Filtering method for leak test cycles and apparatus configured to implement method

Abstract

The invention relates to a filtering method (200) for a leak test cycle of a certain type of object to be tested, for constructing a database capable of determining at least one reference value relating to a leak test of said object type, in particular suitable for an early leak detection method; the invention relates to a method (200) for measuring a test cycle comprising the measurement of a physical quantity related to the degree of leakage in at least one part of an object to be tested or in a peripheral cavity of the object to be tested, said method (200) comprising at least the following steps: determining (E2) a deviation between the measured values of the physical quantity at the starting time and the ending time of the test cycle, referred to as a total deviation; the total deviation is compared (E3) with a total deviation reference value.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of leak detection methods for the containment control of objects, in particular to leak detection methods based on pressure differential. The present application also relates to leak detection devices configured to implement the above-mentioned methods.

[0002] The present application also relates to a filtering method for leak test cycles of a type of object, for building a database able to determine at least one reference value related to the leak test of said type of object, in particular for the early leak detection method of the present application. BACKGROUND

[0003] Various systems and methods for detecting leaks are known, such as the tracer gas method, the soap bubble method, etc. In the present application, the present application more specifically relates to leak detection based on pressure differential (or variation thereof).

[0004] When leak detection is performed using the pressure differential method, the test object whose containment level is to be controlled is subjected to a controlled pressure differential. In particular, this means that a pressure variation (positive or negative) is generated in a specific internal volume of the object (direct method) or in a closed volume surrounding the object (indirect method), for example with respect to the outside of the object.

[0005] After a preset time period, the pressure variation in the specific volume is measured over a preset time period, in order to determine whether the object has a leak, and sometimes to quantify the degree of leak of the test object.

[0006] In fact, a leak causes a pressure variation over a specific time period, and in particular the pressure variation per unit of time can be associated with the degree of leak, for example through the following mathematical relationship: where F is the leak rate, usually expressed in standard cubic centimetres per minute (or scc / min); ΔΡ is the pressure variation measured in the relevant (or specific) volume, expressed in Pascals (Pa); Δt is the time period over which the pressure variation ΔΡ is measured (expressed in seconds); V is the relevant volume to be considered (for example the internal volume of the object), usually expressed in cubic centimetres (cm 3 ); k is a multiplicative constant (expressed in Pa -1 ). It should be noted that the leak rate can be expressed in other ways, for example in the form of mass flow. Therefore, the formula relating the leak rate to a certain physical quantity can have various forms depending on the measurement method used and the physical quantity studied.

[0007] Therefore, it is possible to determine whether the object, or any of its sub-components, has a leak and to determine its containment level (or degree of leak), regardless of the type of object.

[0008] The test object can be an electronic device, a package, a container, etc. The tolerance requirements for the level of tightness (or degree of leakage) can therefore vary greatly depending on the test object itself, its volume, shape and / or function.

[0009] However, when testing the tightness of an object, environmental parameters and / or temporary disturbances of various kinds can make it difficult to track the pressure variations in real time and / or limit the repeatability of such measurements.

[0010] This problem is particularly acute when detecting leaks in an industrial environment, such as a factory, since the temperature and / or pressure in such an environment can vary spatially and / or temporally depending on the operation of the test object or the operation in the vicinity of the leak detection device.

[0011] In addition, the detection device can be designed to test the tightness of objects on an assembly or production line. In this case, it is necessary to have a tightness control that is as fast, accurate and repeatable as possible so as not to disturb (or slow down) the assembly or production line. SUMMARY

[0012] The present invention aims to provide, with the aid of a leak detection device, a new method for early detection of leaks in an object, in order to solve at least one of the problems mentioned above, the method comprising the following steps: - applying a pressure differential to at least one part of the test object or to a closed space surrounding the test object, the pressure differential being respectively relative to the outside of the test object or to at least one part of the test object; - measuring, during a test cycle of predetermined duration, a physical quantity related to the degree of leakage in at least one part of the test object or in a closed space surrounding the test object; - comparing the measured value of the physical quantity with a first reference extreme value; - determining whether the test object has a leak or not, as a function of the comparison between the measured value and the first reference extreme value; - interrupting the test cycle if it is determined that the test object has a leak.

[0013] The method according to the invention thus makes it possible to determine, in advance (i.e. before the end of the full duration of the test cycle), whether the test object has a leak or not, or whether the degree of leakage is above a predetermined threshold. According to experimental observations, the duration of the test cycle can be reduced by between 10% and 70% using the method according to the invention.

[0014] According to one possible feature, the method comprises the following step: - comparing the measured value of the physical quantity with a second reference extreme value; - determining whether the test object is free from leakage based on the comparison between the measured value and the second reference extreme value; - interrupting the test cycle if it is determined that the test object is free from leakage.

[0015] The setting of the second reference extreme value allows to frame the measured value, wherein the first reference extreme value is used to decide in advance whether the object is leaking or not, and the second reference extreme value is used to decide in advance whether the object is free from leakage. Therefore, a shorter average test duration can be obtained to decide whether the object is leaking or not, or whether the sealing requirements defined for the considered object type are met.

[0016] According to another possible feature, the first reference extreme value and the second reference extreme value define a reference value interval, wherein the measured value is compared with said reference interval, and once the measured value exceeds said reference interval, it is decided that the object is leaking; as long as the measured value is between the boundaries of the reference interval, the leakage test is continued until the test cycle is completed.

[0017] According to another possible feature, the first reference extreme value R MAX is defined by the following equation: R MAX = R L + M C + k1 x I D ; wherein R L is a leakage threshold value, M C is a measure of the concentration tendency of the compensation values, I D is a measure of the dispersion, and k1 is a positive real constant.

