Method for measuring a sound signal propagation delay in a fluid by zero-crossing of the said sound signal

Translation of description of equivalent WO 9746854 A1

Method of measuring the propagation time of an acoustic Signai in a fluid by zero crossing of said acoustic signal.

The present mvention provides a method for measuring the propagation time of an acoustic signal in a fluid flow between a first transducer serving as an emitter and a second transducer acting as a receiver and located at a determined distance from the first transducer, the acoustic signal emitted by the first transducer being constituted by at least one pulse emitted at a determined acoustical frequency Fa and the acoustical signal received by the second transducer include an ant an oscillation séπe characteristics whose amplitude is first growing on several péπodes, then decreases over several following péπodes, the envelope of the oscillations character- πstiques having the shape of a spindle, the method comprising sampling the received acoustic signal has a sampling frequency Fe, the signal has numéπser sound sampled received, and to seek, by analysis of the received acoustic sampled and digitized signal, the first meaningful zero-crossing of the characteristic oscillations of the received acoustic signal

It is known for many years to measure flow of a fluid (or volume) flowing in a pipe by using the propagation of acoustic signals transmitted between two acoustic transducers located at spaced points in the direction of flow fluid in principle, a transmitted acoustic signal from the first transducer to the second transducer is received by the second transducer and the propagation time Td of the acoustic signal is measured. Similarly, measuring the propagation time Tu of a transmitted acoustic signal from the second transducer to the first transducer after receipt of said signal by said first transducer.

In a fluid meter, the flow rate can be obtained by combining a measurement of the propagation times of the two acoustic signals transmitted between the two points in opposite directions with a measurement of the induced acoustic phase shifts in each acoustical signal by the propagation of each of the signals in acoustic flow. European Patent Application No. 0426309 décπt an example of such a flow measurement system, wherein the received signal is sampled and converted numéπquement, the measurement of acoustic phase shift being effected by performing synchronous detection on the digitized signal.

When measuring the rate of flow of a gas in a gas meter using two ultrasonic transducers it is necessary to measure the travel time of the ultrasonic wave between the transmission time and the reception time, since the céléπté ultrasonic waves is dependent on the nature of the gas

on the Figure 2 shows the shape of a rectangular pulse signal S \ width T transmitted at a time T_υ by a first ultrasonic transducer disposed in the flow stream of a fluid at a first point, and the shape of the S2 signal constituting the impulse response received echo Tj to a moment by a second ultrasonic transducer disposed in the flow stream of fluid in a separate second item of the first point S2 acoustic signal received by the second transducer is constituted by a caractéπstiques oscillation séπe O_c that increase in amplitude over several péπodes then decrease, the envelope of the characteristic oscillations having the shape of a spindle. The oscillations caractéπstiques O_c S2 signal are preceded and followed by Op parasitic oscillations of low amplitude to determine the instant Ti early caractéπstiques oscillations, it is necessary to identify the first meaningful zero-crossing of caracténstiques oscillations O_c the received acoustic signal S2

There is shown in Figure 3, to a larger scale than in Figure 2, an exemplary acoustic signal S2 received echo to a rectangular pulse S_] transmitted at a determined acoustical frequency Fa

To determine the start of oscillations caracteπstiques O_cAccording to a known method, a threshold voltage V is fixed_s, With respect to which one compares the level of the received acoustic signal S2, the comparison being carried out on a numeπque signal obtained after sampling the analog received acoustical signal at a sampling frequency Fe which is for example a multiple of the acoustic frequency Fa.

In this case, the moment we mark T2 to which the amplitude of the received signal exceeds the threshold voltage V_sAnd then identifies the time of the previous zero crossing (or later) that are considered time T] starting of oscillations O caractéπstiques_c the received acoustic signal S2

Such a measurement method can lead to eπeurs since the oscillations caractéπstiques O_c the received acoustic signal S2 may be more or less amplified as a function of the nature of the gas. Thus, there is shown in the

- Figure 4. a S21 curve that corresponds to the shape of an acoustic signal received for nitrogen (N2) and a S22 curve Φ¹¹ corresponds to the shape of an acoustic signal received to a mixture of carbon dioxide and methane (CO2 / CH4).

