Controlling the temperature of a building

Abstract

The invention relates to a method for operating a temperature control system (5) for a building (1), wherein the temperature control system (1) comprises a plurality of temperature control devices (11, 21) which can be controlled independently of one another, wherein the method comprises the following steps: a) providing a thermal model (M) of the building (1), wherein in the model (M) the building (1) is divided into a plurality of zones (10, 20), to each of which at least one temperature control device (11, 21) is assigned, wherein the thermal model (M) is suitable for predicting a time-dependent room temperature profile for the respective zone (10, 20) depending on a time-dependent operating parameter set (P) of the temperature control system (5), wherein in the thermal model (M) the building parts associated with the respective zone (10, 20) are each represented by at least one separate thermal mass (m) having a heat capacity (C m ), b) providing a target function (Z) for an optimization method, wherein the target function (Z) contains at least one optimization criterion (c2) which depends directly or indirectly on time curves of power consumption levels of the temperature control devices (11, 21), c) extremalizing the target function (Z) on the basis of computer-assisted determined prediction values based of the thermal model (M), whereby an optimized time-dependent operating parameter set (P) is obtained, and d) operating the temperature control devices (11, 21) of the temperature control system (5) according to the optimized time-dependent operating parameter set (P).

Description

[0001] Description

[0002] Temperature control of a building

[0003] The present invention relates to a method for operating a temperature control system for a building, wherein the temperature control system comprises a plurality of independently controllable temperature control devices. Furthermore, the invention relates to such a temperature control system.

[0004] Numerous temperature control systems, and in particular heating systems for buildings, are known from the state of the art, with which different zones of the building can be set to different temperature ranges. In a hydronic central heating system, water is heated in a central heating point and used to transport heat to individual radiators in the areas to be heated. Separate temperature control for these zones is achieved by local thermostats located near the radiators. The radiator thermostats can, if required, be implemented as electronic wireless thermostats, thus enabling wireless temperature control for the individual zones via a higher-level control system. This allows different time-dependent temperature profiles to be specified for different zones of the building depending on their use.Such separate control of the temperatures of individual zones usually leads to more energy-efficient operation of the entire heating system. The central heating points of such central heating systems are often powered by fossil fuels such as heating oil or natural gas, or even wood pellets. As part of the energy transition, heat pumps are increasingly being used as heat generators. As an alternative to such central heating systems, electric direct heating units have recently become more common. These units make it particularly easy to separately control the temperature of individual zones of the building, as the thermal energy is generated locally in the individual building zones and the individual direct heating units can be controlled separately via their respective control units.In this way, a desired temperature profile can be set for each building zone by programming a direct heating device assigned to the zone, without this significantly affecting the temperature control of the other zones. Overall, a comparatively high level of energy efficiency can be achieved for the temperature control of the building. With this type of separate local temperature control, the temporal profiles for the electrical power consumption of the individual heaters result as a direct consequence of the locally specified temperature profiles in combination with the applied control parameters. A disadvantage of this type of temperature control, however, is that the temporal profile of the electrical power consumption usually does not respond to external incentives to shift power consumption due to the local control. However, such temporal flexibility is generally desirable for electrical consumers.This is also becoming increasingly important in the context of the energy transition, as the stabilisation of the electricity grid and the temporal adjustment of consumption profiles to the availability of the electrical power provided (preferably from renewable energy sources) are becoming increasingly important with an increasing number of decentralised producers of renewable electrical energy.

[0005] The object of the invention is therefore to provide an improved method for operating a temperature control system that overcomes this disadvantage. In particular, an operating method is to be provided that allows separate temperature control of individual building zones and simultaneously enables temporal flexibility in the associated power consumption. A further object is to provide a corresponding temperature control system for a building. These objects are achieved by the operating method described in claim 1 and the temperature control system described in claim 15.

[0006] The method according to the invention is used to operate a temperature control system for a building. The temperature control system comprises a plurality of independently controllable temperature control devices.

