WO2012123669A1

WO2012123669A1 - Method and device for the dosimetry of doses of ionizing and non-ionizing radiation

Info

Publication number: WO2012123669A1
Application number: PCT/FR2012/050509
Authority: WO
Inventors: Olivier GILARD; Gianandrea QUADRI; Mathieu BOUTILLIER
Original assignee: Centre National D'etudes Spatiales (Cnes)
Priority date: 2011-03-11
Filing date: 2012-03-09
Publication date: 2012-09-20
Also published as: FR2972538B1; FR2972538A1

Abstract

The invention relates to a method and dosimeter for measuring various values representative of doses of radiation with which an environment is irradiated. The dosimeter includes at least two mixed detectors (11i), each of which outputs at least one electrical signal depending on the dose of ionizing radiation and the dose of non-ionizing radiation received thereby. At least one value DI_sol , representative of a does of ionizing radiation, and at least one value DNI_sol , representative of a dose of non-ionizing radiation, are calculated from electrical signals output by the various detectors by means of the nonlinear modeling of said electrical signals on the basis of said values, said modeling being previously established and parameterized for each detector.

Description

METHOD AND DEVICE FOR DOSIMETRY OF IONIZING AND NON-IONIZING DOSES OF RADIATIONS

The invention relates to a dosimetry method and device for measuring different representative values of ionizing doses and non-ionizing doses of radiation irradiating an environment.

Many solutions have been envisaged in the past to allow the dosimetry of the radiations, that is to say the fluxes of charged particles with high energy (electrons, protons, alpha particles, ...) or electromagnetic radiations of length d wave weaker than the visible range (gamma rays, X-rays, ultraviolet, ...). These radiations are indeed harmful for living beings and / or electronic components. It is therefore useful to be able to perform a dosimetric measurement, for example on board space systems, in aircraft, in nuclear environments (power plants, submarines, ...) ...

One of the problems posed in this context is to be able to perform a measurement that is both representative of all the different potentially harmful radiation, but also allows to discriminate different natures of effects (ionizing and non-ionizing) of radiation. Simultaneously, this measurement must be able to be carried out with simple, robust, transportable equipment, and producing reliable results.

To this end, the natural approach envisaged up to now consists in associating a plurality of detectors, each detector being dedicated to the measurement of one or other of the types of radiation. Nevertheless, besides the detectors dedicated to the measurement of a single radiation are rare and particularly complex in practice, for certain types of radiation, there is no specific detector limited to the measurement of these radiations. For example, FR 2039186 describes a thermoluminescence dosimeter comprising a plurality of different detectors, each detector being sensitive to one or more radiations, the different detectors sensitive to the same common radiation having the same sensitivity, so that the individual measurements of each of the radiation are assumed to be obtainable either by direct reading of a measurement performed by a detector, or by difference between the signals delivered by different detectors. For example, this document estimates that it is possible to obtain a measurement of the neutrons by difference between the signals produced by two detectors of the same sensitivity, one being sensitive to gamma rays, the other being sensitive to both the rays. gamma and neutrons.

The inventors have nevertheless determined that such a method does not give good results in practice. Indeed, firstly, the manufacture of sensors of the same sensitivity is in practice almost impossible. Moreover, even with the same sensitivity, it turns out that the indirect measurement by difference between the signals produced by different detectors is not correct.

In the same sense, US 2010/0102249 discloses a luminescent solid state dosimeter for measuring neutron doses and gamma radiation doses individually by comparing signals from a detector / converter providing a measurement of the neutron and fast gamma rays, by a detector / converter providing a measurement of the dose of neutrons and moderate gamma rays, and by a detector / converter providing a measurement of the dose of gamma rays. According to this document, the doses of fast neutrons and moderate neutrons are obtained by subtracting the dose of gamma rays. But, again, this hypothesis is contradicted by the practical results. Many other documents suggest the same method, erroneous in practice, of comparison or subtraction discrimination (US 6140651, "Silicon Semiconductor Detectors for Various Nuclear Radiations" H. Kitaguchi et al, IEEE Transactions on Nuclear Science, Vol 43, No. 3 June 1996, ...).

The invention therefore aims to solve this general problem of reliably obtaining the measurement of different representative values of radiation doses from the same dosimeter placed in an irradiated environment.

The invention also aims to propose a solution which simultaneously is simple, can be implemented with a low-volume and low-mass dosimeter, at low cost, with low power consumption, capable of covering a wide range of radiations and radiation. energy of radiation. More particularly, the invention aims to provide a solution that can be reliably applied to space systems in low or medium or geostationary orbit, or other space missions, on board aircraft or in nuclear terrestrial environments.

The invention also aims to propose a solution that allows, at least in certain environments, to obtain in a simple manner an evaluation of the differential energy fluence spectrum of radiation.

