Method for Operating State Monitoring of a Heating System, Method for Controlling a Heating System, and a Heating System

The present invention relates to a method for operating state monitoring of a heating system, a method for controlling a heating system, and a heating system.

The operation of heating systems for heating rooms of a building depends at times on outside temperatures of an environment of the building, which not only determine the necessity of heating per se, but also the associated energy consumption.

Thus, in warm summer months with comparatively high outside temperatures, generally no heating of the rooms by the heating system will be necessary, whereas in cold winter months the opposite applies. A possibly unnecessary operation of the heating system in said summer months leads to an unnecessary energy consumption and thus to avoidable energy costs.

Especially in transition regions between periods in which heating would be necessary due to low outside temperatures and those in which no heating would be necessary due to higher outside temperatures, a switching of the heating system from a heating to a non-heating operation, or vice versa, usually takes place too late and results in unnecessary energy costs, which is at times promoted by the usually manual switching of the operating state by an operator of the heating system.

For this purpose, heating systems are known from the prior art whose operating states are monitored to the effect that a respectively advantageous operating state is determined taking into account a detected outside temperature.

As an example, EP 1 988 348 A1 discloses a heating system with a heat pump whose operating state (heating or non-heating) is defined as a function of an outside temperature detected by an outside temperature sensor.

The requirements for such a procedure based on the outside temperature are comparatively high, since due to the volatility of the climate not only strong fluctuations of the outside temperature over the year, but also already strong fluctuations over a single day must be expected, as a result of which often incorrect decisions are made about a switching of the operating state of the heating system.

As a result of such incorrect decisions, increasing energy costs and a loss of comfort perceived negatively by inhabitants of the building occur.

It is therefore an object of the present invention to provide a more efficient option for determining an optimum operating state of a heating system in comparison with the prior art.

To achieve this object, a method for operating state monitoring of a heating system according to claim, a method for controlling a heating system according to claim, and a heating system according to claimare provided.

The respective dependent claims relate to preferred embodiments, which can respectively be provided by themselves or in combination.

According to a first aspect of the invention, a method for operating state monitoring of a heating system of a building is provided, which comprises at least one heating circuit for heating at least one room of the building, wherein the heating circuit can be switched at least between an operating state heating the room and an operating state not heating the room. At it, the method comprises detecting an outside temperature of the building, determining a consumption prediction value as a function of the detected outside temperature, which describes an expected energy consumption of the heating system for heating the at least one room, determining a setpoint value for an operating state parameter of the heating system on the basis of the determined consumption prediction value, wherein the operating state parameter determines an operating state of the heating circuit, and outputting a setpoint value signal corresponding to the determined setpoint value, in particular for further use by the heating system.

In this way, a setpoint value for the operating state parameter based on an estimated energy consumption is provided in the form of the output setpoint value signal, based on which numerous further actions, in particular of the heating system, can be carried out, which can comprise displaying the setpoint value on a display unit of a user interface device or also control actions to be carried out by a control device of the heating system depending on the setpoint value.

The determination of the setpoint value includes an expected energy consumption associated with the heating operation, based on which it can be indicated, inter alia, whether a high energy consumption or a comparatively low energy consumption is to be expected during heating. For example, in combination with an actual value of the operating state parameter, this results in a statement as to whether it would be useful to maintain the operating state or to switch it.

Thus, by using the consumption prediction value, the highly volatile character of the outside temperature is attenuated, in that it is reduced to an energy consumption prediction and thus preferably uses the energy consumption to be used for the day as a decision criterion for an optimum operating state of the heating circuit.

Herein, the consumption prediction value is to be understood as any indication or description of an energy consumption of the heating system and can be indicated as an energy indication, for example in kWh, as a time-period-specific energy indication, for example in kWh/day, and the like, but also as energy costs, for example in €, €/day, etc. In a particularly simple form, the consumption prediction value can also only indicate whether the energy consumption is high or low, wherein this assessment is carried out by comparison with a fixed energy consumption threshold value. In the same way, a division into high, medium or low can also be carried out on the basis of two energy consumption threshold values.

On that account, the outside temperature can be detected, for example, by an outside temperature sensor or on the basis of a retrieval of weather data from a weather service or the like.

In addition, the setpoint value of the operating state parameter only relates to the heating functionality of the heating circuit or circuits and does not include any other functionalities of the heating system, such as, for example, a service water heating.

Herein, the setpoint value signal is to be understood as any electronic or electromagnetic signal which is suitable for data transmission between a transmitter and a receiver and which can be transmitted in a wired and/or wireless manner.

