core history is updated continuously with information from the core calculator 17. information from the core calculator and the data library is also used by the supervision unit 15.

The stability prediction unit 14 comprises, in addition to the core calculator 17 and the data library 18, a dynamic core simulator 19 and a stability-determining device 20. The core simulator 19 comprises a mathematical model of the core and simulates the behaviour of the reactor, that is, how the power, the power distribution and the coolant flow are changed with time in the core. The stability of the reactor core is analyzed for a number of preselected events which may cause the reactor to arrive within the potential non-stable region. Examples of such events are that any of the circulation pumps stops or that any of the preheaters to the feedwater stops functioning.

The core simulator (CS) and the stability-determining device (SD) are suitably implemented in the form of a computer program. The configuration of such a program is shown in Figure 3. In block 31, information about the current state (Di) of the core is obtained:

current measured values of process parameters, for example coolant flow and control rod positions, - a current power distribution from the core c αculator, current information from the data library, for example the burnup of the fuel, the current isotope composition.

Then, the first of the preselected events is simulated (n=l), for example that a preheater has dropped out, resulting in reduction of the temperature of the feedwater, block 32. When the simulation has reached to a new stable operating point, the current state of the core at the new operating point (Pi) is calculated, block 33. Then, the behaviour of the reactor is simulated at the new operating point during a disturbance in some of the process parameters, for example coolant flow or the control rod positions, block 34. The output signal from the simulation may, for example, be the power at a number of points in the core. In block 35, the decay ratio is calcula- ted, that is, the ratio between two consecutive amplitude maxima in the output signal from the simulation or the pre- dominant eigenvalue of the system matrix. If the decay ratio exceeds 0.8, an alarm signal is generated, block 36, otherwise the procedure for the next preselected event is repeated (n=n+l) , which may be, for example, malfunctioning of one of the circulation pumps.

How often the stability prediction may be carried out is limited by the capacity of the computer on which it is imple¬ mented. It may be suitable, during reactor operation, to repeat the prediction once every 24 hours and, in addition, each time any change in the operation has taken place.

The core supervision system 11 also comprises a possibility for the operator to simulate contemplated reactor operations and to predict whether they lead to instability, for example a change of the xenon content or a change of the control rod positions. If the operator, for example, plans a change of the positions of the control rods, he can read the new control rod positions, via unit 21, into the stability prediction unit 14 and also start up a stability prediction. Based on the current state of the reactor and the new control rod positions, the prediction takes place in accordance with the block diagram in Figure 3. In the core calculator 17 a new power distribution is calculated. Instead of current control rod positions, the core simulator 19 uses the new control rod positions during the simulation. In this way, reactor operations which lead to an increased risk of instability may be avoided.

The stability detector unit 13 comprises a computer program which substantially consists of an algorithm for time series analysis of the output signals from the neutron flux detectors and calculation of the decay ratio. Input data to the stabi¬ lity detector unit consist of a subset of input data to the core supervision system. In this embodiment, the stability detector unit 13 comprises part of the core supervision system 11 and uses the same computer and the same presentation system as the rest of the core supervision system. One of the auvαu- tages of both the stability detector unit and the stability prediction unit being included in the core supervision system is that the predicted stability parameters can be verified against the actual stability parameters. The verification takes place on-line.

One advantage with this embodiment of the invention is that it can be integrated with an already existing core supervision system and utilize the computer, the presentation system and the data library of the core supervision system. In another embodiment, of course, a stability analyzer according to the invention can be implemented in a freestanding computer and with presentation equipment of its own.

Claims

1. A method for supervision with respect to the stability of a reactor core in a boiling water reactor plant, wherein - operating data in the form of values of a number of opera¬ ting variables from the reactor plant are continuously sensed, characterized in that, continuously during operation of the reactor,

- the internal state of the reactor core is calculated in dependence on sensed operating data,

- the behaviour of the reactor core for a number of preselec¬ ted events influencing the stability of the core is simulated, in dependence on sensed operating data and the calculated internal state, - a stability determination of the reactor core is carried out based on the simulated behaviour of the reactor plant,

- an alarm signal is generated if a predetermined stability criterion is not fulfilled.

2. A method according to claim 1, characterized in that, based on the simulated behaviour of the reactor plant, the decay ratio of the core is calculated and an alarm signal is generated if the decay ratio exceeds a certain predetermined value.

3. A method according to claim 2, characterized in that the decay ratio of the core is calculated as the ratio between two consecutive amplitude maxima of any of the operating variables in the response to a disturbance.

4. A method according to claim 2, characterized in that the decay ratio of the core is calculated as the predominant eigenvalue of the system matrix.

5. A method according to any of the preceding claims, characterized in that in case of a contemplated change of one or more operating variables, the internal state of the reactor core is calculated, the behaviour of the reactor core is simulated, and a stability determination is carried out in dependence on the contemplated changes.

6. A device for supervision with respect to the stability of a reactor core in a boiling water reactor plant, characterized in that it comprises

- first calculating means (17) for determination of the internal state of the reactor core on the basis of sensed operating data in the form of values of a number of operating variables from the reactor plant,

- simulation means (19) which, depending on the calculated internal state of the reactor and sensed operating data, continuously simulates the behaviour of the reactor core for a number of preselected events influencing the stability of the core,

- second calculating means (20) which, on the basis of the simulated behaviour of the reactor plant, performs a stabili- ty determination of the reactor core and which generates an alarm signal if a predetermined stability criterion is not fulfilled.

7. A device according to claim 6, characterized in that it comprises a stability monitor (13) which detects instability based on oscillations in output signals from a number of neutron flux detectors arranged in the core and which delivers an alarm signal if a predetermined stability criterion is not fulfilled.

8. A device according to claim 6 or 7 , characterized in that it comprises means (21) for initiating a stability determination, in case of a contemplated change in one or more operating variables, based on the simulated behaviour of the reactor plant in dependence on the contemplated changes.

9. A device according to any of the preceding claims, characterized in that it comprises a data library (18) con¬ taining information about the fuel, the core and the history of the core.