Throughout the text, we designate by:

- "ionizing dose" of radiation, the DI value defined by the following relation:

where LET expressed in MeV.cm ² / g is the energy loss in ionizing form (electronic stopping power), E is the energy of the incident radiation, and Φ is the fluence of the radiation expressed in number of particles per centimeter square. LET (E) depends on the nature and energy E of the incident particles forming the radiation as well as the target material receiving the radiation (see "Space Radiation Environment and its Effects on Spacecraft Components and Systems Conf." (SREC04), Cepadues Editions, 2004, in particular pages 83-104; 145-148; 175-183). The ionizing dose is usually expressed in rad (1 J / kg = 1 Gy = 100 rad); in this case, we can then use the following formula:

DI = 1.6. 10 ^"11 LET (E) Φ (krad) when LET is expressed in MeV.cm ² / g

- "non-ionizing dose" (or dose of deterioration by displacement) of radiations, the value DNI defined by the following relation:

DM = NIEL (E) Φ (MeV / g)

where NIEL is the loss of energy in non-ionizing form expressed in MeV.cm ² / g, E is the energy of the incident radiation, and c is the fluence of the radiations in number of particles per square centimeter. NIEL (E) depends on the nature, the energy E of the incident particles forming the radiation, as well as the target material receiving the radiation (see "Space Radiation Environment and its Effects on Spacecraft Components and Systems Conf." (SREC04), Cepadues Editions, 2004, in particular pages 83-104; 145-148; 175-183), - "modeling" of a parameter according to at least one variable, any mathematical expression expressing this parameter as a function of said variable (s).

The invention therefore relates to a method for measuring different representative values of doses of radiation irradiating an environment, in which a dosimeter is placed in said environment comprising a plurality of radiation detectors each delivering at least one electrical signal depending on ionizing doses and / or non-ionizing radiation it receives, the number of detectors being greater than or equal to the number of different doses to be measured, characterized in that a dosimeter comprising at least two detectors, said mixed detectors, each delivering at least one electrical signal dependent on an ionizing dose and a non-ionizing dose of radiation it receives, at least two mixed detectors having properties of variation of the signals they deliver according to said doses which are different from a detector to the other, and in that one calculates at least one value DI _so i representative of an ionizing dose te, and at least one DNI value _so i representative of a non-ionizing dose, from the electrical signals delivered by the different detectors, from a nonlinear modeling of these electrical signals as a function of said values DI _so i DNI _so i, this modeling being previously established and parameterized for each detector.

The invention also extends to a dosimeter allowing the implementation of a method according to the invention.

The invention therefore also relates to a measuring dosimeter of different values representative of doses of radiation radiating an environment, capable of delivering a signal representative of radiation that it receives, comprising a plurality of radiation detectors each delivering at least one dependent electrical signal. ionizing doses and / or non-ionizing radiation it receives, the number of detectors being greater than or equal to the number of different doses to be measured, characterized in that it comprises:

at least two detectors, called mixed detectors, each delivering at least one electrical signal dependent on an ionizing dose and a non-ionizing dose of radiation which it receives, at least two mixed detectors having properties of variation of the signals which they deliver according to said doses that are different from one detector to another,

means for memorizing, for each detector, parameters of a non-linear modeling of the electrical signals delivered by the detector as a function of the values of said ionizing dose and of said non-ionizing dose of radiation received by this detector,

- means for calculating at least one value DI _so i representative of an ionizing dose, and at least one value DNI _so i representative of a non-ionizing dose, from the electrical signals from the different detectors, from a non-linear modeling of these electrical signals as a function of said DI _soh values DNI _soh this modeling being previously established and parameterized for each detector from parameters stored in the storage means.

Thus, contrary to the state of the art in which it is recommended to use separate specific detectors but of the same sensitivities vis-à-vis the same category of radiation, and to deduce a value of dose by difference (this which constitutes a linear modeling), the invention consists on the contrary in using at least two mixed detectors presenting different properties of variation of the signals which they deliver as a function of the ionizing dose and the non-ionizing dose, and a nonlinear modeling to obtain at least one DI value _so representative of the ionizing dose, and at least one DNI value _so i representative of the non-ionizing dose. The inventors have found that this method makes it possible to obtain reliable measurements.

Advantageously and according to the invention, using a dosimeter comprising a plurality of mixed detectors of the same nature but having characteristics of variations of the electrical signals they provide as a function of said received doses which are different. Indeed, nonlinear modeling established for each detector makes it possible to obtain an equation connecting the electric signal (s) delivered by this detector to the values of the ionizing dose and the non-ionizing dose sought. Since the mixed detectors have different characteristics of variations, the equation corresponding to said nonlinear modeling will be different (ie expressed with different parameter values) from one detector to another, and the differences between these different modeling equations are important enough to extract the desired values from the different equations. As such, the inventors have in particular noted with surprise and contrary to what could be thought and sought in the state of the art, that two mixed detectors, even if they are of the same nature, or even from the same manufacturer , in practice have very different response characteristics which, in the context of the invention, is in fact an advantage. It is usually sufficient to use mixed detectors of different technical specifications - in particular different commercial references and / or from different manufacturers. The fact that the variation properties of the signals delivered by the mixed detectors are different from one mixed detector to the other cause that the curves of variation of these signals as a function of the ionizing dose and the non-ionizing dose for these mixed detectors. are intersecting in at least one point of intersection.

