The invention relates to a method for micro-leakage detection in a fluid system. Further on, the invention relates to a micro-leakage detection apparatus suitable for micro-leakage detection in a fluid system.

US 2019/0128762 A1 discloses an apparatus for fluid flow detection. The appa ratus makes use of a signal provided by a pipe temperature sensor and of a signal provided by an ambient temperature sensor. A low flow algorithm may attempt to detect flow leaks such as a dripping tap. If no leak is present during a quiet period, the ambient temperature and pipe temperature will generally tend to be close together. If, on the other hand, a low flow leak is present during a quiet period, there will be generally a noticeable difference between the ambient temperature and pipe temperature.

EP 2 180304 A1 , GB 2 572274 A, US 4336708 A, US 10527516 B2 as well as JP 6611 650 B2 disclose other prior art.

Against this background, a novel method for micro-leakage detection in a fluid system and a novel micro-leakage detection apparatus are provided.

The novel method for micro-leakage detection in a fluid system comprises at least the following steps:

Measure the fluid flow through a fluid pipe by a flow meter.

Measure the pipe temperature of the fluid pipe by at least one pipe temperature sensor. When there is no fluid flow measured by the flow meter, particularly because or when the fluid flow through the fluid pipe is stopped by the fluid valve being closed, analyze the pipe temperature for the micro-leakage detection. The novel method for micro-leakage detection is based both on a flow meas urement by a flow meter and a pipe temperature measurement by the at least one pipe temperature sensor. The novel method allows a very simple and relia ble micro-leakage detection in a fluid system. The micro-leakage which can be detected by making use of the invention is below a measuring resolution or a measuring range of the flow meter.

The pipe temperature of the fluid pipe may be measured by the at least one pipe temperature sensor when the fluid flow through the fluid pipe is allowed and when the fluid flow through the fluid pipe is stopped, wherein the measured fluid flow is analyzed for the micro-leakage detection only if there is no fluid flow measured by the flow meter. In this case, the at least one pipe temperature sensor is active and measures the pipe temperature when fluid flow is meas ured by the flow meter and when no fluid flow is measured by the flow meter. However, the measured pipe temperature is analyzed for micro-leakage detec- tion only under the condition that there is no fluid flow measured by the flow me ter.

Alternatively, the pipe temperature of the fluid pipe may both be measured and analyzed for the micro-leakage detection only when the fluid flow through the fluid pipe is stopped. In this case, the at least one pipe temperature sensor is inactive when fluid flow is measured by the flow meter. The at least one pipe temperature sensor then becomes activated when no fluid flow is measured by flow meter. In this case, the pipe temperature is measured and analyzed for micro-leakage detection only under the condition that there is no fluid flow measured by the flow meter. According to a first embodiment of the method for micro-leakage detection the same has the additional following steps: When there is no fluid flow measured by the flow meter after the fluid flow through the fluid pipe has been stopped, calculate a temporal gradient of the pipe temperature. If the temporal gradient of the pipe temperature differs more than a first threshold from a first reference value, and if there is no flow measured by the flow meter, then detect micro leakage.

According to a second embodiment of the method for micro-leakage detection the same has the additional following steps: Measure the pipe temperature of the fluid pipe by a first pipe temperature sensor and by a second pipe tempera ture sensor being positioned at different locations of the fluid pipe. When there is no fluid flow measured by the flow meter for the defined time interval, calcu late a temperature difference between the pipe temperatures measured by the first and second pipe temperature sensors. If the temperature difference be tween the pipe temperatures differs more than a second threshold from a sec ond reference value, and if there is no flow measured by the flow meter, then detect micro-leakage. The above first and second embodiments are preferred. The same are ambient temperature independent and do not require the measurement of the ambient temperature. Such an ambient temperature independent micro-leakage detec tion is very simple and reliable. It is possible to use the first and second embod iment in combination, meaning that micro-leakage is detected if the temporal gradient of the pipe temperature differs more than the first threshold from the first reference value and/or if the temperature difference between the pipe tem peratures differs more than the second threshold from the second reference value. The novel micro-leakage detection apparatus suitable for micro-leakage detec tion in a fluid system is defined in claim 15. Preferred developments of the invention are provided by the dependent claims and the description which follows. Exemplary embodiments are explained in more detail on the basis of the draw ing, in which:

