Method and system for detecting a leakage in a water piping of a building

Abstract

The present disclosure is directed to a method for detecting a leakage in a water piping of a building (1), wherein the method comprises: - logging (507) data of consumption events (84), wherein the data of each of the consumption events (84) comprises a currently accumulated consumption value during the consumption event (84) and an event identification value during the consumption event (84); - mapping (509), during logging the data of a consumption event (84), the currently accumulated consumption value onto one of a plurality of event identification value bins according to the event identification value; and - detecting (511), during logging the data of a consumption event (84), a leakage by identifying the consumption event (84) as a leakage candidate event if the currently accumulated consumption value of the leakage candidate event exceeds an adaptive maximum consumption value threshold being dependent on the event identification value of the leakage candidate event and being based on a confidence interval about an expectation value of the consumption value, wherein the expectation value is statistically derived from previous consumption events (84) that are mapped onto the event identification value bin of the leakage candidate event.

Description

DescriptionTECHNICAL FIELD

[0001] The present disclosure is directed to a method and system for detecting a leakage in a water piping of a building, in particular a domestic water leakage.BACKGROUND

[0002] Domestic water leakage can occur due to dripping taps, pipe punctures, pipe bursts, loose joints, corrosion, etc. Water leakage is always a waste of water resources and may cause serious damage in the building where it occurs. It is therefore very important for owners and inhabitants of buildings to minimise the damage in the event of a leakage. The first challenge in this respect is to detect a leakage as quickly as possible when it occurs. Once a leakage is detected, proper measures must be taken as quickly as possible to protect against damage caused by the leakage.

[0003] It is a technical challenge to detect leaks without knowing the normal water consumption behaviour in a building. EP 3374 747 B 1 describes a system for detecting non-cyclical leaks using pressure sensing. In order to feed the system with user-defined information about normal water consumption behaviour, EP 3374 747 Bl teaches a user query or interface for user feedback, so that a user can specify dates and / ortimes when the user expects that there will be no water usage, e. g. at night, during working hours or on holiday.

[0004] The known solution has the problem that a user must actively inform the system about normal water consumption behaviour or when the system should expect no consumption. This is not comfortable and depends on the user’s willingness, attendance and care to inform the system.

[0005] It is therefore an object of the present invention to provide a system and method for detecting a leakage in a water piping of a building, wherein the system and method automatically is able to adapt to the normal water consumption behaviour for distinguishing leakage from normal water consumption.SUMMARY

[0006] The method and system according to the independent claims provide a solution to this problem. Preferable embodiments of the invention can be deduced from the dependent claims, the description and the figures.

[0007] According to a first aspect of the present disclosure, a method is provided for detecting a leakage in a water piping of a building, wherein the method comprises: logging data of consumption events, wherein the data of each of the consumption events comprises a currently accumulated consumption value during the consumption event and an event identification value during the consumption event; mapping, during logging the data of a consumption event, the currently accumulated consumption value onto one of a plurality of event identification value bins according to the event identification value; anddetecting, during logging the data of a consumption event, a leakage by identifying the consumption event as a leakage candidate event if the currently accumulated consumption value of the leakage candidate event exceeds an adaptive maximum consumption value threshold being dependent on the event identification value of the leakage candidate event and being based on a confidence interval about an expectation value of the consumption value, wherein the expectation value is statistically derived from previous consumption events that are mapped onto the event identification value bin of the leakage candidate event.

[0008] The inventive idea is here to statistically derive over time for each event identification bin a confidence interval about an expectation value of the consumption value. The maximum consumption value threshold, which may be different for each event identification value bin, is based on that confidence interval to identify a leakage candidate. A consumption value within the confidence interval is considered to be a normal consumption event. Outside of the confidence interval, the current consumption event is a potential leakage event and may cause an action for preventing a further damage caused by a leak.

[0009] The method is preferably implemented in form of a program or algorithm executed by control electronics. In order to use as much as possible existing hardware and already available system infrastructure, it is beneficial to execute the program or algorithm by control electronics of a supply pump assembly, e. g. of a booster pump or a submersible pump being installed for supplying a building with water at a required pressure. Existing supply pump assemblies may already be equipped with communication interfaces for uploading and updating the control electronics, and for communicating with a flow sensor, a shut-off valve, and a user device, e. g. smart devices, smartphone, tablet or PC. Such aninfrastructure can be used to implement the inventive method with as little changes or additions to the hardware as possible.

[0010] For instance, the currently accumulated consumption value may be the total amount of fluid, e. g. water, that has been consumed during a consumption event so far. This may be determined by continuously measuring a fluid flow by a flow sensor or continuously estimating a fluid flow based on a pump operating point, e. g. defined by pump power and pump speed. The measured and estimated fluid flow integrated over time may give the currently accumulated consumption value in terms of a consumed water volume. It should be noted that the currently accumulated consumption value is logged continuously or regularly during the consumption event. A start of a consumption event may be triggered by a pump start, a pump speed increase, a change in pressure and / or a fluid flow increase showing as a flank of the fluid flow over time. An end of a consumption event may be indicated by a pump stop, a pump speed decrease, a change in pressure and / or a fluid flow decrease showing as a fall of the fluid flow over time.

