BE1026828B1 - Boiler performance monitoring and failure prediction system - Google Patents

Abstract

System (1) for monitoring performance and anticipating failure of a boiler (10), comprising a device (20) provided with a set of sensors, comprising at least one of a first pair of sensors temperature, a first temperature sensor (21) for a central heating circuit (CCC) input (12) and a second temperature sensor (22) for a CCC output (13), of a second pair of temperature sensors temperature, a third temperature sensor (23) for a domestic hot water inlet (DHW) (14) and a fourth temperature sensor (24) for a DHW outlet (15), as well as a fifth temperature sensor temperature (25) for a flue gas outlet (11), an accelerometer (26) for said CCC input, and a transmission module (28) for sending data to a control unit (30), configured for receive and analyze data to determine anomalies in boiler components.

Description

Boiler performance monitoring and failure prediction system

Field of the invention

The present invention relates to a system for monitoring performance and predicting failure of a boiler. The invention also relates to a corresponding method for use in this system.

Invention background

Boilers occupy a central place in the central heating systems of buildings and houses. In many countries, heating and domestic hot water (DHW) are both generated using the heat produced by a single boiler. In the Flemish part of Belgium, 67% of households produce DHW with a boiler connected to the central heating system (Eurostatenquete_samenvatting survey, collected on September 25, 2018, on the page https://www.energiesparen.be/marktonderzoedk) .

In a traditional market, the owner of a house is also the owner of the boiler of the house. In this type of market, an owner whose boiler needs to be repaired will call on the services of a boiler maintenance company and will pay according to the service provided, which means that the majority of the services provided are ad services ad hoc. Because repairing a boiler costs much less than installing a new one, many boiler owners prefer to have their boiler repaired until replacement becomes the best economic option. Consequently, a traditional boiler market will be characterized by the fact that a large group of boilers will be made up of fairly old boilers. In the specific case of Belgium, which in 2018 can be identified as such a market

BE2019 / 5870 traditional, it is assumed that 30% of the boilers installed are over 15 years old. According to Belgian law, each boiler must be regularly maintained (for example every year or every two years) by a registered and approved professional. Most professionals are freelancers or SMEs.

Boiler markets in Western European countries are characterized by a variety of market models. There is an emerging trend whereby third parties own and control a fleet of boilers, offering a range of services to their customers, who are the users of individual boilers. In different countries, market consolidation is taking place with energy suppliers and insurance companies entering the boiler market in the role of third-party owners, the former as part of a strategy consisting in proposing a solution total energy costs to customers, including the maintenance of heating equipment in their homes.

The fact that insurance providers are entering the market can be seen as part of a larger overall transition from simply providing financial compensation in the event of damage and first preventing the occurrence of damage. An example of this transition is the development, by insurance-related companies, of systems in which sensors and devices such as cameras are installed in the home, and are connected to a smartphone or some other personal device of the home owner. Sensors and cameras are capable of detecting movement, smoke from water leaks at an early stage, the presence of intruders, open windows or doors, and the like.

BE2019 / 5870

To enter the boiler market, third parties such as energy suppliers and insurance companies create their own local network of qualified technicians, such as electricians or plumbers, or collaborate with an already existing network.

A variant of this economic model is based on the pooling of financial risks, the customer paying subscription fees (for example monthly), covering maintenance and repair costs. In the most popular variant of this model, these subscription fees also cover the purchase and installation of a boiler. As a result, the customer pays for a service, namely the continued use of a boiler, leaving all technical and maintenance concerns to said third party, who purchased and installed the boiler at the customer's premises.

To date, the owners of boilers or service providers relating to boilers are faced with the problem that, whenever a boiler failure occurs, technical assistance is only called after failure detection. Often, a boiler failure is only detected when residents notice a decrease in room temperature or the DHW temperature. Precious time is lost to request assistance, which can result in effects ranging from simple discomfort to potential damage to people living in the building which is heated by the boiler, as well as to goods that are stored there. It should therefore be noted that damage caused by a malfunctioning boiler can lead to claims for compensation. Such events should not be underestimated because the risk of boiler failure can be estimated at around 10%, which

BE2019 / 5870 means that each year around 10% of all boilers installed will fail.

In addition, service providers related to boilers have noticed that customers who have purchased a service, for example with monthly subscription fees, tend to contact the service provider for minor issues that can easily be resolved. solved by the customer himself or which may even be the result of an action by him. For example, the customer may have forgotten that they opened a window somewhere in the building, then noticed a drop in temperature in a room, and asked the service provider for help. Lacking the ability to remotely monitor the performance of the customer's heating system, the supplier will have no choice but to plan a technical intervention, typically on site. These unnecessary technical interventions will obviously affect the profitability of the service provider.

It is clear from the above that the ability to monitor the performance of a boiler enables service providers to offer remote customer assistance and also to keep boiler owners informed of the status of their device. Additionally, the ability to predict and anticipate an imminent boiler failure allows boiler owners to request or service providers to provide assistance in a timely manner, thereby avoiding the inconvenience of failure or failure of the boiler.

Boiler monitoring devices and systems have already been proposed.

Document CN 201177381 proposes a data acquisition device for evaluating operating reliability

BE2019 / 5870 of a boiler and a system for monitoring it. As such, a variety of sensors are introduced into the boiler, the collected data of which is used to provide a better safety analysis and thus decide whether the boiler is operating safely. The device does not provide an evaluation of the performance of the boiler and cannot predict the occurrence of a possible future failure of the boiler.