In a method according to the invention, using a dosimeter comprising a number of mixed detectors sufficient to allow obtaining the different values sought from said nonlinear modeling. In particular, in particular to obtain exclusively the two values DI _soh DNI _so i, advantageously and according to the invention is used a dosimeter consisting of a number of mixed detectors between 3 and 20, in particular of the order of 5 to 10. The fact of using a number of mixed detectors at least equal to 3, these mixed detectors both having properties of different variations makes it possible to ensure that a single solution exists for obtaining the values DI _soh DNI _so i, from said nonlinear modeling. In addition, a larger number of mixed detectors makes it possible to improve the accuracy of the value obtained, and this for a wider range of variation in doses (ionizing and non-ionizing). Indeed, some detectors are more sensitive for low doses, while others are more sensitive for high doses. Accordingly, it is advantageous to use a plurality of mixed detectors to cover a wider range of dose variation.

In principle, any type of mixed detector can be used in a method according to the invention. Nevertheless, advantageously and according to the invention, a dosimeter is used whose detectors comprise solid-state sensitive elements, in particular solid elements formed of electronic components. Indeed, there are many types of mixed detectors having a solid state sensitive element, in particular silicon substrate or gallium arsenide. These detectors are simple, robust, reliable, compact, light, and can be integrated within the same circuit or electronic assembly, or even within the same integrated circuit.

Advantageously and according to the invention, using a dosimeter comprising a plurality of mixed detectors chosen from optoelectronic components: optocouplers, photodiodes, phototransistors, CCD sensors, APS sensors; Bipolar electronic components and BiCMOS electronic components. In a particularly advantageous preferred embodiment, and according to the invention, a dosimeter comprising a plurality of optocouplers as mixed detectors is used. Optocouplers are indeed particularly easy to use and characterize. They do not require additional light source.

In addition, advantageously and according to the invention, nothing prevents the use of a dosimeter further comprising at least one sensitive detector mainly or only at a radiation ionizing dose and / or at least one sensitive detector mainly or only at a dose non-ionizing radiation. As such, it is possible to use a dosimeter further comprising at least one detector which is mainly sensitive to the ionizing dose of the radiation it receives, and substantially insensitive to the non-ionizing dose of the radiation it receives, chosen from among the MOS electronic components. and optical fibers. And one can also use a dosimeter further comprising at least one detector mainly sensitive to the non-ionizing dose of the radiation it receives, and substantially insensitive to the ionizing dose of the radiation it receives, selected from laser diodes, light-emitting diodes, PIN diodes and photovoltaic cells.

In a method and a dosimeter according to the invention, different non-linear modelings can be used. Advantageously and according to the invention, said nonlinear modeling is a model comprising a polynomial function of degree greater than or equal to 2, that is to say a model expressing the values of the signals delivered by each mixed detector according to said desired values DI _soh DNI _so i of the ionizing dose and the non-ionizing dose, this modeling incorporating at least one polynomial function of degree greater than or equal to 2. Advantageously and according to the invention, said nonlinear modeling comprises a polynomial function of degree 2 .

In addition, advantageously and according to the invention, said nonlinear modeling is a model expressing the values of the signals delivered by each mixed detector, and comprises a continuous monotone function of said DI _soh DNI values _so sought, in particular of a logarithmic function - preferably the decimal logarithm of each of these values.

Moreover, in a method and a dosimeter according to the invention it is possible to calculate different representative values of the ionizing dose and the non-ionizing dose of radiation of different natures.

Nevertheless, advantageously and according to the invention, a DI is calculated i _so i representative of the total ionizing dose of all radiation received, and a DNI _so i representative of the total non-ionizing dose of all radiation received.

Advantageously and according to the invention, these values DI _soh DNI _so i are obtained from a modeling by a function which is the compound of a first polynomial function of degree greater than or equal to 2 -notamment equal to 2- and d a second continuous monotonic function of these two values DI _soh DNI _so i. Said second monotonic function may continue for example a linear function or a logarithmic function, preferably the logarithmic decimal, or other. For example, the modeling described in the publication "Implementation of a" Design of Experiments "Methodology for the Prediction of Phototransistor Degradation in a Space Environment" by P. Spezzigu et al. , N ° 4, August 2009. Other nonlinear models can be considered.

The inventors have in particular determined that obtaining the two DI _soh DNI _so i values of the total ionizing dose and the total non-ionizing dose is generally sufficient in most applications, and that the use of such a modeling polynomial of the second degree of the decimal logarithm of these values is particularly faithful. Thus, advantageously and according to the invention is exclusively calculates DI i _n values representative of said total ionizing dose, and _so i DNI representative of said total non-ionizing dose, and using a dosimeter comprising at least 3 mixed detectors.

Advantageously, a dosimeter according to the invention is also characterized by all or some of the characteristics mentioned above in relation to the method according to the invention. A dosimeter according to the invention may be contained in a single housing incorporating both the different detectors, the storage means and the calculation means, as well as the various components and accessories necessary for the operation of these elements. As a variant, nothing prevents the dosimeter according to the invention from being formed into several separate and remote devices, for example a detection device placed in the environment for the measurement of radiation, this detection device communicating remotely with a device for processing the signals delivered by the different detectors, this treatment device being located outside the irradiated environment. For example, a dosimeter according to the invention may be formed of an on-board detection device on board a space system and remotely communicating with a processing device located on the ground and incorporating said storage means and said calculation means.