[0011] The event identification value may be, for instance, an average flow during the consumption event so far. So, the consumption events are categorised by the mapping of the currently accumulated consumption value onto one of a plurality of event identification value bins according to the event identification value. In other words, the consumption events are filled into a histogram having a binning, wherein the binning may be static and pre-defined, or, preferably, adaptive according to the statistics available. For instance, if a certain amount of consumption events is available in a bin, the bin size may be reduced to allow for a better resolution. Vice versa, the bin size may be increased where more consumption events are needed to have meaningful statistics in a bin. A bin may thus be chosen to accommodate at least a certain minimum number of consumption events. The event identification value determines into which histogram bin the consumption event falls.For example, all consumption events with an average fluid flow in the range of 6 - 8 litres / minute may be filled into the histogram bin about 7 litres / minute. The histogram bins may have an overlap, but should not have a distance to each other, so that each consumption event is filled into at least one bin. It is important to note that only the group of consumption events in a certain bin forms the statistics to derive the confidence interval about the expectation value for said bin. The bin size may vary to have sufficient statistics in each bin. For example, high fluid flow events may be less frequent and the bin size for consumption events with an average fluid flow of 10 - 20 litres / minute may be filled into one large bin about 15 litres / minute. And it should be noted that the mapping is performed continuously or regularly during the consumption event, so that a leakage event can be detected any time during an ongoing consumption event.

[0012] As an alternative to the fluid flow, the currently accumulated consumption value may be the total amount of electric energy that has been consumed by a supply pump during a consumption event so far. The average supply pump power during a consumption event may then be used as the event identification value. This could be beneficial if no fluid flow measurement by a flow sensor is available or a flow sensor is to be spared.

[0013] Optionally, the method may further comprise sending a warning signal to a user, e. g. on his / her smartphone, if a leakage candidate event is identified. That is useful, because the normal consumption behaviour may change over time and a user may want to deviate from the normal consumption behaviour on purpose. The user can then decide if or how to react to the warning.

[0014] Optionally, the method may further comprise closing a shut-off valve and / or stopping a water supply pump if a leakage candidate event is identified. That is useful to prevent damage that may be causedby the identified leak. In order not to annoy a user by stopping the water supply during an intentional deviation from the usual normal consumption behaviour, additional criteria may be fulfilled before the water supply is actually stopped. For example, an additional criterion may be that the user actively disapproves the consumption event after receiving a leakage warning, or that the user does not give any feedback to a leakage warning within a pre-determined time window, or that the consumption event occurs at a time when the probability for a consumption event to occur is low, e.g. at night or during holidays.

[0015] Optionally, the method may further comprise initialising the adaptive maximum consumption value threshold by using a baseline threshold being pre-determined for each of the event identification value bins, and regularly updating the adaptive maximum consumption value threshold for one or more of the event identification value bins in batches of consumption events that are mapped onto said one or more of the event identification value bins.Thereby, the leakage identification automatically adapts to changes in the normal consumption behaviour over time. There is no need for a user to proactively inform the program or algorithm about changes in the normal consumption behaviour.

[0016] Optionally, the logged data may further comprise a daytime of a consumption event, wherein the adaptive maximum consumption value threshold is reduced by a factor depending on the daytime of the leakage candidate event. So, the daytime may provide a further dimension of all consumption events to be taken into account irrespective of their event identification value and associated event identification value bin. The program or algorithm may automatically learn over time about the overall likelihood of a consumption event taking place at a certain daytime.

[0017] Optionally, the adaptive maximum consumption value threshold may be a discrete function of the event identification value of the consumption event and / or of the daytime of the consumption event. So, for each event identification value bin the program or algorithm may have an adaptive maximum consumption value threshold. This results in the maximum consumption value threshold being a discrete function of the event identification value. If the daytime of the consumption event is taken into account, the histogram binning may be two-dimensional, i.e. each bin is defined by the event identification value in one dimension and the daytime in the other dimension. The binning in daytime may be an hourly binning or a coarser binning, e. g. two bins: night 22.00-06:00hrs and day 06:00-22:00 hrs. This results in the maximum consumption value threshold being a discrete function of the daytime. Depending on the daytime bin, the adaptive maximum consumption value threshold may be reduced by a factor defined for said daytime bin.

[0018] Optionally, the adaptive maximum consumption value threshold may depend on a selectable detection mode, wherein the detection mode is automatically selected based on at least one pre-determined criterion and / or manually selected by a user input. This is beneficial to reduce complexity of the program or algorithm without losing its functionality. The selectable detection modes may, for instance, be “normal mode”, “sleep mode”, and “away mode”. The pre-determined criterion may be the daytime of the consumption event, e. g. to switch automatically between normal mode and sleep mode. Also, the “away mode” may be automatically selected based on the daytime of the consumption event if it occurs during working hours the user is usually absent. The “away mode” may, for instance, be manually set by a user for longer times of absence, e. g. during holiday.

[0019] Optionally, a consumption event may be added, once said consumption event has finished without exceeding the adaptive maximumconsumption value, to those consumption events that are used to statistically derive the confidence interval about the expectation value of the consumption value for the event identification value bin of said consumption event. Such a finished consumption event is interpreted as a normal consumption event and used to update the statistics that define the normal consumption behaviour, i.e. the confidence interval about the expectation value in the respective bin. If the normal consumption behaviour changes, the program or algorithm shifts over time the expectation value, so that the confidence interval and the maximum consumption value threshold based thereon is shifted to adapt to the new consumption behaviour.