There is therefore a need for a boiler monitoring system for monitoring the performance of a running boiler and predicting future boiler failure. In addition, there is a need to be able to install a boiler monitoring system on a plurality of boilers already in operation, although these boilers may not be of the same age, brand, size, design and the like. In order to install such a system on a plurality of boilers, and for economic reasons, there is also a need to have a boiler monitoring system which is made up of inexpensive components.

Summary of the invention

According to one aspect of the present invention, a system is made available for monitoring the performance of a boiler and anticipating a breakdown of the boiler, said system comprising:

- a device provided with a set of sensors, comprising at least one of:

a first pair of temperature sensors, comprising a first temperature sensor, intended to be applied to a central heating circuit (CCC) input of said boiler, and a

BE2019 / 5870 second temperature sensor, intended to be applied to a CCC outlet of said boiler;

a second pair of temperature sensors, comprising a third temperature sensor, intended to be applied to a domestic hot water inlet (DHW) of said boiler, and a fourth temperature sensor, intended to be applied to an outlet d 'DHW of said boiler;

said device provided with a set of sensors further comprising:

o a fifth temperature sensor, intended to be applied to a smoke evacuation outlet of said boiler;

o an accelerometer intended to be applied to said CCC inlet of said boiler or to said DHW inlet of said boiler; and

a transmission module which is configured to send data representing measurements taken by said set of sensors to a control unit, and

- the control unit which is configured to receive and analyze said data to determine anomalies in boiler components.

An advantage of the embodiments of the present invention is that the performance of an operating boiler can be continuously monitored. Another advantage lies in the fact that the combinations of sensors described above make it possible to anticipate future failures of a boiler, these failures being able to have several causes for origin. Furthermore, it has been found that combinations of sensors as described above make it possible to anticipate such boiler failures,

BE2019 / 5870 that the boiler produces both DHW and central heating or only one of them. Even more, the set of sensors makes it possible to predict a failure of a boiler, regardless of age, brand, power and the design of the plumbing system of which the boiler is a part. As a result, property managers or boiler maintenance companies can install the performance monitoring and fault prediction system of a boiler on existing circuits and thereby upgrade existing plumbing systems. Another advantage lies in the fact that said sensors are inexpensive components, making it economically feasible to equip a large fleet of boilers with said monitoring system.

In one embodiment of the system according to the present invention, said accelerometer is applied to said CCC input of said boiler on the condition that said device provided with a set of sensors comprises said first pair of temperature sensors.

An advantage of this embodiment lies in the fact that the accelerometer is applied to the CCC input of the boiler, in the case where the boiler is connected to a central heating circuit. For other cases, for example when the boiler provides only domestic hot water, the accelerometer will be applied to the DHW input of the boiler.

In one embodiment of the system according to the present invention, said device has a set of sensors further comprising a sensor for measuring electromagnetic interference.

An advantage of this embodiment is that it predicts a wider range of failures.

BE2019 / 5870 boiler. In addition, it has turned out in particular that the fact of including said sensor for measuring electromagnetic interference makes it possible to anticipate boiler failures concerning the supply of domestic hot water.

In one embodiment of the system according to the present invention, said transmission module transmits said data to said control unit using a gateway which is in contact with an Internet router.

An advantage of this embodiment is that the performance of a boiler can be monitored remotely by a control unit.

In a particular embodiment of the system according to the present invention, the Long Range technology (LoRa, for Long Range) is used by said transmission module for sending data to said gateway.

An advantage of this embodiment is that the LoRa technology allows long-range transmissions with low power consumption, so that data can be transmitted to said control unit, preferably by means of said gateway. , even when said boiler and said gangway are separated, for example, by concrete walls.

In one embodiment of the system according to the present invention, said control unit is configured to recover, from said received data, a reference standard for each sensor of said set of sensors, said reference standard being indicative of a boiler in normal running.

An advantage of this embodiment lies in the fact that, for each sensor, a standard is obtained which represents a boiler operating regularly. AT

BE2019 / 5870 As such, the sensor values which are obtained after recovery of this standard can be evaluated against a reference for said sensor as resulting from a normal request from the boiler or from abnormal behavior.

In a particular embodiment of the system according to the present invention, said control unit is configured to analyze a data stream representing measurements taken by said set of sensors to determine anomalies of components of the boiler, in consideration of said reference standards retrieved for each sensor from said set of sensors.

In one embodiment of the system according to the present invention, said control unit is configured to analyze said data flow provided by said set of sensors using machine learning techniques.

An advantage of this embodiment is that past experiences can be applied to get a better idea of the urgency of a technical problem with a boiler. As such, these techniques can be used to define a probability of boiler failure and increase the reliability of these predictions. Preferably, data mining techniques are applied to said data flow, revealing new links between the values of the sensors.

In one embodiment of the system according to the present invention, said control unit is configured to report the results of said comparison to third parties via a communication means.

An advantage of this embodiment is that the warning of a possible future failure of the boiler can

BE2019 / 5870 be brought to the attention of technicians or owners of boilers and the like in due course.

In one embodiment of the system according to the present invention, said sensors are applied in a non-invasive manner on the boiler.

An advantage of this embodiment lies in the fact that the system according to the invention can be mounted on boilers already installed without having to drill holes in the pipes or to carry out any other intrusive operation which could be costly. An existing boiler system can therefore be modernized in a minimum of time.

According to one aspect of the present invention, there is provided a method for monitoring the performance of a boiler and anticipating a breakdown of the boiler, said method comprising the following steps:

- recovery, from data representing measurements taken by a set of sensors of a device equipped with a set of sensors, a reference standard for each sensor of said set of sensors, said reference standard being indicative of a boiler during normal operation, and

- analysis, by means of a control unit, of a data flow representing consecutive measurements taken by said set of sensors to determine anomalies of components of the boiler, in consideration of said reference standards retrieved for each sensor of said set sensors.