The invention also extends to a measurement method implemented in a dosimeter according to the invention. The invention also extends to a measuring method and a dosimeter characterized in combination by all or some of the characteristics mentioned above or below.

Other objects, features, and advantages of the invention will appear on reading the following description which refers to the appended figures in which:

FIG 1 is a functional block diagram illustrating an embodiment of a dosimeter according to the invention,

FIG. 2 is a graphical representation of the various curves representing the variations of the DI and DNI doses for which the responses of the detectors of a dosimeter according to the invention correspond to values measured for the same irradiation,

FIG. 3 is a functional block diagram illustrating in more detail the steps of a method according to the invention making it possible to calculate the dose values that can be extracted from the signals delivered by the detectors of a dosimeter according to the invention,

FIG. 4 is a functional block diagram illustrating in more detail the steps of a method according to the invention making it possible to obtain an energy differential fluence spectrum,

FIG. 5 is a graphic representation of an example of a differential energy fluence spectrum that can be obtained according to the invention.

A dosimeter according to the invention as shown in FIG. 1 comprises a plurality of N detectors 11 _1? 1, 11 _N , collectively referenced I li below, N being a natural integer greater than 2, each of these detectors I li delivering at least one electrical signal representative of a dose of ionizing radiation and / or non-ionizing he receives. A dosimeter according to the invention comprises at least two mixed detectors sensitive to both the ionizing dose and the non-ionizing dose of the radiation it receives, that is to say each delivering at least one electrical signal depending on the ionizing dose and the non-ionizing dose of the radiation it receives. In the preferred embodiment of the invention and as shown in FIG. 1, all the detectors of the dosimeter according to the invention are mixed detectors. And all these mixed detectors may or may not be similar in nature. Advantageously, and according to the invention, a majority of the detectors 1 li is formed of optocouplers (also called photocouplers), that is to say electronic components each comprising an LED 15 and a phototransistor 16 illuminated by the latter. . Such a detector is indeed both sensitive to the ionizing dose and the non-ionizing dose of the different radiations it receives, and provides analog signals simple to obtain and exploit, is also robust and compatible with multiple environments , space or terrestrial, and is small in size and low cost. It is autonomous in its operation and in particular does not require the addition of an additional light source. It is therefore easy to use and to characterize.

Other mixed detectors may, however, be used in a dosimeter according to the invention, for example chosen from optoelectronic components chosen from the group comprising CCD sensors (charge-coupled image sensors), APS sensors (sensors). active pixel images), photodiodes, phototransistors, optocouplers, bipolar transistors, bipolar integrated circuits, integrated circuits in BiCMOS technology ...

Nothing prevents us from providing, in a dosimeter according to the invention, one or more detector (s) sensitive (s) mainly-especially only to the ionizing dose of radiation, and / or one or more sensitive detector (s) ( s) principally -particularly only-to the non-ionizing dose of the radiation, and / or one or more detector (s) sensitive (s) mainly -not only-to a specific nature of radiation (for example selected from neutron radiation, protons, gamma rays, ...). Nothing prevents either to provide in a dosimeter according to the invention several detectors of different natures. In an advantageous embodiment, a dosimeter according to the invention comprises, as detectors I li, a majority of optocouplers, for example five optocouplers, and one or more phototransistor (s), for example two phototransistor. In the example shown in FIG. 1, all the detectors I li are optocouplers s.

The dosimeter according to the invention comprises a bias voltage source 12 and a voltage regulator 13 supplying an analog demultiplexer 14, the outputs of which each make it possible to polarize an emission LED of an optocoupler detector 11 via the intermediary of FIG. a series resistor 17, forming a supply current source of the LED 15. Each output of the analog demultiplexer 14 supplies one and only one emission LED of one and only one optocoupler detector 11 _i .

A bias voltage source 18 and a voltage regulator 19 also feed an analog demultiplexer 20, the outputs of which each make it possible to polarize a phototransistor 16 of an optocoupler detector 11a. The output signal of each phototransistor 16 is transmitted on one of the inputs of an analog multiplexer 21, via a parallel load resistor 22 interposed between this input and ground. Each input of the analog multiplexer 21 receives one and only one output signal (voltage induced by the phototransistor current across the parallel load resistor 22) of one and only one phototransistor 16.

The assembly described above thus constitutes a chain of acquisition of a dosimeter according to the invention.

The digital output of the analog / digital converter 23 is connected to a central bus 24 communicating with a processor 25, with a programmable memory 26, in particular of the EEPROM type, and with a random access memory 27, in particular of the SDRAM type, this assembly constituting a module processing signals from the acquisition chain.