[0020] Optionally, a user may be prompted to give a user feedback if a leakage candidate event is identified, wherein a shut-off valve is closed and / or a supply pump is stopped if the user feedback is actively not approving the leakage candidate event as a normal consumption event. This is particularly useful to avoid further damage caused by a leak that the user disapproves.

[0021] Optionally, the adaptive maximum consumption value threshold may be increased by a pre-determined factor if the user feedback is actively approving the leakage candidate event as a normal consumption event, wherein said normal consumption event is added, after the consumption event has finished, to those consumption events that are used to statistically derive the confidence interval about the expectation value of the consumption value for the event identification value bin of said normal consumption event. This is beneficial for the program or algorithm to learn more quickly from more abrupt changes of the normal consumption behaviour.

[0022] Optionally, the adaptive maximum consumption value threshold may be increased to a shut-off threshold value for as long as no user feedback is received, wherein a shut-off valve is closed and / or a supplypump is stopped if the accumulated consumption value of the leakage candidate event exceeds the shut-off threshold value. The adaptive maximum consumption value threshold may thus be used as warning threshold. The shut-off threshold value may be a pre-determined relative amount higher than the adaptive maximum consumption value threshold, e. g. by 10%.

[0023] Optionally, the event identification value may be an average flow of water during the consumption event, wherein the average flow is determined on the basis of flow sensor measurements and / or on the basis of flow estimations using an operating point of a water supply pump for estimating the flow. Alternatively, the event identification value may be an average power of a supply pump, e. g. a booster pump or a submersible pump, during the consumption event.

[0024] Optionally, the consumption value may be the currently accumulated volume of water consumed during the consumption event. Alternatively, the consumption value may be the currently accumulated energy consumed by a supply pump, e. g. a booster pump or a submersible pump, during the consumption event.

[0025] According to another aspect of the present invention, a system is provided for detecting a leakage in a water piping of a building, wherein the system comprises a pump assembly being installed as a water supply pump at a supply line of the piping for pressurising the water in the piping, wherein the pump assembly comprises a variable speed drive and control electronics for controlling a pump speed, wherein the control electronics of the pump assembly is configured to carry out the previously described method. The previously described method may thus be implemented in form of a program or algorithm executed by control electronics of the pump assembly. Thereby, it is possible to use existing hardware and system infrastructure as much as possible for implementing the pre-sent invention. Existing supply pump assemblies may be updated by uploading a program or algorithm for carrying out the previously described method. The pump assembly may be equipped with communication interfaces for uploading and updating the control electronics, and for communicating with a flow sensor, a shut-off valve, and a user device, e. g. smart devices, smartphone, tablet or PC. Such an infrastructure can be used to implement the inventive method with as little changes or additions to the hardware as possible.

[0026] Optionally, the system may further comprise a flow sensor being installed at the supply line for measuring a flow through the supply line and communicating flow measurements to the control electronics of the pump assembly, wherein the system is configured to use an average flow of water during the consumption event as the event identification value. Using flow measurements of a flow sensor is a very accurate way to determine the fluid flow. If no flow sensor is available or a flow sensor is to be spared for any reason, the fluid flow may be estimated based on the operating point of the pump assembly, wherein the operating point of the pump assembly may be defined by the current power consumption and pump speed.

[0027] Optionally, the system may further comprise a shut-off valve being installed at the supply line for shutting off the supply line when it receives a shut-off command from the control electronics of the pump assembly. This is particularly beneficial to automatically protect against further damage that an identified leakage event may cause. A user does not have to manually shut off the water supply, which would not even be possible when the user is currently absent.

[0028] Optionally, the system may further comprise a stationary or mobile user device configured to run a user application for receiving information about a leakage candidate event from the control electronics of the pump assembly and for sending user feedback to the control electronicsof the pump assembly. This is useful to warn a user and to receive feedback from the user if a leakage event is detected.

[0029] Optionally, the user application may be configured to provide options for a user to manually select user settings for the system and / or a detection mode. This gives an option for a user to influence the program or algorithm.

[0030] Optionally, the system may further comprise an output pressure sensor and a hydraulic accumulator, wherein the output pressure sensor and / or the hydraulic accumulator is integrated into the pump assembly. This can be useful to detect a start and an end of a consumption event based on changes of a pressure value measured by the output pressure sensor. A hydraulic accumulator is useful to smoothen changes in the output pressure.

[0031] The method disclosed herein may be implemented in form of compiled or uncompiled software code that is stored on a computer readable medium with instructions for executing the method. Alternatively, or in addition, the method may be executed by software in a cloud-based system and / or a building management system (BMS).

[0032] The present invention is here a system, a method, and / or a computer program product at any possible technical detail level of integration. The computer program product may include a computer readable storage medium (or media) having computer readable program instructions thereon for causing a processor to carry out aspects of the present invention. The computer readable storage medium can be a tangible device that can retain and store instructions for use by an instruction execution device. The computer readable storage medium may be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storagedevice, a semiconductor storage device, or any suitable combination of the foregoing.

[0033] A non-exhaustive list of more specific examples of the computer readable storage medium includes the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

[0034] Computer readable program instructions described herein can be downloaded to respective computing / processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and / or a wireless network. The network may comprise copper transmission cables, optical transmission fibres, wireless transmission, routers, firewalls, switches, gateway computers and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing / processing device.