In one embodiment of the method according to the present invention, said method further comprises the step of analyzing said data stream using machine learning techniques.

BE2019 / 5870

According to one aspect of the present invention, a kit is made available for monitoring the performance of a boiler and predicting a breakdown of the boiler, the kit comprising at least one of:

a first pair of temperature sensors, comprising a first temperature sensor, intended to be applied to a central heating input (CCC) of a boiler, and a second temperature sensor, intended to be applied to a CCC output said boiler;

a second pair of temperature sensors, comprising a third temperature sensor, intended to be applied to a domestic hot water inlet (DHW) of said boiler, and a fourth temperature sensor, intended to be applied to an outlet d 'DHW of said boiler;

the kit further comprising a fifth temperature sensor, intended to be applied to the smoke evacuation outlet of said boiler; and an accelerometer for application to said CCC inlet or to said DHW inlet of said boiler.

In an embodiment according to the present invention, said kit further comprises a transmission module for sending said data measured by said set of sensors to a control unit.

According to one aspect of the present invention, there is provided a computer program product comprising coding means configured to cause a processor to execute the functions of said system for monitoring performance and predicting failure of a boiler. as described above.

BE2019 / 5870

The technical effects and advantages of the embodiments of the computer program product and the system according to the present invention correspond mutatis mutandis to those of the corresponding embodiments of the system according to the present invention.

Brief description of the drawings

We will now describe in more detail these characteristics and advantages of the embodiments of the present invention, as well as others, with reference to the attached drawings, in which: [FIG. 1] schematically illustrates a performance monitoring system and predicting failure of a boiler according to an embodiment of the present invention;

[Fig.2] schematically illustrates a system for monitoring performance and predicting failure of a boiler according to another embodiment of the present invention;

[Fig.3] schematically illustrates the monitoring of the boiler by the evaluation of the sensor output signals, thus making it possible to observe a future failure of the boiler.

Detailed description of embodiments

Figure 1 schematically illustrates a system 1 for monitoring the performance of a boiler and / or predicting a failure of the boiler according to one embodiment of the invention.

A specific application of the system 1 according to embodiments of the invention is a system making it possible to monitor, preferably continuously, the performance of a boiler 10, and / or to predict a breakdown at

BE2019 / 5870 come from a boiler 10. Typically, said boiler 10 is a boiler installed in a residential building, which is used to provide central heating to the building, to provide domestic hot water (DHW) or for provide a combination of the two. When necessary, the invention will be described with reference to such an application, without limiting the scope of the invention to this description. The system 1 is intended to inform the owner of the boiler or the service provider relating to the boilers concerning the performance of the boiler, typically by providing a general view of the instantaneous performance of the boiler as well as a general view of all recent developments. performance of the boiler. In the context of the present invention, the expression boiler performance refers to the capacity of the boiler to successfully transfer the heat generated to a desired application, such as a central heating circuit of a building or the DHW generation. The performance of the boiler can therefore be evaluated by calculating or measuring the sharing of the energy generated which is actually used for, for example, producing domestic hot water, by comparison with the total amount of energy generated. Typically, such heat transfer can be influenced by the wear of the components of the boiler or by a bad adjustment of the components, resulting in a high consumption of fuel compared to the heat delivered.

In addition, the system 1 is configured to provide information concerning an upcoming event, such as a failure or an upcoming failure of the boiler, the source of which may be, for example, the failure of a component.

BE2019 / 5870

It will be understood that the expression prediction or anticipation of a failure or of a breakdown of a boiler refers to a situation in which a failure has not yet occurred when the system 1, at a certain moment, determines on based on a single event or a combination of events, preferably simultaneous, that there is a significant possibility or probability of failure to come. Such a decision can be made on the basis of past experience. The accuracy of this probability can be evaluated statistically for a large number of predictions made in a large number of systems according to the invention.

According to a preferred embodiment of the invention, the boiler 10 is a boiler serving a dwelling house for a single family. The boiler 10 can also be a central boiler, meeting the needs of a plurality of residential units, such as apartments, or other buildings, such as office buildings, government buildings, public or semi-detached buildings. public, schools, factories, farms and the like.

An advantage of the embodiments of the present invention lies in the fact that a system 1 is made available which can be applied to an existing boiler 10, regardless of the make, age, type or model of the boiler 10. In other words, the system 1 according to embodiments of the invention can be mounted in modernization on an existing boiler. Another advantage lies in the fact that the system 1 can be applied to a boiler 10 in a non-intrusive manner, for example without the need to drill holes in tubes or tanks, and therefore without it be necessary

BE2019 / 5870 to introduce or insert sensors and / or wires inside the plumbing. Another advantage is that the system 1 is composed of a limited number of simple components and sensors, which offers an economically advantageous way of monitoring the performance of a boiler and predicting a breakdown.

In the context of the present invention, the term boiler refers to a device, typically a closed tank, in which a liquid is heated, without producing saturated or superheated vapor. In the context of the present invention, the boiler 10 is used to provide central heating to a building, to provide domestic hot water (DHW) or to provide a combination of the two.

Typically, the boiler 10 is designed to operate at a temperature of at least 35 ° C, preferably at least 40 ° C, more preferably at least 45 ° C, and most preferably at least 50 ° C. It will further be understood that the boiler 10 is designed to operate at a temperature of at most 90 ° C, preferably at most 85 ° C, more preferably at most 80 ° C, even more preferably at most 75 ° C, more preferably at most 70 ° C, and most preferably at most 65 ° C.

