EP2649379A1

EP2649379A1 - Method for controlling the power consumption in a district heating system

Info

Publication number: EP2649379A1
Application number: EP11845259.8A
Authority: EP
Inventors: Thomas Wildig
Original assignee: ECOFECTIVE AB
Current assignee: Kiona Sweden AS
Priority date: 2010-11-30
Filing date: 2011-11-30
Publication date: 2013-10-16
Also published as: SE1051260A1; EP2649379A4; WO2012074478A1; SE535445C2

Abstract

Method for controlling the power in a district heating network to which several properties (2) are connected, and is characterized in that the method comprises the steps a) to establish a certain group (6) of said properties (2), which properties comprise at least one sensor (38) for indoors temperature each and are associated with a respective lower temperature limit for the indoors temperature; b) for each of the properties in the certain group (6), to measure the indoors temperature in the property in question using said sensor (38) for indoors temperature and to establish a certain time value, constituting the expected time until the indoors temperature of the respective property reaches the respective lower temperature limit at a certain decreased heating power, which decreased power is lower than a certain respective normal power level; c) to calculate an expected total power in the district heating network during a future time period; and d) in case the said expected total power exceeds a predetermined value, during said time period only to distribute the respective decreased heating power to one or several of the properties in said group (6) for which the respective time value is larger than the length of the future time period, so that the total power no longer exceeds the predetermined value.

Description

Method for controlling the power consumption in a district heating system

The present invention relates to a method for controlling the power in a district heating network to which several properties are connected.

District heating is a common and efficient way to heat a set of properties, especially apartment buildings, when it is cold outdoors. A central facility can extract and distribute thermal energy to many properties, which also results in environmental advantages. Moreover, it is common that hot tap water is heated by such a district heating system. The use of the properties in such a set of properties follows, to a large extent, a certain periodic rhythm, according to which the heating need varies depending on the time of day, the day of the week, etc. At night, when many people are indoors and emit heat, and there is little need for hot tap water, the heating needs are often relatively small. In the middle of a weekday, on the other hand, the need for heating is relatively large, since fewer persons are typically present indoors. Moreover, meteorological variations, primarily varying outdoors temperatures, give rise to variations in the power necessary in order to maintain a desired indoors temperature. In order to function well under all operation conditions, the central district heating system must therefore be dimensioned to be able to deliver high peak powers. However, high maximum powers in district heating systems are associated with large marginal costs. For this reason, and since the maximum power of the system is typically only used seldom, this problem is conventionally solved by the use of an add-on capacity based upon fossil fuel combustion or in another, environmentally harmful way.

Thus, it would be desirable to achieve a way to operate a district heating plant so that the largest power peaks are smaller than during conventional operation, which would then result in that the maximum capacity of the system can be lower .

Moreover, a method which minimizes thermal losses during operation of such a district heating system would be desirable.

The present invention solves the above described problems.

Hence, the present invention relates to a method for controlling the power in a district heating network to which several properties are connected, and is characterized in that the method comprises the steps a) to establish a certain group of said properties, which properties comprise at least one sensor for indoors temperature each and are associated with a respective lower temperature limit for the indoors temperature; b) for each of the properties in the certain group, to measure the indoors temperature in the property in question using said sensor for indoors temperature and to establish a certain time value, constituting the expected time until the indoors temperature of the respective property reaches the respective lower temperature limit at a certain decreased heating power, which decreased power is lower than a certain respective normal power level; c) to calculate an expected total power in the district heating network during a future time period; and d) in case the said expected total power exceeds a predetermined value, during said time period only to distribute the respective decreased heating power to one or several of the properties in said group for which the respective time value is larger than the length of the future time period, so that the total power no longer exceeds the predetermined value.

In the following, the invention will be described in detail, with reference to exemplifying embodiments of the invention and to the enclosed drawings, wherein: Figure 1 is an explanatory sketch of a set of properties heated according to the present invention by a district heating system; and

Figure 2 is an explanatory sketch of a property being heated according to the present invention.

Figure 1 shows a district heating system comprising a central heat source 1, in the form of a combined power and heat plant, a plant for utilization of thermal waste, a geothermal plant or the like. The heat source 1 is arranged to, during cold weather, distribute thermal energy to a number of properties 2, connected to the source 1. The distribution takes place using a main conduit 3 for a suitable heat carrier, such as hot water. The properties 2 may be of different types, but it is preferred that at least most of them, preferably all of them, are apartment building properties for several families.