It is observed that the S21 curve crosses the threshold voltage V_s a moment that T4 triggers the identification of passage previous zero to a time T3 which is rightly regarded as a mark of the early oscillations O caractéπstiques_c However, we can see that the curve S22_> which is in phase with the S21 curve crosses the threshold voltage V_s at a time Tg postéπeur to time T4 and offset from one of the value of the received signal TR péπode The moment identified for S22 T5 then curve as the point of zero crossing before crossing immediately threshold at time T g, is then regarded as the mark of early caractéπstiques oscillations O_c Curve S22 Cτ_> as can be seen in Figure 4, the curve S22 has a negative lobe which comes qu'affleurer the voltage threshold value V_s no reach or exceed this threshold

As a result of attenuation or amplification of the received signal S2, which vaπent depending on the nature of the gas, the conventional method of measuring time of the first passage has zero caracteπstiques oscillations of the received signal can cause an error of more or less a péπode that did significantly decrease measurement accuracy

The invention aims to overcome the above drawbacks and allow to reduce the method of measuring the sensitivity of the passage of time zero of a received acoustic signal, vis-à-vis the waveform vaπations this signal and vis-à-vis any extéπeures disturbances likely to be detected by a simple threshold crossing monitoring methodology, and generate pouπaient eπonées measures

These objects are achieved according to the invention, by a method of measuring the propagation time of an acoustic signal in a fluid flow between a first transducer serving as an emitter and a second transducer acting as a receiver and located at a determined distance from the first transducer, the acoustic signal emitted by the first transducer being constituted by at least a transmitted pulse has a specific sound frequency Fa and the acoustical signal received by the second transducer comprising a séπe caractéπstiques of oscillations whose amplitude is initially rising on several péπodes, then decreases over several following periods, the envelope of oscillations caractéπstiques having the shape of a spindle, the method comprising sampling the received acoustic signal at a frequency

~ Fe sampling, digitizing the sampled received acoustical signal, and searching, by analyzing the sampled received acoustical signal and numéπsé, the first significant zero-crossing of the received caractéπstiques oscillations of the acoustic signal, caractéπsé in that, to search for the first zero crossing caractéπstiques of significant oscillations of the received acoustical signal, is defined beforehand a first péπode ideal caracteπstique determining the first zero crossing of caracteπstiques oscillations of the received acoustical signal, in this ideal caractéπsant péπode by a théonque amplitudes Δ ratio between the maximum amplitudes Pi - and Pι + of the two lobes of this péπode, is determined for each peπode the acoustic signal received samples and numénsé, the maximum amplitudes P- and P + of the two lobes of the examined péπode compares the ratio of these amplitudes P- and P + the amplitude ratio of the ideal théonque coπespondant péπode then, if the result of the comparison is supéπeur to a threshold value G_s, Consider the peπode examined as a parasite péπode coπespondant to noise while if the result of the comparison is mféπeur this threshold value G_s, Consider the péπode examined as a caracteπstique péπode and then determined the zero crossing between two lobes of this feature peπode, which has zero crossing is considered as the first meaningful zero-crossing of the received acoustic oscillations caractéπstiques signal

The théonque amplitudes Δ ratio between the maximum amplitudes Pi and Pι + of the two lobes of said ideal peπode is determined beforehand from an average for several different gases and different rates, the ratio between the maximum amplitudes P - P + and the observed caracteπstique péπode from the recording of acoustic signals received

The amplitude ratio Δ between the maximum amplitudes Pi- and Pι + of the two lobes of the first péπode ideal caracteπstique of a received acoustic signal is almost constant as a function of the nature of the gas and is independent of the gain factor choosing a basis for comparison of Δ cπtère this report helps greatly reduce the extent of dependence on the nature of the gas, and thus improve the accuracy and reliability of measurements

According to a first embodiment of the invention, the comparison between the ratio of amplitudes P- and P + of the examined péπode théonque and the amplitude ratio Δ coπespondant the ideal péπode is performed by calculating a similarity of G which cπtère is the comparison result compared to the threshold value G_sAnd is defined as follows

G = (P-) - Λ (P +)

(P-)

In a second embodiment of the invention, the comparison between the ratio of amplitudes P- and P + of the review and the report péπode Δ coπespondant theoretical amplitudes of the ideal period is performed by computing a similarity criterion G which constitutes the result of comparison compared to the threshold value G_sAnd is defined as follows:

(P) - A (+?)

