阅读说明：本技术 气体网络和用于同时检测压力或真空下气体网络中泄漏和阻塞的方法 (Gas network and method for simultaneous detection of leaks and blockages in a gas network under pressure or vacuum ) 是由 P·盖恩斯 E·罗劳蒂 于 2019-11-26 设计创作，主要内容包括：提供了用于同时地检测、定位和量化压力或真空下的气体网络(1)中的泄漏(13a)和阻塞(13b)的方法；气体网络(1)包括：-一个或多个压缩气体源或真空源(6)；-一个或多个真空应用或压缩气体的用户(7)或用户区域；-管线(5)或管线(5)的网络(4),用于将压缩气体或真空从所述源(6)输送到用户(7)、用户区域或应用；-多个传感器(9a、9b、9d),提供气体网络(1)内的不同时间和位置的气体的一个或多个物理参数；其特征在于,气体网络(1)还设有多个可控制或可调整的安全阀(10a)、多个可控制或可调整的节流阀(10b)以及可选的能够监测安全阀(10a)和/或节流阀(10b)的状态或状况的一个或多个传感器(9c),并且方法包括以下步骤：-训练阶段(16),其中在第一组传感器(9a、9b、9c、9d)和第二组传感器(9a、9b、9c、9d)的测量结果之间基于这些传感器(9a、9b、9c、9d)的不同测量结果建立数学模型,其中可控制或可调整的安全阀(10a)和可控制或可调整的节流阀(10b)按预定顺序并根据很好地设计的场景被控制以分别产生泄漏(13a)和阻塞(13b)；-操作阶段(17),其中使用在第一组传感器(9a、9b、9c、9d)和第二组传感器(9a、9b、9c、9d)的测量结果之间建立的数学模型来检测、定位和量化气体网络中的泄漏(13a)和阻塞(13b)；其中操作阶段(17)包括以下步骤：在必要时按预定次序并根据很好地设计的场景控制安全阀和节流阀；读出第一组传感器(9a、9b、9c、3d)；-基于这些读出的测量结果,借助数学模型计算或确定第二组传感器(9a、9b、9c、9d)的值；将计算或确定的所述第二组传感器(9a、9b、9c、3d)的值与第二组传感器(9a、9b、9c、9d)的读取值进行比较并确定它们之间的差值；-基于前述差值及其任何衍生值来确定气体网络中是否有泄漏(13a)和/或阻塞(13b)；-在检测到泄漏(13a)或阻塞(13b)的情况下生成警报和/或确定泄漏(13a)和/或阻塞(13b)的位置和/或确定泄漏(13a)的流量和/或阻塞(13b)的阻塞程度和/或生成泄漏和/或阻塞成本。(Methods are provided for simultaneously detecting, locating and quantifying leaks (13a) and blockages (13b) in a gas network (1) under pressure or vacuum; the gas network (1) comprises: -one or more compressed gas or vacuum sources (6); -one or more users (7) or user areas of vacuum application or compressed gas; -a line (5) or a network (4) of lines (5) for conveying compressed gas or vacuum from the source (6) to a user (7) for useA user area or application; -a plurality of sensors (9a, 9b, 9d) providing one or more physical parameters of the gas at different times and locations within the gas network (1); characterized in that the gas network (1) is further provided with a plurality of controllable or adjustable safety valves (10a), a plurality of controllable or adjustable throttle valves (10b) and optionally one or more sensors (9c) capable of monitoring the status or condition of the safety valves (10a) and/or throttle valves (10b) and in that the method comprises the steps of: -a training phase (16) in which a mathematical model is established between the measurements of the first group of sensors (9a, 9b, 9c, 9d) and the second group of sensors (9a, 9b, 9c, 9d) based on the different measurements of these sensors (9a, 9b, 9c, 9d), wherein the controllable or adjustable safety valve (10a) and the controllable or adjustable throttle valve (10b) are controlled in a predetermined sequence and according to well-designed scenarios to produce a leak (13a) and a blockage (13b), respectively; -an operation phase (17) in which leaks (13a) and blockages (13b) in the gas network are detected, localized and quantified using a mathematical model established between the measurements of the first set of sensors (9a, 9b, 9c, 9d) and the second set of sensors (9a, 9b, 9c, 9 d); wherein the operating phase (17) comprises the steps of: controlling the safety valve and the throttle valve in a predetermined order and according to well designed scenarios, if necessary; reading out a first set of sensors (9a, 9b, 9c, 3 d); -calculating or determining the values of the second set of sensors (9a, 9b, 9c, 9d) by means of a mathematical model based on the read-out measurements; comparing the calculated or determined values of the second set of sensors (9a, 9b, 9c, 3d) with the read values of the second set of sensors (9a, 9b, 9c, 9d) and determining the difference between them; -determining whether there is a leak (13a) and/or a blockage (13b) in the gas network based on the aforementioned difference and any derivative thereof; -generating an alarm and/or determining the location of the leak (13a) and/or the blockage (13b) and/or determining the flow of the leak (13a) and/or the blockage (13b) in case a leak (13a) or a blockage (13b) is detectedb) And/or the cost of creating leaks and/or blockages.)

