WO2025056150A1

WO2025056150A1 - Method and system for leak detection based on inverse transient analysis

Info

Publication number: WO2025056150A1
Application number: PCT/EP2023/075032
Authority: WO
Inventors: Alexey ARBUZOV; Ilya Mokhov; Dmitry SAZYKIN
Original assignee: Siemens AG; Siemens Corp
Current assignee: Siemens AG; Siemens Corp
Priority date: 2023-09-12
Filing date: 2023-09-12
Publication date: 2025-03-20
Anticipated expiration: 2026-03-12

Abstract

The present invention concerns a system (100) and a method for detecting and locating a leak (11) in a fluid containing structure (10), the system (100) comprising: a set S1 of N sensors S_i, i=1,...,N, with N ≥ 1, wherein each sensor S_i is configured, when activated, for measuring fluid pressure values in function of the time at a location L^s1i defined with respect to said fluid containing structure; - a set S2 of M transient generation units (hereafter "TGU") G_j, with j = 1,...,M, with M ≥ 2, wherein each TGU G_j is configured, when activated, for generating a transient pressure signal according to a predefined pattern P_j at a location L^s2j defined with respect to said fluid containing structure; the system being configured for successively activating the TGUs of at least one subset of TGUs of said set S2, wherein, within each subset, only one TGU is activated at a time; for acquiring, for each TGU activation and via one or several sensors S_i of the set S1, fluid pressure values resulting from said TGU activation; for each TGU activation, analyzing, for each of said one or several sensors, the fluid pressure values acquired by the concerned sensor for said TGU activation to reference fluid pressure values defined for said concerned sensor with respect to the TGU that has been activated; detecting and locating the leak (11) within the fluid containing structure (10) from said analysis.

Description

Method and system for leak detection based on inverse transient analysis

The present invention relates to pipeline leak detection and location .

More generally speaking, the invention comes within the scope of the detection and location of a leak in a fluid containing structure , e . g . a pipeline network, typically used for fluid distribution and/or transmission . Said fluid might be water, or oil , or gas , or a heating fluid, or any other fluid that needs to be transmitted and/or distributed by means of said structure from one or several source points A to one or several distribution points B . Challenges with respect to leaks in fluid containing structures are mainly related to the detection of the leak and the identi fication of each leak location .

Di f ferent solutions exist for detecting and locating a leak in a fluid containing structure . In particular, a well-known approach is the negative pressure wave (hereafter "NPW" ) approach . The latter proposes to capture a transient wave generated during a burst event ( e . g . a pressure wave resulting from a sudden change in a direction of motion of the fluid inside the fluid containing structure , wherein said change can result in a hydraulic shock, also called water hammer, that can be observed, for instance at the extremities of a pipe ) using high-rate pressure sensors . The NPW approach requires to process the signal captured by the high-rate pressure sensors in order to detect and locate a leak . For said processing, a method called " Inverse Transient Analysis"

( ITA) that relies on the wave propagation along the network of pipes is widely used . The ITA method requires to model the pipe network by a mathematical model that is then fed with pressure data captured by the high-rate pressure sensors . However, developing a mathematical model that can accurately describe the pipe network under leak monitoring is not trivial ( for instance , characteristics of some ( old) parts of the network might not be known, or the flow within some pipes might have degrade due to aging, etc . ) , and might lead to false alerts or might fail to detect a leak .

An obj ective of the present invention is to propose a system and a method that overcome the limitations of the previously described approaches for leak detection and location, and in particular that is able to minimi ze the energy consumed for detecting and locating a leak in a pipe network .

This obj ective is achieved by the measures taken in accordance with the independent claims . Further advantageous embodiments are proposed by the dependent claims .

The general concept of the present invention is to implement a system that generates , at a first location inside the network, a transient signal of a known pattern, and then to receive , at another location of said network, the generated transient signal and analyze it against a known reference that has been obtained for said known pattern during a training phase .