Figure 1 shows a schematic diagram of a fluid flow system; Figure 2 shows a signal flow diagram illustrating a first embodiment; Figure 3 shows a signal flow diagram illustrating a second embodiment; Figure 4 shows a signal flow diagram illustrating a third embodiment, Figure 5 shows a time diagram further illustrating the first embodiment, Figure 6 shows a time diagram further illustrating the second embodi ment, Figure 7 shows a time diagram further illustrating the third embodiment.

Figure 1 shows a schematic diagram of a fluid flow system 10, namely of a po table water system, of a building 11. The fluid flow system 10 comprises a fluid pipe 12 running at least partially inside of the building 11. This fluid pipe 12 is connected to a main water pipe 13 running outside of the building 11. This fluid pipe 12 comprises a fluid valve 14. The fluid valve 14 may be a water tap. A fluid flow through the fluid pipe 12 is stopped when the fluid valve 14 is closed. A fluid flow through the fluid pipe 12 is allowed when the fluid valve 14 is opened. The fluid pipe 12 may be made from a metal like copper or from a plas- tic like polypropylene.

The present invention relates to a method for micro-leakage detection in the fluid system 10 and to a micro-leakage detection apparatus. Figure 1 shows such a micro-leakage detection apparatus 15. The micro-leakage detection apparatus 15 receives at least signals from a flow meter 16 and from at least one pipe temperature sensor 17a, 17b.

The flow meter 16 is assigned to the fluid pipe 12 and measures the fluid flow through the fluid pipe 12.

The flow meter 16 has a measuring range or measuring resolution. The flow meter 16 is configured to measure a fluid flow through the fluid pipe 12 when the fluid valve 14 is opened, meaning that there is a regular fluid consumption across the fluid valve 14. However, when the fluid valve 14 is closed, there may be an irregular fluid consumption caused by micro-leakage. The micro-leakage causes a certain fluid flow being below the measuring range or measuring reso lution of the flow meter 16. So, micro-leakage cannot be detected by the flow meter 16 as such, namely by the flow meter 16 alone.

The at least one pipe temperature sensor 17a, 17b is also assigned to the fluid pipe 12 and measures the pipe temperature of the fluid pipe 12.

Figure 1 shows a first pipe temperature sensor 17a and a second pipe tempera ture sensor 17b. Further, Figure 1 shows an ambient temperature sensor 18 measuring an ambient temperature within the building. The ambient tempera ture sensor 18 may be positioned in the proximity of the fluid pipe 12. Only one of the first and second temperature sensors 17a, 17b and the flow meter 16 are mandatory units for the present invention. The ambient temperature sensor 18 is an optional unit. If an ambient temperature sensor 18 is present, the same is preferably positioned in the proximity of the fluid pipe 12.

A first embodiment of the invention makes use of at least one pipe temperature sensor 17a, 17b and of the flow meter 16 only. A second embodiment of the invention makes use of the first and second pipe temperature sensors 17a and 17b and of the flow meter 16.

The micro-leakage detection apparatus 15 has an interface 15a being config ured to receive signals or data from the flow meter 16 and an interface 15b be ing configured to receive signals or data from the at least one pipe temperature sensor 17a, 17b.

A third embodiment makes use of at least one pipe temperature sensor 17a, 17b, of the flow meter 16 and of the ambient temperature sensor 18. In this case the micro-leakage detection apparatus 15 has an interface 15c being con figured to receive signals or data from the ambient temperature sensor 18.

The method for micro-leakage detection in the fluid system 10 comprises at least the following steps:

Measure the fluid flow through the fluid pipe 12 by the flow meter 16.