A central control device 4 is arranged to control the heating power being distributed to each property 2. For example, the control can take place using a wireless sender 5 communicating with a respective receiver in each property (see below) , which properties in this case are equipped with suitable conventional actuators for controlling the circulation of heat carrier in the property in question, such as the circulation in a radiator circuit. Alternatively, the properties can comprise actuators, which are conventional as such, for controlling a ventilation circuit. For instance, instead of the wireless sender 5, a wired or internet connected signaling system may be used.

Figure 2 shows an exemplary property 20. Hot water from the heat source 1 arrives to a control unit 22 in the property 20 via an incoming conduit 21, and returns to the heat source 1 via an outgoing conduit 23. The control unit 22 is, in turn, connected to a radiator circuit 24, being illustrated principally and arranged to distribute the incoming hot water to radiators arranged in the various rooms in the property. The control unit 22 is also connected to a control device 25 for hot tap water, arranged to circulate the hot tap water in an internal hot water circuit 26 (Swedish: "VVC-krets" ) in the property 22. The control unit 22 comprises a wireless receiver 27, arranged to communicate with the sender 5. Furthermore, a control device 28 for a ventilation system comprises a receiver 29 arranged to communicate with the sender 5. Indoors air is ventilated off from the ventilation system through a chimney 30.

Alternatively, all communication between the control device 4 and the property 20 can take place via the control unit 22, which in turn controls for the example the control devices 25 and 28. Other actuators in the property 20 can be controlled, in a corresponding way, either directly from the control device 4 or from the control unit 22. Sewage water leaves the property 20 through a sewer conduit 31.

Thus, the energy balance in the property is essentially determined by the following energy flows :

• The difference in thermal energy between incoming hot water 32 and outgoing, cooled water 33.

• Departing thermal energy via hot sewage water 35.

• Indoors air 34 being ventilated away, possibly after recapture of thermal energy in the exhaust air.

• Thermal losses 36 via the walls, foundation and roof of the property.

• Supplied thermal energy originating from the use of the property, from various internal sources, exemplified in the property 20 by persons and a computer 37. Cooling inside the property 20 of used hot tap water also constitutes such an internal energy source.

Furthermore, the property 20 comprises at least one sensor 38 for indoors temperature, arranged to measure the indoors temperature in the property 20 and to communicate the measurement value to the central control device 4 in a way which is conventional as such, for example wirelessly such as described above for the communication between the control device 4 and the control unit 22, whereby the sender 5 also constitutes receiver for wireless signals from the sensor 38. The communication may also take place via the control unit 22. It is preferred that the property 20 comprises several such sensors for indoors temperature, preferably arranged in several different rooms in the property 20.

According to the invention, at first a certain group 6 (see figure 1) of the properties 2 being connected to the heat source 1 is selected. Which ones of the properties 2 that are selected depend on the specific operation applications, but all properties 7 in the group 6 comprise at least one sensor 38 each for indoors temperature, arranged to communicate the measurement value to the control device 4. Properties 8 which are not selected may for example be such properties that lack such indoors temperature sensors connected to the control device 4; that lack the possibility to be controlled from the central control device 4; or which for any other reason lack prerequisites to take part in the method described herein. Other properties 9, which are not selected, may for example have specific operation requirements, such as that they must keep a certain constant indoors temperature, which for example may be the case for hospitals, certain archives and so on .

Furthermore, each property 7 in the group 6 is associated with a respective lower temperature limit for the indoors temperature, below which the indoors temperature of the prop- erty must not fall.

The method according to the present invention considers the power control of the properties 2 in the district heating network during a certain future time period. Which future time period to use depends on the specific application, but may be a period which commences immediately or later, and may for example be a certain part of a day, such as between 18:00 and 24:00 the coming evening. For each of the properties 7 in the group 6, the indoors temperature is measured and communicated to the control device 4. Thereafter, for all properties 7 a certain time value is established, which constitutes an estimation of the expected time until the indoors temperature of the respective property has fallen from the current indoors temperature down to the respective lower temperature limit in case only a certain decreased heating power is distributed to the property in question starting immediately, which decreased power is lower than a certain respective normal power level. Thus, the time value is the length of a certain time period.