More precisely, the present invention concerns a system for detecting and locating a leak in a fluid containing structure , said system comprising : - a set S I of N sensors S±, with N > 1 , wherein each sensor S± is configured, when activated, for measuring fluid pressure values in function of the time at a location L^sli defined with respect to said fluid containing structure ;

- a set S2 of M transient generation units (hereafter "TGU" ) Gj , with j = l , ..., M, with M > 2 , wherein each TGU Gj is configured, when activated, for generating a transient pressure signal according to a predefined pattern Pj at a location L^s2j defined with respect to said fluid containing structure . Preferentially, the TGUs and the sensors are installed along said fluid containing structure so that between two directly successive TGUs , there is at least one sensor ; the system being configured for

- successively activating the TGUs of at least one subset of TGUs of said set S2 , wherein, within each subset , only one TGU is activated at a time . Said subset may comprise all TGUs of the set S2 ( in such a case , the subset is equal to the set S2 ) ; for acquiring, for each TGU activation and via one or several sensors S± of the set S I , fluid pressure values resulting from said TGU activation; for each TGU activation, analyzing, for each of said one or several sensors , the fluid pressure values acquired by the concerned sensor for said TGU activation with respect to reference fluid pressure values defined for said concerned sensor with respect to the TGU that has been activated;

- detecting and locating a leak within the fluid containing structure from said analysis . The present invention concerns also a method for detecting and locating a leak in a fluid containing structure , the method comprising :

- successively activating the TGUs of at least one subset of TGUs of said set S2 , wherein, within each subset , only one TGU is activated at a time ;

- acquiring, for each TGU activation and via one or several sensors S± of the set S I , fluid pressure values resulting from said TGU activation; for each TGU activation, analyzing, for each of said one or several sensors , the fluid pressure values acquired by the concerned sensor for said TGU activation to reference fluid pressure values defined for said concerned sensor with respect to the TGU that has been activated;

- detecting and locating a leak within the fluid containing structure from said analysis .

Preferentially, the system comprising a predefined schedule used for implementing the successive activation of the TGUs and acquisition of the fluid pressure values by the sensors , said predefined schedule being configured for defining for each of the TGUs , and optionally for each of the sensors , a start activation time and an end activation time .

According to the present invention, the reference fluid pressure values have been measured by the concerned sensor during a training phase of the system, during which each TGU has been activated for generating said transient pressure signal , and for each TGU activation, said one or several sensors have been activated for acquiring each, reference fluid pressure values , wherein said reference fluid pressure values result from said predefined pattern . In other words , in order to enable detection, the system needs first to be trained during said training phase for determining for each sensor and with respect to at least some of the TGUs , said reference fluid pressure values , that are then used, during normal operation of the system, for detecting and locating leaks .

Preferentially, the set S2 of TGUs is divided into one or several subsets , wherein several TGUs of the system might then be active at the same time , as long as each of said several TGUs belong to a di f ferent subset of the set S2 , and as long as a predefined distance separates each of the di f ferent subsets . This enables to shorten the time required for monitoring potential leak of the fluid containing structure by activating several TGUs at the same time .

According to the present invention, i f said analysis , e . g . a comparison, between the measured fluid pressure values and the reference fluid pressure values , that have both been acquired by the same sensor for a same TGU, results in a di fference exceeding a predefined threshold, then the system concludes that a leak is detected, and i f the leak has been detected from fluid pressure values acquired by a sensor directly neighboring the TGU whose activation led to said acquired fluid pressure values , then the system is configured for locating the leak between said sensor and said TGU . Also , if a leak is detected by a first sensor and by a second sensor that are directly neighboring sensors , notably when considering said one or several sensors , then, i f for the first sensor, the leak is detected from fluid pressure values resulting from a transient pressure signal generated by a first TGU that is closest to the second sensor than to the first sensor, and, for the second sensor, the leak is detected from fluid pressure values resulting from a transient pressure signal generated by a second TGU that is closest to the first sensor than to the second sensor, then the system is configured for locating said leak between the first sensor and the second sensor .