[0007] The method comprises the following steps: a) Providing a thermal model of the building, wherein the building is divided in the model into a plurality of zones, each of which is assigned at least one temperature control device, wherein the thermal model is suitable for predicting a time-dependent room temperature profile for the respective zone as a function of a time-dependent set of operating parameters of the temperature control system, wherein the building parts belonging to the respective zone are each represented in the thermal model by a separate thermal mass with an effective heat capacity, b) Providing an objective function for an optimization method, wherein the objective function contains at least one optimization criterion which depends directly or indirectly on the temporal profiles of power consumption of the temperature control devices,c) Extremizing the objective function based on computer-aided prediction values ​​derived from the thermal model, thereby obtaining an optimized time-dependent set of operating parameters, and d) operating the temperature control devices according to the optimized time-dependent set of operating parameters. The temperature control system serves to regulate the temperature of the individual zones of the building according to zone-specific specifications. Either constant temperatures or time-variable temperature profiles (each within certain setpoint limits) can be specified for the respective zone. These specified profiles are also referred to as temperature bands. The temperature control devices can be heating devices, cooling devices, or combined heating and cooling devices. Since heating is generally more important than cooling, at least in the middle latitudes,The temperature control system advantageously comprises at least one heating device for each zone. For independent control, each temperature control device can be assigned at least one dedicated control device. Generally, the term "control" also includes regulation. The individual control devices can be components of the respective assigned temperature control device. In any case, they are advantageously local control devices, each of which is assigned to exactly one temperature control device and located in its area.

[0008] Within the thermal model, the building is divided into individual, separately temperature-controlled zones. These zones can, in particular, be individual rooms and / or groups of rooms. It is also possible that some zones correspond only to parts of rooms. However, the air circulation between the individual zones is advantageously weaker than within the zones, so that at least a partial thermal separation of the zones is achieved and separate temperature control according to zone-specific temperature ranges is possible. At least one temperature control device is assigned to each of the individual zones of the model. This "assignment" results from the fact that the temperature control device is located in the area of ​​the respective zone in the real building and is thus suitable for temperature-regulating the room air of the zone according to the corresponding specifications.However, the aforementioned "division" of the building into individually temperature-controlled zones should not generally exclude the possibility that, in addition to these zones, the building may contain additional passive areas (e.g., attics, basements, or storage rooms) in which no temperature control devices are installed. The only important factor is that there are several separately temperature-controlled zones, which are individually considered in the model.

[0009] The thermal model is suitable for predicting the success of temperature control for the individual zones. In other words, time-dependent temperature curves for the individual zones can be predicted depending on the operating parameters of the individual temperature control devices. The predicted temperature curves can depend on numerous parameters, as described below. At the very least, in addition to the operating parameters of the temperature control systems, they also depend on the heat capacities of the associated building components of the respective zones. These are represented in the model as separate heat capacities for each of the zones under consideration. In particular, the building components of each zone can be represented by a single effective value of the associated heat capacity in order to keep the modeling effort as low as possible. Alternatively, multiple heat capacities for the respective zone can be included in the model.The only essential requirement is that each zone is included in the thermal model with at least one heat capacity for the associated building components. The building components belonging to the respective zone form a thermal mass, which determines the thermal inertia of the respective zone. The information about this thermal inertia is used within the scope of the invention to determine possible temporal flexibility in the operation of the individual temperature control devices, broken down by zone. The term "building components" refers not only to the wall areas, doors, windows, floor and ceiling areas belonging to the zone, but also to the other fixed components (e.g., furniture, built-in appliances, books) that are typically not moved significantly over the course of the day or week and have a significant influence on the thermal mass of the zone.

[0010] The available temporal flexibility is utilized based on an optimization of the objective function provided in step b). This objective function can have one or more optimization criteria, in other words, it can be a function of these criteria. The objective function can therefore have multiple terms (e.g., summands), each representing an optimization criterion. In the individual terms, the optimization criteria can, if necessary, be weighted with weighting factors to enable prioritization of the respective criteria. At least one of the existing optimization criteria should depend on the temporal progression of the power consumption of the individual temperature control devices and, in particular, should be influenceable by temporal shifts with constant total power consumption. This dependency can be direct or indirect.In other words, the criterion is not the temporal progression itself, but a variable that can be influenced by temporal shifts in power consumption. This variable can, for example, be related to the temporal stabilization of a power grid from which electrical power for the temperature control devices is drawn. It can also be related to the adjustment of power consumption to the temporal availability of renewable energy. More generally, the optimization criterion can be a variable that reflects the best possible match between the time-dependent power consumption of the energy system and an at least approximately available time-dependent target power profile.

[0011] The provided objective function is, in particular, a "superordinate objective function" in the sense that it depends on the operating parameters of several (and in particular, all of the temperature control devices present in the building). By extreme-calculating these parameters in step c), a superordinate operating parameter set is obtained, which contains individual sub-operating parameter sets for the respective temperature control devices. The sub-operating parameter sets are time-dependent and, in particular, contain time-dependent curves for the power consumption of the individual temperature control devices. In step d), the individual temperature control devices are then operated with the time-dependent sub-operating parameter sets optimized in this way.