[0035] Computer readable program instructions described herein can be downloaded to respective computing / processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and / or a wireless network. The network may comprise copper transmission cables, optical transmission fibres, wireless transmission, routers, firewalls, switches, gateway computers and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer readable program instructions from the network and forwards the computer readable program instructions for storage in a computer readable storage medium within the respective computing / processing device.

[0036] Computer readable program instructions for carrying out operations of the present invention may be assembler instructions, instruction- set-architecture (ISA) instructions, machine instructions, machine dependent instructions, microcode, firmware instructions, state-setting data, configuration data for integrated circuitry, or either source code or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C++, or the like, and procedural programming languages, such as the "C" programming language or similar programming languages. The computer readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet Service Provider). In some embodiments, electronic circuitry including, for example, programmable logic circuitry, field-programmable gate arrays (FPGA), orprogrammable logic arrays (PLA) may execute the computer readable program instructions by utilizing state information of the computer readable program instructions to personalize the electronic circuitry, in order to perform aspects of the present invention.SUMMARY OF THE DRAWINGS

[0037] Embodiments of the present disclosure will now be described by way of example with reference to the following figures of which:Fig. 1 shows schematically a water piping of a building that is equipped with an embodiment of the system according to the present invention;Fig. 2 shows schematically how the system shown in Fig. 1 communicates with a user;Fig. 3a, b show different embodiments of the system according to the present invention;Fig. 4 shows schematically more detailed an embodiment of the system according to the present invention;Fig. 5a shows an example of a consumption event;Fig. 5b shows a flowchart of steps of an embodiment of the method according to the present invention;Fig. 6 an example of a Flow-Volume-diagram;Fig. 7 shows another flowchart of steps of an embodiment of the method according to the present invention; andFig. 8 an example diagram of a consumption event probability for each hour of the day.DETAILED DESCRIPTION

[0038] Fig. 1 shows a household building 1 with a water piping for supplying different water consumption devices with fresh water, such as a washing machine 3, a bathtub 5, a shower 6, a kitchen sink 7, a toilet flushing cistern 9, a bathroom tap 1 1 , and / or a garden hose 12 for irrigation. Any of these water consumption devices could be leaky or dripping. If it is leaking into a drainage system 13 of the building 1 , the damage caused by a leak is a higher water consumption. If it is leaking somewhere else, even a very small leak may cause serious damage to the building 1 . For instance, a pipe of the piping system may be punctured by a bore, or burst by freezing, corrosion, or ageing of seals. It is therefore very important for owners and inhabitants of the building 1 to minimise the damage in the event of a leakage. The first challenge in this respect is to detect a leakage as quickly as possible when it occurs. Once a leakage is detected, proper measures must be taken as quickly as possible to protect against damage caused by the leakage.

[0039] The building 1 of Fig. 1 is equipped with a pump assembly 15 in form of a water supply pump being here a booster pump of the type Grundfos “SCALA® 2”. The water supply pump 15 is predominantly used to provide the water piping of the building 1 with fresh water at a desired pressure. An inlet of the water supply pump 15 may be connected to water supply network 17 to receive fresh water from. An outlet side of the water supply pump 15 is connected to the water piping of the building 1 to feed fresh water at desired stable water pressure into the water piping of the building 1 . As will be described in more detail below, the water supply pump 15 is used as a part of an inventive system19 for detecting a leakage in the water piping of the building 1 . The inventive leakage detection system 1 here further comprises a leakage protection unit 21 installed downstream of the outlet side of the water supply pump 15. It should be noted that the water supply pump 15 may alternatively be a submersible pump for retrieving fresh water from a well, a tank, or reservoir.

[0040] The hardware of the water supply pump 15 is known, but the software installed on it contains an inventive program or algorithm, which makes use of existing hardware and functionalities of the water supply pump 15. For example, as shown in Fig. 2, the water supply pump 15 is able to wirelessly communicate with a user device 23, e. g. a smartphone having an appropriate app installed thereon. The communication between the user device 23 and the water supply pump 15 is bi-directional, i.e. the water supply pump 15 is able send information and / or queries 25 to the user device 23, and the user device 23 is able to set 26 parameters, configurations, or user feedback to the water supply pump 15.

[0041] Figs. 3a, b show two possible embodiments of the inventive leakage detection system 19 in more detail. It should be noted that other embodiments of the inventive leakage detection system 19 are also possible. The water supply pump 15 in form of a Grundfos “SCALA® 2” booster pump is in this case in fact more than just a pure pump. It combines a pump unit 27, a check valve 29, an internal hydraulic accumulator 31 , and an outlet pressure sensor 33 into one water supply pump device 15. The leakage protection unit 21 comprises in this case a mo- tor-controlled shut-off valve 35 and a flow sensor 37. It is convenient here, but not mandatory that the shut-off valve 35 and the flow sensor 37 are integrated into one leakage protection device 21 . The difference between the embodiments shown in Figs. 3a, b is the position of the leakage protection unit 21 relative to the water supply pump 15 either downstream (Fig. 3a) or upstream (Fig. 3b) of the water supplypump 15. An advantage of the downstream embodiment of Fig. 3a is that the leakage protection unit 21 does not have to be attached to normed DN pipe size. An advantage of the upstream embodiment of Fig. 3b is that a leakage in or at the water supply pump 15 can be detected and damage caused by it prevented. It should be noted that both embodiments here show an external hydraulic accumulator 39 downstream of the inventive leakage detection system 1 . The external hydraulic accumulator 39 can be installed anywhere in the water piping of the building 1 to smoothen the water pressure within the water piping of the building 1 . Manual shut-off valves 41 may be useful to be installed as shown between the different devices 15, 19, 39 for plumbing and maintenance.