Further aspects of the present invention will be better understood through the following drawings , wherein like numbers designate like obj ects :

Figure 1 schematic illustration of a system according to the invention;

Figure 2 flowchart of a preferred embodiment of a method according to the invention .

A preferred embodiment of a system 100 according to the invention is illustrated in Figure 1 . Said system 100 is configured for detecting and locating a leak 11 in a fluid containing structure 10 . The latter might be a fluid transmission and distribution network made of a plurality of interconnected pipes or ducts or the like , which may form or comprise one or several interconnection nodes , so that the created network be appropriate for the transmission of said fluid from one or several sources or points towards one or several other points , where said fluid can typically be distributed . Such fluid containing structures , like water and distribution network, oil or gas distribution networks , are well-known to the skilled persons and are not the subj ect of the present invention .

The system 100 comprises a set S I of N sensors S±, i=l , ..., N, with N > 1 , and a set S2 of M TGU Gj , with j = l , ..., M, with M > 2 . Each sensor S± is configured for measuring fluid pressure values in function of the time at a location L^sli defined with respect to a fluid containing structure 10 . Said sensors are typically pressure sensors capable of measuring the pressure of the fluid inside the fluid containing structure at the location where it is installed . Each of said sensors might be further configured for recording said fluid pressure values in function of the time , said recording being then used by the system for further analysis in order to detect and locate the leak 11 .

Each TGU Gj is configured for generating a transient pressure signal according to a predefined pattern Pj at a location L^s2j defined with respect to said fluid containing structure 10 . Typically, the TGUs are components of the fluid containing structure 10 that might be remotely controllable by the system according to the invention and that are capable of transiently modi fying the pressure of the fluid at the location where it is installed . It is for instance an automated washout valve , a remotely controllable hydrant ( e . g . fire hydrant ) , a discharge valve , or any other component that can, by its activation, generate said transient pressure signal according to a predefined pattern . Said predefined pattern is for instance a single pulse of predefined pressure amplitude and duration (pulse width) , notably characteri zed by a predefined shape , and which is typically obtained by controlling a washout valve ( e . g . the time between an opening and a closing of the washout valve defines the duration of said pulse , optionally its shape , and a reached maximum opening si ze of the washout valve during said opening defines the amplitude of the single pulse . The fluid pressure values that are measured by the sensors depend on the characteristics of said predefined pattern of the transient pressure signal . Preferentially, L^sli

L^s2j for any i and j, i.e. V i,j, the sensors and the TGUs are installed along the length of the fluid containing structure, e.g. a pipe network, at different locations. In particular, the TGUs are installed along said fluid containing structure 10 so that at least one sensor is comprised between two directly successive TGUs. Therefore, different configurations are possible when installing the sensors and the TGUs as shown in Fig. 1: for instance, four sensors, namely Sg-Sg, might be installed between two successive TGUs, namely G4 and G5, while between the TGUs G2 and G3, only one sensor S4 is installed. A same configuration for installing the sensors might be applied to the whole fluid containing structure 10, or different configurations (i.e. succession of TGUs and sensors) might be used depending on the characteristics of the fluid containing, e.g. in function of the network of pipes (taking into account for instance junctions of pipes and components of the pipeline comprising said pipes) , pipe diameter, length, etc. The maximum distance separating two successive sensors or two successive TGUs or a sensor from a TGU is typically comprised between 0.5-2 km in order to ensure that propagated signals can be worked out for detecting a leak..