The boiler 10 as mentioned in the present invention typically comprises a boiler (not shown in the figure), in which a fuel is transformed into heat, and a heat exchanger (not shown in the figure) for transferring said heat produced to a liquid such as water. The heat exchanger is typically located inside the boiler 10 and transfers the heat generated to the liquid, which is then heated.

Depending on demand, fuel is drawn into the boiler 10 and directed to the boiler burner (s). The fuel burned by the boiler 10 can be

BE2019 / 5870 natural gas (containing mainly methane), or a petroleum-based product, such as fuel oil. It will however be clear that the system 1 according to embodiments of the invention is independent of the type of fuel source. Carbon dioxide and steam are produced during the combustion of the fuel. The boiler is therefore provided with a smoke evacuation outlet, a duct or channel 11 for transporting said exhaust gases or the fumes to the outside air, and also with at least one fan (not shown). in the figure) to expel said smoke from the duct.

A central heating system (not shown in the figure) comprises a first closed circuit, usually containing water or a water-based composition which can flow through said closed circuit. It will be clear that other liquids can also be used for this purpose. The water contained in said first closed circuit passes through the boiler 10, where it can be heated, and delivers said heat acquired further in the circuit, typically via radiators. The boiler 10 therefore plays an important role in the central heating system.

According to embodiments of the present invention, the boiler 10 provides central heating to a building. The boiler 10 then comprises a central heating circuit (CCC) inlet 12, for returning water from said first closed circuit to the boiler 10, and a central heating circuit (CCC) outlet 13, for entering the first closed circuit the water which may have been heated by the activity in the boiler 10.

According to embodiments of the present invention, the boiler 10 provides DHW.

BE2019 / 5870

As shown in Figure 1, boilers can be in the form of a single compact unit, without the need for a separate hot water tank. For the production of DHW in a configuration of this type, water will typically be drawn into the boiler 10 from a main water supply, will receive heat via the heat exchanger to reach a certain temperature. and will leave the boiler 10 towards its place of use. The advantage of this compact unit is that it saves space on the property.

The boiler then comprises a domestic hot water inlet (DHW) 14A for bringing water into the boiler 10 coming from a main water supply (not shown in the figure), and a domestic hot water outlet (DHW) 15A to conduct the water heated by the boiler 10 to its place of use in the building (not shown in the figure).

Referring now to Figure 2, the boiler 10 can also be connected to a tank 100 or a storage tank of the DHW (such a combination is sometimes known as a combined boiler).

According to a preferred configuration, water can be drawn from the main water supply to said separate tank, in which it is heated by a heating fluid which circulates in a second closed circuit between the boiler 10 and the separate tank 100 A representation of this configuration is shown in Figure 2.

In such a configuration, a domestic hot water inlet (DHW) 14B can be defined which brings water into said separate tank 100 from a main water supply (not shown in the figure), and a

BE2019 / 5870 domestic hot water outlet (DHW) 15B to conduct the heated water in tank 100 to its place of use in the building.

Alternatively, water is drawn into the boiler 10 from a main water supply through a DHW inlet 14C, heated in the boiler 10 via the heat exchange and then transported to said separate tank , which is typically thermally insulated to reduce heat loss. The heated water is then stored in the separate tank until it is used, when it is transported to its place of use by an outlet for DHW 15C.

In all configurations, a DHW inlet 14 can be defined to bring water from a main water supply into a boiler or a tank to heat the water before use, as well as an outlet for 'DHW 15 to conduct the heated water from the boiler or the tank to a place of use. In the context of the invention, boiler 10 will be called both a single unitary boiler and a combination of a boiler and a separate tank 100 as described above.

According to preferred embodiments, a boiler 10 provides both heat for a central heating system and domestic hot water (DHW). Priority will normally be given to the production of DHW. These boilers will then have a CCC 12 inlet, a CCC 13 outlet, a DHW 14 inlet and a DHW 15 outlet as described above.

The boiler 10 is further provided with at least one circulation pump (not shown in the figure), for example for pumping water or a liquid heated through the first closed circuit of the central heating system or through the

BE2019 / 5870 second closed circuit. Such a pump is typically electrically powered.

The boiler 10 can also be in contact with an expansion tank (not shown in the figure) to protect the DHW equipment from excessive pressure caused by thermal expansion as well as pressure surges.

The water demand is typically regulated by means of valves (not shown in the figure), which are present on at least the outlet of CCC 12 and the outlet of DHW 14.

According to one aspect of the invention, a system 1 is provided for monitoring the performance of a boiler and / or predicting a breakdown of the boiler. To this end, said system 1 comprises a device 20, provided with a set of sensors.

According to embodiments of the invention, the set of sensors of said device 20 comprises at least one of:

a first pair of temperature sensors, comprising a first temperature sensor 21, intended to be applied to a central heating circuit (CCC) input 12 of a boiler 10, and a second temperature sensor 22, intended to be applied to a central heating circuit (CCC) output 13 of the boiler 10; and

a second pair of temperature sensors, comprising a third temperature sensor 23, intended to be applied to a domestic hot water inlet (DHW) 14 of the boiler 10, and a fourth temperature sensor 24, intended to be applied to a domestic hot water outlet (DHW) 15 from the boiler 10.

The set of sensors of said device 20 further comprises a fifth temperature sensor 25, intended to be

BE2019 / 5870 applied to the smoke evacuation outlet 11, and an accelerometer 26.