(P-) -z

where z represents an offset term for making the similarity criterion G slightly dependent on the maximum amplitude (P) of the negative lobe of the characteristic period examined.

According to this second embodiment, it reduces the risk to detect waveforms resembling the similarity criterion G or template, to be embedded in noise.

To further reduce the probability of detecting similar periods of low amplitude, according to a third embodiment of the invention, the comparison between the ratio of the amplitudes P- and P + of the reporting period and the theoretical amplitude ratio of coπespondant ideal period is performed by computing a similarity criterion G which is the comparison result compared to the threshold value G_sAnd is defined as follows:

G = (P-) - _Δ [(P +) -z ']

(P-) -z

where z represents an offset term for making the similarity criterion G slightly dependent on the maximum amplitude (P) of the negative lobe of the examined characteristic period, and z 'represents an offset term for making the similarity criterion G slightly dependent on the maximum amplitude (P +) of the positive lobe of the characteristic period examined.

To avoid eπeurs in determining the maximum amplitudes P- and P + of the two lobes of each reporting period, eπeurs which may be due to a phase shift between the received acoustic signal and the sampling signal, we can use an algorithm approximating the maximum (in absolute value) from the sample points near each maximum for best accuracy on the values ​​P- and P +. Always to avoid these same mistakes when the received acoustic signal is repeatable, several successive measurements are performed respectively on consecutive acoustic signals by shifting the acoustic signal for each sampling signal relative to the considered acoustical signal, in order to obtain several sampled points offset in the vicinity of the maximum.

According to one embodiment also to reduce the probability of detecting similar periods of low amplitude, is performed for each reporting period, two or three computations of the similarity criterion G with values ​​of different neighboring theoretical amplitudes of the report report theoretical predetermined amplitude and not deviating from it by more than 10%, and we consider the review period as a characteristic period if the computed similarity criterion G is in any case less than the value of threshold G_s.

To determine the threshold value G_s the similarity criterion G as well as the value of the offset term z in advance is plotted, for different gases and different flow rates, by varying the phase difference between the sampling frequency Fe and the received acoustical signal and by applying a factor gain varying between 0.45 and 1.5, an array of first curves representing the maximum values ​​of similarity criterion G for the observed characteristic period and a network of second curves representing the minimum values ​​of the similarity criterion G for parasites periods observed, according to different possible values ​​of the offset term z, and one chooses the threshold value G_s and that of the offset term z as a function of the margin of safety possible in the intermediate region between the first network and the network of curves second curves.

For example, the threshold value G_s compnse can be between 0.7 and 1.7 and the value z offset of the term may be between 0.21 and 0.25 V. The invention is advantageously applicable to a method of measuring the flow velocity of a fluid between two transducers arranged at measurement points spaced apart in the direction of a fluid flow according to which the value of the flow velocity to be measured is obtained by combining a measurement of the respective propagation times of the two signals sound emitted between the two points in opposite directions by the transducers with a measurement of the acoustic phase shifts induced respectively in each acoustical signal by the propagation of each of said acoustic signals in the flow, the measurement of the respective propagation times of the two signals

~ acoustic emitted between the two points of measurement being performed according to the method according to the present invention. Other features and advantages of the invention emerge from the following descπption of particular embodiments, with reference to the accompanying drawings, on which ^•

- Figure 1 is a representation, in the form of a block diagram, of the whole of a known gas metering device which is applicable to the present mvention,

- Figure 2 is a diagram showing the principle of a known method of determining the propagation time of an acoustic signal in a fluid medium, - Figure 3 is an enlarged diagram of a portion of Figure 2 showing the principle of a known method of determining the propagation time of a sound signal by comparing the received signal with a threshold value,

- Figure 3a is a diagram similar to that of Figure 3 showing how it operates on a sampled signal according to a known method of determining the propagation time of an acoustic signal by the zero crossing,

- Figure 4 is a set of two diagrams similar to that of Figure 3 showing the πsques of eπeur in a known method of determining the propagation time of an acoustic signal using a comparison of received signals with a value of threshold,

- Figure 5 is a diagram showing the implementation of pπncipe the process according to the invention for determining the propagation time of an acoustic signal in a fluid medium, and - Figure 6 is a graph in which families of curves for choosing useful parameter values ​​for carrying out the method according to the invention Figure 7 is a flowchart showing an example of various steps of the method according to the invention various types of electronic circuits can be used poui ensure a counting fluid by an ultrasonic method for measuring the fluid flow rate Figure 1 shows the pπncipe diagram of one example of such electronic circuits which are associated with two ultrasonic transducers 1, 2