In the following, and according to the present invention, a first device, e.g. a first TGU or respectively a first sensor, and a second device (of the same type as the first device) , e.g. a second TGU, or respectively a second sensor, are directly successive devices, e.g. directly successive TGUs, or respectively directly successive sensors, if there is a continuous portion or section or segment of said fluid containing structure that connects the location at which the first device is installed, e.g. the location of the first TGU, or respectively the location of the first sensor, to the location at which the second device is installed, e.g. the location of the second TGU, or respectively the location of the second sensor, and that between said first device and said second device, e.g. between the first TGU and the second TGU, or respectively between the first sensor and the second sensor, no third device (of the same type) , e.g. no third TGU, or respectively no third sensor, is installed. Typically, G1 and G2 are directly successive TGUs and S3 and S4 are also directly successive sensors according to the invention, while G2 is a direct neighbor for both S3 and S4. Indeed, according to the present invention, a first device, e.g. a TGU or a sensor, is directly neighboring a second device, e.g. another TGU or sensor, if there is a continuous portion or section or segment of said fluid containing structure that connects the first device and the second device (i.e. that goes from the first device to the second device) , and that between the first device and the second device, no third device, e.g. no other TGU or sensor, is installed.

Preferentially, the system comprises a predefined schedule, for instance stored in a memory of the system, configured for defining for each of the TGUs, and preferentially for each of the sensors, a start activation time and an end activation time. Typically, for each sensor, the system is configured for waking-up the concerned sensor at the start activation time defined for said concerned sensor and for putting said concerned sensor in a standby mode at said end activation time defined for said concerned sensor, the concerned sensor remaining in an active mode between said start activation time and end activation time. In said active mode, i.e. when it is activated, the sensors according to the invention are configured for measuring or acquiring said fluid pressure value at the location where it is installed. Similarly, for each TGU, the system is configured for waking-up the concerned TGU at the start activation time defined for said concerned TGU and for putting said concerned TGU in a standby mode at said end activation time defined for said concerned TGU, the concerned TGU remaining in an active mode between said start activation time and end activation time . In the active mode , i . e . when it is activated, the TGU is configured for generating said transient pressure signal . The TGU and the sensors are typically battery powered devices , and therefore , their temporary activation according to the predefined schedule enables to spare energy, avoiding thus early replacement of the batteries . Typically, the TGUs and sensors will be sleeping most of the time , and wake up preferably at night for minimi zing an impact on the fluid distribution network operations of said fluid containing structure .

Preferentially, the system 100 further comprises a Leak Monitoring System (hereafter LMS ) 101 that is configured for controlling the TGUs and the sensors , notably in function of said predefined schedule in order to activate and deactivate them at scheduled times , and for collecting the fluid pressure values measured by the sensors , e . g . by collecting recorded fluid pressure values , in order to analyze the acquired fluid pressure values with respect to reference fluid pressure values . Preferably, the LMS may comprise a processing unit , e . g . a processor, and a memory for enabling the processing of the collected fluid pressure values and their analysis . Said memory may store the predefined schedule . The LMS 101 might be remotely installed with respect to the fluid containing structure 10 . Typically, the LMS might be in charge of launching a leak alert in case of a detection of a leak 11 from the fluid pressure values acquired by the sensors , and, i f determined, for providing an indication of the detected leak position 12 , for instance as located between a TGU and a sensor that are direct neighbors , or as located between two directly neighboring sensors . Typically, the sensors and the TGU may communicate with the LMS using a known in the art wireless communication method . For instance , each sensor and each TGU, as well as the LMS may comprise a communication module configured for enabling a direct communication and/or transmission of data between said sensor or TGU and the LMS . The communication module , e . g . a radio transmitter and/or a radio receiver, preferentially a transceiver, is configured for enabling a direct communication between the sensor or TGU and the LMS , said communication being preferably a wireless communication . Typically, the communication module uses radio communication techniques based on low power, wide area, network communication, such as LoRa, LPWan, etc .

The general concept of the invention is further summari zed by the preferred method 200 illustrated by the flowchart of Fig . 2 . Let ' s consider a system 100 according to the invention comprising a set S I of sensors and a set S2 of TGUs as described in Fig . 1 .