The normal power level is selected to represent a heating power which is distributed to the property in question during normal operation under the current operation conditions . The normal power level may for instance be the currently distributed heating power to the property under the current operation state, or the calculated heating power to be distributed to the property in question during the future time period, given information regarding expected weather and control parameters for the heat distribution to the property.

The certain lower power level is constituted by a suitable power level which is lower than the normal power level, and preferably lower than an expected lowest possible power level which will be able to maintain the prevailing indoors temperature during the above indicated future time period and given the then expected operation prerequisites in terms of outdoors temperatures and such. According to a preferred embodi- ment, the lower power level corresponds to that the distribution of thermal energy to the property in question is completely shut off, alternatively that the distribution of thermal energy to the radiator circuit of the property is shut off but that the distribution of thermal energy to the hot tap water in the property is maintained at a normal level.

Moreover, an expected total power in the district heating network is calculated for the future time period. It is pre- ferred that this total power represents an expected maximum, instantaneously distributed power during the future time period, and that it is calculated based upon available data regarding the construction and use of the properties, meteorological forecasts and so on. See below for a more detailed description of this.

According to the invention, a predetermined, maximum power is established for the heat source 1. According to a preferred embodiment, this maximum power is the actual maximum power for the system, more preferably the actual maximum power minus a certain predetermined security margin.

In case the above mentioned expected total power exceeds this predetermined maximum power, according to the present invention the control device 4 will, during the said time period, control the distribution of thermal energy so that one or several certain of the properties 7 in the group 6 are only receiving the above described lower heating power (s) . By in this way sufficiently decreasing the heating power to at least one of the properties 7, it may be achieved that the total distributed power during the future time period does not exceed the predetermined value. Thus, the power peak which, during conventional operation of the system, would have followed upon the decreased outdoors temperature or the other conditions forming the basis for the forecast high total power, may be eliminated, by temporarily cutting down on the heating power distributed during the period to one or several properties.

According to the invention, the above described calculated respective time value is used for each property 7 in the group 6 in order to decide to what or which properties the respective lower power is to be distributed. More specifical- ly, one or several of the properties 7 in the group 6 are selected to form an additional group 10, in which all properties in the group 10 has a respective time value which is larger than the future time period. The said selection is also made so that the expected total power distributed from the source 1, when the group 10 is operated at the respective lower powers, no longer exceeds the predetermined value during the future time period. Since the respective time period of the properties belonging to the group 10 is longer than the future period, their respective indoors temperatures will not fall below the respective lower temperature limit during operation at the respective lower power during the future period . According to a preferred embodiment, the or those properties in the group 6 having the largest respective time value is or are selected to the group 10.

Moreover, it is preferred that the single property which has the largest time value is selected in a first step, after which the property having the second largest time value is selected, and so on, until the total expected power does not exceed the predetermined value, whereby the thus selected properties constitute the group 10.

Since the control device 4 in this way is arranged to distribute a respective lower heating power to a limited share 10 of the connected set 2 of properties, it is hence achieved that the coming power peak can be decreased, which results in that the system can be dimensioned with a lower maximum capacity and still be able to deliver thermal energy at an expected service level. Furthermore, it has been found that in most sets of properties connected to district heating, there is in general at a given point in time properties for which the supplied district heating energy can be temporarily decreased without their respective indoors temperature risking to fall below a contractually regulated lowest temperature. Actually, according to the invention, during operation the respective "thermal inertia" of the properties, that is their volumetric thermal capacity, is exploited in order to level out the power peaks loading the heat source 1.

Moreover, the thermal losses in the district heating network decrease as a whole, since these are proportional to the distributed power and therefore decrease with decreasing variance .

Further, it is possible that the operational prereguisites are changed during the future time period. Also, it may be so that other properties than those in the group 10 some time into the future time period prove to be more suitable for a decreased energy supply. Therefore, it is preferred that, while the future time period is running, repeatedly update what properties that at the moment are to take part in the group 10 for lower power distribution. According to a pre- ferred embodiment, such updating follows the same rules as the original selection, and preferably takes place periodically, such as at least once every hour, alternatively continuously .

In other words, the indoors temperature in that or those properties to which the respective lower heating power at the moment is distributed is measured, and the calculated time values for the properties in the certain group are updated. Then, one or several of the properties which presently have a respective calculated time value which is larger than the time which remains of the period, is or are heated at its or their respective lower power level. The selection is coordinated so that the total distributed power to all connected properties 2 does not exceed the predetermined value, while the rest of the properties in the group 6, except those taking part in the group 10, among which possibly properties which earlier during the period were heated at their respective lower power level, are heated with a respective normal power level.