[0042] Fig. 4 shows the inventive leakage detection system 19 in more detail. The water supply pump 15 comprises control electronics 43 on a printed circuit board, PCB, 45 arranged within a housing of the water supply pump 15. There is a wired or wireless bidirectional communication connection 47 between the control electronics 43 of the water supply pump 15 and the leakage protection unit 21 , wherein the communication connection 47 is mainly used to receive flow measurements from the flow sensor 37 and to send a shut-off signal or opening signal to the motor-controlled shut-off valve 35.

[0043] The hardware of the control electronics 43 on the PCB 45 may be the same as for an ordinary existing booster pump of the type Grundfos “SCALA® 2”, wherein a pump control logic 49 is part of an embedded software 51 being executable by the control electronics 43 of the of the water supply pump 15. The hardware of the control electronics 43 on the PCB 45 may as shown comprise a processor 53, a power factor correction (PFC) / variable frequency drive (VFD) 54 for controlling the speed of the pump unit 27, a real time clock 55, a WIFI module 57, a multi standard radio 59, a human-machine interface (HMI) 61 , a power supply unit (PSU) 60, an electromagnetic compatibility module (EMC)63, a configurable l / O-module (CIO) 65, and an outlet pressure module 67. It further comprises internal inputs / outputs 69 and external in- puts / outputs 71 . A user 73 may use the HMI 61 to read or input infor- mation / commands / settings, or smartphone 23 having an app installed thereon to read or input information / commands / settings. The app on the smartphone 23 may include a pump control interface 75 for monitoring and / or influencing the pump control logic 49. The smartphone 23 may use a common WIFI network to communicate with the control electronics 43 of the water supply pump 15 on the PCB 45 via the WIFI module 57.

[0044] The inventive leakage detection method is here implemented by an additional software functionality in form of a leakage detection program or algorithm 77 as part of an updated embedded software 51 being executable by the control electronics 43 of the water supply pump 15. Furthermore, the app on the smartphone 23 may be updated to include an leakage detection interface 79 for monitoring and / or influencing the leakage detection program or algorithm 77.

[0045] Fig. 5a shows a typical consumption event in a diagram showing the water flow q currently measured by the flow sensor 37 as a function of time t. A start of an event may be triggered when the flow q exceeds minimum start flow qstart, e. g. qstart = 0.1with a positive time-derivative of the flow q, i.e. q > qstartand> 0. An end of an event may be indicated by the flow q falling below a maximum end flow qend, e. g. qend= 0.08 l / m, with a negative time-derivative of the flow. The values for qstartand qendmay depend on the minimum flow that the flow sensor 37 is able to reliably measure and on the prevailing noise that may hamper the flow measurement.

[0046] The shape of the consumption event does have to be rectangular as shown in Fig. 5a. For example, the number of open taps 3, 5, 7, 9,1 1 may change during a consumption event, or the opening degree of a tap changes during a consumption event. This will lead to other shapes of a consumption event. For the inventive leakage detection method, however, an average flow qavgmay be used as an event identification value, wherein qavgis the flow averaged from the start of the consumption event to the end of the consumption event. So, once the consumption event has ended, the consumption event is effectively treated here as if it had an essentially rectangular shape with a constant average flow qavg. It should be noted, however, that the average flow qavgmay change during an ongoing consumption event.

[0047] Fig.5b shows a flowchart with method steps of an event handler 81 and an event analysis 83 as part of the leakage detection program or algorithm 77. In step 501 of the event handler 81 , an updated array of measured flow values q = qk>qk-i><lk-2>is considered to detect a start and / or an end of a consumption event, wherein N is the array size, e.g. N=6, and k is an array index. The larger the array size is, the more latency is added to the leakage detection program or algorithm 77. The lower the array size is, the higher the risk for false positive leakage detections is. In step 503, the flow edge is read on the basis of the array q = (qk, qk-i, qk-2, -> qk-N)-

[0048] If the event is opened or already open in step 505, i.e. q > qstart, the event analysis 83 is started in logging step 507 by iteratively calculating the cumulative volume Veventof water consumed during the on- going consumption event, e. g. as Vevent=+ Veventin litres if the flow qkis measured in litres per minute. The cumulative volume Veventof water consumed during the ongoing consumption event is here logged as the currently accumulated consumption value. Furthermore, the average flow qavgis determined and logged, e. g. qavg=Vevent■ 60 in litresper minute if the cumulative volume Veventis determined in litres and thecurrent duration Teventof the consumption event is measured in seconds. The average flow qavgis here logged as the event identification value.