According to embodiments of the present invention, the boiler 10 is a boiler intended to supply heat in the form of a heated liquid to the central heating of a building. It will be clear that, for these embodiments, the device 20 will not contain the second pair of temperature sensors 23, 24 or else this pair will not be installed and / or activated, which means that this second pair remains on standby .

According to embodiments of the present invention, the boiler 10 is a boiler intended to supply domestic hot water (DHW) to a building. It will be clear that, for these embodiments, the device 20 will not contain the first pair of temperature sensors 21, 22 or else this pair will not be installed and / or activated, which means that this first pair remains on standby .

According to embodiments of the present invention, the boiler 10 is a boiler intended to supply both heated water to the central heating of the building and domestic hot water (DHW). In these cases, the device 20 will contain the two pairs of temperature sensors and the two pairs will be installed and activated.

In the context of the present invention, an accelerator 26 is a device intended to measure an acceleration or to detect and measure vibrations. The accelerometer 26 can measure vibrations in three dimensions and as such can be used to measure the effect of water flowing through a pipe or a ventilator ventilating the combustion chamber.

The accelerometer 26 can in principle be applied to any of the inputs or outputs 12-15. According to forms of

BE2019 / 5870 embodiment of the invention, the accelerometer 26 is applied to the CCC 12 input of the boiler 10 in the case where the boiler 10 is connected to a central heating circuit. Preferably, the device 20 comprises a casing (not shown in the figures), which casing further houses said accelerometer 26. If the boiler supplies only DHW, the accelerometer 26 is preferably applied to the input d 'DHW 14.

According to preferred embodiments, the set of sensors of said device 20 further comprises a sensor for measuring electromagnetic interference 27 (hereinafter called EMI sensor). When used, said EMI sensor 27 is preferably applied to the output of DHW 15. The inventors have discovered that the fact of including said EMI sensor 27 allows a more complete prediction of the boiler failures associated with the DHW supply. As such, the fact of including the EMI sensor 27 in the device 20 makes it possible to further extend the range of predictable boiler failures.

According to preferred embodiments, the sensors of said assembly are waterproof.

According to embodiments of the present invention, the device 20 is electrically powered in a manner known in the state of the art. Preferably, the device 20 is supplied electrically using a standard micro USB cable.

According to embodiments of the present invention, the device 20 further comprises a transmission module 28, which is connected to the set of sensors by means known in the state of the art, such as communication protocols wired (not shown in the figure) or wireless, and which receives data supplied by said set of sensors.

BE2019 / 5870

Said transmission module 28, which may or may not be housed in said casing of said device 20, is configured to transfer the data supplied by said assembly to a control unit 30.

The data measured by said set of sensors can be communicated to the control unit 30 by said transmission module 28 by means of communication protocols known to those skilled in the art.

According to a preferred embodiment, a gateway 29 is used which is in contact with a router (not shown in the figure) connected to the Internet and which can send said data to the control unit 30 via said Internet connection. Transmitters according to the IEEE 802.11 (Wi-Fi) standard can be used to transfer data from the transmission module 28 to said gateway 29. However, it has been found that, in many cases, Wi-Fi signals are not not recommended due to a limited range, which may also be hampered by the presence, for example, of concrete walls near the boiler 10. Preferably, wireless communication technologies such as LoRa ( Long Range) are used. LoRa technology is a digital wireless data communication technology that enables long-range transmissions with low power consumption.

The control unit 30 is configured to receive data measured by said set of sensors and, preferably, transmitted by said transmission module 28.

According to embodiments of the present invention, the control unit 30 is also configured to analyze said data supplied by said set of sensors. Preferably, said analysis is carried out on a continuous basis

BE2019 / 5870 thanks to the constant input of measurements provided by the set of sensors.

The inventors have found that, for a boiler operating normally, the constant data flow which is provided by said set of sensors reveals a regular behavior pattern for each individual sensor, said pattern being a consequence of the demands of the users of the boiler or of 'automatic operations. In other words, the combination of sensors in said set of sensors allows close monitoring of the usual actions of the boiler 10.

These actions of the boiler are defined by a limited number of requests (for example switching on or off the central heating; demand for domestic water from a limited number of taps), said requests often being associated with specific time slots (for example a request in hot water from the tap of a shower can take place in the morning or at the end of the day). Consequently, said scheme is defined by recurring automatic operations performed by the boiler 10 to generate hot water for heating or for sanitary use, as well as by the actions of the user of the building, such as requesting hot water by opening a water tap or requesting an increase or decrease in the temperature of a room by modifying the thermostat.

Said operations and actions can influence more than one sensor, so it may be necessary to observe measurements taken by combinations of sensors to determine a certain action of the boiler. As such, the values measured by the sensors are a function of the action which is carried out at this time by the boiler 10.

BE2019 / 5870

It turned out that the specific and limited combination of sensors described in this document made it possible to deduce and define said regular pattern.

According to embodiments of the invention, the control unit 30 is further configured to recover, from said received data, a reference standard indicating normal behavior of the boiler and its components. A boiler in operation and under normal load will be characterized by recurring and regular patterns for the measured data of each individual sensor. Consequently, to be able to compare any anomaly with usual behavior, it is necessary to define a reference standard for each sensor of said plurality. Typically, the control unit is configured to determine such a reference standard for each individual sensor, for example when starting or applying system 1 on the particular boiler, which means that, for a limited period of time after the application of system 1 on a particular boiler, data will be recorded in order to deduce a form of standard behavior for each sensor in system 1. Such an initialization phase usually takes several days, typically 5 days, but not more than 10 days. However, it is assumed that, during this initialization phase, the boiler 10 will operate in a normal manner and will be requested accordingly.