In this way, the above described advantages can be achieved, at the same time as the variations of indoors temperature in individual properties may be decreased.

It is preferred that the control device 4 calculates the above described estimation of the total power distributed from the source 1 during the future time period based upon available information of the system, the individual connected properties 2 as well as additional available information regarding the expected operation conditions during the future period .

According to a preferred embodiment, the calculation is at least partially based upon historical data regarding the total power distributed to the connected properties 2 during different types of operation conditions. For example, different historically measured total powers during periods of different outdoors temperatures may be available in tabulated form, whereby the calculation can take place by interpolation or in some other suitable, conventional way.

According to another preferred embodiment, the calculation is at least partly based upon a measured value for the outdoors temperature near at least one of the connected properties 2 in the certain group. The colder the outdoors temperature, the higher heating power is required to maintain a desired indoors temperature.

According to an additional preferred embodiment, the calculation is based at least partly upon a weather forecast covering the future time period and at least one of the properties 2. Even if the present outdoors temperature gives a good indication as to the power requirement during a future period, especially if the future period soon commences, a weather forecast gives better precision during the calculation . In case an outdoors temperature measurement or a weather forecast is only relevant to some of the connected properties 2, for instance due to local meteorological differences, an estimation, which is carried out in a way which is conventional as such, of the operation conditions for the rest of the properties may be used.

It is preferred that the control device 4 is arranged to control the distribution of thermal power to connected properties to which the power is not decreased using a control algorithm, which is known beforehand and which may be conventional as such. In this case, it is preferred that the expected total power distribution is calculated based at least partly upon the known properties of this control algorithm. For example, a weather forecast covering a certain property may form the basis for a simulation of how the control algorithm will control the heating power to the property in question, given an initial indoors temperature and the lower temperature limit for the property, and based upon such a simulation the total heating power across the time period can be calculated for the property. Thereafter, the total heating power for all connected properties 2 can be calculated based thereupon . Moreover, it is preferred that the total distributed power is calculated at least partly on historical data regarding periodical patterns in the use 37 of the connected properties 2, and the additional heating power which such use is expected to give rise to within the properties during the fu- ture time period. Different types of uses, such as presence of persons and animals inside the properties; use of technical equipment, lighting and hot tap water; airing and so on constitute either heat sources or heat sinks affecting the operation of the heating system. Since such activities are largely periodical, and to some extent predictable in their nature, they may be measured and thereafter used during the calculation in order to give a more unbiased value of the total power during the future period. It is also preferred to use a priori knowledge regarding at least one, preferably all thermal characteristics of the properties, especially its thermal capacity, as a basis for the calculation of the total power distribution during the future period. In other words, in an initial step information is collected concerning the thermal leakage through the shell 37 of the property and the ventilation 34, which information for instance can be collected from available type descriptions of different properties, and be supplemented with information regarding heat recovery systems for the ventila- tion, any installed additional insulation, and so on, alternatively with actual measurements on the spot. Then, this information together with corresponding data describing additional energy sinks such as the sewage water 31, can be used to create a model of the property as a body having a certain thermal capacity. Finally, the model can be used in order to calculate the expected power requirement during the future period, especially in combination with historical data regarding the use and/or the used control algorithm.

It is possible to, in different ways and depending on available information, combine the different above described ways to calculate the expected total energy need during the future period, with the overarching purpose of achieving an estima- tion which is as accurate as possible.

In a way similar to the above described with respect to the calculation of the expected total power for the set of properties, it is preferred that the time value for at least one, preferably all, of the properties 7 in the group 6 is calculated based upon the volumetric thermal capacity of the property in question as modeled using available information regarding the energy flows 34, 35, 36 and so on, according to the above. For example, using such a thermal model, suitable conventional differential equations may be set up to describe the indoors temperature of the property 20 as a function of time, supplied energy and other operation conditions. Such equations may then be solved in order to calculate the expected time until the indoors temperature has fallen to the lower limit.