^"spaced apart from one another in a pipe 3 in which flows a fluid such as a gas. The two transducers 1 and 2 are connected to a switching unit 4 which comprises two switches 5, 6 and allows the use of each transducer alternately as a transmitter and receiver A module transmission 14 and a receiving module 17 are connected respectively to the switches 5, 6 of the switch block 4. The transmission unit 14 comprises an operational amplifier 16 and a numéπque-analog converter 15. The reception module 17 comprises at least one amplifier 18 and an analog-converter 19 which numenque numéπse and simultaneously samples the received signal. An energy source electπque 7 and a module 8 of électπque power management are linked in particular to transmission units 14 and receiving 17 and the switching unit 4 and a microcontroller 10 includes a microcontroller 10 quartz clock 9, an arithmetic logic unit, the memory circuits and ROM, and may cooperate with display circuitry 13, a ROM 12 and a link remscnptible séπe January 1 type RS 232

Gas meters intended to equip each subscriber of a network of distπbution must be both accurate, reliable and best walking These constraints can impose to avoid the use of expensive components, to implement small piles of long term but medium capacity power for reasons of secunte and perform calculations according ensuring a measuring method has both precision and reliability while remaining simple enough to be energy efficient the measuring method of the the invention which will be explained later is thus advantageously applied to a device for measuring the flow rate of a gaseous fluid, to ensure counting of the consumption of the fluid, by using two ultrasound transducers prepared to distance from the other has intéπeur a pipe in the direction of fluid flow as an example, the ultrasonic transducers can be operated at an acoustic frequency Fa of the order of 40 kHz and the sampling frequency Fe of the received acoustic signals is advantageously 320 kHz, that is to say, eight times higher than the frequency Fa

The sampling frequency Fe is advantageously equal to a multiple of the sound frequency Fa However, the frequency Fe can also take other values ​​when the values ​​(multiples of Fa or not multiple) of

Fe are too low, it is preferable to use such a algoπthme approximate maximum of the neighborhood (in absolute value) of each peπode

^~ the acoustic signal, or even to perform several successive measurements by shifting to each new measurement the sampling signal relative to the acoustic signal considered, in order to obtain greater accuracy of the values of the maximum amplitudes (absolute value). These methods will be explained later.

When the Fe values ​​are high enough the use of such methods is not necessary. The acoustic signal emitted by each transducer is constituted for example by a pulse transmitted to the determined sound frequency Fa. This pulse is for example rectangular. Several pulses pouπaient also be issued.

In general, the fluid flow in a duct can be written:

Φ = (SL / 2) (Tu - Td) / where Tu.Td

S is the average passage section offered to the flow between the two transducers,

L designates the distance between the transducers,

Td is the propagation time of the acoustic signal emitted in the direction of flow by the first transducer (upstream) to the receipt of the acoustic signal by the second transducer (downstream), Tu is the propagation time of the acoustic signal emitted in the opposite direction of flow by the second transducer to receiving this acoustic signal by the first transducer (upstream).

If the measurement of this flow also involves a measurement of acoustic phase shifts induced in each acoustic signal through the propagation of each of the acoustic signals emitted by the transducers in the flow, the fluid flow in a duct can be written:

Φ = (SL / 4πFa) (2π [Fa (Tu - Td)] + (<pd - (pu)) / Tu.Td

where S, L, Td, Tu have the meaning indicated above, and φd and (could represent the acoustic phase shifts induced in each acoustic signal due to the signal propagation and F represents the frequency of said acoustic signals. From specifically, Tu and Td propagation time can be written respectively:

T = L / (cv) Td ≈ L / (c + v) where c and v respectively represent the propagation speed of the acoustic signal and the velocity of the gas.

For a given gas composition, such as the propagation velocity c depends pπncipalement temperature at which vaπe little flow measurements, the term Tu.Td vaπe too little and this term can be recalculated with a relatively low frequency. However, since it is the absolute measurement of the propagation times, the precision obtained for this measurement is directly transferred to the measurement accuracy of the flow. Thus, it is essential to avoid a shift on this measure.