During the initialization phase and when the boiler is assumed to be operating normally, said control unit 30 is configured to identify and store data and reference diagrams deduced from this phase in the form of a reference standard for each sensor of said set of sensors. Reference standards are stored in a database and indicate normal operation

BE2019 / 5870 of the boiler. The database therefore contains values and diagrams for individual sensors which can be measured when the boiler 10 is operating normally.

Advantageously, the system 1 makes it possible to deduce regular patterns from the measurements taken by sensors applied to boilers 10 operating normally, regardless of the type or manufacturer of the boiler. It has been found that said diagrams, reflecting normal behavior, are different between the boilers 10.

In particular, it has been found that diagrams reflecting normal behavior can differ between different types of boiler 10 for the same manufacturer, while boilers 10 from different manufacturers can have similar diagrams. However, it will be clear that system 1 can recover, from data measured by the set of sensors, regular patterns for each individual sensor, reflecting that the boiler is operating normally.

According to embodiments of the invention, the control unit 30 is further configured to analyze consecutive data supplied by said set of sensors in order to search for anomalies of components of the boiler by comparing said consecutive data with reference standards. recovered. In fact, after establishing a reference standard indicating normal operating behavior for each individual sensor of said plurality of sensors, consecutive measurements for said set of sensors, measurements which are obtained after determination of the reference standard, can be weighed vis-à-vis these reference standards. As such, the control unit 30 is configured to detect abnormal operation or anomalies by comparing the instantaneous measurements of a sensor.

BE2019 / 5870 individual to said regular patterns, or by comparing instantaneous measurements of combinations of sensors to normal patterns or to combinations of such patterns.

According to embodiments of the invention, the control unit 30 is further configured to calculate from said consecutive measurements, with respect to the recovered reference standards, instantaneous overall performance of the boiler 10. These overall performance of the boiler, which can also indicate performance changes, can be expressed using a numeric value, color code and the like.

It has unexpectedly been found that the approximate values of the sensors 21-26 are sufficient to notice a deterioration in the performance of the boiler and can serve as a basis for calculating a probability of failure of the boiler. For the simplest cases, a direct link can be established between a sensor measurement and the occurrence of a fault (for example a temperature drop measurement for an outgoing flow to central heating or the DHW indicates a decrease in the boiler output). In addition, it may happen that a breakdown is not due to an abrupt change but to a progressive evolution.

According to preferred embodiments of the invention, said control unit 30 is further configured to analyze, in consideration of reference standards retrieved for each sensor of said plurality, the data flow provided by said set of sensors using machine learning techniques.

In the context of the invention, the term machine learning techniques relates to techniques and methods for examining large pre-existing databases in order to generate new information.

BE2019 / 5870

Furthermore, according to preferred embodiments of the invention, said control unit 30 is further configured to analyze, in consideration of reference standards retrieved for each sensor of said plurality, the data flow supplied by said set of sensors using machine learning techniques and / or pattern recognition models.

The machine learning approach brings into play the calculation step, starting from the incoming data flow of sensor measurements and possibly using insights obtained by data mining, an algorithm or a mathematical function which represents a probability of boiler failure.

Said machine learning approaches may include, but are not limited to, supervised or unsupervised analysis: classification techniques (e.g. Naive Bayes, Linear Discriminant Analysis, Quadratic Discriminant Analysis, Neural Network, Tree-Based Approaches, Vector Machines Support, Nearest neighbor approaches), regression techniques (e.g. Linear regression, Multiple regression, Logistic regression, Probit regression, Ordinal logistic regression, Ordinal probit regression, Poisson regression, Negative binomial regression, Multinomial logistic regression, Regression truncated), grouping techniques (eg grouping by k means, hierarchical grouping, PCA), adaptations, extensions, and combinations of the approaches mentioned above.

According to preferred embodiments of the invention, said machine learning approach uses the Bayesian method of breaking point detection, which aims

BE2019 / 5870 to identify changes in the probability distribution of sensor measurements.

It has been found that the combination of the sensors described above, with a continuous analysis of data measured by the control unit 30, preferably using data mining and / or machine learning techniques, allows to observe changes in boiler performance and also to provide a reliable probability of a boiler failure.

According to embodiments of the invention, the control unit 30 is further configured to communicate the results of said analysis to a network by means of communication means (not shown in the figure).

If the analysis carried out by the control unit 30 deduces that all the measurements taken by the set of sensors reveal normal behavior of the boiler, the results of the analysis are stored by the control unit 30 and no action is not implemented. Preferably, said analysis is carried out continuously. If the result of the analysis highlights an upcoming failure of a boiler or a reduction in boiler performance beyond a certain predefined level, said result is also communicated to a network via 'a means of communication. Said network may contain the owner of the boiler, a boiler technician and / or other third parties.

According to embodiments of the invention, the control unit 30 is further configured to recover, from the analysis, a set of possible causal factors for the fault, the breakdown or the drop in performance, and communicate the said diagnosis to the owner of the boiler,

BE2019 / 5870 to the boiler technician and / or to any other third party through said means of communication.

Consequently, intermittent boiler failures can be noticed for boilers of different types and manufacturers, which means that the system of the invention can be applied independently of the brand of the boiler 10. Independence from -vis any brand and the possibility of modernizing existing boiler installations are part of the advantages of system 1.