[0018] According to another possible feature, the second reference extreme value R MIN is defined by the following equation: R MIN = R L + M C - k2 x I D ; wherein R L is a leakage threshold value, M C is a measure of the concentration tendency of the compensation values, I D is a measure of the dispersion, and k2 is a positive real constant.

[0019] It should be noted that the leakage threshold value R L corresponds to a set (or preset) leakage value used to decide whether the object is sealed or leaking. In the context of a regular test cycle (i.e. without the early leakage detection method), after a preset duration and regardless of transient phenomena that can affect the measured value, the variation of the physical quantity measured per unit of time corresponds (directly or indirectly) to the degree of leakage of the test object and is determined, and then compared with the leakage threshold value R LThe comparison is made to determine whether the object is sealed.

[0020] It should also be noted that the measured values of the physical quantity, for example determined as a degree of leakage or a rate of leakage, for example in Pascals per second, Pa / s, can be subject to drift, for example due to the influence of environmental parameters (temperature, humidity, season, etc.), which can vary over the course of a day or over the course of a year. Therefore, when determining the first reference extreme value and / or the second reference extreme value, these environmental parameters must be taken into account and compensated for in order to determine as accurately as possible whether the test object has a leak, whether the degree of leakage is higher than a preset threshold, and / or, conversely, whether the test object can be considered to be sealed (or at least has a degree of leakage lower than a preset value or a leakage threshold).

[0021] According to another possible characteristic, the central tendency measure and the dispersion indicator are determined on the basis of a plurality of tests carried out on objects of the same type as the test object, i.e. identical or similar objects.

[0022] Advantageously, the central tendency measure and the dispersion indicator are empirical values, determined on the basis of a large number of identical or structurally similar objects, and / or on the basis of the behaviour of these objects in the method for detecting leaks. This is designed to determine the central tendency measure and the dispersion indicator as accurately as possible, ensuring that the method of the application is able to determine in advance, with high precision and accuracy, whether the test object has a leak or is sealed.

[0023] According to another possible characteristic, the central tendency measure is the arithmetic mean or the median, while the dispersion indicator is the standard deviation or the interquartile range, respectively.

[0024] According to the type of distribution of the possible measured values of the physical quantity, the central tendency measure is the arithmetic mean or the median, while the dispersion indicator is the standard deviation or the interquartile range, respectively.

[0025] In particular, when the distribution of the measured values of the physical quantity is a normal distribution, i.e. the statistical distribution of the values complies with a normal distribution, the central tendency measure is the arithmetic mean and the dispersion indicator is the standard deviation. Conversely, if the distribution of the measured values of the physical quantity is a non-normal distribution, the central tendency measure is the median and the dispersion indicator is the interquartile range.

[0026] According to another possible characteristic, the constants k1 and k2 are equal to each other and are comprised between 1 and 5, preferably equal to 3.

[0027] The values of the constants k1 and k2 make it possible to set the level of confidence regarding the decision as to whether there is a leak or whether the seal is intact; generally speaking, the higher the values of k1 and k2, the more accurate the method of the application, but at the expense of the lengthening of the test cycle brought about by the method. It should be noted that the higher the values of k1 and k2, the less the instantaneous phenomena (such as air flow, human intervention, etc.) are taken into account or have an impact on the early decision, but at the expense of the lengthening of the test cycle brought about by the method.

[0028] According to another possible feature, if the measured value of the physical quantity at the start of the test is substantially equal to one of the last measurement values of the previous test cycle (preferably the last measurement value of the previous test cycle, the selected measurement value being sufficient to be representative of the measurement value at the end of the test cycle), the comparison of the measurement values with the one or more threshold values is interrupted and the test cycle is continued until completion, at which time the degree of leakage is determined and compared with the preset threshold value.

[0029] During a test process in an industrial environment, such as a production line, various unexpected situations can occur, resulting in the same object being tested twice. In this case, it is necessary to deactivate the early leak detection method and to complete the test cycle in order to determine whether the test object has a leak or whether the degree of leakage is higher than the preset threshold value (or the leak threshold value R L ), in particular because the test object is in a stable state and its measurement values can no longer be effectively compared with the reference threshold value.

[0030] According to another possible feature, the method comprises the following steps: - calculating a function on the basis of two different measurement values of the physical quantity; - comparing the calculated function with a first reference threshold value; - depending on the result of the comparison of the calculated function with the first reference threshold value, either continuing the early leak detection method or completing the test cycle, for example with an alarm indicating a fault.

[0031] The calculation of a function on the basis of two measurement values of the physical quantity, successive or not but different from one another, and the comparison of this function with a reference function make it possible to determine whether the test cycle (and the leak test of the object) is proceeding normally and without any fault that could cause the early leak detection method to fail. Advantageously, the calculated function is related to or indicates a parameter relating to the shape of the curve of the physical quantity, i.e. to the temporal evolution of the measurement values of the physical quantity.

[0032] Otherwise, the early leakage detection method is deactivated and the test cycle is completed, with an alarm indicating a fault, so that the operator in charge of monitoring the leakage detection method is aware that an inspection is required and possibly retest the object later (otherwise, the object can be qualified as not meeting the preset tightness requirements).

[0033] According to another possible feature, the computed function is compared with the second reference extreme value function; if the comparison of the computed function with the second reference extreme value function is true, the early leakage detection method is continued, otherwise the test cycle is completed, for example with an alarm indicating a fault.