At step 201 , in particular during normal operation of the system 100 , and notably according to said predefined schedule , the system 100 , for instance its LMS 101 , is configured for successively putting in the active mode the TGUs of at least one subset of said set S2 , said subset comprising preferentially at least two TGUs , the TGUs of the subset being thus in the active mode the one after the other so that only one TGU of said subset is in the active mode at a time . For this purpose , the system 100 , e . g . the LMS 101 , is configured for successively sending an activation signal to the TGUs of the subset , notably according to said predefined schedule , said activation signal being configured for activating the TGU receiving the signal . Preferentially, one or several sensors of the set S I are also activated, for instance by the LMS 101 sending an activation signal to each of said one or several sensors , notably according to said predefined schedule , or alternatively, in function of the location of the activated TGU . Indeed, and preferentially, said one or several sensors are selected by the system 100 , e . g . its LMS 101 as follows : the system 100 , or LMS 101 , is configured for putting in the active mode all sensors comprised between an activated TGU and any TGU that is a directly successive TGU to the activated TGU . In such a case , the system, e . g . its LMS , stores a map of the fluid containing structure that makes it possible to determine the position of each sensor and the position of each TGU . Alternatively, for each subset of S2 , a memory of the system or LMS defines a sensor subset of S I whose sensors needs to be activated each time a TGU of the S2 subset is activated . Alternatively, the system, e . g . its LMS , simply applies the activation times provided by the predefined schedule for activating the sensors . In other words , for each TGU activation, fluid pressure values might be acquired by a subset of S I comprising said one or several sensors that have been activated, wherein said S I subset might be defined in function of the location of the activated TGU or in function of said predefined schedule . As explained earlier, in in its active mode , the TGU is configured for generating said transient pressure signal at its location, and, in its active mode , the sensor is configured for acquiring fluid pressure values measured at its location in function of the time and that result from the activation of the TGU, i . e . from the propagation of the transient pressure signal in the fluid containing network from the location of the activated TGU to the location of the sensor .

Preferentially, each subset of the set S2 comprises TGUs installed on a predefined portion of the fluid containing structure , so that the fluid containing structure 10 might be split in several portions . Advantageously subsets that are separated by a distance greater than a predefined distance measured along the fluid containing structure might have identical or similar predefined schedule , so that two TGUs , each of a di f ferent subset , might be in the active mode as long as said two subsets are suf ficiently distant the one from the other . This prevents generated transient pressure signals to interfere with each other, while enabling to shorten the time required for detecting and locating leaks since several TGUs might be activated at the same time . Said di f ferent portions might be typically defined in function of the geometry and characteristics of the fluid containing structure .

At step 202 , the system 100 , e . g . its LMS 101 , is configured for acquiring, for each TGU activation and via said one or several sensors S± of the set S I that have been activated, fluid pressure values resulting from said TGU activation . In other words , each time a TGU of the S2 subset is activated, a transient pressure signal is generated by the activated TGU and said transient pressure signal propagates in both directions , i . e . upward and downward, with respect to the flow, i . e . fluid motion direction, of the fluid contained in the fluid containing structure 10 . Said transient pressure signal will then reach the sensors comprised within a certain ( or predefined) distance from the activated TGU and that correspond to said one or several sensors activated by the system 100 or LMS . Indeed, given signal distortion increases with the distance , only a subset of the set S I of sensors need to be activated, namely said one or several sensors S±, that are the closest , i . e . within said predefined distance , from the activated TGU .

At step 203 , and for each TGU activation, the system, for instance the LMS 101 , is configured for analyzing, for each of said one or several sensors , the fluid pressure values acquired by the concerned sensor for said TGU activation with respect to reference fluid pressure values measured during a training phase by said concerned sensor for said TGU that has been activated . Optionally, said analysis might be reali zed by the sensor itsel f . In such a case , the sensor may comprise a processing unit configured for performing said analysis , and for instance detecting a leak from said analysis , e . g . a correlation or comparison, of its measured fluid pressure values with its reference fluid pressure values defined for the TGU that has been activated and for which it measured, and optionally recorded, said fluid pressure values . Said analysis may comprise determining a time of arrival of the transient pressure signal at the sensor, analyzing the shape of the received transient pressure signal ( e . g . amplitude of a pulse of the received transient pressure signal , width of a pulse , or number of pulses comprised in the received transient pressure signal , which are typically determined from the measured fluid pressure values ) and/or determining an amplitude of the received transient pressure signal that can be determined from said measured fluid pressure values , and comparing said arrival time and/or shape and/or amplitude to said reference fluid pressure values , which may comprise a range of values and/or a mean value for said arrival time , and/or a range and/or mean values characteri zing said shape , and/or a range of values and/or a mean value for said amplitude .