The inventors have found that analysis of the measurements of a limited selection of sensors applied to the boiler 10 at different specific positions can lead to a reliable prediction of a failure of the boiler, which allows the owner of the boiler or a service provider relating to boilers to anticipate said breakdown by calling technical assistance in due time. It has been noted that signals indicating an impending or upcoming boiler failure can be seen up to 48 hours before the failure occurs. For some problems, these signals can be seen 96 hours before the actual failure.

The inventors have further found that it was possible to obtain said diagram for normal boiler behavior as well as for a future breakdown of the boiler by bringing said set of sensors into contact with elements of the boiler 10, such as plumbing, in a non-invasive manner vis-à-vis existing equipment. Sensors 21-27 can be attached to plumbing or to tanks using means known to those skilled in the art. These means include pliers, clamps, tape, glue and the like. No sensor is therefore inserted in any channel, tube or tank and it is not necessary to make any modification to these

BE2019 / 5870 elements, such as drilling a hole. Although the exact temperature of the water in any of the inlets and outlets is largely unknown and simply estimated, it has been found that these temperature estimates are sufficient to infer an upcoming boiler failure. The non-invasive mounting has the consequence that the system of the invention can be applied to older or already existing boiler installations. Existing boilers can therefore be modernized with the system of the invention.

Furthermore, no digital or electrical connection needs to be made between the device 30 and the boiler 10. The only condition is that certain specific parts of the boiler 10 must be accessible for a technician, such as plumbing and the like. .

According to preferred embodiments of the invention, the control unit 30 is not present in the residential building housing the boiler 10. The unit 30 is typically a unit placed in the central position, configured to monitor a plurality of boilers 10, having different ages, brands, types, and the like.

An advantage of the present invention lies in the fact that, for each brand of boiler 10, a standard diagram can be defined which reflects normal behavior of the boiler and which then makes it possible to determine the performance of the boiler and to predict a breakdown. coming from the boiler.

System 1 includes components such as sensors which are economical. An advantage of the present invention therefore lies in the fact that a system is provided for predicting a future breakdown of the boiler and for monitoring the performance of the boiler in an economically advantageous manner.

BE2019 / 5870

According to preferred embodiments of the invention, the system 1 is further provided with a communication means, said means being configured to communicate the analysis results as well as the set of recovered causal factors to the users of the network. using data transfer protocols known in the art.

According to preferred embodiments of the present invention, a quantified report is provided on a regular basis to persons such as the owner of the boiler according to the measurements taken by said set of sensors.

The results of the analysis performed by the control unit 30 can be shared by the communication means to a plurality of people, such as the owner of the boiler, the boiler technician and the like.

According to embodiments of the present invention, said means of communication thus uses an application programming interface (API). API stands for Application Programming Interface and is used to allow interaction between software applications and to integrate different software applications.

More particularly, an API is used to transfer data contained in a software program A to another environment under the control of a second processor which executes a software program B and to store said data in a second non-volatile memory.

Advantageously, an API can be used to communicate said data to a software program used by a boiler technician. Typically, technicians will

BE2019 / 5870 from one customer to another during the day using a mobile device 50 provided with a GUI interface providing information concerning the repairs which must be carried out.

As used herein, the term mobile device generally refers to any computer device designed to be portable. Mobile devices include laptops, mobile phones, personal digital assistants (PDAs), tablet computers, digital cameras, video cameras and the like.

Using an API, system 1 can report an upcoming failure and also suggest certain causes for the fault, which can save valuable time for the technician, regardless of the program or software system the technician is using.

According to preferred embodiments of the present invention, said communication means are further configured to communicate with a software program available to the owner of the boiler, thereby informing the owner of the boiler of the performance of the boiler and of any future failure of the boiler. Preferably, this information is transferred to a mobile device 60 of the owner of the boiler. The information provided by the media can help the customer to resolve the problem himself. Alternatively, the program may offer to reserve an intervention from a technical team.

A specific date can be offered to the owner of the boiler for repair of the boiler. And the owner of the boiler may be asked to explicitly confirm his agreement by replying to this message.

BE2019 / 5870

According to another aspect of the invention, a method is provided for monitoring the performance of a boiler and anticipating a breakdown of the boiler, said method comprising the following steps:

recovery, from data representing measurements taken by a set of sensors 21-26 of a device as described above, of a reference standard for each sensor of said set of sensors 21-26, said standard of reference being indicative of a boiler in normal operation, and

- analysis, by means of a control unit 30 as described above, of a data stream representing consecutive measurements taken by said set of sensors 21-26 to determine anomalies of components of the boiler, in consideration said reference standards retrieved for each sensor from said set of sensors.

According to embodiments of the method, the set of sensors further comprises an electromagnetic interference sensor 37.

According to preferred embodiments of the method, said method further comprises the step of analyzing said data stream using machine learning techniques.

According to another aspect of the invention, a kit is made available, said kit comprising the following elements:

at least one of:

a first pair of temperature sensors, comprising a first temperature sensor 21, intended to be applied to a circuit input of

BE2019 / 5870 central heating (CCC) 12 of a boiler 10, and a second temperature sensor 22, intended to be applied to an outlet of CCC 13 of said boiler 10;

a second pair of temperature sensors, comprising a third temperature sensor 23, intended to be applied to a domestic hot water inlet (DHW) 14 of said boiler 10, and a fourth temperature sensor 24, intended to be applied at a DHW outlet 15 from said boiler 10;

the kit further comprising:

- A fifth temperature sensor 25, intended to be applied to the smoke evacuation outlet 11 of said boiler 10; and

an accelerometer 26 intended to be applied to said inlet of CCC 12 or to said inlet of DHW 14 of said boiler 10.