1. A method for simultaneously detecting, locating and quantifying leaks (13a) and blockages (13b) in a gas network (1) under pressure or vacuum; the gas network (1) comprises:

-one or more compressed gas or vacuum sources (6);

-one or more users (7) or user areas of vacuum application or compressed gas;

-a line (5) or a network (4) of lines (5) for conveying compressed gas or vacuum from the compressed gas or vacuum source (6) to a user (7), a user area or an application;

-a plurality of sensors (9a, 9b, 9d) providing one or more physical parameters of the gas at different times and locations within the gas network (1);

characterized in that the gas network (1) is further provided with a plurality of controllable or adjustable safety valves (10a), a plurality of controllable or adjustable throttling valves (10b) and possibly one or more sensors (9c) capable of monitoring the status or condition of said safety valves (10a) and/or said throttling valves (10b) and in that the method comprises the steps of:

-a training phase (16) in which a mathematical model is established between the measurements of the first (9a, 9b, 9c, 9d) and second (9a, 9b, 9c, 9d) sets of sensors based on the different measurements of these sensors (9a, 9b, 9c, 9d), wherein the controllable or adjustable safety valve (10a) and the controllable or adjustable throttle valve (10b) are controlled in a predetermined order and according to well-designed scenarios to produce a leak (13a) and a blockage (13b), respectively;

-an operating phase (17) in which leaks (13a) and blockages (13b) in the gas network are detected, localized and quantified using the mathematical model established between the measurements of the first set of sensors (9a, 9b, 9c, 9d) and the second set of sensors (9a, 9b, 9c, 9 d);

wherein the operating phase (17) comprises the steps of:

-controlling the safety valve and the throttle valve, if necessary in a predetermined order and according to well-designed scenarios;

-reading out the first set of sensors (9a, 9b, 9c, 9 d);

-calculating or determining the values of the second set of sensors (9a, 9b, 9c, 9d) by means of the mathematical model based on these read-out measurements;

-comparing the calculated or determined values of the second set of sensors (9a, 9b, 9c, 9d) with the read values of the second set of sensors (9a, 9b, 9c, 9d) and determining the difference between them;

-determining whether there is a leak (13a) and/or a blockage (13b) in the gas network based on the aforementioned difference and any derivative thereof;

-generating an alarm in case of detection of a leak (13a) or a blockage (13b) and/or determining the location of a leak (13a) and/or a blockage (13b) and/or determining the flow of a leak (13a) and/or the degree of blockage of a blockage (13b) and/or generating a leak and/or a blockage cost.

2. Method according to claim 1, characterized in that the first set of sensors (9a, 9b, 9c, 9d) comprises a plurality of pressure sensors (9b), a plurality of flow sensors (9a), optionally a plurality of sensors (9c) capable of determining the state of a safety valve (10a) and/or a throttle valve (10b) and optionally one or more differential pressure sensors (9d) at different locations in the gas network (1), and that the second set of sensors (9a, 9b, 9c, 9d) comprises a plurality of flow sensors (9a) at different locations in the gas network (1) and a sensor (9c) capable of determining the state of the throttle valve.