It is furthermore preferred that the time value for at least one, preferably all, properties 7 in the group 6 are calculated based upon a measured value for the outdoors tempera- ture near the property in question and/or a weather forecast covering the said future time period and the property in question, analogously to the above described for the expected total power. Moreover, and also analogously to the above described for the expected total power, it is preferred that the said time value for at least one, preferably all, of the properties 7 in the group 6 are calculated based upon a control algorithm which is known as such and which controls the heating power which is distributed to the property in question, with the purpose of maintaining an indoors temperature therein which is above the lower allowable level, and/or historical data regarding periodical patterns in the use of the property and the additional heating power that this use is expected to give rise to in the property during the future time period.

The above described, different types of information can be combined in many ways in order to maximize the reliability in the estimation of the time value.

The district heating network can be designed so that, for at least one of the properties 7 in the group 6, thermal energy can selectably be distributed for heating of either indoors air and/or hot tap water. This may, for instance, take place as shown in figure 2, where a control unit 22 is arranged to direct hot water to either a radiator circuit 24 and/or an internal hot water circuit 26. In this case, it is preferred that the availability of hot tap water is prioritized. As long as a need for additional hot tap water is present in a certain one of the properties which may be controlled selectably regarding air and hot water, it is then preferred that an adjustment down of the heating power distributed to the property in question only bears upon the heating of indoors air and not upon the heating of hot tap water in the property. In other words, in this case the energy supply to such a property is not shut down completely, only all or part of the energy supply to the radiator circuit 24. This will lead to that the thermal inertia in the house body can be utilized according to the purposes of the invention without negatively affect the availability of hot tap water.

In order to decide when a need for additional hot tap water is present, it is preferred that the temperature in the hot tap water circuit is measured, in a way which is conventional as such, and that it is communicated to the control unit 22, the control device 4 or to another control device which is arranged to control the heating power to the internal hot water circulation in question. In this case, a need for additional hot tap water is said to exist when the measured temperature falls below a predetermined value, such as 55°C.

A method according to the present invention thus achieves that a smoother operation can be achieved without need for additional heat sources in the form of for example fossil fuel combustion in order to manage power peaks. The calcula- tions of the inventors have shown that a plant can be designed with only 70-85% of the maximum capacity which would be required during conventional operation. In many cases, this may also be achieved using software installations in an existing plant, without additional extra equipment than com- munication links between various sensors, which are often already installed, and the control device 4.

Alternatively, using the invention, more properties may be connected to one central heat source 1, since the thermal energy can be exploited at a smoother and higher level than before, when the maximum capacity was only used during power peaks . Above, preferred embodiments have been described. However, it is obvious to the skilled person that many modifications can be made to the described embodiments without departing from the idea of the invention.

In view of a future expected power peak, the heat energy stores in one or several of the connected properties 2 may for example be increased before the period in question commences. Thus, the indoors air of a property can be heated one or a couple of °C above the normal temperature before the period commences, which increases the time value of the property and therefore the possibilities to adjust down the energy supply to the property in question during the future time period without falling below the lower temperature limit. In a corresponding way, the hot tap water in the property can be overheated before the commencing of the period, so that less heating of the hot tap water is required during the period.

It may also be the case that it is known that the need for hot tap water is large during a future period during which also the outdoors temperature is expected to fall, such as the period at night between 18:00 and 24:00. In this case, the indoors temperature can be raised in the property in question during the day, to a higher temperature than what would have been the case during conventional operation, and thereafter the heating power to the radiator circuit in the property can be adjusted down during the period 18:00 - 24:00 at the same time as the heating of hot tap water is maintained during these hours.

As a complement or alternative to the historical data regarding the operation of the property which may form the basis for the calculation of the expected total heating power and the time values of the individual properties, an adaptive method may advantageously be used, according to which operation data is compiled by a central device, such as the control device 4, and where the parameters of an as such conventional energy model of the whole system or individual proper- ties continuously are adjusted based upon actual results of different operational situations. It is especially preferred that energy effects from the use of the properties are quantified this way, by measuring the actual indoors temperature given supplied heating power, the outdoors temperature and the volumetric thermal capacity of the building body. Such a method will be able to calculate ever more unbiased values for both expected total powers and the above described time values . Thus, the invention is not to be limited to the described embodiments, but may be varied within the scope of the enclosed claims.