According to embodiments according to the present invention, the kit further comprises a transmission module 28 configured to receive data measured by said set of sensors and to send said data to a control unit 30.

According to embodiments according to the present invention, the kit also comprises an electromagnetic interference sensor 37.

The present invention also relates to a computer program, possibly stored on a computer-readable medium, comprising coding means configured to cause a processor to execute the functions of the control unit 30 as described above.

BE2019 / 5870

Although the invention has been described above with reference to specific embodiments, this description is intended to illustrate the invention and not to limit it, the scope thereof being defined by the appended claims. Those skilled in the art will readily understand that combinations of characteristics different from those described here are possible without departing from the scope of the claimed invention.

Examples

A Junkers TOP 30-3 ZWB boiler was fitted with a system 1, said system having a device 20 provided with first and second pairs of temperature sensors. The boiler was then monitored for a period of 55 days. Figure 3 illustrates the data flow provided by the set of sensors. An analysis carried out by the control unit 30 indicated at an instant A (leftmost vertical line) an increased probability of imminent failure, leading to alerting the owner of the boiler. While it has been authorized to continue operating without significant change for the user of the boiler, an actual failure of the boiler has been measured at time B (rightmost vertical line).

Claims (15)

Claims 1. System (1) for monitoring the performance of a boiler (10) and anticipating a breakdown of the boiler (10), said system (1) being characterized in that it comprises: - a device (20) provided with a set of sensors, comprising at least one of: a first pair of temperature sensors, comprising a first temperature sensor (21), which is applied to a central heating circuit input (12) of said boiler (10), and a second temperature sensor (22), which is applied to a central heating circuit output (13) of said boiler (10); a second pair of temperature sensors, comprising a third temperature sensor (23), which is applied to a domestic hot water inlet (14) of said boiler (10), and a fourth temperature sensor (24), which is applied to a domestic hot water outlet (15) of said boiler (10); said device (20) being provided with a set of sensors further comprising: o a fifth temperature sensor (25), which is applied to a smoke evacuation outlet (11) of said boiler; o an accelerometer (26) which is applied to said central heating circuit inlet (12) of said boiler (10) or to said domestic hot water inlet (14) of said boiler (10); and - a transmission module (28) which is configured to send data representing measures taken by BE2019 / 5870 said set of sensors to a control unit (30), and - the control unit (30) which is configured to receive and analyze said data to determine anomalies in components of the boiler. 2. The system (1) according to claim 1, wherein said accelerometer (26) is applied to said central heating circuit input (12) of said boiler (10) on condition that said device (20) provided with a set of sensors includes said first pair of temperature sensors. 3. System (1) according to claim 1 or claim 2, said device (20) being provided with a set of sensors further comprising a sensor (27) for measuring electromagnetic interference. 4. System (1) according to one of the preceding claims, wherein said transmission module (28) transmits said data to said control unit (30) using a gateway (29) which is in contact with an Internet router. 5. System (1) according to claim 4, in which the Long Range LoRa technology is used by said transmission module (28) for sending data to said gateway (29). 6. System (1) according to one of the preceding claims, in which said control unit (30) is configured to recover, from said received data, a reference standard for each sensor of said set of sensors, said reference standard being indicative of a boiler (10) in normal operation. 7. The system (1) according to claim 6, in which said control unit (30) is configured to analyze a data stream representing measurements. BE2019 / 5870 taken by said set of sensors to determine anomalies in boiler components, in consideration of said reference standards retrieved for each sensor from said set of sensors. The system (1) according to claim 7, wherein said control unit (30) is configured to analyze said data stream provided by said set of sensors using machine learning techniques. 9. System (1) according to one of the preceding claims, in which said control unit (30) is configured to report the results of said comparison to third parties via a means of communication. 10. System (1) according to one of the preceding claims, wherein said sensors (21-27) are applied in a non-invasive manner on the boiler. 11. Method for monitoring the performance of a boiler (10) and anticipating a breakdown of the boiler (10), said method being characterized by the following steps: - recovering, from data representing measurements taken by a set of sensors of a device (20) according to claims 1-3, a reference standard for each sensor of said set of sensors, said reference standard being indicative a boiler in normal operation (10), and - analysis, by means of a control unit according to claims 1-10, of a data stream representing consecutive measurements taken by said set of sensors to determine anomalies of components of the boiler, in consideration of said standards of BE2019 / 5870 reference retrieved for each sensor from said set of sensors. The method of claim 11, wherein said method further comprises the step of analyzing said data stream using machine learning techniques. 13. Kit necessary for monitoring the performance of a boiler and anticipating a boiler failure, characterized in that it comprises at least one of: a first pair of temperature sensors, comprising a first temperature sensor (21), which is applied to a central heating circuit input (12) of a boiler (10), and a second temperature sensor (22) , which is applied to a central heating circuit output (13) of said boiler (10); a second pair of temperature sensors, comprising a third temperature sensor (23), which is applied to a domestic hot water inlet (14) of said boiler (10), and a fourth temperature sensor (24), which is applied to a domestic hot water outlet (15) of said boiler; the kit further comprising a fifth temperature sensor (25), which is applied to the smoke evacuation outlet (11) of said boiler (10); and an accelerometer (36) which is applied to said central heating circuit inlet (12) or to said domestic hot water inlet (14) of said boiler (10). 14. The kit of claim 13, wherein said kit further comprises a transmission module (28) for sending said measured data. BE2019 / 5870 5 by said set of sensors to a control unit (30). 15. A computer program product comprising coding means configured to cause a control unit (30) to perform the functions according to one of claims 1-14.