GB2589893A

GB2589893A - Heater monitoring

Info

Publication number: GB2589893A
Application number: GB1918200.5A
Authority: GB
Inventors: Voglitsis Kostas
Original assignee: Centrica PLC
Current assignee: Centrica PLC
Priority date: 2019-12-11
Filing date: 2019-12-11
Publication date: 2021-06-16
Anticipated expiration: 2039-12-11
Also published as: GB2589893B; GB201918200D0

Abstract

A method of boiler status monitoring, comprising sensing the input electrical power to the boiler (402, Fig.4) and determining, based on the sensed electrical power, whether the boiler is in a burn state. Data indicative of the burn state history is stored, based on the determined burn state. A maintenance status of the boiler is determined, based on the burn state history. The burn state may be one of a plurality of boiler states including a circulating state and an idle state. The boiler maintenance status may be used to determine a service indicator or interval for the boiler. A boiler status monitoring apparatus comprises an input for receiving a measure of the electrical power supplied to the boiler, and a processor (430, Fig.4) to identify periods where the boiler is in a burn state, based on the received measure of electrical power, and to thereby determine a cumulative burn time. The apparatus further comprises an output for outputting a boiler maintenance status, based on the cumulative burn time. The input may comprise a current sensor (420, Fig.4) which may be a retrofit current sensor.

Description

Heater monitoring The present invention relates to monitoring of heaters, and particularly but not exclusively to monitoring boilers.

Heaters employing combustion of a fuel, such as domestic gas boilers or hot air furnaces for example, typically require regular checking and maintenance to ensure continued safe operation. This is typically performed by a qualified engineer at regular service intervals, which are typically determined by the heater manufacturer. The service interval is usually one year, and is a compromise between the inconvenience and cost of a service, and safety and efficiency factors.

It is an object of certain aspects and embodiments of the present invention to provide improved heater monitoring.

According to a first aspect of the invention there is provided a method of boiler status monitoring, comprising sensing the input electrical power to the boiler; determining, based on the input current, whether the boiler is in a burn phase; storing data indicative of the burn phase history; and determining, based on the burn phase history, a maintenance status of the boiler.

By monitoring the burn, or combustion state in this way, a more accurate picture can be obtained of the use of a boiler, and the use of particular components in the boiler which may require greater maintenance (eg due to high wear or greater criticality). This in turn gives a greater understanding of the maintenance status of the boiler, in terms of whether it is appropriate and/or necessary for the boiler to be serviced.

In embodiments the boiler maintenance status is used to determine a service indicator or interval for the boiler. Thus a simple flag or indicator could be provided, indicating whether a service is required. This could be a simple visual output on the boiler such as a light, or a message or notification could be sent to a mobile device for example. In more sophisticated embodiments an output could be an expected or predicted interval (eg a number of weeks or months) until, or the date of, the next service required.

In a further embodiment a service is requested or instructed automatically, based on the determined maintenance status. For example an engineer visit could be automatically scheduled for a particular day, via communication or an interface with a service provider. Alternatively a recommended day could be proposed to a boiler owner, and an engineer scheduled subject to confirmation.

The stored data indicative of burn phase (or combustion phase) history may be, or at least include, the cumulative total burn time of the boiler, in other words, the total amount of time for which the boiler has actually been burning, since a reference time or date. The reference time is typically the last boiler service, at which point the total is reset to zero. Alternatively or additionally, the total burn time since installation or manufacture of the boiler may be stored. Other historical burn phase data which can be stored includes the pattern or distribution of burn times, eg the number of cycles to and/or from a burn phase (a burn phase being a substantially uninterrupted period during which the boiler is in a burn state), and the ratio of burn time to non-burn time (or total elapsed time).

The maintenance status can be determined in simple embodiments by comparing the cumulative burn time to a threshold time. For example, a total burn time before a service is required might be 500hrs, 750hrs, 1000hrs, 1500hrs, or 2000hrs. By monitoring a burn trajectory or trend, eg the current or historical number of hours of burn time per day, a service date or interval can be predicted, based on typical use. In more sophisticated embodiments, patterns of use can alternatively or additionally be used. For example long uninterrupted burn periods may be considered to put less stress on a boiler than the equivalent time made up of a number of shorter cycles of burn times.