[0034] The setting of the first and second reference extreme values enables to define an interval within which the computed function is qualified as accurate or normal and no fault is present that would cause the leakage test to fail.

[0035] According to another possible feature, the first reference extreme value function is computed by the following formula: F MAX = F m + k3 x I F ; where F m is a measure of central tendency of the computed function, I F is a measure of dispersion of the computed function, and k3 is a positive real constant.

[0036] According to another possible feature, the second reference extreme value function is computed by the following formula: F MIN = F m - k4 x I F ; where F m is a measure of central tendency of the computed function, I F is a measure of dispersion of the computed function, and k4 is a positive real constant.

[0037] As mentioned above, the measure of central tendency of the computed function is the arithmetic mean or the median (for the function), and the measure of dispersion of the computed function is the standard deviation or the interquartile range, respectively. The choice of the type of measure of central tendency and the type of measure of dispersion advantageously depends on the type of distribution of the computed function (normal or non-normal).

[0038] Moreover, the constants k3 and k4 are advantageously equal to each other, for example, according to a desired level of precision or accuracy, the value of which is at least equal to 2, preferably at least equal to 3, 4 or 5. It should be noted that the higher the value of k3 and k4, the less the influence of transient phenomena (such as air flow, human intervention, etc.) is taken into account or has on the early decision, but at the expense of the time reduction effect brought by the method.

[0039] According to another possible feature, said calculation function is the difference and / or the ratio between the two measurements of said physical quantity.

[0040] Advantageously, the function calculated on the basis of the two measurements of said physical quantity, consecutive or not but different from each other, is related to the derivative of said curve or to the time evolution of said physical quantity measurements.

[0041] The present application also relates to a method for the selection of leak test cycles for a specific type of test object, for the construction of a database able to determine at least one reference value related to the leak test of objects of that type, in particular for the early leak detection method of the present application described above. Wherein a test cycle comprises the measurement of a physical quantity related to the degree of leak of at least one portion of the test object or of the closed space surrounding the test object, said selection method comprises at least the following steps: - calculation of the difference between the measurements of the physical quantity at the beginning and at the end of the test cycle, called total difference; - comparison of the total difference with a reference total difference, followed by the optional determination of whether the test cycle can be included in said database for the determination of at least one reference value.

[0042] By calculating the difference between the measurements of the physical quantity at the beginning and at the end of the test cycle, the selection method of the present application allows to quickly eliminate all the anomalous leak test cycles, thus avoiding the creation of a database that does not comply with the normal and / or average test conditions of said objects.

[0043] It should be noted that the beginning and the end of the test cycle refer to representative values of the physical quantity measured at the beginning and at the end of the test; these values can be the values measured at specific points in time (for example after having empirically determined their stability and representativeness) or can be the average of the first and last measurements.

[0044] According to a possible feature of said selection method, the reference value of the total difference is a measure of central tendency calculated from the test cycles of said database.

[0045] Advantageously, the measure of central tendency and the dispersion indicator are empirical values, based on a large number of objects of the same type or of similar structure and / or on the performance of these objects in the leak detection method. This design choice aims to determine the measure of central tendency and the dispersion indicator as accurately as possible, ensuring that said selection method is able to generate a database that effectively selects test cycles and acquires reference values that ensure the correct operation of the early leak detection method described above.

[0046] According to another possible feature of said selection method, the dispersion indicator is a reference value for the total difference, calculated from the test cycles of said database.

[0047] According to another possible feature of the screening method, the comparison of the total difference values of the test periods is implemented by verifying whether the total difference values are within an interval delimited by a central tendency measure and a dispersion indicator calculated.

[0048] According to another possible feature of the screening method, the central tendency measure of the total difference values is the arithmetic mean or the median, while the associated dispersion indicator corresponds to the standard deviation or the interquartile range, respectively.

[0049] According to the type of distribution of the possible measurement values (normal or non-normal) and of the total difference values, the central tendency measure of the total difference values is the arithmetic mean or the median, while the associated dispersion indicator corresponds to the standard deviation or the interquartile range, respectively.

[0050] According to another possible feature of the screening method, the method further comprises the steps of: - calculating at least two difference values between successive different measurement values of the physical quantity during the test period, referred to as intermediate difference values; - comparing each of the intermediate difference values with a corresponding reference value, the reference value corresponding to each intermediate difference value; and then, optionally, determining whether the test period can be included in the database for determining at least one reference value.

[0051] Advantageously, by cumulating several criteria of judgment, it is possible to accurately assess the relevance of the inclusion of a test period in the database.

[0052] According to another possible feature of the screening method, the reference value of an intermediate difference value is a central tendency measure calculated from the test periods of the database.

[0053] According to another possible feature of the screening method, for each intermediate difference value, a dispersion indicator is calculated from the test periods of the database.

[0054] According to another possible feature of the screening method, the comparison of the intermediate difference values of the test periods is implemented by verifying whether each of the intermediate difference values is within an interval delimited by a corresponding calculated central tendency measure and dispersion indicator.

[0055] According to another possible feature of the screening method, the central tendency measure of each intermediate difference value is the arithmetic mean or the median, while the associated dispersion indicator corresponds to the standard deviation or the interquartile range, respectively.

[0056] According to the type of distribution of the possible measurement values (normal or non-normal) and of the intermediate difference values, the central tendency measure of each intermediate difference value is the arithmetic mean or the median, while the associated dispersion indicator corresponds to the standard deviation or the interquartile range, respectively.