[0012] The core idea of ​​the present invention is to exploit the temporal flexibility caused by the thermal inertia of the individual zones in order to achieve a related optimization goal, such as a reduction in peak loads and / or an increase in the proportion of renewable energy generated. To this end, an optimization of an objective function is carried out at a higher level, which depends on the operating parameters of several temperature control devices. In other words, the optimization to exploit the temporal flexibility in power consumption is not carried out individually for each temperature control device, but at a higher level for the entire building. The result of this higher-level optimization can then be transmitted to the control devices of the individual temperature control devices and implemented locally in the device control system.By separately controlling the individual temperature control devices, the individual zones of the building can be temperature-controlled according to their local requirements. The local requirements can be incorporated into the higher-level optimization, for example as boundary conditions. In order to be able to precisely predict the local temperature profiles of the individual zones, it is important that the thermal masses of the individual zones are represented as separate heat capacities in the thermal model. When optimizing the operating parameters based on the thermal model, the existing temporal flexibility due to the different thermal inertias is utilized. The optimization takes place at a higher level and leads to an overall optimum. This has the advantage that the optimization can be carried out in just one step and is therefore more resource-efficient than if a separate optimization were carried out for each zone.The optimization can, for example, be carried out in a central processing unit. Furthermore, the joint optimization of all operating parameters by extremalizing a higher-level objective function leads to the utilization of synergies between the operating parameters of the individual zones. This is because, for example, desired temporal shifts can be implemented where the greatest flexibility exists within the existing boundary conditions and depending on the different thermal inertias. In this way, the higher-level optimization can, in particular, find a better trade-off between the highest possible overall energy efficiency and the best possible utilization of the available temporal flexibility. The temperature control system according to the invention is used to control the temperature of a building. It comprises a plurality of independently controllable temperature control devices.Furthermore, it comprises a central control unit designed to carry out the method according to the invention. The advantages of the temperature control system according to the invention arise analogously to the advantages of the method according to the invention described above.

[0013] Advantageous embodiments and further developments of the invention emerge from the claims dependent on claim 1 and the following description. The described embodiments of the method can also be implemented in the temperature control system, and vice versa.

[0014] Thus, the extremalization in step c) can generally advantageously be carried out under the boundary condition that a specified time-dependent room temperature profile is maintained for the respective zone, in particular within specified limits. In other words, a temperature range to be maintained can be specified for each of the zones. This ensures that the user requirements regarding the temporal temperature profiles are met for each zone. However, the permitted ranges can advantageously be chosen to be comparatively broad in order to enable greater exploitation of possible temporal shifts in power consumption. In this way, comparatively large deviations from the usual room temperature can be permitted, particularly for times when the respective zones are not in use (e.g. offices outside of normal working hours).In addition to this boundary condition for the target temperature control, further boundary conditions can optionally be taken into account during the extremalization, such as power limits for the operation of the temperature control devices.

[0015] According to a further generally advantageous embodiment, the temperature control system can comprise an associated control unit for each of the temperature control devices. Furthermore, it can comprise a higher-level control system for controlling the individual control devices. The optimized time-dependent overall operating parameter set, which is determined during the extremalization, can then in particular comprise a plurality of time-dependent partial operating parameter sets for the respective temperature control devices. These can be transmitted from the higher-level control system to the associated control units of the individual temperature control devices. The higher-level control system can also have a higher-level computing unit in which the optimized parameters are determined centrally. The partial operating parameter sets therefore comprise setpoints or temporal setpoint trajectories for the individual temperature control devices.The variables defined therein can, for example, be temporal trajectories for power consumption. In this case, implementing the trajectories in the individual control units is comparatively easy, since local control is no longer required. This design is particularly advantageous in connection with direct electric heating devices, where the heat output is essentially linear to the electrical power consumption. In a hydronic heating system, the transmitted setpoints for the individual devices can be specified, for example, by (possibly time-varying) target temperatures for local radiator thermostats.

[0016] Generally advantageously, the objective function can contain a first optimization criterion that represents energy efficiency in achieving specified room temperature profiles for the respective zones. This means that this energy efficiency can either appear as a term in the objective function itself, or it can contain a term that is directly related to such energy efficiency. To achieve the goal of exploiting temporal flexibility, the objective function in this case expediently has at least one additional optimization criterion that depends directly or indirectly on the temporal profiles (and thus also on temporal shifts) of the power consumption of the temperature control devices.In other words, by considering at least two decoupled optimization criteria in the objective function, a favorable trade-off between energy-efficient operation and additional use of temporal flexibility can be found. Prioritizing these potentially competing objectives can be achieved by appropriately weighting the individual optimization criteria in the objective function, for example, using weighting factors or another type of weighting function.