The analog multiplexer 21 has an output connected to the input of an analog / digital converter 23, the output of which delivers digital signals representative of the output voltage across the parallel resistor 22 of each phototransistor 16, in turn, this voltage the output varies depending on the ionizing dose and the non-ionizing dose of radiation received by each detector li. More exactly, in an optocoupler, this is the current transfer rate (referred to as CTC for "current transfer coefficient"), i.e., the ratio of the LED supply current to the output current of the phototransistor, which is representative of the doses ionizing and non-ionizing radiation received by this optocoupler 1 li.

For each optocoupler detector 11 i, the CTCi normalized current transfer rate with respect to its value before irradiation can be expressed by a nonlinear analytical mathematical modeling as a function of a value DI representative of an ionizing dose, and a value DNI representative of a non-ionizing dose of radiation received by the optocoupler detector 1 1 ^.

This nonlinear modeling can be the subject of different variants, depending in particular on the chosen detectors, their response characteristics, the nature of the main expected radiations in the considered environment ... The nonlinear modeling adopted must also allow the extraction of the _desired values DI _soh DNI _so i.

That being the case, the inventors have determined that, particularly in the case of detectors formed of optocouplers and / or phototransistors, advantageously and according to the invention, this nonlinear modeling is advantageously a modeling by a function which is composed of a first polynomial function of degree greater than or equal to 2-in particular equal to 2- and a second continuous monotone function of these values DI, DNI, this second continuous monotone function being able to be for example chosen from a linear function, an affine function, an exponential function, a logarithmic function, a conic (parabola or hyperbola), ... but advantageously and preferably a logarithmic function, for example the logarithmic decimal function (log).

In particular, and advantageously and according to the invention, excellent results have been obtained, surprisingly, with a polynomial modeling of the second degree of the logarithmic decimal (log) of the values DI, DNI, according to the following formula (I):

The coefficients B _{0 b} B _lb B _{2 b} B _{3 b} B _{4 b} B _{5 b} are real numbers independent of the values DI, DNI, which vary from one detector 1 li to another, and can be determined by preliminary calibration of each detector 1 li as shown below.

In the following, we consider that the nonlinear modeling used is that corresponding to the formula (I).

The general method that can be implemented in a dosimeter according to the invention is shown schematically in FIG.

In a first preliminary configuration phase, not shown in FIG. 2, which can be executed once and does not have to be repeated for each measurement, the dosimeter according to the invention is calibrated, the various coefficients B _0> B _lb B _{2 b} B _{3 b} B _{4 b} B _{5 b} being determined for each detector 11 i.

In order to do this, the dosimeter is exposed in the laboratory to different or reference radiations (where j is the number of different radiations used) comprising radiations capable of producing ionizing and / or non-ionizing effects and the DI, DNI values of the dose. ionizing and non-ionizing doses are known. Following each of these exposures oj, the CTCi (oj) values of the current transfer rates delivered by the different detectors I li are measured and recorded. We then obtain, for each detector I li, a system of equations from which we can extract the coefficients B _{0 b} B _lb B _{2 b} B _{3 b} B _{4 b} B _{5 i} , since j is greater than or equal to six and that the system determinant is non-zero.

Advantageously, nine experiments are preferably carried out, with nine different radiations, for example as described in the publication "Implementation of a" Design of Experiments "Methodology for the Prediction of Phototransistor Degradation in a Space Environment" P. Spezzigu et al IEEE Transactions on nuclear science, Vol 56, No. 4, August 2009. And for each optocoupler detector I li is extracted the six coefficients B _{0> b} B _lb B _{2 b} B _{3 b} B _{4 b} B _{5 i} , of the nine equations, by example by the standard least squares method as also described in this publication. The different coefficients B _{0> b} B _lb B _{2 b} B _{3 b} B _{4 b} B _{5 i} , for the various detectors I li are recorded in the EEPROM memory 26.

It should be noted that the calibration performed in this first phase can be specifically adapted according to the environment in which the dosimeter is intended to be used later, in particular according to the nature of the radiation present in this environment. In particular, the different reference radiations used to obtain the coefficients B ₀ , B _1h B ₂ , B _{3 i} , B ₁ , B ₅ , may be chosen so as to correspond as closely as possible to the radiation present in the environment. of future use of the dosimeter according to the invention.

When using the dosimeter thus calibrated, the latter is placed in the environment undergoing radiation, and the values of the output voltages of the optocoupler detectors I li provided by the different detectors 1 li at the output of the converter 23 are recorded in SDRAM memory 27.

From these different values of output voltages, the value of the load resistor 22 and the values of the supply voltage delivered by their voltage regulator 13 and the series resistance 17 of supplying the LED 15, the processor 25 calculates, for each optocoupler detector 1 li on the one hand the intensity of the supply current of the LED 15, on the other hand the intensity of the output current of the phototransistor, and the value of the transfer rate of the CTCi current. These different CT values are stored in the EEPROM 26 during the acquisition step 31.

N equations E _t with two unknowns are thus obtained: the values of the doses DI, DNI.