It has been found to be advantageous to distinguish not only a burn state (and a non-burn state), but also other states of boiler operation. Doing this allows tracking of the boiler state more closely, and in certain embodiments can allow the burn time to be tracked more accurately and reliably. In embodiments therefore, boiler operation is classified into one of a plurality of states. Preferably there are at least three states, and more preferably these states are a burn state, a circulation state where hot water is circulating (for example through a hot water heating system) but the boiler is not burning, and an idle state, where the boiler is in effect in a standby, or sleep state.

In embodiments, a burn state is determined by monitoring statistical variations in the input electrical power. In other words, the state of the boiler can be determined based on relative changes in the electrical power, rather than absolute values. Such changes may be compared to a predetermined threshold, which threshold can be constant for all makes and models of boiler.

An advantage of certain aspects of the invention is that they can be boiler agnostic -that is, they can successfully determine the boiler maintenance status for a wide range of different makes and models of boiler, and for a range of different ages and maintenance histories. In embodiments the determination of burn state is made without knowledge or reference to any manufacturer or model specific data, and preferably without access to sensor or control data from the boiler's internal or proprietary processing and/or control logic. Additionally embodiments do not require any boiler specific customisation, calibration or installation.

Sensing the electrical power may comprise measuring the current drawn by the boiler, optionally via a retrofitted sensor such a current sensor clamped to an electrical cable, or a smart plug. In other words, in embodiments sensing of the electrical power can be performed without accessing an internal boiler components, and can be achieved without any direct electrical contact, eg using a Hall effect sensor.

In a further aspect, there is provided a boiler status monitoring apparatus comprising an input for receiving a measure of the electrical power supplied to the boiler; a processor adapted to identify periods where the boiler is in a burn state, based on the received measure of electrical power, and to determine a cumulative burn time, and an output for outputting a boiler maintenance status, based on the cumulative burn time.

In embodiments the input comprises a current sensor, preferably a retrofit current sensor for example a device intended to be affixed to or around the electrical supply cable of the boiler. Alternatively a smart plug could be used. In some embodiments, the electrical power can be sensed at a higher level, for example to a room of a property, or to a whole property, and disaggregation algorithms could be employed to extract the boiler power or current. Such algorithms are known in the art.

The invention also provides a computer program and a computer program product for carrying out any of the methods described herein and/or for embodying any of the apparatus features described herein, and a computer readable medium having stored thereon a program for carrying out any of the methods described herein and/or for embodying any of the apparatus features described herein.

The invention extends to methods, apparatus and/or use substantially as herein described with reference to the accompanying drawings.

Any feature in one aspect of the invention may be applied to other aspects of the invention, in any appropriate combination. In particular, features of method aspects may be applied to apparatus aspects, and vice versa.

Furthermore, features implemented in hardware may generally be implemented in software, and vice versa. Any reference to software and hardware features herein should be construed accordingly.

Preferred features of the present invention will now be described, purely by way of example, with reference to the accompanying figures in which: FIG. 1, which illustrates boiler electrical power FIG. 2, which shows boiler electrical power and a boiler status signal FIG. 3, which is a flowchart for determining a boiler status FIG. 4, which shows a boiler and associated apparatus.

Figure 1 shows a trace 102 of the current of the electrical supply powering a boiler, against time. This can be measured directly using a smart plug for example, or may be calculated or inferred from a measure of the electrical supply to a house using disaggregation.

It has been found that using the electrical signal, it is possible to identify those time periods during which the boiler is burning fuel, typically gas. In figure 11 these are from Ti to T2, T3 to T4, and T5 to T6. The duration of these periods can be logged, and a record kept of the cumulative total burn time. This might be the cumulative time since a last designated reference time, such a last boiler service, and/or a total time since commissioning/installation of the boiler. Additionally or alternatively the total burn time per month or week for example could be determined. Typical burn times per month for a domestic property might be 150hrs in the UK in winter.