Measure the pipe temperature of the fluid pipe 12 by at least one pipe tempera ture sensor 17a, 17b.

When there is no fluid flow measured by the flow meter 16 because or when the fluid flow through the fluid pipe 12 is stopped by the closed fluid valve 14, ana lyze the pipe temperature for the micro-leakage detection.

The pipe temperature of the fluid pipe 12 may be measured by at least one pipe temperature sensor 17a, 17b when the fluid flow through the fluid pipe 12 is al lowed and when the fluid flow through the fluid pipe 12 is stopped, wherein the measured fluid flow is analyzed for the micro-leakage detection only when there is no fluid flow measured by the flow meter 16. In this case, the at least one pipe temperature sensor 17a, 17b is active and measures the pipe temperature when fluid flow is measured by the flow meter 16 and when no fluid flow is measured by the flow meter 16. However, the measured pipe temperature is analyzed for micro-leakage detection only under the condition that there is no fluid flow measured by the flow meter 16.

Alternatively, the pipe temperature of the fluid pipe 12 may both be measured and analyzed for the micro-leakage detection only when the fluid flow through the fluid pipe 12 is stopped by the fluid valve 14.

In this alternative case, the at least one pipe temperature sensor 17a, 17b is inactive or becomes inactivated when fluid flow is measured by the flow meter 16. The at least one pipe temperature sensor 17a, 17b is active or becomes activated when no fluid flow is measured by flow meter 16. In this case, the pipe temperature is both measured and analyzed for micro-leakage detection only under the condition that there is no fluid flow measured by the flow meter 16.

The method for micro-leakage detection is based both on a flow measurement by the flow meter 16 and pipe temperature measurement by the at least one pipe temperature sensor 17a, 17b. The method allows a very simple and relia ble micro-leakage detection in the fluid system 10. The invention allows to de tect micro-leakage that causes a fluid flow being below the measuring range or measuring resolution of the flow meter 16. The micro-leakage detection apparatus 15 is configured to execute the above method steps. The interface 15a of the micro-leakage detection apparatus 15 is configured to receive signals or data from the flow meter 16 measuring the fluid flow through the fluid pipe 12. The interface 15b of the micro-leakage detection apparatus 15 is configured to receive signals or data from the at least one pipe temperature sensor 17a, 17b measuring the pipe temperature of the fluid pipe 12. A processor 15d of the micro-leakage detection apparatus 15 is configured to detect micro-leakage by analyzing the pipe temperature provided by the at least one pipe temperature sensor 17a, 17b when there is no fluid flow measured by the flow meter 16. The micro-leakage detection apparatus 15 further comprises a memory 15e.

If the pipe temperature of the fluid pipe 12 is both measured and analyzed only when there is no fluid flow measured by the flow meter 16, then the processor 15d of the micro-leakage detection apparatus 15 is configured to active the least one pipe temperature sensor 17a, 17b under the condition that there is no fluid flow measured by the flow meter 16.

As mentioned above, a first embodiment of the invention makes use of at least one pipe temperature sensor 17a and/or 17b and of the flow meter 16 only. In the following description of the first embodiment it is presumed that the pipe temperature sensor 17a is used for the measurement of the pipe temperature. In this first embodiment a temporal gradient of the pipe temperature measured by the pipe temperature sensor 17a is calculated when there is no fluid flow measured by the flow meter 16 after the fluid flow through the fluid pipe 12 has been stopped. The condition that the fluid flow has been stopped can be de tected on basis of the signal provided by the flow meter 16, namely when there is fluid flow measured by the flow meter 16 and subsequently no fluid flow measured by the flow meter 16. The temporal gradient is also often called gra dient over time. If the temporal gradient of the pipe temperature differs more than a first threshold from a first reference value, and if there is no flow meas ured by the flow meter 16, then micro-leakage is detected.