[0057] According to another possible feature of the screening method, the method further comprises the step of: - comparing the measurement of the physical quantity at the beginning of the current test cycle with the measurement of the physical quantity at the end of the previous test cycle, said test cycles being consecutive test cycles; - determining whether the current test cycle can be included in the database for the determination of at least one reference value; wherein, if the two measurements are substantially equal, the test cycle is not included in the database.

[0058] Advantageously, by checking a plurality of parameters it is possible to determine the relevance of the test cycles to be included in the database, with the aim of avoiding repeated tests on the same object. In this scenario, it is necessary to avoid that both test cycles on the same object are included in the database, thus guaranteeing the relevance of the database constructed.

[0059] According to another possible feature of the screening method, said comparison of the measurements of the physical quantity between two consecutive test cycles is implemented on the basis of a measure of central tendency and a measure of dispersion associated with each of said measurements.

[0060] According to another possible feature of the screening method, the measure of central tendency used for the comparison is the arithmetic mean or the median, while the measure of dispersion associated with the comparison of the measurements between two consecutive test cycles is the standard deviation or the interquartile range.

[0061] The present application also relates to a leak detection device configured to implement the early leak detection method described above.

[0062] The present application also relates to an electronic device, such as a computer, or a leak detection device configured to implement the screening method of the leak test cycles described above. BRIEF DESCRIPTION OF DRAWINGS

[0063] The present application will be better understood with the following description of specific embodiments thereof, given by way of example only and with reference to the accompanying drawings. Said embodiments are not limitative, and reference is made to the attached drawings, in which: Figure 1 , denoted by Figure 1 , is a highly simplified schematic view of a leak detection device according to the present application; Figure 2 , denoted by Figure 2 , is a graph showing an example of pressure variation during a leak detection process by means of the device shown in Figure 1 ; and Figure 3 , denoted by Figure 3 , is a flow chart showing the leak detection method according to the present application. Figure 4 , denoted as Figure 4 , is a flow chart illustrating the leak detection method according to an embodiment of the application; Figure 5 , denoted as Figure 5 , is a measurement curve of the physical quantity as a function of time during a continuous test period; Figure 6 , denoted as Figure 6 , is a flow chart illustrating the leak detection method according to another embodiment of the application; Figure 7 , denoted as Figure 7 , is a flow chart illustrating the test period filtering method according to the application; Figure 8 , denoted as Figure 8 , is a flow chart illustrating the test period filtering method according to another embodiment of the application. DETAILED DESCRIPTION

[0064] Figure 1 is a highly simplified schematic view of a leak detection device 1 for detecting the tightness of an object 10. The test object can be any object whose tightness is to be detected: a package, a heat exchanger, a cell phone, a power battery for driving a car, etc.

[0065] The device 1 thus comprises: - a pressurization or vacuum system 5 for controlling a characteristic volume with respect to the test object 10, i.e. this characteristic volume can be the internal volume of the test object (direct method) or an enclosed volume around the object (indirect method); - a first pressure sensor 7 configured to measure the pressure variations of the characteristic volume, the sensor 7 enabling the detection of the tightness of the test object 10; - a second pressure sensor 17, optionally, configured to measure the pressure applied by the pressurization or vacuum system 5 to the characteristic volume; - a gas duct connection 11, e.g. a pipe or a hose, configured to connect the pressurization system 5 to the test object 10 and to a reference 13; - an electronic unit 15, e.g. an electronic circuit, connected to each pressure sensor 7, 17 and configured to acquire the pressure values measured by the sensors 7, 17.

[0066] Advantageously, the pressurization or vacuum system 5 comprises a pressure (or depression) source 51, which can be, for example, a pump, a compressor, a compressed air or gas supply source.

[0067] ​The first pressure sensor 7 is preferably a differential pressure sensor, while the second pressure sensor 17 is advantageously an absolute pressure sensor. In a variant not shown, the first pressure sensor 7 is an absolute pressure sensor, and the leak detection device is not provided with a reference piece 13.

[0068] The leak detection of the object 10 by means of the device 1 shown in Figure 1 is considered as a test cycle C L , which is divided into four main steps, as shown in Figure 2 : Filling step I: the test object's characteristic volume is filled with compressed air (or any gas, preferably an inert gas, such as nitrogen), so that the pressure rises to a target pressure value P1; Stabilization step II: after the test object's characteristic volume is pressurized, it is necessary to return it to a state of thermal and mechanical equilibrium, usually for a predetermined duration, also known as stabilization time, in order to reduce the interference of transient phenomena on the leak measurement; Test step III: at the end of this step, the measurement of the pressure variation over time in the characteristic volume is taken, and thus the degree of leak of the test object is determined; Evacuation step IV: the pressurized characteristic volume of the test object is returned to ambient pressure.

[0069] It should be noted that the leak detection can also be carried out in a vacuum environment (or low pressure conditions), i.e. the first step I is no longer to raise the pressure, but to lower the pressure in the test object's characteristic volume to a predetermined value. Steps II and III remain unchanged, while the fourth step (step IV) is to raise the pressure in the test object to a pressure value that matches the ambient pressure. Thus, in the pressure and vacuum leak detection processes, the filling step I and the evacuation step IV can be considered as "inverted".

[0070] The present application is a differential pressure early leak detection method based on the device 1, which has significant advantages, especially in the stabilization step II and / or the test step III (detailed below) of the test cycle C L . One of the purposes of the present application is to shorten the test cycle duration and achieve as fast as possible leak detection, and thus to determine as early as possible and accurately whether the test object has a leak.