3. The method according to claim 2, characterized in that at least a part of the flow sensor (9a) is placed in the vicinity of the safety valve (10 a).

4. Method according to any of the preceding claims, characterized in that the aforementioned sensor (9a, 9b, 9d) is capable of measuring one or more of the following physical parameters of the gas: flow, pressure, differential pressure, temperature, humidity, gas velocity, etc.

5. Method according to any one of the preceding claims, characterized in that it comprises, for the training phase (16), a starting phase (15) in which the aforementioned sensors (9a, 9b, 9c, 9d) are calibrated before use.

6. Method according to claim 5, characterized in that during operation at least the second set of sensors (9a, 9b, 9c, 9d) is calibrated by in-situ calibration or self-calibration.

7. Method according to any of the preceding claims, characterized in that the operating phase (17) is temporarily interrupted or stopped at certain times, after which the training phase (16) is resumed in order to redefine the relation or mathematical model between the measurements of the different sensors (9a, 9b, 9c, 9d) before resuming the operating phase (17).

8. Method according to any one of the preceding claims, characterized in that the operating phase (17) steps are repeated sequentially at given time intervals.

9. The method according to any one of the preceding claims, wherein the safety valve (10a) is formed by a bleed valve.

10. Method according to any of the preceding claims, characterized in that at least some of the sensors (9a, 9b, 9c, 9d) are integrated in one module together with a safety valve (10a) or a throttle valve (10 b).

11. Method according to any of the preceding claims, characterized in that a sensor (9a, 9b, 9c, 9d) is provided in the gas network (1) in the vicinity of each safety valve (10a) and/or throttle valve (10b) and/or a safety valve (10a) and/or throttle valve (10b) is provided in the vicinity of each sensor (9a, 9b, 9c, 9 d).

12. The method according to any of the preceding claims, characterized in that the mathematical model is a black box model.

13. Method according to any of the preceding claims, characterized in that the aforementioned mathematical model takes the form of a matrix and/or a non-linear vector function with parameters or constants, wherein the output or 'target' of the mathematical model is monitored for changes during the operating phase (17).

14. The method according to any of the preceding claims, characterized in that the gas is air, oxygen or nitrogen, or another non-toxic and/or harmless gas or gas mixture.

15. Method according to any of the preceding claims, characterized in that a differential pressure sensor (9d) above the throttle valve (10b) is used as a status sensor (9c) which can determine the status or condition of the throttle valve (10 b).

16. A gas network under pressure or vacuum, said gas network (1) being provided with at least:

-one or more compressed gas or vacuum sources (6);

-one or more users (7) or user areas of vacuum application or compressed gas;

-a line (5) or a network (4) of lines (5) for conveying compressed gas or vacuum from a compressed gas or vacuum source (6) to a user (7) or user area;

-a plurality of sensors (9a, 9b, 9d) providing one or more physical parameters of the gas at different times and locations within the gas network (1);

characterized in that the gas network (1) is further provided with:

-a plurality of controllable or adjustable safety valves (10a) and a plurality of controllable or adjustable throttle valves (10 b);

-optionally one or more sensors (9c) able to register the status or condition of one or more safety valves (10a) and one or more throttle valves (10 a);

-a data acquisition control unit (11) for collecting data from said sensors (9a, 9b, 9c, 9d) and for controlling or regulating the aforementioned safety valve (10a) and throttle valve (10 b);

-a computing unit (12) for performing the method according to any one of the preceding claims.

17. The gas network according to claim 16, characterized in that the safety valve (10a) is formed by a relief valve.

18. The gas network according to any of the preceding claims 16 to 17, characterized in that at least some of the sensors (9a, 9b, 9c, 9d) are integrated in one module together with a safety valve (10a) or a throttle valve (10 b).

19. Gas network according to any of the preceding claims 16 to 18, characterized in that in the gas network (1) there is a sensor (9a, 9b, 9c, 9d) in the vicinity of each safety valve (10a) and/or throttle valve (10b) and/or a safety valve (10a) and/or throttle valve (10b) in the vicinity of each sensor (9a, 9b, 9c, 9 d).