[0049] In mapping step 51 1 , the currently accumulated consumption value, i.e. Vevent, is mapped onto one of a plurality of event identification value bins according to the event identification value, i.e. qavg- The consumption event is thus categorised according to the event identification value, i.e. qavg- For example, if the consumption event has an average flow qavgin the range between 6 litres / minute and 8 litres / mi- nute, the consumption event is associated to the bin about 7 litres / minute. Depending on the statistics, the binning can be fine or coarse. The binning may be here every litre / minute with a bin size of 2 litres / minute. The bin size may be chosen larger where less statistics is expected, e. g. for high average flows. Fig. 6 shows the result of the mapping in a volume-flow diagram.

[0050] In detection step 51 1 of the event analysis 83 shown in Fig. 5b, the consumption event is checked for being a leakage candidate event if the currently accumulated consumption value, i.e. Vevent, exceeds an adaptive maximum consumption value threshold. The method is reiterated and continues with the next repetition at step 501 if the event handler 81 by updating the data array<7 =with new flow measurements. If the event is not opened anymore in step 505, i.e. q < qstart, the event handler 83 checks in step 513 if the event end condition q < qendis fulfilled. If not, the data array is again updated in step 513. If the event has ended, the event handler is reset in step 515.

[0051] The volume-flow diagram of Fig. 6 shows how the consumption events 84 (shown as small circles) are distributed after the mapping step 509 of the event analysis 81 . The maximum consumption valuethreshold may initially be set to the dashed line 85. When the accumulated consumption value, i.e. Vevent, exceeds the maximum consumption value threshold, the consumption event is identified as a leakage candidate event and a warning may be sent to the user 73 via the leakage detection interface 79 in the app on the user’s smartphone 23. The dashed line 85 may thus be a warning threshold. The maximum consumption value threshold may then be set 10% higher to the solid line 87 in Fig. 6. If the solid line 87 in Fig. 6 is exceeded, the leakage detection program or algorithm 77 commands the motor-controlled shutoff valve 35 to shut off the water supply in order to protect against further damage caused by the identified leakage. It has shown, however, that static maximum consumption value thresholds rarely fit to the individual consumption behaviour of the user 73. Therefore, the maximum consumption value threshold is adaptive and changes after the initial setting for each average flow bin individually based on a confidence interval about an expectation value of the consumption value, wherein the expectation value is statistically derived from previous consumption events that are mapped onto the same average flow bin. The confidence interval is shown in the shaded area of Fig. 6 and the adaptive maximum consumption value threshold is the discrete function 89 shown by “+" signs in Fig. 6. As shown in Fig. 6, a linear interpolation between the maximum consumption value thresholds may result in a steady threshold function over the range of average flows. This is beneficial if the bin size (e.g. here 2 litres per minute) is larger than the bin distance (here 1 litre), so that a consumption event may fall into two overlapping bins. The linear interpolation between the neighbouring maximum consumption value thresholds may be used as the maximum consumption value threshold at the given average flow of the consumption event.

[0052] Fig. 7 shows a flowchart with certain steps of the inventive method. It begins with starting step 701 , in which the learning of the normal consumption behaviour is started. In initialising step 703, the initial baseline thresholds 83, 85 are set as maximum consumption value threshold to start with. In step 704, an event batch is cleared to learn from scratch. In step 705, the currently accumulated consumption value, i.e. Vevent, the currently determined average flow qaVg, and the time of the day is constantly updated and provided during an ongoing consumption event. The updated information about the average flow qaVg and the time of the day in used in step 708 to determine the adaptive maximum consumption value threshold Vmax. The adaptive maximum consumption value threshold Vmaxis set in step 708 according to the detection mode, which may be manually set by the user 73 and / or automatically set based on the time of the day. The detection modes may, for instance, be “normal mode”, “sleep mode” and “away mode”. The average flow qavgis used in step 708 to determine the average flow bin for determining the adaptive maximum consumption value threshold Vmax. The adaptive maximum consumption value threshold Vmaxmay be set in step 708 for the relevant average flow bin by Vmax= kwarning■ kmode■ Vci, wherein Vcimay be an expectation value in that average flow bin plus two standard deviations (95% confidence interval) and kwarningis a factor for increasing the maximum consumption value threshold Vmaxif a warning signal was sent to the user 73. For example, kwarning= 1.1 if warning signal was sent to the user 73 and / cwarnmfl= 1-0 otherwise. kmodeis a factor that depends on the detection mode, e. g. kmode= 1.0 in normal mode, kmode= 0.5 in sleep mode, and kmode = °.25 in away mode. So, the adaptive maximum consumption value threshold Vmaxis stricter for the sleep mode and the away mode, i.e. when the probability for a consumption event to occur is low. The user 73 may have the possibility to set, e.g. by means of the user device 23, the confidence interval around the expectation value. For instance, if there are too many false positive leakage detections, the user 73 mayset the confidence interval to 99%, i.e. to three standard deviations from the expectation value.

[0053] The currently accumulated consumption value Veventis compared in step 707 with the adaptive maximum consumption value threshold Vmax. If Vevent< Vmax, the consumption event is normal and not a leakage candidate event. Otherwise, it is a leakage candidate event that triggers warning step 709, in which the user 73 is prompted for a user feedback to actively approve the leakage candidate event as a normal consumption event or not. The user feedback is processed in step 71 1 . If the user actively approves the consumption event as normal (Y), the process continues with step 712, in which the adaptive maximum consumption value threshold K„azis set to a maximum for the rest of that ongoing consumption event to exceptionally allow a maximum volume consumption. Then, the process continues with the event handling step 705, after which the consumption event is, once the consumption event has ended, stored in step 713 to a batch of events, wherein the all events of the batch belong to the same average flow bin. So, each average flow bin has its own batch of events. The ended normal consumption event is stored to the batch of events as long as a pre-defined batch size is not reached in step 715. Once the batch size is reached for that bin, an expectation value for the bin is determined from the batch of events in step 717. The confidence interval for all bins is updated in step 719. Furthermore, in steps 721 and 723, a date-time model is computed on the basis of the event batch, e. g. to yield an updated consumption event probability for each hour of the day as shown in Fig. 8. The date-time model may be used to set the detection mode automatically.

