DE2603094A1 - METHOD FOR LOCATING DEFECTIVE FUEL ELEMENTS IN NUCLEAR REACTORS - Google Patents

Description

26.1.1 976 Kl / 24,290 .3 260 3094

INTERATOM

Internationale Atomreaktorbau GmbH 506 Bensberg

Method for the localization of defective fuel elements in nuclear reactors

-Z- 609836/0259

The invention relates to a method for the unambiguous localization of i-simultaneous occurring defects on fuel rods of reactor fuel assemblies, the rods of each fuel assembly having a characteristic Mixture of several n-different stable noble gas isotope pairs contain and during During the operation of the reactor, the gas plenura of the pressure vessel is continuously examined for the presence of these stable noble gas isotopes will. This method, known as the tag gas method, is used to locate defective fuel elements all fuel rods of an element by a certain isotope mixture of noble gases, the tag gas (tag for short), which in small amounts are introduced into the cracked gas space, individually marked. In the event of a defect in the fuel rod cladding In addition to the filling gas and the fission gases that have already formed, tag gas also enters the primary cooling circuit and further into the protective gas space (gas plenum), from where it is fed to a mass spectrometer, which shows the significant Measures isotope concentrations of the day gas of the defective fuel element.

The main advantage of tag systems over other leak detection systems is that they can be used at the same time Detection and localization of fuel element damage. One advantage, however, is the simultaneous occurrence is called into question by two or more leaks on different fuel assemblies, namely when the new one Tag mix, which is composed of the tags of the defective elements, no longer clearly identifies the type of error present and indicates the fuel assembly number.

So far, methods have been described that the

Allow optimal evaluation of the information content of the new tag mix, but only probability statements can be made about the type of error actually present. Another way is to construct the tags in such a way that

609836/0259

that in the case of two or more simultaneous leaks, different types of errors are largely unambiguous remain. This is primarily intended to mean the number of fuel assemblies that are leaking at the same time will. In this context, only methods are known which allow the tags to be constructed in such a way that that a clear distinction can be made between first-order and higher-order defects (several defects), However, it remains open which type of higher-order error is involved. With the known methods however, it is not possible to locate several defective fuel assemblies as clearly as possible.

The procedure described here permits

it is to design tag systems unambiguously with regard to higher-order error types. The maximum uniqueness of a tag system with regard to possible types of errors depends on the dimension of the underlying system. The dimension of a day system means the number of isotope pairs, each of which forms an isotope ratio for itself. The η dimensions of an n-dimensional tag system span the reduced isotope space. A certain tag gas composition is represented by a point (its position vector, in the following Course called isotope vector) shown in the reduced isotope space.

The method according to the invention is therefore

characterized in that in the n-dimensional isotope space the points characterizing each individual fuel element are selected in this way be that with each new point to be selected all linear combinations of i-points from the now existing Set of points match with as few linear combinations as possible of i other points from the same set of points.

The difference vectors between any two

Isotope vectors define the set of all relative vectors. A tag system should now be unique with regard to i leakages

609836/02 5 9

(i.e. that an i-th order error, clearly of must distinguish between all other error types of the i-th order), any 2i-1 relative vectors must be linearly independent be. This must apply to all possible combinations of relative vectors to the (2i-1) -th class. This can be done in Realize n-dimensional tag systems if 2i-1 = η is. The following overview is obtained

System uniquely into linearly independent relative vectors

with respect to i leaks dimension of the syst. η = 2i-1

i = 1 η = 1

2 3

3. 5

4 -_- .7

If, however, the number of dimensions of the underlying tag system for errors of the i-th order is less than 2i-1, then only a limited uniqueness are required and you get the overview

j dimensions less system clearly up to

than the required 2i-1

j = 1 finite number ^ of points

2 one-dimensional point manifold

3 two-dimensional point manifold

, 4 three-dimensional point manig

wrinkles

On the basis of these considerations, different things can be made depending on the dimensions of the tag system and the required degree of clarity formulate strict uniqueness conditions. In general * uniqueness conditions are requirements for linear independence of 1.2, .... maximum η relative vectors. The larger the class of linearly independent relative vectors, the more stringent the condition is. The uniqueness conditions can also be formulated in a geometrically clear manner (see Table 1).

'60983 6/0259 ι

- 5 table 1

systems

Conditions that can be demanded in the reduced icotope roughness /'.ir/.ahl dor
linearly independent
(lu) relative vectors

Ma; <iiiiaIor theorc— tiueher ϊηίίοπηίΐ-tion.shalt · of a system with the above requirements

1-dim.

Zv / ei points must not be identical.

1 leak can be clearly identified.

2-dim.

No third point is allowed on the route lie between 2 points. Tightened: may ^ a normal distance £ not fall below.