20. Gas network according to any of the preceding claims 16-19, characterized in that the gas network (1) is further provided with a monitor (14) to display or signal the location of leaks (13a) and blockages (13b), leak flow, blockages, leak costs, blockages, leaks (13a) and blockages (13 b).

21. Gas network according to any of the preceding claims 16 to 20, characterized in that the sensor (9c) capable of recording the status or condition of the user (7) is part of the user (7) itself.

22. The gas network according to any of the preceding claims 16 to 21, characterized in that the computing unit (12) is a cloud-based computing unit (12) which may or may not be wirelessly connected to the gas network (1).

Detailed Description

The gas network 1 in fig. 1 mainly comprises a network 4 of a source side 2, a consumer side 3 and a pipeline 5 between the two.

In this case, the gas network 1 is a gas network 1 under pressure. The gas may be air, oxygen or nitrogen, or any other non-toxic and/or harmless gas or gas mixture.

The source side 2 comprises a plurality of compressors 6, in this case three, which generate compressed air. The user side 3 comprises a plurality of users 7 of compressed air, in this case also three.

The compressor 6 may also comprise a compressed air dryer.

It is not excluded that there may also be a compressor 6 downstream of the gas network 1. This is called a "booster compressor".

Compressed air is delivered from a compressor 6 to a user 7 through a network 4 of lines 5.

In most cases, this network 4 is a very complex network of pipelines 5.

Fig. 1 shows this network 4 in a very schematic and simplified manner. In most real cases, the network 4 of lines 5 comprises a large number of lines 5 and couplings connecting users 7 in series and in parallel with compressors 6. It is not excluded that a part of the network 4 adopts or comprises a ring structure.

This is because the gas network 1 is typically expanded over time with additional users 7 or compressors 6, whereby new pipelines 5 have to be laid between existing pipelines 5, which causes tangling of the pipelines 5.

The gas network 1 may also be provided with a pressure vessel 8, wherein all compressors 6 are in front of this pressure vessel 8.

It is not excluded that there may be one or more pressure vessels 8 downstream of the gas network 1.

In addition, components 18, such as filters, separators, atomizers and/or regulators, can also be provided in the gas network 1. These components 18 may be present in various combinations and may be present both near the buffer vessel 8 and near the individual users 7.

In the example shown, the component 18 is arranged behind the buffer container 8 and in the vicinity of the respective user 7.

The network 4 also comprises a plurality of sensors 9a, 9b, 9c, which are located at different locations in the network 4.

In this solution, two flow sensors 9a have been installed, one of which immediately after the aforementioned pressure vessel 8, it will measure the total flow q provided by all the compressors 6.

It is not excluded that the flow of the compressor 6 is calculated or measured by itself.

In addition, the figure shows four pressure sensors 9b, which measure the pressure at different locations in the network 4.

It is also proposed that the pressure sensor 9b measures the pressure in the pressure vessel 8 to correct the "mass input-mass output" principle for large concentrated volumes.

It is clear that more or less than four pressure sensors 9b may also be provided. In addition, the present invention does not limit the number of flow sensors 9 a.

In addition to the flow sensor 9a or the pressure sensor 9b, additionally or alternatively, the sensors 9a, 9b may be used to determine one or more of the following physical parameters of the gas: differential pressure, temperature, humidity, gas velocity, etc.

According to the invention, the gas network 1 is also provided with a plurality of safety valves 10a which can blow gas out of the gas network 1. The safety valve 10a is adjustable or controllable, which means that the amount of gas it vents can be set or adjusted.

The safety valve 10a may be formed by a relief valve, which is normally provided as a standard in the gas network 1. Such a relief valve may be controlled as the relief valve 10 a.

According to the invention, the gas network 1 is also provided with a plurality of throttles 10b, which are installed at various positions in the pipeline 5. The choke 10b may partially close the line 5 to simulate a blockage as if it were an original blockage. They are adjustable or controllable, which means that the extent to which they close the relevant line 5 can be set or controlled.