Figure 2 shows a signal flow diagram for the first embodiment of the invention. In step 20 the flow meter 16 measures the fluid flow though the fluid pipe 12. In step 21 the pipe temperature sensor 17a measures the pipe temperature of the fluid pipe 12.

In step 22 it is determined if the flow meter 16 measures a fluid flow through the fluid pipe 12. If it is determined in step 22 that the flow meter 16 measures a fluid flow through the fluid pipe 12, then the method goes back to step 20. If it is determined in step 22 that the flow meter 16 measures no fluid flow through the fluid pipe 12, then the method goes to step 23.

In step 23 is determined if the flow meter 16 measures no fluid flow through the pipe. If this is not the case, the method goes back to step 20. If this is the case, the method goes to step 24. In step 24 the temporal gradient - also often called gradient over time - of the pipe temperature measured by the pipe temperature sensor 17a is calculated.

Then, in step 25 it is determined if the temporal gradient of the pipe temperature differs more than a first threshold from a first reference value or not.

If the temporal gradient of the pipe temperature does not differ more than the first threshold from the first reference value, no micro-leakage is detected in step 26. If the temporal gradient of the pipe temperature differs more than the first threshold from the first reference value, and if there is still no fluid flow measured by the flow meter 16, then in step 27 micro-leakage is detected.

In connection with the first embodiment, alternatively the pipe temperature sen sor 17b may be used for the measurement of the pipe temperature. Further on, both pipe temperature sensors 17a, 17b may be used and an average value may be calculated for the pipe temperature. The first reference value for the temporal gradient of the pipe temperature may be determined as follows: If there is no fluid flow measured after the fluid flow through the fluid pipe has been stopped, then calculate and store the temporal gradient of the pipe temperature. Calculate an average value from the stored temporal gradients. Determine the first threshold from this average value.

The average value may be multiplied by a security-factor to determine the first reference value.

The above method is executed by the micro-leakage detection apparatus 15 in or at a defined sampling rate. The calculation of the temporal gradient may take place at each sampling time of the sampling rate. However, the calculated tem poral gradient may not be stored at each sampling time of the sampling rate. It is possible that the calculated temporal gradient is only stored example given every 10 times or every 20 times or every 50 times or every 100 times after cal culation of the same. These calculated temporal gradients may be stored in a ring buffer of the memory 15e of the micro-leakage detection apparatus 15. The ring buffer may have a defined buffer size. If the ring buffer is completely filled, then the average value from the stored temporal gradients may be calculated. If the the ring buffer is not completely filled, then the average value may not be calculated. If the ring buffer is completely filled and if a newly calculated tem poral gradient is to be stored, then the oldest one of the stored temporal gradi ents becomes replaced by the newly calculated temporal gradient and the aver age value is newly calculated.

The calculated temporal gradient may only be stored and used to calculate the average value if the absolute value of a difference between the calculated tem poral gradient and a previously calculated temporal gradient or the absolute value of a difference between the calculated temporal gradient and an average value of previously stored temporal gradients is below a respective threshold. Figure 5 shows a time diagram further illustrating the first embodiment of the invention. Figure 5 shows as a function of the time t a fluid flow rate 50 and a pipe temperature 51 measured by the pipe temperature sensor 17a.

At point of times t1 , t3 and t5 a respective fluid flow 50 through the fluid pipe 12 starts. At point of times t2, t4 and t6 the respective fluid flow 50 through the fluid pipe 12 stops because of a closed the fluid valve 14.

After the fluid flow through the fluid pipe 12 has been stopped at the point of times t2, t4 and t6, the temporal gradient 52 of the pipe temperature 51 is calcu lated. If the calculated temporal gradient 52 of the pipe temperature 51 differs more than the first threshold from the first reference value, and if there is no flow measured by the flow meter 16, then micro-leakage is detected.

In Figure 5, the temporal gradients 52 calculated at point of times t2, t4 do not differ more than the first threshold from the first reference value. So, no micro leakage is detected at point of times t2, t4. The temporal gradient 52 calculated at point of times t6 differs more than the first threshold from the first reference value. So, micro-leakage 53 is detected at point of times t6. The first reference value may correspond to the average of the temporal gradients 52 calculated at point of times t2, t4.