The aim of the analysis is to determined similarities between the measured fluid pressure values and the reference fluid pressure values. Said analysis might be a comparison or correlation of measured fluid pressure values with reference fluid pressure values that are defined for each couple (Gj,S±) , wherein Gj belong to said S2 subset and S± belongs to said one or several sensors of the set SI. In other words, the system, e.g. its LMS, may store for each couple (Gj,S±) said reference fluid pressure values that are used in said analysis .

According to the present invention, the reference fluid pressure values have been obtained during said training phase of the system 100. For said training phase, the assumption is made that the fluid containing structure comprises no leak. Then, for said training phase, the steps 201 and 202 previously described are performed for all S2 subsets, i.e. successively activating the TGUs of each subset of TGUs of the set S2, wherein, within each subset, only one TGU is activated at a time, and acquiring, for each TGU activation and via said one or several sensors S± of the set SI, fluid pressure values resulting from said TGU activation. The fluid pressure values acquired during the training phase serve then as reference values when the system is controlled during its normal operation. Given that the fluid containing structure 10 may degrade with time, new reference fluid pressure values might be acquired, e.g. periodically, for use in the analysis of step 203 for reducing false positive leak alerts. At step 204, the system 100, e.g. its LMS 101, is configured for detecting and locating a leak within the fluid containing structure 10 from said analysis. As explained earlier, the detection of a leak in the vicinity of a sensor might be realized by the sensor itself, in which case the latter analyses currently measured fluid pressure values acquired for a TGU with respect to reference fluid pressure values defined for said TGU, and then sends, e.g. to the LMS, an alert indicating a presence of a leak. Alternatively, the sensors having acquired fluid pressure values for an activated TGU may send said acquired fluid pressure values to the LMS which performs the analysis. In each case, the system is configured for detecting or signaling a presence of a leak in a vicinity of a sensor if a difference between measured fluid pressure values acquired for a TGU of an S2 subset by said sensor and reference fluid pressure values measured for said TGU by said sensor exceeds a predefined threshold. Indeed, after the successive activation of the TGUs of the S2 subset, the system knows, in function of the activated TGU, which sensors measured fluid pressure values that led to a detection of a leak, i.e. for which the measured fluid pressure values differ from the reference fluid pressure values, which indicates the presence of a leak 11. Remain to locate said leak.

For this purpose, the system may implement a decisional process. Preferably, if a leak is detected from fluid pressure values acquired by a sensor and resulting from the transient pressure signal generated by a TGU that is a directly neighboring TGU of said sensor (i.e. no other sensor or no other TGU is located between the concerned sensor and TGU) , then the system is configured for locating the leak between said sensor and said TGU. And if a leak is detected from fluid pressure values measured by a first sensor and from fluid pressure values measured by a second sensor, wherein said first and second sensor are directly neighboring sensors and belong notably to the S I subset comprising said one or several sensors , then, i f for the first sensor, the leak is detected from fluid pressure values resulting from a transient pressure signal generated by a first TGU that is closest to the second sensor than to the first sensor, and, for the second sensor, the leak is detected from fluid pressure values resulting from a transient pressure signal generated by a second TGU that is closest to the first sensor than to the second sensor, then the system is configured for locating said leak between the first sensor and the second sensor .