[0071] The method according to the present application, as shown in Figure 3 , comprises the following steps: S1 Connection: connecting the test object 10 to the leak detection device 1; S2 pressure regulation and differential establishment: the pressure within at least one portion of the test object or within the closed space surrounding the test object 10 is changed, and a pressure differential is established between a portion of the test object 10 and the outside of the object, or between the closed space surrounding the test object 10 and at least one portion of the test object 10; S3 physical quantity measurement: within at least one portion of the test object 10 or within the closed space surrounding the test object 10, a physical quantity R u related to the degree of leakage F is measured; MAX S4 numerical comparison: the measured value of the physical quantity R MIN is compared with a first reference extreme value R MAX and / or with a second reference extreme value R u ; S5 leakage determination: on the basis of the comparison between the measured value of the physical quantity R MAX and the first reference extreme value R u , it is determined whether the test object 10 has a leakage F MIN ; and / or on the basis of the comparison between the measured value of the physical quantity and the second reference extreme value R MIN , it is determined whether the test object has no leakage; S6 test interruption: if it is determined that the test object has a leakage or has no leakage (i.e. it is considered to be leak-tight), the test cycle is interrupted.

[0072] It is specified that the physical quantity can be a pressure, a flow rate or a pressure variation over time (in particular a pressure variation per unit of time).

[0073] According to an embodiment variant not shown in the figures of the present application, the physical quantity variation over time is measured by means of an adapted leak detection device, for example the device 1, i.e. a leak detection device configured to measure these physical quantities, for example comprising pressure sensors, flow meters, etc.

[0074] The first reference extreme value R MAX and the second reference extreme value R MIN together define a reference value interval I R . In which the measured value of the physical quantity R R is compared with said reference value interval I R : as soon as the measured value R R goes beyond said reference value interval I MAX , it is determined whether the test object has a leakage; if the measured value R MIN is between the limits R MAX and R MIN of the reference value interval I R , the leakage test continues until the end of the test cycle.

[0075] Therefore, the first reference extreme value R MAX and the second reference extreme value R MIN are parameters for defining the measured value R MAX and are defined respectively by the following formulas: The first reference extreme value R MAX is defined by the following formula:MAX : R MAX = R L + M C - k1 x I D ; wherein R L is a leakage threshold value, M C is a measure of the concentration tendency of the compensation value, I D is a measure of the dispersion, and k1 is a positive real constant; a second reference extreme value R MIN : R MIN = R L + M C - k2 x I D ; wherein R L is a leakage threshold value, M C is a measure of the concentration tendency of the compensation value, I D is a measure of the dispersion, and k2 is a positive real constant.

[0076] In particular, the leakage threshold value R L is a leakage threshold level (predefined according to the desired level of tightness), i.e. the test object 10 is considered to be leaking when it reaches this threshold value. That is, if the degree of leakage of the test object is higher than the predefined leakage threshold value R L , the test object is determined to be leaking; and if the degree of leakage of the test object is lower than the predefined leakage threshold value R L , the test object is determined to be not leaking (or, in other words, to be tight).

[0077] The measured value of the physical quantity R can be defined as the degree of unfitness or the leakage rate. For example, it is in Pa / s, but this measured value can have an offset. For example, due to the influence of environmental parameters (temperature, humidity, etc.), this offset can change within a day; it can also simply be that the pressure (which has changed) has not yet stabilized.

[0078] Therefore, when setting the first reference extreme value R MAX and / or the second reference extreme value R MIN , the offset needs to be taken into account and compensated for by means of the compensation value M C , in order to determine as precisely as possible whether the test object is leaking, i.e. the degree of leakage is higher than the predefined threshold value R L , or, on the contrary, whether the test object can be considered to be tight, i.e. at least the degree of leakage is lower than the predefined leakage threshold value R L .

[0079] It should be noted that the measurement step S3 of the physical quantity R in relation to the degree of leakage F u can be repeated several times, for example periodically, as long as the physical quantity R is in the first reference extreme value RMAX With the second reference extreme value R MIN This applies when the measured value R is insufficient to determine whether the test object is properly sealed or if there is a leak. Therefore, each measured value R in measurement step S3 needs to be considered. i Determine the first reference extreme value R respectively MAXi With the second reference extreme value R MINi .

[0080] In this way, for example through empirical methods, it is possible to determine the optimal value for each measurement R. i Determine the first reference extreme value R MAXi With the second reference extreme value R MINi The component parameters, namely the central tendency measure M that determines the compensation value. Ci and the associated dispersion index I Di .

[0081] To complete this determination process, multiple sealing tests must be performed on objects similar to or identical to the test object (where "similar object" refers to objects that behave similarly or identically to the test object in the leakage detection method) throughout a complete test cycle to obtain the most probable value of the measurement at a specific time point, and then determine the central tendency measure M of the compensation value. C And the dispersion index I associated with the measurement step S3 at a specific time point. D .

[0082] It should be noted that the central tendency measure M C The measured value can be the arithmetic mean or the median, while the associated dispersion index I... D These correspond to the standard deviation or interquartile range, respectively.

[0083] In fact, when establishing the measured value R in measurement step S3 i When accessing the database, if the measured value R i If the statistical distribution of the compensation value is normal (i.e., the distribution is normal), then the central tendency measure M of the compensation value is... C It is the arithmetic mean, and the dispersion index I D The standard deviation is denoted as .

[0084] However, if the measured value R i If the statistical distribution of the compensation value is non-normal, then the central tendency measure M is... C It is the median, and the dispersion index I D It is the interquartile range.

[0085] Meanwhile, the constants k1 and k2 are advantageously equal to each other, and the values ​​of the constants k1 and k2 range from 1 to 5, preferably 3.