[0017] Various parameters can be considered for the optimization criterion that reflects the utilization of temporal flexibility. For example, this optimization criterion can generally be particularly advantageous in reflecting the correspondence between the time-dependent power consumption of the energy system and a specified and / or at least approximately known target power profile.

[0018] A predefined time-dependent target power profile can be present, for example, if the optimization is controlled by a so-called aggregator (operator of a virtual power plant). In other words, the aggregator can then specify as an optimization criterion which time behavior of the power consumption should be achieved during operation of the temperature control system. The corresponding optimization criterion contained in the objective function can then be a match (to be maximized) between the actual time-dependent power consumption and this predefined ideal time behavior.

[0019] As an alternative to such a complete target curve, the time-dependent target power profile can also be known only approximately and / or in sections. For example, the desired target power profile can be calculated and / or estimated for specific time periods. It is also possible for this information to be presented in the form of details about optimization directions for specific time periods. For example, the specification can include that the power consumption should be as high as possible in a specific time period and / or as low as possible in another time period. Or, so-called target power bands can be specified, i.e. time-dependent power ranges (with upper and lower limit curves) within which the power consumption should move. According to a further alternative, the information about the performance profile to be achieved can also be provided generally via an incentive.This can in particular be a financial incentive which creates an incentive for particularly low power consumption (e.g. through a high purchase price for supplied electrical energy) or for particularly high power consumption (e.g. through a correspondingly low purchase price) in a specific period of time. Therefore, the optimization criterion associated with the utilization of temporal flexibility can also include and / or depend on a financial variable. Such a financial incentive, e.g. in the form of time-dependent stock exchange prices, can contribute to stabilizing a power grid and / or increasing the proportion of renewable energy generated, and thus to reducing climate-damaging emissions, particularly through the resulting utilization of existing temporal flexibility in the operation of the temperature control system.“Generation” is generally understood here as the conversion of one form of energy into another.

[0020] Accordingly, the building can be connected to a public power grid, and the optimization criterion can represent or include the provision of so-called ancillary services for the power grid. Such ancillary services include, for example, a contribution to grid frequency and / or voltage control. Such a contribution can be achieved through targeted power consumption according to a time specification, for example, through a short-term reduction in power consumption to compensate for another consumption peak, or through a short-term increase in power consumption to compensate for a supply peak (or vice versa). According to a further advantageous embodiment, the building can also be connected to a public power grid, and the optimization criterion can include a purchase price for the purchase of electrical energy required to operate the temperature control system.In particular, the total price for purchasing required energy can be minimized. If the price for traded electrical energy quantities on an electricity exchange varies over time (e.g., depending on the temporal course of solar radiation and / or wind), this can be an incentive for maximizing the share of renewable electricity generated in the total energy consumption within the power grid (electricity mix). In other words, in this embodiment, the optimization criterion can also reflect or include a reduction in climate-damaging emissions during the operation of the power grid as a whole.

[0021] Similarly, in a further advantageous embodiment, the building can be connected to a local generator of electrical energy. For example, the building can comprise a photovoltaic system and / or a wind turbine or be connected to such a system. In a first variant of this embodiment, the optimization criterion can then comprise self-consumption of locally generated electrical energy. This self-consumption can, in particular, be maximized. Alternatively or additionally, the optimization criterion can comprise a price for the sale of locally generated electrical energy or be linked to such a price. Greenhouse gas emissions from the operation of the entire power grid can also be minimized by offering and trading locally generated renewable energy accordingly. A time-varying price can, in turn, be a suitable incentive for measures that achieve this.The use of existing temporal flexibility in the operation of a temperature control system represents such a measure.

[0022] According to a generally advantageous embodiment, the thermal model can be based on a balanced thermal energy balance for the thermal mass of the respective zone. In other words, such a balance can be formulated as a boundary condition, which is taken into account during the extremalization in step c). The balanced thermal energy balance reflects that the law of conservation of energy applies overall to the transfer and / or storage of thermal energy and that, therefore, the change in the thermal energy content of the thermal mass is equal to the difference between the thermal energy introduced into this thermal mass and the thermal energy released by it (i.e., the difference between heat input and heat release).Such an energy balance is advantageously established for a plurality of thermal zones of the model, and in particular for each separately temperature-controlled zone within the building, resulting in a corresponding number of such boundary conditions. From such an energy balance, a temporally discretized analysis by dividing by the length of the considered time step yields a corresponding power balance. In general, such conditions can be formulated either as a function of energy values ​​or power values, so that the descriptions of thermal energy and thermal power can be considered equivalent in this sense.