[0054] If the user does not approve the leakage candidate event as a normal consumption event by actively disapproving, i.e. a negativefeedback (N) is received from the user 73, the leakage detection program or algorithm 77 commands the motor-controlled shut-off valve 35 in step 725 to shut off the water supply in order to protect against further damage caused by the identified leakage. A manual restart 727 may then be needed to proceed. In case no user feedback is received in step 71 1 , the warning factor kwarningis set in step 729, e. g. / cwarnmfl= 1.1. Thereby, the threshold is increased by 10%. If the warning factor kwarningis already set next time the user feedback is evaluated in step 71 1 , the leakage detection program or algorithm 77 commands the motor-controlled shut-off valve 35 in step 725 to shut off the water supply even without active disapproval, because the shut-off threshold was exceeded then.

[0055] Fig. 8 shows an example diagram of a consumption event probability for each hour of the day. Thereby, the leakage detection program or algorithm 77 can automatically set the detection mode to “sleep mode” when the probability of a normal consumption event for this hour is low, e. g. the five hours with the lowest consumption event probability. Also, the “away mode” may be automatically set, e. g. when there are no consumption events within 48 hours.

[0056] Where, in the foregoing description, integers or elements ore mentioned which hove known, obvious or foreseeable equivalents, then such equivalents are herein incorporated as if individually set forth. Reference should be made to the claims for determining the true scope of the present disclosure, which should be construed so as to encompass any such equivalents. It will also be appreciated by the reader that integers or features of the disclosure that are described as optional, preferable, advantageous, convenient or the like are optional and do not limit the scope of the independent claims.

[0057] The above embodiments are to be understood as illustrative examples of the disclosure. It is to be understood that any feature described in relation to any one embodiment may be used alone, or in combination with other features described, and may also be used in combination with one or more features of any other of the embodiments, or any combination of any other of the embodiments. While at least one exemplary embodiment has been shown and described, it should be understood that other modifications, substitutions and alternatives are apparent to one of ordinary skill in the art and may be changed without departing from the scope of the subject matter described herein, and this application is intended to cover any adaptations or variations of the specific embodiments discussed herein.

[0058] In addition, "comprising" does not exclude other elements or steps, and "a" or "one" does not exclude a plural number. Furthermore, characteristics or steps which have been described with reference to one of the above exemplary embodiments may also be used in combination with other characteristics or steps of other exemplary embodiments described above. Method steps may be applied in any order or in parallel or may constitute a part or a more detailed version of another method step. It should be understood that there should be embodied within the scope of the patent warranted hereon all such modifications as reasonably and properly come within the scope of the contribution to the art. Such modifications, substitutions and alternatives can be made without departing from the spirit and scope of the disclosure, which should be determined from the appended claims and their legal equivalents.

[0059] Where the Figures contain flowcharts or block diagrams, they illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard,each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical functions). In some alternative implementations, the functions noted in the blocks may occur out of the order noted in the Figures. For example, two blocks shown in succession may, in fact, be accomplished as one step, executed concurrently, substantially concurrently, in a partially or wholly temporally overlapping manner, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustration, and combinations of blocks in the block diagrams and / or flowchart illustration, can be implemented by special purpose hardware- based systems that perform the specified functions or acts or carry out combinations of special purpose hardware and computer instructions.

[0060] List of reference numerals:I building3 washing machine5 bathtub6 shower7 kitchen sink9 toilet flushing cisternI I bathroom tap12 garden hose13 drainage system15 pump assembly / water supply pump17 water supply network1 system21 leakage protection unit23 smartphone25 sending to user device26 receiving from user device27 pump unit29 check valve31 internal hydraulic accumulator33 pressure sensor35 motor-controlled shut-off valve37 flow sensor39 external hydraulic accumulator41 manual shut-off valves43 control electronics45 PCB47 communication connection49 pump control logic51 embedded software53 processor54 PFC / VFD55 real time clock57 WIFI module59 multi standard radio60 PSU61 HMI63 EMC65 CIO67 outlet pressure module69 internal inputs / outputs71 external inputs / outputs73 user75 pump control interface77 leakage detection program or algorithm79 leakage detection interface81 event handler83 event analysis84 consumption event85 dashed line / initial warning threshold87 solid line / initial shut-off threshold89 discrete function of adaptable threshold501 updating data array503 reading flow edge505 checking if event is open(ed)507 logging509 mapping51 1 detecting513 checking if event is close(d)513 resetting701 starting703 initialising704 clearing705 event handling707 checking if threshold is exceeded708 setting threshold709 asking for user feedback71 1 check the user feedback713 storing event in batch715 checking if batch size is reached717 computing an expectation value719 updating the confidence interval721 computing date-time model723 updating date-time model725 shutting off water supply727 manual restart729 setting warning factor