[0086] First reference extreme value R MAXWith the second reference extreme value R MIN A numerical interval can be defined as a confidence interval (i.e., the range of the confidence interval is determined by the values ​​of constants k1 and k2), within which the probability that the measured value R is the true value meets a preset percentage. Therefore, when constants k1 and k2 are equal to 3, the probability that the measured value R is true and reliable is 99.73%; while when constants k1 and k2 are equal to 4, the probability that the measured value R is true and reliable is 99.993%.

[0087] In method 100, such as Figure 4 In the illustrated implementation variant, the method further includes step S7: placing the test object, i.e., period C i The physical quantity value R measured in the initial stage of the test is compared with the previous test object, i.e., the period C. i-1 The physical quantity value R measured at the end of the test is compared, where the period C i With period C i-1 It consists of two consecutive cycles.

[0088] Step S7 uses a schematic diagram of two consecutive cycles in [ Figure 5 The text specifically illustrates that if the period C... i The measured physical quantity value R and period C i-1 If the measured physical quantity values ​​R are basically equal, that is, the difference between the two values ​​is less than 10%, preferably less than 5%, then the comparison of measurement values ​​based on one or more extreme values ​​is interrupted, i.e., step S1, and the long test cycle C is completed. L After a specific (or preset) time period, determine the leakage level of the test object and compare it with a preset threshold.

[0089] Therefore, the early leak detection method 100 was discontinued, and the test cycle C was completed. L To determine whether the test object has a leak, or whether the degree of its leak exceeds a preset threshold R. L It should be noted that step S7 is advantageously performed during test cycle C. L Complete it as early as possible to maximize time savings if the test subject fails.

[0090] In another embodiment of the early leak detection of the present invention, the method includes at least: Step S8 (Function Calculation Step): Calculate the function F between two different measured values ​​of the physical quantity R; Step S9 (Function Comparison Step): Compare the calculated function F with at least one reference function F associated with function F. MIN and / or F MAX Compare them.

[0091] Implementation variants of this method 100'' are illustrated in the flowchart [ Figure 6are shown in detail in the description below; therefore, in addition to the steps of the method 100' shown in the figure, this variant comprises steps S8 and S9. Figure 5

[0092] said calculated function F(R i ,R i-1 ) is, for example, the difference between two measurements R i ,R i-1 of the physical quantity R, which can be consecutive or not, but distinct from each other. Advantageously, the function F(R i ,R i-1 ) is proportional to the derivative of the curve of the measurements of the physical quantity R or to the trend over time. For example, the function F(R i ,R i-1 ) can be of the form k(R i -R i-1 ), where k is a multiplicative constant.

[0093] The calculation step S8 of the function F can be based on the same measurement data of the physical quantity R, provided that at least two measurements of the physical quantity R have been made (i.e. one iteration of step S3 has been performed); or, alternatively, on additional measurements of the physical quantity R, i.e. through additional measurements in parallel to determine whether the test object 10 is leaking (in accordance with steps S4 and S5).

[0094] The calculated function F(R i ,R i-1 ) is compared with the first reference extreme function F MAX and / or the second reference extreme function F MIN by means of step S9.

[0095] It is specified that said first reference extreme function is calculated from the formula: F MAX =F m +k3x I F ; where F m is a measure of the central tendency of the calculated function, I F is an index of dispersion with respect to the calculated function and k3 is a positive real constant.

[0096] At the same time, said second reference extreme function is calculated from the formula: F MIN =F m -k4x I F ; where F m is a measure of the central tendency of the calculated function, I F is an index of dispersion with respect to the calculated function and k4 is a positive real constant.

[0097] ​As mentioned above, the concentration measure F of the calculation function m may be the arithmetic mean or the median, while the dispersion indicator I of the calculation function F corresponds to the standard deviation or the interquartile range, respectively. The concentration measure F of the calculation function m and the dispersion indicator I F are chosen depending on the type of distribution of the calculation function (normal or non-normal).

[0098] Furthermore, the constants k3 and k4 are advantageously equal, and have a value of at least 2, for example, and preferably at least 3, depending on the desired accuracy.

[0099] According to the result of the comparison between the calculation function F(R i ,R i-1 ) and the first reference extreme value function F MAX and / or the second reference extreme value function F MIN , either the early leak detection method 100'' is continued, or the test cycle C L is completed, and an indicative alarm is emitted, for example, to an operator.

[0100] In particular, if the calculation function F(R i ,R i-1 ) is within the reference interval [F MIN ; F MAX ], it is considered that no abnormal situation has occurred that would cause the leak test to fail, and the leak detection method is continued. If, on the other hand, the calculation function F(R i ,R i-1 ) is outside the reference interval [F MIN ; F MAX ], it is considered that an abnormal situation has occurred during the leak test, and a long test cycle C L must be performed on the test object in order to ensure the accuracy of the leak test.

[0101] It should be noted that the steps S3 to S5, S7 and / or S8 and S9 can be performed simultaneously or sequentially, depending on the environmental conditions of the leak test and / or the environmental conditions of the test object.

[0102] The application also relates to a filtering method 200 for a leak test cycle of a type of test object, which makes it possible to determine at least one reference value related to the leak test of said type of object, by constructing a database.

[0103] In particular, the filtering method 200 makes it possible to construct a database BDD in order to determine (or calculate) reference extreme values R MAX ,R MIN and / or reference extreme value functions F MAX ,FMIN , thereby supporting the optimized implementation of the early leak detection method 100, 100' and 100'' described above.