In addition to the aforementioned sensors 9a and 9b, which measure physical parameters of the gas, there are a plurality of sensors 9c or "condition sensors 9 c", which are located at the safety valve 10a and the throttle valve 10 b.

The state sensor 9c at the safety valve 10a will be able to measure the open/closed state of the safety valve 10a, while the state sensor 9c at the safety valve 10b will be able to measure the valve opening, i.e. the relative increase or decrease of the blockage thereby produced. The condition sensor 9c in the vicinity of the throttle valve 10b can be replaced by a differential pressure sensor 9d, which determines the pressure drop over the throttle valve 10 b.

Although not explicitly indicated in fig. 1, it cannot be excluded that in the gas network 1, in the vicinity of the compressor 6 and the user 7, there are additional status sensors 9c determining the on/off status of these components. Preferably, these status sensors are part of the user 7 itself.

Then, the additional status sensor 9c (e.g. on/off of the compressor 6) is intended to significantly reduce the cross-sensitivity of the model during the training phase 16 and the operation phase 17, as explained below.

Sensors 9a, 9b measuring the pressure or flow of gas at the safety valves 10a and 10b may also be used. Sensors measuring the temperature of the gas at the safety valve 10a and the throttle valve 10b may also be used.

Preferably, at least a part of the flow sensors, pressure sensors, temperature sensors and/or condition sensors 9a, 9b, 9c should be located in the vicinity of the relief valve 10a and the throttle valve 10 b.

In this scheme, each condition sensor 9c is located near the relief valve 10a or the throttle valve 10b, one flow sensor 9a is located near the relief valve 10a, one pressure sensor 9b is located near the relief valve 10a, and three pressure sensors 9b are located near the throttle valve 10 b.

This will enable the state sensor 9c to be used to determine the state (i.e., open or closed) of the relief valve 10a and the state of the throttle valve 10b, and the valve opening degree of the throttle valve 10 b. In this scenario, the available condition sensor 9c will measure the relative increase or decrease in the blockage of the associated throttle valve 10b, which will allow the degree of blockage to be quantified. In addition, with the flow sensors 9a, the flow of the respective safety valves 10a will be measured, which will make it possible to quantify the leakage rate.

Although there is a great freedom in choosing which sensor 9a, 9b, 9c will be placed or not at the safety valve 10a or throttle valve 10b, it is preferred that there is a sensor 9a, 9b, 9c in the gas network 1 near each safety valve 10a or throttle valve 10b and/or vice versa, i.e. near each sensor 9a, 9b there is a safety valve 10a or throttle valve 10 b.

It is also possible that at least a part of the sensors 9a, 9b, 9c is integrated in one module together with the safety valve 10a or the throttle valve 10 b.

This will simplify and accelerate the mounting or integration of the sensors 9a, 9b, 9c and the safety valves 10a and 10 b. In addition, it can be ensured that the correct and suitable sensors 9a, 9b, 9c for the safety valve 10a and the throttle valve 10b are placed together in one module.

In this solution, and preferably, the condition sensors 9c are each integrated in one module together with the respective safety valve 10a or throttle valve 10 b.

The aforementioned differential pressure sensor 9d is preferably placed above the filter, separator, atomizer and/or regulator component 18. In the present solution, four differential pressure sensors 9d are included in the gas network 1. Differential pressure sensor 9d may also be placed above throttle valve 10b and then take over the role of status sensor 9 c.

On the other hand, the aforementioned humidity and temperature sensors should preferably be installed on the inlet/outlet of the compressor 6 and the user 7. In the example shown, these additional sensors are not all included in the gas network 1, but of course this is also possible. Such sensors may be used, in particular, in more extensive and complex gas networks 1, as well as in networks that only measure volumetric flow, not mass flow.

According to the invention, the gas network 1 is also provided with a data acquisition control unit 11 to collect data from the aforementioned sensors 9a, 9b, 9c, 9d and also to control the safety valve 10a and the throttle valve 10 b.

In other words, the sensors 9a, 9b, 9c, 9d determine or measure physical parameters of the gas of the safety valve 10a and the throttle valve 10b and send this data to the data acquisition control unit 11, and the data acquisition control unit 11 will control or check whether the safety valve 10a and the throttle valve 10b are opened or closed and the degree of opening or closing to simulate a leak or form or simulate a blockage by blowing gas.