As mentioned above, a second embodiment of the invention makes use of the first and second pipe temperature sensors 17a, 17b and of the flow meter 16.

In this second embodiment the pipe temperature of the fluid pipe 12 is meas ured by the first pipe temperature sensor 17a and by the second pipe tempera ture sensor 17b being positioned at different locations of the fluid pipe 12. The pipe temperature sensors 17a and 17b have a different distance to the fluid valve 14. The pipe temperature sensor 17b is positioned closer to the fluid valve 14 than the pipe temperature sensor 17a. The distance between the pipe tem perature sensors 17a and 17b may be at least 20 cm.

When there is no fluid flow measured by the flow meter 16 for a defined time interval, a temperature difference between the pipe temperatures measured by the first and second pipe temperature sensors 17a and 17b is calculated. If the temperature difference between these pipe temperatures differs more than a second threshold from a second first reference value, and if there is no flow measured by the flow meter 16, then micro-leakage is detected.

Figure 3 shows a signal flow diagram for the second embodiment of the inven tion.

In step 30 the flow meter 16 measures the fluid flow though the fluid pipe 12. In step 31 the first pipe temperature sensor 17a measures the pipe temperature of the fluid pipe 12. In step 32 the second pipe temperature sensor 17b measures the pipe temperature of the fluid pipe 12.

In step 33 it is determined if the flow meter 16 measures no fluid flow through the pipe. If this is not the case, the method goes back to step 30. If this is the case, the method goes to step 34.

In step 34 it is determined if the flow meter 16 measured no fluid flow through the fluid pipe 12 for a defined time interval after the fluid flow has been stopped. If this is not the case, the method goes back to step 30. If this is the case, the method goes to step 35.

In step 35 the temperature difference between the pipe temperatures measured by the first and second pipe temperature sensors 17a and 17b is calculated. Then, in step 36 it is determined if this temperature difference differs more than a second threshold from the second reference value or not. If this temperature difference does not differ more than the second threshold from the second reference value, no micro-leakage is detected in step 37.

If this temperature difference differs more than the second threshold from the second reference value, and if there is still no fluid flow measured by the flow meter 16, then in step 38 micro-leakage is detected.

The second reference value is determined as follows: If there is no fluid flow measured for the defined time interval, then calculate and store the temperature difference between the pipe temperatures measured by the first and second pipe temperature sensors 17a, 17b. Calculate an average value from the stored temperature differences. Determine the second threshold from this average val ue.

The average value may be multiplied by a security-factor to determine the sec- ond reference value.

The calculation of the temperature difference between the pipe temperatures measured by the first and second pipe temperature sensors 17a, 17b may take place at each sampling time of the sampling rate. However, said temperature difference may not be stored at each sampling time of the sampling rate. It is possible that said calculated temperature difference is stored example given every 10 times or every 20 times or every 50 times or every 100 times after cal culation of the same. Said calculated temperature difference may be stored in a ring buffer of the memory 15e of the micro-leakage detection apparatus 15. The ring buffer may have a defined buffer size. If the ring buffer is completely filled, then the average value may be calculated. If the ring buffer is not completely filled, then the average value may not be calculated. If the ring buffer is com pletely filled and if a newly calculated temperature difference between the pipe temperatures measured by the first and second pipe temperature sensors 17a, 17b is to be stored, then the oldest one of the stored temperature differences becomes replaced by the newly calculated temperature difference and the av erage value is newly calculated.

The calculated temperature difference may only be stored and used to calculate the average value if the absolute value of a difference between the calculated temperature difference and a previously calculated temperature difference or if the absolute value of a difference between the calculated temperature differ ence and an average value of previously stored temperature differences is be low a threshold. Figure 6 shows a time diagram further illustrating the second embodiment of the invention. Figure 6 shows as a function of the time t a fluid flow rate 60 and pipe temperatures 62, 63 measured by the pipe temperature sensors 17a, 17b.