A method of identifying burn periods from the electrical signal will now be described. The current signal is a time series of current data with, for example a sampling interval of 1 second. A rolling sample window is defined, which is continuously updated with new samples. A statistical measure of the sample window is taken, and each new value compared to that measure, to determine significant differences. An appropriate threshold is defined to distinguish a change which is determined to be indicative of a change to or from a burn period. In one embodiment the mean value of samples in the rolling window is obtained and updated, and a Z-score, or standard score for new values defines the relationship (level of similarity or difference) between that new value and the previous values in the window.

An extension of the method described above is illustrated in Figure 2, where the current value 202 is again shown. A second trace 204 is also shown, and is a signal having one of three possible values indicating three possible boiler states. A first value shown up until time Ti and from time T4 onwards is a quiescent state, indicating the boiler is in an idle state. A second value from Ti to T2 is a burn state, where the boiler is identified as being in a burn phase. A third value, from T2 to T3 is where the boiler is identified as being in circulating (but not burning) state. The states are determined based on logic and statistical analysis of the current values, as will be explained below, with reference to Figure 3.

In step 302, initial algorithm parameters are set. These may be based on empirical testing or domain knowledge for example. Examples of the initial parameters include the size of the rolling window (ie the number of samples, or corresponding length of time), the quiescent lag (which is the length of time or number of samples the algorithm waits for before comparing new values with the statistical measure of the sample window) and the threshold(s) for distinguishing a change of state. A quiescent value for current data can also be set, which is the value below which the boiler is determined to be idle.

In step 304, a new current data value is obtained, and at step 306, it is determined whether the quiescent lag period has elapsed. If it has not, the current data value is appended to the window and the window statistical properties updated at steps 308 and 310 respectively. The process the returns to step 304 to read a new data value. If however at step 306 it is determined that the quiescent lag period has elapsed, it is determined in step 312 whether the Z-score for the new value is greater than the threshold. In other words, it is determined whether the number of standard deviations by which the value is above or below the mean of the window, exceeds a threshold set in step 302.

If the determination in step 312 is negative (ie the difference is not sufficiently great) the boiler state is unchanged, and the current data value is appended to the rolling window in step 308, the window properties are updates in step 310, and the process returns to step 304 to read a new data value.

If the determination in step 312 is positive (ie the difference is sufficiently great) the process proceeds to step 314, where it is determined whether the current value is greater than the quiescent value. If not, then the boiler status is set to the idle state, and the process returns to step 304, via steps 308 and 310 as before. If the current value is greater than the quiescent value, then is checked in step 316 whether the current value is greater than the window mean (ie the mean of data values in the current window). If the result of step 316 is negative (current value not greater than mean), then the boiler status is set to the circulating state. If the result of the step 316 is positive (current value greater than mean) then the boiler state is set to the burn state. In both cases, after step 316 the process returns to step 304, via steps 308 and 310 as before.

Returning to step 302, setting of the quiescent value may be based on domain knowledge (eg based on known values for particular makes and models of boiler) but as mentioned there are over 5000 different models known to be in operation in the UK, so in embodiments it is preferred to store current data over an extended period, such as 12 or 24 hours, and filter out any values over a threshold current (for example 200mA, which in practise should be greater than the quiescent value of most boilers) . The median value of the remaining samples is taken, and a buffer added and subtracted to provide a quiescent range. The upper value of the range is used as the quiescent (threshold) value.

Additional logic based on domain knowledge may also be used to assist in determining the boiler state. For example, certain state transitions may not be allowed, such as a transition directly from an idle state to a circulation state, as this never occurs in a properly functioning boiler.

The output of the process of Figure 3 is the boiler state, represented as a signal in one of three possible states. Further processing/logic can be applied to store a history of the output state, and/or process the output to provide a burn phase history, such as cumulative burn time, and other statistics eg transitions between states. Still further steps can utilise the output state and/or the history of the output state to perform an action relating to servicing of the boiler. Such actions include indicating to a user that the boiler needs servicing, or proposing or arranging a time to service the boiler in the future. Therefore the process may interact with the user, and or with a maintenance provider.