Claims

C L A I M S

1. Method for controlling the power in a district heating network to which several properties (2) are connected, c h a r a c te r i z e d i n that the method comprises the steps

a) to establish a certain group (6) of said properties (2), which properties comprise at least one sensor (38) for indoors temperature each and are associated with a respective lower temperature limit for the indoors temperature;

b) for each of the properties in the certain group (6), to measure the indoors temperature in the property in question using said sensor (38) for indoors tempera- ture and to establish a certain time value, constituting the expected time until the indoors temperature of the respective property has fallen from the current indoors temperature and down to the respective lower temperature limit at a certain decreased heating power, which decreased power is lower than a certain respective normal power level;

c) to calculate an expected total power in the district heating network during a future time period; and d) in case the said expected total power exceeds a pre- determined value, during said time period only to distribute the respective decreased heating power to one or several of the properties in said group (6) for which the respective time value is larger than the length of the future time period, so that the to- tal power no longer exceeds the predetermined value, after which the indoors temperature in the property or those properties to which the respective lower heating power presently is being distributed is measured, the calculated time values for the properties in the group (6) are updated, and one or several of the properties that presently has or have a respective calculated time value which is larger than the remaining time of the period are caused to be heated at its or their respective decreased power level, so that the total power does not exceed the predetermined value, while the other properties in the group (6) are caused to be heated at a respective normal power level,

and in that the control in step d) of the size of the power which is distributed to said one or several of the properties in the group (6) takes place using a control device, locally installed in each such respective property, which is caused to control the circulation in a likewise locally arranged heating and/or ventilation circuit.

2. Method according to claim 1, c h a r a c t e r i s e d i n that, in step d) , the respective lower heating power is only distributed to the or those of the properties in said group (6) for which the respective time value is largest.

3. Method according to claim 1 or 2, c h a r a c t e r i s e d i n that the expected total power in step c) is calculated based upon historical data regarding the total power during different types of operation conditions.

4. Method according to any one of the preceding claims, c h a r a c t e r i s e d i n that the expected total power in step c) is calculated based upon a measured value for the outdoors temperature near at least one of the connected properties (2) .

5. Method according to any one of the preceding claims, c h a r a c t e r i s e d i n that the expected total power in step c) is calculated based upon a weather forecast covering the said future time period and at least one of the connected properties (2) .

6. Method according to any one of the preceding claims, c h a r a c t e r i s e d i n that the expected total power in step c) is calculated based upon a control algorithm which is known per se, and which controls the size of the heating power being distributed to the connected properties (2) with the purpose of maintaining an indoors temperature in each of the said properties (2) above a certain lowest allowable level .

7. Method according to any one of the preceding claims, c h a r a c t e r i s e d i n that the expected total power in step c) is calculated based upon historical data regarding periodical patterns in the use (37) of the connected properties (2) and the additional heating power, which this use (37) is expected to give rise to inside the properties (2) during the future time period.

8. Method according to any one of the preceding claims, c h a r a c t e r i s e d i n that the time value for at least a certain one (20) of the connected properties (2) is calculated based upon the thermal capacity of the property (20) .

9. Method according to any one of the preceding claims, c h a r a c t e r i s e d i n that the time value for at least one certain property (20) in the certain group (6) is calculated based upon a measured value for the outdoors temperature near the property (20) .

10. Method according to any one of the preceding claims, c h a r a c t e r i s e d i n that the time value for at least one property (20) in the certain group (6) is calculated based upon a weather forecast covering the said future time period and the property (20) .

11. Method according to any one of the preceding claims, c h a r a c t e r i s e d i n that the time value for at least one property (20) in the certain group (6) is calculated based upon a control algorithm, which is known as such and which regulates the heating power which is distributed to the property (20) with the purpose of maintaining an indoors temperature therein which is above a certain lowest allowable level .

12. Method according to any one of the preceding claims, c h a r a c t e r i s e d i n that the time value for at least one property (20) in the certain group (6) is calculated based upon historical data regarding periodical patterns of the use (37) of the property (20) and the additional heating power that this use (37) is expected to give rise to therein during the future time period.

13. Method according to any one of the preceding claims, c h a r a c t e r i s e d i n that the district heating network can distribute thermal energy for heating of both in- doors air and hot tap water to at least one of the properties (7) in the certain group (6), and in that, as long as a need for additional hot tap water is present in the property, a downwards adjustment of the heating power distributed to the property in question only concerns the heating of indoors air and not the heating of hot tap water in the property.

14. Method according to claim 13, c h a r a c t e r i s e d i n that the temperature in the internal hot water circulation circuit (26) is measured, and in that a need for addi- tional hot tap water is present when the measured temperature falls below a predetermined value.