At point of times t1 , t3 and t5 a respective fluid flow 60 through the fluid pipe 12 starts. At point of times t2, t4 and t6 the respective fluid flow 60 through the fluid pipe 12 stops.

When there is no fluid flow measured by the flow meter 16 for a defined time interval after the fluid flow through the fluid pipe 12 has been stopped by closing the fluid valve 14 at the point of times t2, t4 and t6, a temperature difference 63 between the pipe temperatures 61 , 62 measured by the first and second pipe temperature sensors 17a, 17b is calculated. The condition that the fluid flow has been stopped can be detected on basis of the signal provided by the flow meter 16, namely when there is fluid flow measured by the flow meter 16 and subse- quently no fluid flow measured by the flow meter 16. If the temperature difference 63 between the pipe temperatures 61, 62 differs more than the second threshold from the second reference value, and if there is no flow measured by the flow meter 16, then micro-leakage 64 is detected.

In Figure 6, the temperature differences 63 calculated a defined time interval after the point of times t2, t4 do not differ more than the second threshold from the second reference value. So, no micro-leakage is detected. The temperature difference 63 calculated a defined time interval after the point of time t6 differs more than the second threshold from the second reference value. So, micro leakage 64 is detected. The second reference value may correspond to the av erage of the temperature differences 63 calculated the defined time interval af ter the point of times t2, t4.

The above first and second embodiments are preferred. The same do not re quire the measurement of the ambient temperature. Such an ambient tempera ture independent micro-leakage detection is very simple and reliable.

It is possible to use the first and second embodiment in combination. So, micro leakage may be detected if the temporal gradient of the pipe temperature differs more than the first threshold from the first reference value or if the temperature difference between the pipe temperatures differs more than the second thresh old from the second reference value.

A third embodiment makes use of the ambient temperature sensor 18.

In this third embodiment, the ambient temperature is measured by the ambient temperature sensor 18. When there is no fluid flow measured by the flow meter 16 for a defined time interval, then a temperature difference between the pipe temperature and the ambient temperature is calculated. If the temperature difference between the pipe temperature and the ambient temperature differs more than a third threshold from a third reference value, and if there is still no flow measured by the flow meter 16, then micro-leakage is de tected.

Figure 4 shows a signal flow diagram for the third embodiment of the invention.

In step 40 the flow meter 16 measures the fluid flow though the fluid pipe 12. In step 41 at least one of pipe temperature sensors 17a, 17b measures the pipe temperature. In step 42 the ambient temperature sensor 18 measures the am bient temperature.

In step 43 it is determined if the flow meter 16 measures no fluid flow through the fluid pipe 12. If this is not the case, the method goes back to step 40. If this is the case, the method goes to step 44.

In step 44 it is determined if the flow meter 16 measured no fluid flow through the fluid pipe 12 for a defined time interval. If this is not the case, the method goes back to step 40. If this is the case, the method goes to step 45.

In step 45 the temperature difference between the pipe temperature measured by the respective pipe temperature sensor 17a, 17b and the ambient tempera ture measured by the ambient temperature sensor 18 is calculated.

Then, in step 46 it is determined if this temperature difference differs more than a third threshold from the third reference value or not. If this temperature differ ence does not differ more than the third threshold from the third reference val ue, no micro-leakage is detected in step 46. If this temperature difference differs more than the third threshold from a third reference value, and if there is still no fluid flow measured by the flow meter 16, then in step 47 micro-leakage is de tected. The third threshold for the temperature difference between the pipe temperature and the ambient temperature may be determined as follows: If there is no fluid flow measured for the defined time interval, then calculate and store the tem perature difference between the pipe temperature and the ambient temperature. Calculate an average value from the stored temperature differences. Determine the first threshold from this average value. The average value may be multiplied by a factor to determine the third reference value.