[0023] In the thermal energy balance, the following contributions can be taken into account, particularly for the thermal mass of the respective zone:

[0024] - Heat transfers to the corresponding thermal masses of neighboring zones,

[0025] - heat transfer by conduction to the building's surroundings,

[0026] - heat transfer to the indoor air of the respective zone,

[0027] - a proportional heat input by at least one associated temperature control device,

[0028] - a proportional heat input through thermal radiation from outside,

[0029] - a proportional heat input from other heat sources within the respective zone and

[0030] - a change in the thermal energy contained in the thermal mass.

[0031] The boundary condition formulated in the form of a power balance can thus be calculated for a given zone with the corresponding effective heat capacity C m the thermal mass as follows:

[0032] The index z runs over selected zones adjacent to the respective zone under consideration. These can, for example, be certain zones directly adjacent to the respective zone in the building, such as neighboring rooms on the same floor and / or rooms located directly below or above on neighboring floors. The term 2z[^tr,z • ßm ~ 0 z)] represents the heat transfer to the corresponding thermal masses of neighboring zones. 6 m the temperature of the respective zone under consideration, and 6 Z is the temperature of a neighboring zone z. H tr zis the corresponding heat transfer coefficient. This summation can also include the corresponding heat transfer to the outside air in the building's surroundings. In this case, the index z in the summation also includes the building's surroundings as a separate zone (even if it is not part of the building). Similarly, a represents the temperature of the indoor air in the respective zone under consideration, and H ma stands for the heat transfer coefficient for heat transfer between the thermal mass of the respective zone and its interior air. Q heat represents the heat output of the temperature control devices assigned to the respective zone in question. h indicates the relative proportion of the total released heat output that is transferred into the thermal mass of this zone. Similarly, Q sotfor the thermal output of the thermal radiation incident on the building from outside (especially solar radiation), and ^ m , soi represents the relative proportion of this energy that is incorporated into the thermal mass. Q int represents the heat output of other heat sources within the respective zone, and ^m,int ste ht for the corresponding relative proportion that is introduced into the thermal mass under consideration. These additional heat sources include, for example, people in the room and other consumers such as electrical equipment. C m ■ represents the change in thermal energy contained in the thermal mass (here expressed as a change in power). The partial derivative of the corresponding temperature 6 mThis thermal mass after time t can be advantageously represented in the model by a discretized approach, i.e., by decomposing it into individual time steps. In the energy or power balance formulated above, the sign of the individual terms depends on the transfer direction. Heat inputs into the thermal mass and heat outputs from it are therefore included in the balance with opposite signs. Thus, the heat input $ m h ■ Q heat by the associated temperature control device, for example, be positive or negative, depending on whether this device is operated as a heating device or as a cooling device.

[0033] According to a further generally advantageous embodiment, the thermal model can be based on a balanced thermal energy balance for the indoor air of the respective zone. In other words, such a balance can be formulated as a (particularly further) boundary condition, which is taken into account during the extremalization in step c). Here, too, the balanced thermal energy balance reflects the law of conservation of energy. Therefore, the change in the thermal energy content of the indoor air of the respective zone is set equal to the difference between the thermal energy introduced into the indoor air and the thermal energy released by it. This energy balance is also advantageously established for a plurality of thermal zones of the model and in particular for each thermal zone within the building, resulting in a corresponding number of such boundary conditions.In contrast to the energy balance of the corresponding thermal mass described above, the heat capacity of the air can typically be neglected when considering the energy balance of indoor air. Thus, the corresponding balance can be formulated as an equation of heat input and heat output (relative to the indoor air). The boundary condition, formulated as a power balance, can thus be as follows for the indoor air of a given zone: Where 0 e the temperature of the ambient air outside the building (outside air), and H ve is the corresponding coefficient for the heat transfer between the indoor air and the outdoor air, which takes place in particular through air exchange (ventilation). a h , ^ a soi and a int in turn indicate the relative shares of the corresponding heat outputs Q heat , Q soi and Q intinto the indoor air of the respective zone under consideration. By neglecting the heat capacity of the indoor air, the corresponding set of boundary conditions can be handled even more easily than the first set of boundary conditions for thermal mass described above. However, it should not be ruled out that the heat capacity of the indoor air (and thus the change in the thermal energy contained therein) is also taken into account as a term in the energy or power balances for the indoor air.