For instance , and referring to Fig . 1 , i f a subset comprises the TGU Gi and G2 , then when G2 is activated, S4 will detect the presence of the leak 11 , and, when G3 is activated, S4 will detect no leak . In conclusion, the system will deduce that a leak is present between G2 and S4 . Let ' s say now that a leak is present between Se and S7 , and wherein the S2 subset comprises the TGU G4 and G5 . When G4 is activated, then S5 and Se will detect no leak, while S7-S8 will detect the presence of a leak . Then G5 is activated, and this time S7 and Ss detect no leak, but S5 and Se detect or report the presence of a leak . In this case , considering the direction of propagation of the transient pressure signal generated by G4 and then G5 , the system can easily deduce that the leak is located between Se and S7 , since it is the only solution satis fying the obtained detection scheme .

Once a leak is detected, the system 100 may send or display a leak alert comprising for instance an indication of the location of the sensor ( s ) for which the measured fluid pressure values conducted to the detection of the leak 11 . I f determined, the leak position 12 with respect to the fluid containing structure 10 might be part of the leak alert .

To conclude , the present invention proposes a system 100 and a method 200 for detecting and locating a leak 11 , that advantageously do not require any model of the fluid containing structure for detecting and locating said leak, which are able to save energy by activating sensors and TGUs only at predefined times , reducing thus power consumption signi fi- cantly .

Claims

1. System (100) for detecting and locating a leak (11) in a fluid containing structure (10) , said system (100) comprising :

- a set SI of N sensors S±, i=l,...,N, with N > 1, wherein each sensor S± is configured, when activated, for measuring fluid pressure values in function of the time at a location L^sli defined with respect to said fluid containing structure;

- a set S2 of M transient generation units (hereafter "TGU") Gj, with j = l,...,M, with M > 2, wherein each TGU Gj is configured, when activated, for generating a transient pressure signal according to a predefined pattern Pj at a location L^s2j defined with respect to said fluid containing structure; the system (100) being configured for

- successively activating the TGUs of at least one subset of TGUs of said set S2, wherein, within each subset, only one TGU is activated at a time; for acquiring, for each TGU activation and via one or several sensors S± of the set SI, fluid pressure values resulting from said TGU activation; for each TGU activation, analyzing, for each of said one or several sensors, the fluid pressure values acquired by the concerned sensor for said TGU activation with respect to reference fluid pressure values defined for said concerned sensor with respect to the TGU that has been activated;

- detecting and locating said leak (11) within the fluid containing structure (10) from said analysis.

2. System (100) according to claim 1, wherein the TGUs and the sensors are installed along said fluid containing structure ( 10 ) so that between two directly successive TGUs , there is at least one sensor .

3 . System ( 100 ) according to claim 1 or 2 , comprising, for enabling the successive activation of the TGUs and acquisition of the fluid pressure values by the sensors , a predefined schedule configured for defining for each of the TGUs and, optionally for each of the sensors , a start activation time and an end activation time .

4 . System ( 100 ) according to claim 3 , configured for : for each sensor, waking-up the concerned sensor at the start activation time defined for said concerned sensor and for putting said concerned sensor in a standby mode at said end activation time defined for said concerned sensor, the concerned sensor remaining in an active mode between said start time and end time ; for each TGU, waking-up the concerned TGU at the start activation time defined for said concerned TGU and for putting said concerned TGU in a standby mode at said end activation time defined for said concerned TGU, the concerned TGU remaining in an active mode between said start time and end time .

5. System ( 100 ) according to claims 1-4 , wherein said reference fluid pressure values have been measured by said concerned sensor during a training phase of the system, during which each TGU has been activated for generating said transient pressure signal , and for each TGU activation, said one or several sensors have been activated for acquiring each, reference fluid pressure values , wherein said reference fluid pressure values result from said predefined pattern .

6. System ( 100 ) according to one of the claims 1-5 , wherein several TGUs of the system are configured for being active at the same time , each of said several TGUs belonging to a di f ferent subset of the set S2 , wherein a predefined distance separates each of the di f ferent subsets .