The calculation of the temperature difference between pipe temperature and the ambient temperature may take place at each sampling time of the sampling rate. However, said temperature difference may not be stored at each sampling time of the sampling rate. It is possible that said calculated temperature differ ence is stored example given every 10 times or every 20 times or every 50 times or every 100 times after calculation of the same. Said calculated tempera ture difference may be stored in a ring buffer of the memory 15e of the micro leakage detection apparatus 15. The ring buffer may have a defined buffer size. If the ring buffer is completely filled, then the average value may be calculated. If the ring buffer is not completely filled, then the average value may not be cal culated. If the ring buffer is completely filled and if a newly calculated tempera ture difference between the pipe temperature and the ambient temperature is to be stored, then the oldest one of the stored temperature differences becomes replaced by the newly calculated temperature difference and the average value is newly calculated.

The calculated temperature difference may only be stored and used to calculate the average value if the absolute value of a difference between the calculated temperature difference and a previously calculated temperature difference or if the absolute value of a difference between the calculated temperature differ ence and an average value of previously stored temperature differences is be low a threshold. Figure 7 shows a time diagram further illustrating the third embodiment of the invention. Figure 7 shows as a function of the time t a fluid flow rate 70 and a temperature difference 71 between the pipe temperature measured by one of the pipe temperature sensors 17a, 17b and the ambient temperature.

At point of times t1 , t3 and t5 a respective fluid flow 70 through the fluid pipe 12 starts. At point of times t2, t4 and t6 the respective fluid flow 70 through the fluid pipe 12 stops by closing the fluid valve 14. When there is no fluid flow meas- ured by the flow meter 16 for a defined time interval after the fluid flow through the fluid pipe 12 has been stopped at the point of times t2, t4 and t6, the tem perature difference 71 is determined. If the temperature difference 71 between the pipe temperature and the ambient temperature differs more than the third threshold from the third reference value, and if there is no flow measured by the flow meter 16, then micro-leakage 73 is detected.

In Figure 7, the values V1, V2 of temperature difference 71 determined a de fined time interval after the point of times t2, t4 do not differ more than the third threshold from the third reference value. So, no micro-leakage is detected. The value V3 of the temperature difference 71 determined a defined time interval after the point of time t6 differs more than the third threshold from the third ref erence value. So, micro-leakage 72 is detected. The third reference value may correspond to the average of the temperature difference values V1 , V2 calcu lated the defined time interval after the point of times t2, t4.

It is possible to use the third embodiment in combination with the first and/or second embodiment. So, micro-leakage may be detected if the temporal gradi ent of the pipe temperature differs more than the first threshold from the first reference value or if the temperature difference between the pipe temperature and the ambient temperature differs more than the third threshold from the third reference value. Further on, micro-leakage may be detected if the temperature difference be tween the two pipe temperatures differs more than the second threshold from the second reference value or if the temperature difference between the pipe temperature and the ambient temperature differs more than the third threshold from the third reference value.

The invention allows micro-leakage detection in the magnitude of less than 1 liter per hour. Such a micro-leakage flow rate is below a measuring resolution or a measuring range of the flow meter 16.

List of reference signs

10 fluid flow system

11 building 12 fluid pipe

13 main water pipe

14 fluid valve

15 micro-leakage detection apparatus 15a interface 15b interface

15c interface 15d processor 15e memory

16 flow meter 17a pipe temperature sensor

17b pipe temperature sensor 18 ambient temperature sensor 20 step 21 step 22 step

23 step

24 step

25 step

26 step 27 step

30 step

31 step

32 step

33 step 34 step

35 step 36 step

37 step

38 step 40 step 41 step

42 step

43 step

44 step

45 step 46 step

47 step

48 step

50 fluid flow rate

51 pipe temperature 52 temporal gradient

53 micro-leakage 60 fluid flow rate 61 pipe temperature 62 pipe temperature 63 temperature difference

64 micro-leakage

70 fluid flow rate

71 temperature difference

72 micro-leakage