[0034] The provision of the thermal model in step a) may in particular include the specification of values ​​for one or more of the following variables for the respective zone under consideration:

[0035] - Coefficients (H tr z ) for the heat transfer between the thermal mass of the respective zone and the corresponding thermal masses of neighbouring zones and / or the outside air,

[0036] - a coefficient (H ma ) for the heat transfer from the thermal mass of the respective zone to the indoor air of the respective zone,

[0037] - a relative share for the heat input by at least one associated temperature control device into the thermal mass of the respective zone,

[0038] - a relative proportion (f m , S0 () for the heat input by thermal radiation from outside into the thermal mass of the respective zone,

[0039] - a relative proportion (^ m ,m t ) for the heat input by other heat sources within the respective zone into the thermal mass of the respective zone,

[0040] - an effective heat capacity (C m ) for the thermal mass of the respective zone,

[0041] - a coefficient (H ve ) for heat transfer to the outside air through air exchange,

[0042] - a relative proportion (^ h ) for the heat input by at least one associated temperature control device into the indoor air of the respective zone,

[0043] - a relative proportion (^ soi ) for the heat input by thermal radiation from outside into the indoor air of the respective zone and

[0044] - a relative share for the heat input from other heat sources within the respective zone into the indoor air of the respective zone.

[0045] Overall, the corresponding values ​​can then be specified for a plurality of separately temperature-controlled zones and in particular for each of these zones. In general, they can be determined as a sub-step of step a) or in an additional step of the method according to the invention. The variables listed above are coefficients and relative proportions, i.e. variables which can remain constant at least over certain periods of time during operation of the temperature control system or can change between certain discrete values. Therefore, these variables or the possible discrete values ​​of these variables can be determined once in an initialization step, and these values ​​can be kept fixed during ongoing operation. An example of a variable that is not constant but changes between discrete values ​​is the coefficient for heat transfer through air exchange between the inside air of a zone and the outside air.In the simple case of a room with one window, for example, this coefficient can change between two limit values ​​depending on whether the window is fully open or fully closed. Intermediate values ​​are of course generally also possible with partial opening, and in zones with multiple windows and / or external doors, various combinations can occur. The opening states, from which the corresponding values ​​of the coefficient are derived, are generally easy to determine using appropriate window and door sensors. In an analogous way, such a correlation between the condition of the building and the value of a coefficient or relative component also applies to the other variables mentioned. In this case, different sensors may simply be used to determine the relevant building condition.

[0046] The temperature control system can therefore generally comprise a plurality of sensors, and the above-mentioned values ​​can be determined based on measured values ​​from these sensors. These sensors can, in particular, each include one or more status sensors for doors and / or windows (including interior doors), presence detectors for determining the number of people present in the zones, sensors for the device status of the equipment included, sensors for measuring solar radiation, and sensors for measuring the temperature of the indoor or outdoor air.

[0047] In general, the aforementioned variables can be determined, for example, using an initialization routine at the start of operation of the temperature control system. Above all, the possible values ​​of the coefficients and / or the relative components can then be kept constant, at least over certain periods of operation. Alternatively, continuous or at least occasional adjustment to the respective measured sensor values ​​is also possible. In this way, even medium-term changes in the thermal boundary conditions, such as those caused by the moving of furniture, can be captured in the thermal model. The aforementioned values ​​can be determined, for example, using a predefined algorithm or using an artificial intelligence method (e.g. using an artificial neural network trained using machine learning).In addition to the coefficients and relative components listed above, other variables whose values ​​fluctuate more significantly over time or whose values ​​are less predictable can also play a role in determining the values ​​of the objective function. One such "flexible variable" is solar radiation, for example, which can also be easily recorded using corresponding sensor measurements. Other relevant time-dependent variables are the target temperature ranges described above for the individual zones, which can be dynamically changed at short notice by the building's occupants if necessary, as well as the time-dependent target power specifications described above, as well as, if applicable, prices for the purchase or sale of electrical energy. Time-dependent input variables can be transmitted to the model for all of these values, which are then taken into account during the extremalization process. Short-term updates to these flexible variables are also possible in the model if necessary.

[0048] It is generally advantageous to implement a temporal discretization within the thermal model when modeling the time dependence of the predicted variables. In other words, the determination of the predicted values ​​can be based on a temporal discretization of the temporal behavior of the heat transfer and / or the thermal energy content. This is particularly useful for modeling the partial derivative that occurs in the term for the change in the thermal energy contained in the thermal mass. Since the remaining terms describe only simple, linear dependencies, the modeling and thus also the extremalization of the resulting objective function is comparatively simple in terms of computational effort.