Claims

Claims1 . A method for detecting a leakage in a water piping of a building ( 1 ), wherein the method comprises: logging (507) data of consumption events (84), wherein the data of each of the consumption events (84) comprises a currently accumulated consumption value during the consumption event (84) and an event identification value during the consumption event (84); mapping (509), during logging the data of a consumption event (84), the currently accumulated consumption value onto one of a plurality of event identification value bins according to the event identification value; and detecting (51 1 ), during logging the data of a consumption event (84), a leakage by identifying the consumption event (84) as a leakage candidate event if the currently accumulated consumption value of the leakage candidate event exceeds an adaptive maximum consumption value threshold being dependent on the event identification value of the leakage candidate event and being based on a confidence interval about an expectation value of the consumption value, wherein the expectation value is statistically derived from previous consumption events (84) that are mapped onto the event identification value bin of the leakage candidate event.

2. The method according to claim 1 , further comprising sending a warning signal to a user (73) if a leakage candidate event is identified.

3. The method according to claim 1 or 2, further comprising closing a shut-off valve (35) and / or stopping a water supply pump (15) if a leakage candidate event is identified.

4. The method according to any of the preceding claims, further comprising initialising the adaptive maximum consumption value threshold by using a baseline threshold being pre-determined for each of the event identification value bins, and regularly updating the adaptive maximum consumption value threshold for one or more of the event identification value bins in batches of consumption events (84) that are mapped onto said one or more of the event identification value bins.

5. The method according to any of the preceding claims, wherein the logged data further comprises a daytime of a consumption event (84), wherein the adaptive maximum consumption value threshold is reduced by a factor depending on the daytime of the leakage candidate event (84).

6. The method according to any of the preceding claims, wherein the adaptive maximum consumption value threshold is a discrete function (89) of the event identification value of the consumption event (84) and / or of the daytime of the consumption event (84).

7. The method according to any of the preceding claims, wherein the adaptive maximum consumption value threshold depends on a selectable detection mode, wherein the detection mode is automatically selected based on at least one pre-determined criterion and / or manually selected by a user input.

8. The method according to any of the preceding claims, wherein a consumption event (84) is added, once said consumption event (84) has finished without exceeding the adaptive maximum consumption value, to those consumption events (84) that are used to statistically derive the confidence interval about the expectation value of the consumption value for the event identification value bin of said consumption event (84).

9. The method according to any of the preceding claims, wherein a user is prompted to give a user feedback if a leakage candidate event is identified, wherein a shut-off valve (35) is closed and / or a supply pump (15) is stopped if the user feedback is actively not approving the leakage candidate event as a normal consumption event (84).

10. The method according to claim 9, wherein the adaptive maximum consumption value threshold is increased by a pre-determined factor if the user feedback is actively approving the leakage candidate event as a normal consumption event, wherein said normal consumption event is added, after the consumption event (84) has finished, to those consumption events (84) that are used to statistically derive the confidence interval about the expectation value of the consumption value for the event identification value bin of said normal consumption event.1 1 . The method according to claim 9 or 10, wherein the adaptive maximum consumption value threshold is increased to a shut-off threshold value for as long as no user feedback is received, wherein a shut-off valve is closed and / or a supply pump is stopped if the accumulated consumption value of the leakage candidate event exceeds the shut-off threshold value.

12. The method according to any of the preceding claims, wherein the event identification value is an average flow of water during theconsumption event (84), wherein the overage flow is determined on the basis of flow sensor measurements and / or on the basis of flow estimations using an operating point of a water supply pump ( 15) for estimating the flow.

13. The method according to any of the preceding claims, wherein the consumption value is the currently accumulated volume of water consumed during the consumption event (84).

14. A system (19) for detecting a leakage in a water piping of a building ( 1 ), wherein the system ( 1 ) comprises a pump assembly ( 15) being installed as a water supply pump at a supply line of the piping for pressurising the water in the piping, wherein the pump assembly (15) comprises a variable speed drive and control electronics (43) for controlling a pump speed, wherein the control electronics (43) of the pump assembly (15) is configured to carry out the method of any of the preceding claims.

15. The system (19) of claim 14, further comprising a flow sensor (37) being installed at the supply line for measuring a flow through the supply line and communicating flow measurements to the control electronics (43) of the pump assembly (15), wherein the system ( 19) is configured to use an average flow of water during the consumption event (84) as the event identification value.

16. The system (19) of claim 14 or 15, further comprising a shut-off valve (35) being installed at the supply line for shutting off the supply line when it receives a shut-off command from the control electronics (43) of the pump assembly (15).

17. The system (19) of any of the claims 14 to 16, further comprising a stationary or mobile user device (23) configured to run a user application for receiving information about a leakage candidate eventfrom the control electronics (43) of the pump assembly (15) and for sending user feedback to the control electronics (43) of the pump assembly (15).

18. The system (19) of claim 17, wherein the user application is configured to provide options for a user (73) to manually select user settings for the system ( 1 ) and / or a detection mode.

19. The system (19) of any of the claims 14 to 18, further comprising an output pressure sensor (33) and a hydraulic accumulator (31 ), wherein the output pressure sensor (33) and / or the hydraulic accumulator (31 ) is integrated into the pump assembly (15).

Images