However, the inventors have determined that, although the detectors I li are similar (for the majority or all consist of optocouplers), if the technical specifications of these different detectors are different (which may be the case for example by choosing different commercial references and / or different manufacturers), these different equations are in practice all sufficiently different from each other so as to allow the extraction of a common solution DI _soh DNI _so i for the values DI, DNI. This is particularly illustrated in Figure 2 which shows that the different representative curves respectively different equations may have multiple points of intersection, but that there is in fact a common intersection area, which may correspond to values representative of the DI _soh values DNI _so i real sought. In FIG. 2, each of the curves corresponds to one of the equations E _t polynomial of degree 2 obtained for a detector 1 li.

Thus, contrary to the state of the art in which it is desired to use detectors different from one another and of the same sensitivity and properties of variations, and to proceed by difference of the responses provided by the different detectors, the invention consists in use all similar detectors, and find that in practice all these similar detectors actually have different responses, allowing the extraction of DI _soh values DNI _so i from a nonlinear modeling.

Mathematically, this extraction is reduced to solving a system of N equations (N being the number of different references of optocouplers used) to two unknowns: DI and DNI.

We have :

This overdetermined and non-linear system can be solved by minimizing the sum of the quadratic differences between the measurement results and the polynomial predictions. This is explained by :

To find the minimum of ε (ΌΝΙ, Όΐ), the following system must be solved:

being the solution sought.

It should be noted that, strictly speaking, condition (IV) is not sufficient to rule on the nature (minimum, maximum, collar, etc.) of the critical point

This would require the following conditions to be certain

to have a local minimum of the function ε:

But, as long as there is, in principle, a local minimum, we can content ourselves with solving the system (IV) and then checking, a posteriori, that the previous inequalities are well satisfied.

The development of equation (III) makes it possible to obtain the following form:

Equations (IV) can be rewritten in the form

Equation (Vll-a) can be solved using the method of

Tartaglia-Cardan applicable to polynomials of degree 3. We thus obtain an analytic relation between

By identifying the equation (VII- a) with an equation of the form ax ³ + bx ² + ex + d = 0 and using the usual notation of the method of

Tartaglia-Cardan, it comes the system (VIII):

The solutions of equation Vll-a are given by

According to the sign of the 3 expressions of x _i can then

be injected in turn into (VH-b) to obtain the following equation (X):

This transcendental equation (X) can be solved numerically to determine DNI _sol for each x _i for example by the methods of Ridder and Brent. In practice, the equation (X) can have several roots DNI _sol to be determined. Knowing these different roots, the associated _soil DIs can be calculated from the previously established IX-a, IX-b and IX-c relationships. The pair {DNI _sol , DI _sol ) sought is the one that minimizes the function

Thus, in the method according to the invention schematically represented in FIG. 3, during the first step 31, the different values CTC _i delivered by the different detectors 11i are stored in memory 26. From these different values and coefficients B _{0 i} , B _lb B _{2 i} , B _{3 i} , B _{4 i} , B _{5 i} , also stored in memory 26, the processor calculates the various parameters AQ to Aj4 from equations (VI) in step 32. In subsequent step 33, the processor calculates the values a, b, c, d, p , q, Δ of the system (VIII). In the subsequent step 34, the processor calculates DNI _so i by solving the transcendental equation (X). In the subsequent step 35, the processor calculates DI _so i from the relationships (IX) ·

It should be noted that other methods of extracting the desired values DI _so i DNI _so i are possible from the nonlinear modeling according to the invention.

For example, it is possible to search for the different points of intersection of the different pairs of curves corresponding to the formula of the selected modelization. In other words, for each pair of equations, that is to say for each pair of detectors 11 _i7 lt _j DI values, DNI which simultaneously satisfy the two equations are searched and stored in the EEPROM 26. It calculates first the roots of the polynomial of a first equation E _t and this value is substituted in a second equation E _j whose roots are then calculated. We then obtain different DNI values _k (i, j) of DNI capable of satisfying the two equations E _b E _j . For each of these values, the possible values of DI are then calculated. Each pair of values thus obtained is stored in the EEPROM memory 26. This calculation is repeated for all the pairs of detectors. Once these different calculations are executed, a common value is determined common to the different values obtained on the pairs of detectors, possibly after elimination of singular values out of predetermined limits of validity.

Moreover, the CTC current transfer rate is used as a signal delivered in the case of a mixed detector formed of an optocoupler. In the case of a mixed detector formed by another type of component, the signal variable supplied by the detector in the nonlinear modeling, instead of this current transfer rate, uses the relevant variable variable according to the ionizing dose and the non-ionizing dose of this mixed detector. Thus, for example, in the case of a CCD or APS sensor, the relevant parameter as delivered signal is the current of darkness; in the case of a photodiode or a phototransistor, the relevant parameter as delivered signal is the dark current or the photocurrent; in the case of a bipolar electronic component or BiCMOS, the relevant parameter as the delivered signal is the current gain. Moreover, in the case of a laser diode or a light-emitting diode, the relevant parameter as delivered signal is the transmitted optical power (proportional to the power supply current); in the case of an optical fiber the relevant parameter as delivered signal is the attenuation (ratio of the output power to the input power); in the case of a MOS electronic component, the relevant parameter as the delivered signal is the threshold voltage.