According to the invention, the gas network 1 is further provided with a calculation unit 12 for processing data from the sensors 9a, 9b, 9c, 9d, wherein the calculation unit 12 will be able to perform the method according to the invention for detecting and quantifying leaks 13a and blockages 13b in the gas network 1, as described below.

The aforementioned calculation unit 12 may be a physical module, which is a physical part of the gas network 1. It cannot be excluded that the computing unit 12 is not a physical module, but a so-called cloud-based computing unit 12, which may or may not be wirelessly connected to the gas network 1. This means that the computing unit 12 or the software of the computing unit 12 is located in the "cloud".

In this solution the gas network 1 is also provided with a monitor 14 for displaying or signaling the leaks 13a and blockages 13b detected using the method.

The operation of the gas network 1 and the method according to the invention is very simple and as follows.

Fig. 2 schematically illustrates a method for simultaneously detecting a leak 13a and a blockage 13b in the gas network 1 of fig. 1.

In a first phase 15, i.e. a start-up phase 15, the sensors 9a, 9b, 9c, 9d are calibrated before use, if necessary. Of course, if there are other sensors, they may also be calibrated before use.

This operation takes place once when the sensors 9a, 9b, 9c, 9d are placed in the gas network 1. Of course, the sensors 9a, 9b, 9c, 9d may be recalibrated over time.

Preferably, at least the second set of sensors 9a, 9b, 9c, 9d should be calibrated during operation or by in situ self-calibration. This means that the sensors 9a, 9b, 9c, 9d in the gas network 1 are calibrated, i.e. after they have been installed. "in operation" or "in the field" means that calibration is performed without removing the sensors 9a, 9b, 9c, 9d from the network 1.

Of course, all sensors 9a, 9b, 9c, 9d and thus the first set of sensors 9a, 9b, 9c, 9d may be calibrated in operation or in the field by self-calibration.

In this way it can be ensured that the placement and/or possible contamination of the sensors 9a, 9b, 9c, 9d will not affect their measurement results, since only after the placement of the sensors 9a, 9b, 9c, 9d can calibration be performed or repeated for a period of time.

Then, the second stage 16 or training stage 16 begins.

In this phase, a mathematical model is created between the measurements or "features" of the first set of calibrated sensors 9a, 9b, 9c, 9d and the measurements or "targets" of the second set of calibrated sensors 9a, 9b, 9c, 9 d.

Preferably, the first set of sensors 9a, 9b, 9c, 9d comprises a plurality of pressure sensors 9b, a plurality of flow sensors 9a and possibly one or more sensors 9c at different locations in the gas network, and the second set of sensors 9a, 9b, 9c, 9d comprises a plurality of flow sensors 9a and status sensors 9c at different locations in the gas network.

In this scheme, a part of the flow sensors 9a, the pressure sensors 9b, and a part of the condition sensors 9c form a first group of sensors, and the remaining flow sensors 9a and the condition sensors 9c form a second group of sensors.

For the sake of completeness, it is stated herein that the invention is not limited thereto. For the first and second sets of sensors, a random selection from the sensors 9a, 9b, 9c, 9d can be made, the only limitation being that the sensors in the first set are not allowed to be in the second set and vice versa.

The aforementioned mathematical model is based on various measurements of the sensors 9a, 9b, 9c, 9d, wherein the adjustable safety valve 10a is controlled to produce a leak and the adjustable throttle valve 10b is controlled to produce a blockage.

In other words, the data acquisition control unit 11 collects data or measurements from the sensors 9a, 9b, 9c, 9d, wherein the data acquisition control unit 11 will control the safety valves 10a so as to open them such that a leak forms in the gas network 1, and wherein the data acquisition control unit will control the throttling valves 10b so as to close them such that a form blockage forms in the gas network 1, such that data can be collected from the sensors 9a, 9b, 9c, 9d when one or more leaks 13a or blockages 13b occur in the gas network 1.