It can be clearly determined whether it is 1 or 2 Leaks. With 2 leaks the system is up to a finite number of Points (the possible intersection points between 2 points) clearly. St «ocNc.-v»

3-dim,

Two lines between 2 points must not cross each other. Tightened; must not fall below a normal distance £. • If there are 2 leaks, the system is clear. It is noticeable whether there are 1 or 2 leaks or 2 or 3 leaks. If there are 3 leaks, the system is down to »T-dim. Point infiltration (the possible intersections of two Δ-surfaces) clearly.

Ί-dim.

.-dim.

.

Distance between 2 points. May not Δ-area, pierce.

Δ surfaces must not intersect.

8 367,025,9

It is clear whether it is 1, 2 or 3 Error. ■ With 3 leaks the system is up to a finite number of Points (the O-dirn. Sections between the A-surfaces) · unambiguous.

3 leaks can be clearly identified will. ■ '

-SAD

The construction of unique tags in the reduced isotope space can be done in two ways:

1. In the reduced isotope space there is a rigid, largely asymmetrical one Point grid set. Each point corresponds to a possible isotope vector, i.e. a possible tag gas composition. Now, according to the trial and error principle, the cheapest corner (natural isotopic composition of the noble gases used) of the reduced isotope space only those isotope vectors are selected that meet the uniqueness conditions.

2. The individual isotope vectors become neighboring vectors, taking into account the uniqueness conditions and the minimum distance successively developed from the cheapest corner of the reduced isotope space.

Construction of the point grid and definition of the cheapest corner

The individual components are in an n-dimensional tag system of the isotope vectors by the isotope ratios of the sequences

609836/02 59

260309A

r / r

;, l't

given. By an n-tuple of values

an isotope vector is completely determined in the reduced isotope space. All representable isotope ratios at all of a dimension lie in an interval whose upper and lower bound by the concentrations in the "representational Components are given. For example

_ώ τ

A, B ... isotopes ν-, ν «.. isotope volumes

c ·. .... Concentration of the isotope j in the representing component for the isotope i.

If the various significant isotope ratios i, '' ^ '/ ^ ^{1 are} always defined in such a way that the representative component of the isotope Λ. Cheaper than that of the isotope B., then that of 609 8.36 / 02 corresponds to 59

defined isotope vector of the cheapest corner. Used as a performing. Component of A. the technically pure noble gas with
If its natural isotope ratios are selected, the initial isotope ratios result from the natural isotope ratios (I..) and a safety margin f

> Xj Πα U '

S X

In this case, the natural corner of the reduced
Isotope space to the cheapest.

The development of the isotope vectors from the cheapest corner of the reduced isotope space is a useful method for cost optimization d <xc. - <■■ --—'--- ··. · -

609 8 36 / .02 69

Claims (1)

January 26, 1976 24.290.3 Patent claims: 1. Procedure for clear localization of i-simultaneous defects in fuel rods of reactor fuel assemblies, the rods of each fuel assembly a characteristic mixture of η different * stable noble gas isotope pairs included, and during the reactor operation the gas plenum of the pressure vessel continuously after the presence of this stable Noble gas isotopes is examined, characterized in that each individual fuel element in the n-dimensional isotope space characterizing points are chosen so that with each new point to be selected all linear combinations of i points from the existing point set with as few linear combinations of i other points as possible match from the same number of points. 2. The method according to claim 1, characterized in that that the localization of i simultaneously occurring leaks by the linear independence of all possible Combinations of difference vectors of any two isotope vectors (relative vectors) to the (2i-1) th class is achieved. 3. The method according to claim 1 _f, characterized in that a η = 21-1 dimensional tag system by the linear independence of all possible combinations of its relative vectors for. (2i-1) th class, the clear localization of i simultaneously occurring leaks is achieved. ^{4. The} method according to claim 1, characterized in that an n = 2i-2 dimensional tag system due to the linear independence of all possible combinations of its relative vectors to the (2i-2) -th class, which except for a final 609836 / 02S9 borrowed number of points in isotope space unique localization achieved by i simultaneously occurring leaks. 5. The method according to claim 1, characterized in that a η = 2i-3 dimensional tag system by the linear independence of all possible combinations of its relative vectors to the (2i-3) -th class, except for a one-dimensional point manifold in the isotope space unambiguous localization of i simultaneously occurring leaks achieved. 6. The method according to claim 1, characterized in that a η = 2i-4 dimensional tag system by the linear independence of all possible combinations of its relative vectors to the (2i-4) -th class, except for a two-dimensional Diversity of points in the isotope space unique localization of i simultaneously occurring leaks achieved. 7. The method according to claim 1, characterized in that that a η = 2i-5 dimensional tag system through the linear independence of all possible combinations of its Relative vectors to the (2i-5) th class, which are unique except for a three-dimensional point manifold in the isotope space Localization of i simultaneously occurring leaks achieved. 6098 3 6/0259