7 . System ( 100 ) according to one of the claims 1- 6 , configured for detecting a leak i f said analysis results in a di f ference between the measured fluid pressure values and the reference fluid pressure values exceeding a predefined threshold .

8 . System ( 100 ) according to claim 7 , wherein i f the leak is detected by a sensor from acquired fluid pressure values resulting from a transient pressure signal generated by a TGU that is a directly neighboring TGU of said sensor, then the system is configured for locating the leak between said sensor and said TGU, and i f a leak is detected by a first sensor and by a second sensor that are directly neighboring sensors , then, i f for the first sensor, the leak is detected from fluid pressure values resulting from a transient pressure signal generated by a first TGU that is closest to the second sensor than to the first sensor, and, for the second sensor, the leak is detected from fluid pressure values resulting from a transient pressure signal generated by a second TGU that is closest to the first sensor than to the second sensor, then the system is configured for locating said leak between the first sensor and the second sensor .

9. System (100) according to one of the claims 1 to 8, comprising a Leak Monitoring System configured for being remotely located with respect to the fluid containing structure and configured for controlling the TGUs and collecting the fluid pressure values acquired by the sensors .

10. Method (200) for detecting and locating, by means of a system (100) , a leak (11) in a fluid containing structure (10) , wherein said system (100) comprises:

- a set S2 of M transient generation units (hereafter "TGU") Gj, with j = l,...,M, with M > 2, wherein each TGU Gj is configured, when activated, for generating a transient pressure signal according to a predefined pattern Pj at a location L^s2j defined with respect to said fluid containing structure; the method (200) comprising the steps:

- successively activating (201) the TGUs of at least one subset of TGUs of said set S2, wherein, within each subset, only one TGU is activated at a time;

- acquiring (202) , for each TGU activation and via one or several sensors S± of the set SI, fluid pressure values resulting from said TGU activation; for each TGU activation, analyzing (203) , for each of said one or several sensors, the fluid pressure values acquired by the concerned sensor for said TGU activation with respect to reference fluid pressure values defined for said concerned sensor with respect to the TGU that has been activated;

- detecting and locating (204) said leak (11) within the fluid containing structure from said analysis.

11. Method (200) according to claim 10, comprising installing along said fluid containing structure the TGUs and the sensors so that between two directly successive TGUs, there is at least one sensor.

12. Method (200) according to claim 10 or 11, comprising measuring the reference fluid pressure values by said concerned sensor during a training phase of the system, wherein, during said training phase, each TGU has been activated for generating said transient pressure signal, and for each TGU activation, said one or several sensors have been activated for acquiring each, reference fluid pressure values, wherein said reference fluid pressure values result from said predefined pattern.

13. Method (200) according to one of the claims 10 to 12, comprising activating several TGUs of the system so that they are active at a same time, wherein each of said several TGUs belongs to a different subset of the set S2, wherein a predefined distance separates each of the different subsets.

14. Method (200) according to one of the claims 10 to 13, wherein a leak (11) is detected if said analysis results in a difference between the measured fluid pressure values and the reference fluid pressure values exceeding a predefined threshold.

15 . Method ( 200 ) according to one of the claims 10 to 14 , wherein i f the leak ( 11 ) is detected by a sensor from acquired fluid pressure values resulting from a transient pressure signal generated by a TGU that is a directly neighboring TGU of said sensor, then the system is configured for locating the leak ( 11 ) between said sensor and said TGU, and i f a leak ( 11 ) is detected by a first sensor and by a second sensor that are directly neighboring sensors , then, i f for the first sensor, the leak ( 11 ) is detected from fluid pressure values resulting from a transient pressure signal generated by a first TGU that is closest to the second sensor than to the first sensor, and, for the second sensor, the leak ( 11 ) is detected from fluid pressure values resulting from a transient pressure signal generated by a second TGU that is closest to the first sensor than to the second sensor, then the system is configured for locating said leak ( 11 ) between the first sensor and the second sensor .