Preferably, the method is carried out continuously over time at predetermined time intervals in order thus to ensure a continuous operating state monitoring of the heating system.

In a preferred embodiment, the operating state parameter can be represented at least as a first value and a second value, wherein the first value corresponds to the heating operating state and the second value corresponds to the non-heating operating state, for example the values “on” and “off” or “0” and “1” etc., wherein the determination of the setpoint value for the operating state parameter is carried out in such a way that the setpoint value corresponds to the first value if the determined consumption prediction value is greater than a predetermined first threshold value, and the setpoint value corresponds to the second value if the determined consumption prediction value is less than a predetermined second threshold value.

In this way, an assignment is provided which can be implemented in a comparatively simple manner and on the basis of which the setpoint value can be determined from the determined consumption prediction value.

Preferably, the second threshold value is less than the first threshold value, so that a brief undershooting of the first threshold value following an overshoot does not immediately lead to a switchover of the setpoint value again. The same applies to the second threshold value.

In this way, a particularly robust and disturbance-insensitive procedure is provided which reduces the risk of a rapidly successive change of the setpoint value.

In a preferred embodiment, the method comprises providing a consumption prediction function, which describes the expected energy consumption of the heating system for heating the at least one room at least as a function of the outside temperature of the building, and wherein determining the consumption prediction value is carried out on the basis of the provided consumption prediction function and the detected outside temperature.

Herein, the consumption prediction function is to be understood as any type of mathematical assignment, by means of which a target value from a target quantity is assigned to a value of an input variable, or a combination of values of a plurality of input variables. Examples herein comprise, and not limitedly, graphical assignments via characteristic maps, assignment tables, but also equation-based assignments.

In this way, the method is extended by a function-based description of the expected energy consumption as a function of the detected outside temperature, which allows a comparatively fast, robust and reproducible evaluation of the consumption prediction value.

In addition, the consumption prediction function can be adapted in a comparatively simple manner to specific characteristics of the heating system or can also be updated in a simple manner using the detected outside temperatures themselves, for example by adapting coefficients in the case of an equation-based assignment.

In a preferred embodiment, a target set of the consumption prediction functions further comprises at least a first value, which describes an expected energy consumption that is equal to or greater than a defined energy consumption limit value, and a second value, which describes an expected energy consumption that is less than the defined energy consumption limit value.

In this way, the statement of the consumption prediction limit value is reduced to the two cases of a low and a high expected energy consumption, wherein the two ranges “low” and “high” are separated by said energy consumption limit value and the consumption prediction value assigns the energy consumption to one of these two ranges.

In the case of an expected low energy consumption, it can be assumed that no heating by the heating circuit would be necessary, as a result of which the setpoint value is set to the second value, whereas for an expected high energy consumption heating by the heating circuit would be necessary, as a result of which the setpoint value is set to the first value.

Preferably, the defined energy consumption limit value is between 0.1 and 5 kWh per day, preferably between 0.5 and 2.5 kWh per day and particularly preferably this is 1 kWh per day.

In addition, in a preferred embodiment, the consumption prediction function describes a probability that the expected energy consumption of the heating system is equal to or greater than a defined energy consumption limit value, wherein a target set of the consumption prediction functions comprises a plurality of continuously distributed probability values for this purpose, in particular in the value range between 0 and 1.

As described previously, this procedure allows a classification of a low or a high expected energy consumption, wherein, in contrast to the discrete case of the previous embodiment, a statement is made about a probability.

This allows a more precise estimation of the expected energy consumption, wherein ultimately no statement is made about the actual level of the expected energy consumption, but a statement is made about how probable it is that the energy consumption is greater than the energy consumption limit value. Such a procedure allows, inter alia, the simple definition of transition regions for switching the setpoint value.

Thus, for example, the setpoint value can be set to the first value if a probability described by the consumption prediction function is 20% or less, whereas the setpoint value can be set to the second value if the probability described by the consumption prediction function is 60% or more. The intermediate range of 20 to 60% serves as a transition region in which, starting from the respective previous probability, no change of the setpoint value takes place.

Herein, the indication of the probability is yet not limited to the case outlined above. Thus, the probability indicated by the consumption prediction function can also indicate how probable it is that the expected energy consumption of the heating system is less than the defined energy consumption limit value.