For example, if the desired accuracy of the flow rate is 1%, the measurement accuracy of the propagation times must be close to 0.5%, which coπespond a mféneure eπeur 2 microseconds on the propagation time with the common gases and usual lengths L between two transducers (example: L = 150 mm and the gas being methane)

According to a method for measuring propagation times Tu and Td, known as the passage of method name is zero, measuring the propagation time (Tu pz) and (Td pz) of the acoustic signals emitted by each transducer until the detection by the zero crossing and is deduced the actual propagation times Tu and Td, with T = (Tu pz - to) and Td = (Td pz - to) where to is a constant is subtracted to obtain the propagation time in the gas, to depending only on the nature of the transducers and electronics

Figure 3a illustrates the measurement method by zero crossing performed on each received ultrasonic signal according to a known method, in association with a device such as that of Figure 1.

The received signal is amplified by the amplifier 18 with a gain gives then converted numéπque and sampled simultaneously in the converter 19 In the example, the sampling is performed at a frequency Fe, for example equal to 8 F or 320 kHz if Fa is 40 kHz. Instructions for performing the steps, contained in a ROM of the microcontroller 10, are used to compare the values ​​of the amplitude of the received signal and the sampled points are stored in a RAM, a predetermined value to a threshold coπespondant V_sAnd we search points sampled amplitude values ​​following the first point A, which has a value supéπeure amplitude threshold.

From that point, we search the next zero crossing of the curve representative of the ultrasonic signal. To do this, two are determined consecutive sampled points B and C that frame the point where the amplitude of the curve passes through zero and is carried by linear interpolation between B and C to determine the point D and thus measure the passage of time has zero TPZ.

According to an embodiment variant, illustrated in Figures 3 and 4 discussed above, it is also possible to search the last zero crossing of the curve representing the ultrasound signal before said curve passes through the threshold

The present invention aims to improve this method of determining a signal propagation time by measuring the time of zero crossing. As mentioned above with reference to Figures 3 and 4, the simple comparison between the voltage amplitude values ​​to the sampled points of a received sound signal and a threshold voltage value V_s can lead to eπeurs in determining the passage of time has zero TPZ and therefore in the propagation time of the acoustic signal If in Figure 4 the passage of time has zero T3 for S21 curve relating to a gas coπespondant signal constitutes nitrogen is coπect, however the time of zero crossing T5 for the S22 curve relating to a gas coπespondant signal is a mixture of carbon dioxide and methane, is incoπect and excessive.

The method according to the invention avoids such eπeur and, in general, reduce the method of measuring the sensitivity of the zero crossing time of waveforms as well as vaπations possible external disturbances, if they were detected, would generate measurement eπeurs

According to the invention, to find the first passage leading zero of the oscillations O caractéπstiques_c the received acoustic signal S2 is defined beforehand a first péπode ideal caractéπstique distinct parasitic oscillations of low amplitude noise component, which determines the first zero crossing of caractéπstiques oscillations O_c the received signal S2 This ideal penod is caractéπsée by a Δ amplitude ratio between the amplitudes of the maximum values ​​and Pi + Pι positive and negative lobes of the ideal penod

Upon receipt of an acoustic signal S2 to be analyzed, which is sampled and numéπsé conventionally, is determined for each péπode the sampled received acoustical signal and numéπsé, the amplitudes of the values

^~ Maximum P- and P + of negative and positive lobes péπode examined For this, it is not necessary to examine all the sampled points of the péπode, and determination of maximum values ​​(in absolute value) and P- P + the negative and positive lobes of the period under review can be made by a single research vicinity of these extreme points.

However, it should be noted that there is a phase difference between the received sound signal and the sampling signal. Thus, depending on the phase difference, the sampled point represent more or less faithful maximum values ​​(in absolute values) P- and P + of negative and positive lobes.

To remedy this problem, it is possible to use a maximum of approximation algorithm (in absolute value) from the sampled points at most of the neighborhood in order to obtain an estimate of the real value of this maximum regardless phase difference between the received sound signal and the sampling signal.

For example, one can use a known polynomial approximation algonthme

The previously mentioned problem pouπait also be solved by increasing the sampling frequency.

However, when the method according to the invention is implemented in a fluid meter that works on batteries, it is essential to minimize the energy consumption of the process and this requirement is not compatible with an increase in the frequency sampling.