In this way, the entire data or set of measurements may be collected, as well as information from the safety valve 10a and the throttle valve 10b, i.e. the location and size of the leak 13a and the location and extent of the blockage 13 b. The calculation unit 12 will build a mathematical model based on all this information. The mathematical model is preferably a black box model or a data driven model. The model typically contains a number of estimated parameters or coefficients, also referred to as 'weights'.

The black box model takes the form of, for example, a matrix, a non-linear mathematical vector function, or the like.

The mathematical model is not based on any assumptions.

The training phase 16 should preferably be performed during operation of the gas network 1 or while the gas network 1 is operational.

The mathematical model is used in an operational stage 17 to detect and quantify leaks 13a and blockages 13b in the gas network 1. Although not usual, it cannot be excluded to control the safety valve 10a in a predetermined sequence during an operating phase to localize the leak 13 a. It should be noted that control according to scenario 000 … is also possible. It cannot be excluded that the adjustable throttle valve 10b is controlled in a predetermined sequence during an operating phase to position the choke 13 b.

Also during this phase the data acquisition control unit 11 will collect different data from the sensors 9a, 9b, 9c, 9d and the calculation unit 12 will perform the necessary calculations using the mathematical model established in the previous phase 16.

The operating phase 17 starts with reading the first set of sensors 9a, 9b, 9c, 9 d.

Using these read measurements, the values of the second set of sensors 9a, 9b, 9c, 9d, also referred to as 'predicted targets', are determined or calculated by the calculation unit 12 using a mathematical model.

The determined or calculated values of the second set of sensors 9a, 9b, 9c, 9d are compared with the read values of the second set of sensors 9a, 9b, 9c, 9d and the difference between them is determined.

Based on the above difference, the calculation unit 12 determines whether there is a leak 13a or a blockage 13b and, if necessary, locates the leak 13a or the blockage 13b in the gas network 1.

For this purpose it will be checked whether the difference exceeds a certain threshold value, which will then indicate a leak 13a or a blockage 13b in the gas network 1.

The threshold may be predetermined or may be selected empirically.

When a leak 13a or blockage 13b is detected, an alarm will be generated and the corresponding location, leak rate, blockage level, and/or leak and blockage costs may be generated together. In this scenario, this may be accomplished using a monitor 14 that displays an alarm.

The user of the gas network 1 will notice this alarm and can take appropriate steps.

Preferably, the steps of operating phase 17 are repeated sequentially and cyclically at specific time intervals.

Thus, for example, the leak 13a and the blockage 13b may be detected during the entire operation of the gas network 1 and more than once during or shortly after start-up of the gas network 1.

The aforementioned time interval may be selected and set according to the gas network 1. It cannot be excluded that the time interval may vary with time.

In a preferred variant of the invention, at some point, the operating phase 17 will be temporarily interrupted or stopped, after which the training phase 16 will be resumed, in order to reestablish the mathematical relationship between the measurements of the different sensors 9a, 9b, 9c, 9d before resuming the operating phase 17.

"at a specific moment" is to be understood herein as a preset moment, for example once a week, month or year, or a moment selectable by the user.

This will update the mathematical model to account for possible time-varying behavior of the system. These time-varying behaviors are behaviors that are not captured by the mathematical model during the training phase 16 when the mathematical model is trained under different scenarios.

This may include, for example, a change in the topology of the gas network 1 or the addition of new components to the gas network 1.

Although in the example of fig. 1 it is a gas network 1 under pressure, it may also be a gas network 1 under vacuum.

Then, the source side 2 includes a plurality of vacuum sources, i.e., vacuum pumps or the like.

In this scenario, the user 7 has been replaced by an application requiring vacuum.

Furthermore, the method is the same as described above, taking into account that the leak 13a now introduces ambient air into the gas network 1. Preferably, other thresholds will be set to generate the alarm.

Also in this solution, the safety valve 10a introduces ambient air into the gas network 1, instead of blowing real air. Therefore, the safety valve 10a is more likely to be an intake valve. However, the principle remains unchanged.

The invention is in no way limited to the embodiments described by way of example and shown in the drawings, but the method and the gas network according to the invention can be implemented in various modifications without departing from the scope of the invention.