EXAMPLE:

The reliability of the method according to the invention has been demonstrated by an example using a dosimeter composed of five optocoupler detectors, namely four hardened (that is to say radiation-resistant) optocouplers, references Mii 66168, MH66227, Mii 66223-RT2, Mii 66223-RT3, marketed by the company Micropac (Dallas, USA), and a standard optocoupler Mii 66092 marketed by the company Micropac (Dallas, USA), and two detectors formed of phototransistors marketed by the company OPTOi (Trento, ITALY).

This dosimeter was subjected in the laboratory to a reference proton beam having a known differential fluence spectrum corresponding to the curve C1 of FIG. 5, with a known ionizing dose DI = 4.2 krad, and with a known non-ionizing dose DNI = 7.95 95 ⁷ MeV / g.

By applying the method according to the invention described above, a measured value of the ionizing dose DI _so i = 4.27 krad and a measured value of the non-ionizing dose DNI is obtained _. = 7.96 10 ⁷ MeV / g. As can be seen, the values obtained according to the invention are very close to reality (2% error on the ionizing dose, 0.2% on the non-ionizing dose), thus demonstrating the excellent and surprising reliability of the dosimetry according to the invention.

Furthermore, the inventors have determined that in the case of an environment mainly irradiated with proton radiation, which is common in the space environments of low Earth orbiting Earth satellites, it is even possible to obtain a relatively good evaluation of the differential energy fluence spectrum through a dosimeter according to the invention.

Indeed, the inventors have determined that it turns out that the energy differential fluence spectra in such a situation can be assimilated in reality to a normal logarithmic function of standard deviation equal to 1, that is to say according to the following formula (XI):

Thus, in this hypothesis, the parameters K and μ can be obtained according to the following formulas (XIII) and (XIV):

In formulas (XI) to (XIV):

-ln is the natural logarithm,

-E is the energy,

the LET function is the linear energy transfer function of the ionizing radiation in the constituent material of the detectors, which is known and tabulated in the EEPROM memory 26,

the function NIEL is the function of loss of non-ionizing energy in the constituent material of the detectors, which is also known and tabulated in the EEPROM memory 26, -K is a scale factor factor,

-μ is the average value of the normal logarithmic function,

-δΦ / δΕ is the differential energy fluence.

Thus, in the method shown in FIG. 4, after the steps 34, 35 providing the values DI _soh DNI _so i according to the invention, the processor calculates the ratio DI _so it DNI _so [at the step 41, then, from of the formula (XIII), the function DI _so it DNI _so i (μ) being tabulated in EEPROM memory 26, determines in step 42 the value μ corresponding to this ratio DI _so i I DNI _soh and records this value in the EEPROM memory 26 during step 43.

In the example given above:

DI _soil the _soil DNI = 5.36 x 10 ^"8 krad / MeV / g

and we get μ = 3.41

Knowing the value of μ, that of K is calculated in step 44 by the processor from one of the equations (XII), for example by applying the formula (XIV) and this value is stored in the EEPROM 26 at step 45.

In the example, we find K = 1.897 × ¹⁰ cm ^-2 . It should be noted that the integration terminals of formula (XIII) were taken equal to the cutoff energies of the reference proton beam, namely 8 MeV. and 190 MeV respectively, corresponding to the cutoff energies specified for the reference proton beam.

From these two values of μ and K, we can plot the spectrum of the differential fluence according to the formula (XI). FIG. 5 represents the curve C2 corresponding to the differential energy fluence spectrum obtained in the example mentioned above. As can be seen, the result provides a very good estimate of reality.

It goes without saying that the invention may be subject to numerous variants with respect to the preferred embodiments and to the example described above and shown in the figures. In particular, the number of detectors can vary as well as their nature. In addition, as noted above, other forms of non-linear modeling may be considered. A dosimeter according to the invention is particularly simple, compact, energy-saving, can be easily calibrated and reconfigured according to the different applications. It can be used in many different applications, on board space satellites, in low orbit, medium or high or geostationary, aboard any other space system, or on the ground, for medical applications, for the safety of people in nuclear environments, for calibration of reference radiation sources, or others.

It should be noted in particular that the qualities of the invention make it possible to envisage the application of a method and a dosimeter according to the invention to reliably evaluate the variations of the ionizing and non-ionizing doses over time . For example, it is possible to use a method and a dosimeter according to the invention for a first measurement campaign in a predetermined environment, then, after a lapse of time without measurement, to reuse the same dosimeter according to the invention for a second measurement campaign in the same environment or even in a different environment. Also, one can evaluate the flux that irradiated an environment in a time interval between two instants, by measuring the doses at each of these two instants and by calculating the variations of the ionizing dose and the non-ionizing dose. When the two moments tend to get closer, we obtain an evaluation of the flow. There is nothing to prevent the addition of a dosimeter according to the invention to a specific module which makes it possible to evaluate the ionizing flux and the non-ionizing flux by calculating the derivative with respect to the time of the ionizing dose, respectively the non-ionizing dose.