Figure 4 shows a typical domestic boiler 402. The boiler receives control signals from a boiler control module 404, which may in turn be controlled by a thermostat and/or a remote device (eg a smartphone via an internet connection). The control signal. To the boiler is typically a demand for central heating and/or a demand for hot water. The boiler's internal logic then controls the various boiler and associated components (fan, pump, fuel flow, ignition etc) in response to such signals. The boiler receives power from a power source 406, which is typically a wall socket in a domestic arrangement. Plumbing connections to the boiler are shown as heating flow 408, heating return 410 and domestic hot water 412.

A sensor 420 is shown measuring the current drawn from power source 406. This may be a clip on sensor attached to a power cable, or alternatively a smart plug may be used. Sensors shown schematically as 422 may optionally be provided to measure the temperature of the various plumbing connections to the boiler. A processor 430 receives data from sensor 420 and optionally the sensors 422. The processor can output a boiler state, such as whether or not the boiler is in a burn phase, or alternatively it can provide a more processed output, such as cumulative burn time or a maintenance state or signal. The temperature data from the sensors 422 can optionally be used to augment the current data, together with supplementary logic to determine the burn state. For example they can help provide context to the boiler operation which can in turn provide greater certainty or differentiation between states. In one embodiment, it has been found that the temperature gradients of at least one of the heating flow, heating return, and/or domestic hot water can usefully be used.

It will be understood that the present invention has been described above purely by way of example, and modification of detail can be made within the scope of the invention.

Although the invention has been described primarily in relation to boilers (ie hot water heaters) embodiments are equally applicable to hot air furnaces, or any other type of heater periodically employing a combustion or burn phase to produce heat.

Each feature disclosed in the description, and (where appropriate) the claims and drawings may be provided independently or in any appropriate combination.

Claims

CLAIMS1. A method of boiler status monitoring, comprising: sensing the input electrical power to the boiler; determining, based on the sensed electrical power, whether the boiler is in a burn state; based on the determined burn state, storing data indicative of the burn state history; and determining, based on the burn state history, a maintenance status of the boiler.
2. A method according to Claim 1, wherein the boiler maintenance status is used to determine a service indicator or interval for the boiler.
3. A method according to Claim 1 or Claim 2, further comprising instructing or requesting a service to be performed on the boiler, in dependence on the determined maintenance status.
4. A method according to any preceding claim, further comprising determining the boiler state from one of a plurality of, preferably three or more, predetermined boiler states, including a burn state.
5. A method according to claim 4, wherein said plurality of states includes an idle state.
6. A method according to claim 4 or claims, wherein said plurality of states includes a circulating state.
7. A method according to any preceding claim, wherein the stored data indicative of burn state history includes the cumulative total burn time of the boiler.
8. A method according to claim 7, wherein the cumulative total burn time is determined since the last boiler service.
9. A method according to any preceding claim, wherein the stored data indicative of burn state history comprises the cumulative number of state transitions to and/or from a burn state.
10. A method according to any preceding claim, wherein sensing the electrical power comprising measuring the current drawn by the boiler.
11. A method according to any preceding claim, wherein a burn state is determined by monitoring relative variations in the input electrical power.
12. A boiler status monitoring apparatus comprising an input for receiving a measure of the electrical power supplied to the boiler; a processor adapted to identify periods where the boiler is in a burn state, based on the received measure of electrical power, and to thereby determine a cumulative burn time; and an output for outputting a boiler maintenance status, based on the cumulative burn time
13. An apparatus according to claim 12, wherein said input comprises a current sensor.
14. An apparatus according to claim 13, wherein said current sensor is a retrofit current sensor.
15. A computer system or computing device having means for performing a method according to any of claims 1 to 11, the means optionally comprising one or more processors with associated storage storing software code executable by the processor(s) to perform the method.