In addition, a higher sampling rate requires the use of a more expensive clock.

An interesting solution to solve the aforementioned problem within the constraints of energy consumption and costs would be (when the acoustic signal is repeatable over time in successive steps) to perform several successive measurements respectively on consecutive acoustic signals in the same propagation direction by shifting the sampling signal relative to the acoustic signal when performing a new measurement in order to obtain, after these different measurements several sampled points offset in the vicinity of the maximum amplitudes P- and P + of the positive and negative lobes .

For example, the first measurement is performed normally sampling the received acoustic signal and the second measurement is performed on the second signal ^~ sound by shifting the sampling signal relative to said second acoustic signal to a predetermined value l / 2Fe-

It is thus possible to perform several successive phase shifts (1 / nFe value for n phase shifts) on the second acoustic signal and the signals following acoustical to get more points sampled in the vicinity of the maxima and have an even greater accuracy these maximum amplitudes P- and P +

It is also advantageous to combine the use of a maximum approximating algoπthme (in absolute value) with this solution to further increase accuracy.

This is followed by a comparison of the relative amplitudes of these maximum values ​​P- and P + Δ coπespondant the report of the ideal péπode.

If the result of the comparison is supéπeur to a threshold value G_s, Consider the penod considered as a parasite péπode coπespondant to noise, and continue the analysis on the following péπode

If the result of the comparison is less than the threshold value G_s, Considering the period under review as a caractéπstique péπode.

reliably then we can determine the zero-crossing partner has examined this péπode. This zero-crossing considered the first meaningful zero-crossing of the received acoustic oscillations caracténstiques signal may advantageously be the zero crossing between the two positive and negative lobes caractéπstique péπode.

Figure 5 shows a first oscillation caracteπstique O_it penod of DT with a positive lobe whose amplitude has a maximum value P + and a negative lobe whose amplitude has a maximum value in absolute value P- on associated zero crossing in this péπode .DELTA.T a first oscillation caractenstique O_it defines a zero-crossing time to time coπespondant

T₇ The amplitude ratio between the amplitudes of the maximum values ​​P and

P + of the two lobes of an examined péπode the received acoustic signal is an independent comparison criterion of the nature of the gas. Thus, the electronics aging can result in gain variations which will not affect the amplitude ratio between the maximum amplitudes P- and P + The pattern recognition of the received acoustic signal can be performed in a simple manner which reduces energy consumption, since the comparison is between the only maximum amplitudes P- and P + lobes and théonque Δ amplitude ratio between the maximum amplitudes Pi and Pι + of the two lobes of a penod

^"caractéπstique ideal. The amplitude ratio théonque Δ is predetermined from an average for several different gases and at different rates, the report between the maximum amplitudes P- and P + of the characteristic period observed from the recording of acoustic signals received.

In a particular embodiment, the comparison between the ratio of the amplitudes P- and P + of the reporting period and the amplitude ratio coπespondant the ideal period is performed by computing a similarity criterion G which constitutes the result of the comparison compared to the threshold value G_sAnd is defined as follows:

G = (P-) - _Δ (P +) (P)

By performing comparison taking into account the maximum amplitude ratios of the lobes of a period of the received acoustic signal, the Δ theoretical reference amplitude ratio is not affected by the nature of the gas or the gain of the amplifier chain of the received signal. This helps to facilitate the measurements.

However, pouπait be among the noise signal periods that constitute noise waveforms resembling to that defined by the comparison criterion G. To make the detection of characteristic periods less sensitive to risk to take into account forms resembling waves drowned in noise and does not constitute a first period O_it the characteristic oscillations O_c the received signal, the similarity criterion can be made slightly dependent on the amplitude P-.

Thus, according to another particular embodiment, a comparison between the ratio of amplitudes P- and P + of the reporting period and coπespondant amplitude ratio of the ideal period is performed by computing a similarity criterion G which constitutes the result of comparison compared with the threshold value G_sAnd is defined as follows

G = (P-) - _Δ (P +) (P) -z

where z represents an offset term for making the similarity criterion G slightly dependent on the amplitude of the maximum value (P-) of the negative lobe of the characteristic period examined. _ In this case, the denominator of the criterion G is greatly reduced when the amplitude P is close to the offset term z. If the offset term z has a value slightly greater than the maximum amplitude of the noise (e.g. of the order of 0.20 to 0.25 volts), this allows faster a noise rejection signal periods received. To determine the value of G_s which should be compared the similarity cπtère G, and the value of the offset term z, draw up curves first network representing the values ​​of G resemblance cπtère for caractéπstiques péπodes observed and a network of second curves representing values minimum similarity criterion G for the observed spurious péπodes, according to different possible values ​​of the offset term z, and one chooses the threshold value G_s and that of the offset term z, and one chooses the threshold value G_s and that of the offset term z as a function of the margin sécuπte possible in the intermediate region between the first network and the network of curves second curves.