Furthermore, it should be noted that if a dosimeter according to the invention can be fully integrated in the same portable housing, nothing prevents to provide on the contrary other architectures more complex, for example by placing the acquisition chain of signals, that is to say mainly the detectors I li in the environment in which the radiation doses must be measured, to associate with this acquisition chain means of communication at a distance, in particular wireless, for example radiofrequency, capable of transmitting the signals delivered by the different detectors 11 i, and of deporting outside the irradiated environment the signal processing chain, comprising in particular the processor 25, the memories 26, 27 and the bus 24, also associating remote communication means, including wireless, for example radiofrequency, allowing the reception of signals from the acquisition chain. For example, the acquisition chain can be placed on board a space satellite and the processing chain can be placed on the ground.

Claims

1 / - A method for measuring different representative values of radiation doses irradiating an environment, wherein is placed in said environment a dosimeter comprising a plurality of detectors (1 li) of radiation each delivering at least one electrical signal dependent on ionizing doses and and / or non-ionizing radiation it receives, the number of detectors (110 being greater than or equal to the number of different doses to be measured, characterized in that a dosimeter comprising at least two detectors, called detectors (110 mixed, each delivering at least one electrical signal dependent on an ionizing dose and a non-ionizing dose of radiation it receives, at least two mixed detectors having properties of variation of the signals they deliver according to said doses which are different a detector to the other, and in that at least calculates a value _n i DI representative of an ionizing dose, and at least one DNI value _so i representative of a non-ionizing dose, from the electrical signals delivered by the different detectors (110, from a nonlinear modeling of these electrical signals as a function of said DI values _so i DNI _soh this modeling being previously established and parameterized for each detector (11 _i ).

21 - Process according to claim 1, characterized in that a dosimeter comprising a plurality of detectors (110 mixed of the same nature and having characteristics of variations of the electrical signals they provide according to the received doses which are different.

3 / - Method according to one of claims 1 or 2, characterized in that a dosimeter whose detectors (1 10 comprise elements (15,16) sensitive solid state.

4 / - Method according to one of claims 1 to 3, characterized in that a dosimeter is used whose detectors (110 mixed are selected from optoelectronic components: optocouplers, photodiodes, phototransistor, CCD sensors, sensors APS, bipolar electronic components and BiCMOS electronic components. 5 / - Method according to one of claims 1 to 4, characterized in that said nonlinear modeling is a model comprising a polynomial function of degree greater than or equal to 2.

6 / - Method according to one of claims 1 to 5, characterized in that said nonlinear modeling is a model expressing the values of the signals delivered by each mixed detector, and comprises a continuous monotone function of said values DI _soh DNI _so i sought .

II - Process according to claim 6, characterized in that said continuous monotone function is a logarithmic function.

8 / - Method according to one of claims 1 to 7, characterized in that calculates a value DI _so i representative of the total ionizing dose of all radiation received, and a DNI _so i representative of the non-ionizing dose total of all received radiation.

91 - Method according to one of claims 1 to 8, characterized in that a dosimeter consists of a number of detectors (1 li) mixed between 3 and 20, in particular of the order of 10.

10 / - dosimeter for measuring different representative values of radiation doses radiating an environment, capable of delivering a signal representative of radiation that it receives, comprising a plurality of detectors (1 li) of radiation each delivering at least one dependent electrical signal ionizing doses and / or non-ionizing radiation it receives, the number of detectors (11 being greater than or equal to the number of different dose values to be measured,

characterized in that it comprises:

at least two detectors, called detectors (11 mixed), each delivering at least one electrical signal dependent on an ionizing dose and a non-ionizing dose of radiation that it receives, at least two mixed detectors having variation properties signals they deliver according to said doses that are different from one detector to another,

means (26) for storing, for each detector (I li), parameters of a non-linear modeling of the electrical signals delivered by the detector (l li) as a function of the values of said ionizing dose and of said non-ionizing dose of radiation received by this detector (I li),

-means (25) for calculating at least one value DI _so i representative of an ionizing dose, and at least one representative DNI i _n value of a non-ionizing dose, from the electrical signals delivered by the different detectors (l li), from a non-linear modeling of these electrical signals as a function of said DI _soh values DNI _so i, this modeling being previously established and parameterized for each detector (I li) from parameters stored in the storage means (26).

11 / - Dosimeter according to claim 10, characterized in that it comprises a plurality of detectors (l li) mixed of the same nature and having characteristics of variations of the electrical signals they provide according to the doses received which are different.

12 / - Dosimeter according to one of claims 10 or 11, characterized in that the detectors (I li) comprise elements (15, 16) sensitive to solid state.

13 / - dosimeter according to one of claims 9 to 12, characterized in that the detectors (l li) mixed are selected from optoelectronic components: optocouplers, photodiodes, phototransistors, CCD sensors, APS sensors; Bipolar electronic components and BiCMOS electronic components.

14 / - Dosimeter according to one of claims 10 to 13, characterized in that it consists of a number of detectors (l li) mixed between 3 and 20, in particular of the order of 10.

15 / - dosimeter according to one of claims 10 to 14, characterized in that said means (25) for calculating are adapted to calculate a value DI _so i representative of the total ionizing dose of all radiation received, and a DNI value _so i representative of the non-ionizing dose of all received radiation.