[0104] The method 200 comprises at least the following steps: Selection / study El : on the test period C L The selection / study is performed; Determination E2: calculation of the total difference Δ T , said total difference Δ T between the physical quantity R measurement value R L at the initial instant and the physical quantity R measurement value R ini at the end of the test period C fin ; Comparison E3: comparison of the total difference Δ T with a reference total difference Δ Tref , preferably with a first reference extreme value Δ Tref1 and a second reference extreme value Δ Tref2 ; Decision step: determination of whether the test period C L can be included in the database for determining at least one reference value.

[0105] It is specified that the initial instant and the end of the test period C L refer to representative values of the physical quantity R at the beginning and at the end of the test; these representative values can be measurement values at specific time points (for example after having determined their stability and representativeness by experience) or can be the average of the first and last measurement values.

[0106] In particular, based on the test periods C L in the database BDD, the following parameters are calculated: reference total difference Δ Tref : this parameter is a measure of central tendency; dispersion indicator I Δ : this indicator is a dispersion parameter calculated for the reference total difference Δ Tref .

[0107] The measure of central tendency Δ Tref may be the arithmetic mean or the median, while the corresponding dispersion indicator I Δ is the standard deviation or the interquartile range, respectively (depending on the type of distribution of the total differences).

[0108] The comparison is therefore advantageously used to verify that the total difference Δ T of the test period is within the range defined by the calculated measure of central tendency ΔTref and the dispersion indicator I Δco-defined interval, i.e. Δ Tref1 = Δ Tref -k e x I Δ ≤ Δ T ≤ Δ Tref +k e x I Δ = Δ Tref2 , with k e being a positive real constant. Said constant k e is comprised between 1 and 5, and is preferably equal to 1.

[0109] Therefore, if the total difference Δ T of the test period C L is comprised in the above defined interval, the test period C L (i.e. the set of all the measured points) will be included in the database BDD for the evaluation of the partial reference values of the database and / or for the data support of the early leakage detection decision method 100, 100' or 100" for a specific object or object type.

[0110] Otherwise, the test period C L and the measured values of the physical quantity R associated thereto will not be included in the construction process of the database BDD and a further test period will be performed by the filtering method 200, for example selecting a recorded test period.

[0111] [ Figure 8 ] shows a first implementation variant of the filtering method shown in [ Figure 7 ]; the filtering method 200' shown in [ Figure 8 ] further comprises the following steps, based on the method 200 shown in [ Figure 7 ]: determining E4: within the test period C L , calculating the difference Δ i1 and Δ i2 between at least two groups of consecutive and mutually different measured values R i and R i-1 of the physical quantity R (called intermediate differences); comparing E5: comparing said at least two intermediate differences Δ i1 and Δ i2 with at least one reference value Δ ref1 and Δ ref2 corresponding to each intermediate difference Δ i1 and Δ i2 , respectively.

[0112] It is to be noted that steps E4 and E5 can be performed repeatedly in a sliding manner on the measured values of the physical quantity R, in order to verify the absence of disturbing factors in the test period C L which would prevent it from being included in the database BDD.

[0113] Specifically, based on the test period C in the database BDD L The following parameters were calculated: Intermediate differences Δ i1 Δ i2 The corresponding reference value Δ ref1 Δ ref2 The reference values ​​are each a measure of central tendency; Dispersion index I i1 I i2 The indicators mentioned are for each intermediate difference Δ i1 Δ i2 The corresponding reference value Δ ref1 Δ ref2 The calculated dispersion parameter.

[0114] Intermediate differences Δ i1 or Δ i2 Corresponding measure of central tendency (reference value Δ) ref1 or Δ ref2 The value can be the arithmetic mean or the median, and the corresponding dispersion index I... i1 and I i2 These are the standard deviation or interquartile range, respectively.

[0115] Specifically, the comparison E5 is advantageously used to verify: test cycle C L The median difference Δ i1 and Δ i2 They are respectively located in the corresponding central tendency measures (Δ) obtained by calculation. ref1 Δ ref2 ), and the dispersion index of the association (I) i1 I i2 ) and positive real constants (k) i1 k i2 Within the interval jointly defined by Δ ref1 -k i1 ×I i1 ≤Δ i1 ≤Δ ref1 +I i1 , and Δ ref2 -k i2 ×I i2 ≤Δ i2 ≤Δ ref2 +I i2 ; Where, k i1 k i2 All are positive real constants.

[0116] Therefore, if each intermediate difference Δ i1 Δi2 is within the above defined interval, then the test period C L (i.e. the set of all the measurements) will be included in the database BDD for the evaluation of the partial reference values of the database, for the optimization of the filtering method (200, 200') and / or for the provision of data support to the early leakage detection decision method 100, 100' or 100" for a specific subject or subject type.

[0117] At the same time, the constant k i1 is advantageously equal to k i2 , which takes values between 1 and 5, preferably equal to 1.

[0118] The filtering method 200' can further comprise the following additional steps: Comparison E6: the measurements of the physical quantity R R i at the initial instant of the test period C ini are compared with the measurements of the physical quantity R R i-1 at the final instant of the previous test period C fin , wherein said test period C i is compared with C i-1 is a consecutive test period; Decision E7: it is determined whether the test period C i can be included in the database BDD for the determination of at least one reference value.

[0119] If the measurements are substantially equal, then the test period C i will not be included in the database BDD.

[0120] In particular, the comparison step E6 between the measurements of the physical quantity R R C and R D for the consecutive test periods C i and C i-1 can be performed also on the basis of a measure of central tendency M C associated with a measure of dispersion I D .

[0121] The measure of central tendency M C for the comparison of the measurements between two consecutive test periods can be the arithmetic mean or the median; while the measure of dispersion I D associated with it can be the standard deviation or the interquartile range.