By way of example, curves of the families 1 1 1 and 121 of Figure 6 were obtained by vaπer the phase difference between the sampling frequency and the signal and applying a gain factor between 0.45 and 1 Vanant 5, the gas pπs account being methane and a mixture of 20% carbon dioxide and 80% nitrogen, the measures were made at zero flow and 7m-vTime.

Networks curves 1 1 1 and 121 are used to select the values ​​of the z offset term and the threshold value G_s depending on the desired sécuπté margin on these values. If we consider Figure 6, it is seen that one can choose for G_s for example compnse a value between about 0.7 and 1, 7 and V, to the offset term z, for example compnse a value between 0.21 and 0.25 V

If, for example G is chosen_s = 2 and z = 0.22 V; margin sécunté compnse z is between 0.20 and 0.25 V, while the margin for sécuπté G_s compnse is between 1 and 4 practice, with this example, a review period will be considered caracteπstique if G <G_s = 1.5 Other vanantes embodiment can further increase the safety and reliability of the process.

Thus, in one particular embodiment, the comparison between the ratio of amplitudes P- and P + of the reporting period and coπespondant amplitude ratio of the ideal péπode is performed by calculating a similarity cπtère G which is the result of comparison compared to the threshold value G_sAnd is defined as follows

G = (P-) - _Δ [(P +) -z '] (P-) -z

where z represents an offset term for making the similarity of cntère G slightly dependent on the amplitude of the maximum value (P-) of the negative lobe of the caractéπstique peπode examined, and z 'represents an offset term for making the similarity criterion G slightly dependent on the amplitude of the maximum value (P +) of the positive lobe of the caractéπstique péπode examined

This reduces the sensitivity of resembling periods of low amplitude and, without substantially increasing the computational complexity, can decrease the gain of the amplifier chain to values ​​as low as 0.35, which saves energy.

According to another vaπante, and whatever the formula chosen for the similarity criterion G is carried out for each examined péπode, two or three computations of the similarity criterion G with values ​​of the ratio of different amplitudes of neighboring Δ report amplitude théonque predetermined and does not deviate therefrom by more than 10%, that we consider the péπode examined as a péπode caracteπstique if the similarity calculated cntère G is in all cases the value inféπeur threshold G_s Figure 7 is a flowchart showing an exemplary implementation process of the method according to the invention.

If we consider a sampling frequency Fe has a, for example 320 kHz which is eight times the sound frequency Fa of the transducers, the sampled signal will comprise 8 points penod signal received Insofar as the received acoustic signal has the péπodiques form of oscillations, successive minima and maxima are known or, avoiding to look at all the sampled points

We begin after the initialization phase 201, a look at a step 202, the next local minimum (P- of amplitude) During this step 202, if one applies a G resemblance cπtere involving an offset term z, we retain only the first local mimmum whose amplitude P- is supéπeure in absolute value to offset term z as the first local minimum was found, it proceeds to step 203 which is to search in the vicinity of half péπode T / 2 previous, the value of the maximum amplitude P +

In the next step 204, one calculates the value of G cntère resemblance with the values ​​previously find the minimum P and the maximum P + of the examined péπode which has at least a minimum whose absolute value is

^"supéneure the term z A shift step 205, is performed a comparison test between the calculated cπtere G and the threshold value G_s If G> G_s for the review period, we consider that it is a parasite period, and there is a return to step 202 to search for a next local minimum.

If G <G_s for the period under review, it is considered a characteristic period and proceeds to step 206 to determine the next zero crossing of the amplitude of the received Signai which will be the zero crossing between the local maximum P + and the minimum local P and will be the true starting point of characteristic oscillations O_c the received acoustic signal.

The same process is repeated moments later for a new measurement for determining the propagation time of an ultrasonic acoustic signal.