[0049] According to a generally advantageous embodiment, at least a subset of the temperature control devices present in the building can be electrical temperature control devices, i.e. in particular electrical heating, cooling, or combined heating and cooling devices. In general, an electrical temperature control device is understood to be a device whose primary energy source for temperature control is electrical energy. The heating or cooling energy used should therefore come at least predominantly from electrical energy. This is in contrast to fuel-based temperature control systems, in which electrical energy is only used to operate electrical control units. Particularly advantageously, all existing temperature control devices can even be designed as electrical temperature control devices.In general, the use of electrical temperature control devices is particularly advantageous within the scope of the invention because they result in a comparatively high electrical energy requirement, so that the utilization of existing temporal flexibility represents a major lever overall. It is particularly preferred if at least a subset of the temperature control devices is provided by electrical direct heating devices and / or electrical direct cooling devices. These are temperature control devices in which the conversion of electrical energy into thermal energy takes place locally in the area of ​​the respective heating device. This is in contrast to the operation of central heat pumps, in which electrical energy is also the essential energy source (at least to be used by the operator), but in which the transfer to the local temperature control devices in the individual zones takes place via a water circuit.A particular advantage of electrical direct temperature control devices is that the amount of heat introduced into or removed from the zone is essentially proportional to the amount of electrical energy converted. This results in an essentially linear relationship, which significantly facilitates the description of the temporal developments by the thermal model and the resulting extremalization of the objective function under the boundary condition of existing temperature specifications. In principle, however, the advantages of the invention also apply to a hydronic temperature control system, e.g. based on a central heat pump or another electrically operated central heating or cooling source. Here, adjustable thermostats can be used for the separate temperature control of the individual zones, which are controlled based on the operating parameter sets according to the invention.

[0050] The invention will now be described by way of some preferred embodiments with reference to the attached drawings, in which:

[0051] Figure 1 shows a schematic representation of a building,

[0052] Figure 2 shows a schematic representation of the heat transfer paths described in the thermal model,

[0053] Figure 3 shows a schematic flow diagram of an operating procedure and Figure 4 shows an expanded representation of a building.

[0054] In the figures, identical or functionally identical elements are provided with the same reference symbols.

[0055] Figure 1 shows a schematic overview sketch of a building 1 with a temperature control system 5. The building 1 has a local generator of electrical energy, in this example a photovoltaic system 2. It is also connected to the public power grid 3. Electrical energy is therefore available from different sources in order to supply electrical equipment in the building 1. In this example, the building 1 has two separately temperature-controlled zones 10 and 20, which can correspond to individual rooms or groups of rooms or separate room areas. The first zone 10 is assigned its own temperature control device 11, and the second zone 20 is also assigned its own temperature control device 21. In the example shown, these are heating devices. The presence of additional optional electrical equipment 12 is indicated for the first zone by the socket symbol only as an example.In addition to these separately temperature-controlled zones, building 1 can also have one or more passive areas 8, as symbolized here by the roof structure. Such a passive area 8 does not have an associated temperature control device. Therefore, its internal temperature is adjusted depending on the temperature of the environment 7 and the adjacent temperature-controlled zones 10 and 20.

[0056] Within the scope of the present invention, operating parameters for such or a similar temperature control system 5 are determined on the basis of a thermal model. Figure 2 shows a schematic representation of relevant heat transfer paths T1 to T11, which can be taken into account in such a thermal model. To simplify the model, however, some of the heat transfer paths shown can be neglected, or optionally, additional ones can be taken into account. Figure 2 shows possible heat transfer paths for a selected separately temperature-controlled zone 10. For this zone 10, the thermal mass m of the associated building parts and the interior air a (room air) of the zone are modeled as separate components, each with associated temperatures 6. m or 6 a . These temperatures 6 m or 0 aare each modeled with their time course. In addition, they can optionally also be considered spatially resolved. However, within the scope of the present invention, modeling an effective temperature value for 6 m or e aper magazine considered. Schematically, only one neighboring zone z is shown next to zone 10, and T 1 is the heat transfer between the zone 10 under consideration and this neighboring zone z. In the model, the corresponding heat transfers T1 to all immediately neighboring zones are conveniently considered, which is reflected in the sum in the first term of equation (1). T2 denotes the heat transfer by heat conduction to the environment 7 of the building, primarily through the outer wall 15 of zone 10, but also through existing windows 14 and external doors when closed. T3 denotes the heat transfer between the thermal mass m and the indoor air a of the zone 10 under consideration. T4 denotes the proportional heat input by one or more temperature control devices 11 of the zone 10 under consideration. T5 denotes the proportional heat input by heat radiation from an external radiation source 16, in particular by solar radiation.T6 denotes the proportional heat input from other heat sources 13 within the considered zone 10, for example, from people and other electrical equipment. These heat transfer paths T1 to T6, which directly affect the thermal mass m, appear as summands in the power balance of equation (1), together with the change in thermal energy within the thermal mass m. Such an equation (1) can be established, in particular, within the model for each separately temperature-controlled zone and, if necessary, also in a modified form (without the contribution T4) for optionally present passive areas.