[0122] In particular, each measurement R ini and R fin can be interval-limited in the following way: M C - k x I D ≤ R ini ≤ M C + k x I D; M C -k×I D ≤R fin ≤M C +k×I D ; wherein k is a positive real constant (for example between 1 and 5, preferably equal to 1); M C and I D are parameters calculated on the basis of the values of the physical quantity R recorded in the database BDD, as defined above.

[0123] By comparing the interval limits of the measured value R ini and R fin , it is determined whether there is an overlap; if there is an overlap, the test period C i is not included in the database BDD; if there is no overlap, the test period C i is included in the database BDD for the evaluation of the partial reference values of the database BDD, for the optimization of the filtering method (200, 200') and / or for the provision of data support for the early leak detection decision method 100, 100' or 100" for a specific object or object type.

[0124] Steps E2 and E3, E4 and E5, E6 can be performed sequentially or simultaneously; however, the test period C L analyzed / tested with the filtering method (200, 200') of the application must meet the comparison criteria of steps E3, E5 and / or E7 to be included in the database BDD by the decision of step E7.

[0125] It should be noted that the present application also relates to an electronic device, for example a leak detection device 1 or a computer, configured to perform the filtering method 200, 200' of the leak test periods as described above.

Claims

1. A filtering method (200; 200') for a leakage test cycle of a test object (10) type, said method being used to construct a database (BDD) capable of determining at least one reference value related to leakage testing of said object type, particularly suitable for early leakage detection methods; wherein, A test cycle (C) L This includes measuring at least one part of the object under test (10) or in the peripheral cavity of the object under test (10) in relation to the degree of leakage (F). u The method (200; 200') includes at least the following steps related to the physical quantity (R): Determine (E2) the test cycle (C) L The difference (Δ) between the measured values ​​of the physical quantity (R) at the start and end times. T This difference is called the total difference; Compare the total difference with the total difference reference value (Δ) Tref1 Δ Tref2 ) for comparison (E3).

2. The method (200; 200') according to the preceding claims, characterized in that, The total difference reference value (Δ) Tref ) is based on the test cycle (C) in the database (BDD). L The central tendency measure calculated from the above.

3. The method (200; 200') according to claims 1 and 2, characterized in that, Based on the test cycle (C) in the database (BDD) L ), for the total difference reference value (Δ Tref The dispersion index (I) is calculated. Δ ).

4. The method (200; 200') according to claims 2 and 3, characterized in that, For the test cycle (C) L The total difference (Δ) T The comparison is performed by checking the total difference (Δ). T Whether it is within the calculated measure of central tendency (Δ) Tref ) and dispersion index (I Δ It is carried out within the interval defined by ).

5. The method (200; 200') according to claim 4, characterized in that, The central tendency measure of the total difference (Δ) Tref The arithmetic mean or median is the denominator, while the corresponding dispersion index (I0) is the median. Δ These are the standard deviation or interquartile range, respectively.

6. The method (200') according to any of the preceding claims, characterized in that, The method (200') further includes the following steps: Determine at least two differences (Δ) for (E4). i1 Δ i2 The difference mentioned above is called the intermediate difference, which is the test period (C). L The difference between two consecutive and distinct measurements of a physical quantity (R) within a given domain; The at least two intermediate differences (Δ) i1 Δ i2 Each difference in the equation is associated with a corresponding intermediate difference (Δ). i1 Δ i2 Reference value (Δ) ref1 Δ ref2 ) for comparison (E5).

7. The method (200; 200') according to the preceding claims, characterized in that, Intermediate difference (Δ) ref1 Δ ref2 The reference value is a measure of central tendency calculated based on the test period in the database (BDD).

8. The method (200; 200') according to the preceding claims, characterized in that, Based on the test period in the database (BDD), for each intermediate difference reference value (Δ) ref1 Δ ref2 The dispersion index (I) was calculated separately. i1 I i2 ).

9. The method (200; 200') according to the preceding claims, characterized in that, For the test cycle (C) L The comparison of the intermediate differences is done by checking each intermediate difference (Δ). i1 Δ i2 Whether it is within the corresponding calculated measure of central tendency (Δ) ref1 Δ ref2 ) and dispersion index (I i1 I i2 It is carried out within the interval defined by ).

10. The method (200; 200') according to the preceding claims, characterized in that, Intermediate differences (Δ) i1 Δ i2 The measure of central tendency (Δ) ref1 Δ ref2 The arithmetic mean or median is the denominator, while the corresponding dispersion index (I0) is the median. i1 I i2 These are the standard deviation or interquartile range, respectively.

11. The method (200') according to claim 10, characterized in that, The method (200') further includes the following steps: The test cycle (C) i The measured value of the physical quantity at the start time (R) ini ) and the previous test cycle (C i-1 The measured value of the physical quantity at the end time (R) fin The test period (C) is compared with the test period (C). i C i-1 () refers to a continuous testing cycle; Determine the test cycle (C) i Whether the measured value (R) can be included in the database (BDD) to determine at least one reference value, wherein if the measured value (R) ini R fin If the test cycles (C) are basically equal, then the test cycle (C) is... i ) are not included in the database (BDD).

12. The method (200') according to the preceding claim, characterized in that, For two consecutive test cycles (C) i C i-1 The measured values ​​of each physical quantity R between ) ini R fin The comparison is based on the measure of central tendency (M). C ) and the measured values ​​(R) ini R fin Related dispersion index (I) D It was done by ).

13. An electronic device, such as a computer, or a leak detection device (1), configured to implement a filtering method (200; 200') for the leak test cycle.

Images