In a preferred embodiment, the method further comprises detecting a plurality of additional outside temperatures, which are detected at time intervals from one another, and wherein determining the consumption prediction value further comprises determining an average outside temperature from a set of values, comprising the detected outside temperature and the detected plurality of additional outside temperatures, and determining a minimum outside temperature from the set of values, wherein the provided consumption prediction function describes the expected energy consumption as input variables as a function of an average outside temperature and a minimum outside temperature, and wherein determining the consumption prediction value is carried out on the basis of the provided consumption prediction function and the determined average outside temperature and the determined minimum outside temperature as input variables of the consumption prediction function.

In this way, not only fluctuations of the outside temperature are compensated by the averaging, but also an additional evaluation is carried out on the basis of the minimum outside temperature. Thus, not only an average outside temperature is used for estimating the energy consumption, but also a minimum outside temperature. This is advantageous in particular for scenarios in which, although a comparatively high average outside temperature is present, due to strong cooling in the morning and/or evening hours of a day, also very low outside temperatures are reached, which would necessitate heating by the heating circuit, but would possibly not take place without taking into account the minimum outside temperature.

By said procedure, in the course of determining the consumption prediction value, thus also a fluctuation around the average outside temperature is taken into account, which proves to be advantageous in particular for the case of cool morning and/or evening hours described above by way of example.

In a preferred embodiment, the detection of the plurality of additional outside temperatures is carried out over N days, where N≥1, wherein a plurality of outside temperatures are respectively detected for each of the N days, wherein determining the average outside temperature in turn comprises determining respective daily averages of the outside temperatures for each of the N days, determining a first average value from the determined daily averages of the N days and outputting the determined first average value as the average outside temperature, and wherein determining the minimum outside temperature in turn comprises determining respective daily minima of the outside temperature for each of the N days, determining a second average value from the determined daily minima of the N days and outputting the determined second average value as the minimum outside temperature.

In this way, not only is the robustness of the method further increased, since fluctuations of the outside temperature are further attenuated by the averaging, but also a tendency of the last days can be taken into account, so that, for example, a short-term heating period does not immediately lead to the switching off of the heating circuit. Herein, preferably, the number N of days is 2 to 14, or 3 to 7 and particularly preferably N=7.

In an additional preferred embodiment, providing the consumption prediction function further comprises providing a plurality of heating systems, which are respectively assigned to a building and respectively comprise at least one heating circuit for heating at least one room of the respective building, setting the energy consumption limit value, detecting operating data and environmental data of the provided plurality of heating systems, which in turn comprises, for each heating system of the plurality of heating systems, detecting a plurality of outside temperatures of the respective building, which are respectively detected at predetermined points in time over a predetermined period of time, and detecting a plurality of energy consumption values, which respectively describe the energy consumption of the heating system for heating the at least one room of the respective building and are detected at the predetermined points in time, as well as determining the consumption prediction function on the basis of the outside temperatures detected during the detection of operating data and environmental data and energy consumption values as well as the defined energy consumption limit value.

In this way, the consumption prediction function can be provided on the basis of a plurality of data of a wide variety of heating systems, which allows a particularly reliable statement about an expected energy consumption as a function of the outside temperature.

Further, preferably, the consumption prediction function is determined on the basis of the detected operating data and environmental data by means of a method for machine learning (machine learning), for example, using a regression method, a decision tree or an artificial neural network.

In particular, during the determination of the consumption prediction function, a model function is set with one or more function parameters still to be determined, the values of which are determined on the basis of the machine learning. Herein, it is attempted, by selecting the function parameters, to map the determined values for the desired input variables as accurately as possible onto the associated determined values of the output variable of the consumption prediction function to be mapped thereon.

At it, preferably, only a differentiation of the cases with an energy consumption greater or less than the defined energy consumption limit value is carried out for the output variable, which allows a relatively simple machine learning to be carried out rapidly and ultimately leads to the consumption prediction function already described above, which only differentiates between a low and a high energy consumption (either discretely or via a probability, depending on the model function set).

In a preferred embodiment, detecting the outside temperature further comprises measuring an outside temperature via an outside temperature sensor of the heating system.

In addition, in a preferred embodiment, the method also comprises retrieving weather data for a region of the building, wherein detecting the outside temperature comprises outputting a temperature measurement value contained in the retrieved weather data as the detected outside temperature. The region herein can be (not limitedly) understood as a region in a perimeter of the building of 0 to 250 km, preferably of 0 to 100 km and particularly preferably of 0 to 25 km.

In this way, the heating system is not dependent on its own outside temperature sensor, but can resort to external data sources for detecting the outside temperature in the course of the method.