[0057] In addition, the model can take into account the heat transfer to and from the indoor air a. T7 denotes the heat transfer from the indoor air a to the outside air of the environment 7 through air exchange, e.g. when windows and / or doors are open or through the operation of an automatic ventilation system. T3 has already been described. T8 denotes the heat input into the indoor air a by the associated temperature control device 11. T9 denotes the heat input into the indoor air a by thermal radiation. T10 denotes the heat input by other internal heat sources 12. T11 denotes the heat transfer through (direct) heat conduction to the environment 7. This last contribution can typically be neglected. This is reflected, for example, in the fact that only the transfer paths T7, T3, T8, T9 and T10 are represented as corresponding summands in equation (2).

[0058] Figure 3 shows a schematic flow diagram for an operating method of a temperature control system according to an embodiment of the invention. Thus, in step a), a thermal model M is provided. In this thermal model M, in particular, the respective separately temperature-controlled zones 10, 20 are each represented by assigned thermal masses with corresponding effective heat capacities. Furthermore, the model describes in particular at least a part of the heat transfer paths T1 to T11 shown in Figure 2 from / to the thermal mass m and from / to the interior air a of the respective zone. Thus, the thermal model is suitable for determining a time-dependent room temperature profile (i.e., a time behavior for e a ) to predict.

[0059] In step b), an objective function Z is provided, on the basis of which the operating parameters of the temperature control system can be optimized. This objective function can have several objective terms containing several optimization criteria c1, c2, .... These optimization criteria c1, c2, ... can be weighted in the individual objective terms with associated weighting functions w1, w2, for example, by multiplying them by associated weighting factors. The first optimization criterion can, for example, be energy efficiency in achieving specified room temperature profiles or can contain and / or represent such energy efficiency. The second optimization criterion can be independent of this energy efficiency term (i.e., represent a separate dimension) and represent the utilization of temporal flexibility in the operation of the temperature control system.In other words, it can depend directly or indirectly on temporal shifts in the power consumption of the temperature control devices. The second optimization criterion can also react to shifts with constant total energy consumption. This means that the value for the additional optimization criterion changes even if power consumption is only shifted in time without the total energy consumed by the temperature control devices changing. This additional optimization criterion can, for example, represent the provision of system services for the power grid, e.g., a contribution to the temporal stabilization of the grid frequency and / or grid voltage in the power grid. Alternatively or additionally, the further optimization criterion can be an increase in the proportion of renewable electrical energy generated locally in the building.According to a further embodiment, the optimization criterion can be a financial criterion, i.e., the minimization of a purchase price for electrical energy and / or the maximization of a sales price for electrical energy. Time-varying exchange prices can create incentives for increasing the share of renewable energy in the entire power grid. Accordingly, financial optimization in response to such incentives can lead to a reduction in the overall greenhouse gas emissions caused by the operation of the power grid. By appropriately weighting these different optimization criteria c1 and c2, a favorable trade-off can be found between overall energy-efficient operation and the use of available temporal flexibility.

[0060] In step c) of the method, an extremalization of this objective function is carried out, i.e., a minimization or maximization, depending on the sign of the objective terms it contains. The operating parameters of the temperature control system 5 are varied, and the expected values ​​of the objective function Z are calculated using computer support on the basis of the thermal model M. This prediction can be based on temporal discretization, i.e., a division into discrete time steps. Due to the comparatively simple and predominantly linear relationships of the underlying equations (1) and (2), the optimization can be carried out with relatively little computational effort. For example, a mixed-integer optimization method, in particular mixed-integer linear optimization, can be used. This method is well known in the field and is used particularly efficiently to solve such simple optimization problems.The result of the optimization is an overall operating parameter set P for the entire temperature control system 5, which contains partial operating parameter sets for the individual temperature control devices 11, 21 included. In step d), the temperature control system is operated with the higher-level operating parameter set P determined in this way. The sequence of the individual steps a) to d) does not have to be as strictly sequential as shown in Figure 3. In particular, step c) can be carried out repeatedly in order to calculate optimal operating parameter sets P for certain future time periods in advance, which are then implemented for these time periods in step d). Accordingly, steps c) and d) are carried out repeatedly and sometimes even overlapping in time. Step a) can optionally include a determination of the underlying model parameters such as the heat transfer coefficients and relative proportions of the heat inputs.These values ​​can be determined (as a substep of a) or in an additional step) based on sensor data and updated during operation if necessary. Other variables such as solar radiation and actual temperature trends can also be used. a and 6 m can be re-determined continuously or at least occasionally through measurements during operation. These measurements, which are not shown separately here, therefore lead to ongoing adjustments of parameters in the thermal model M.

[0061] Figure 4 shows an expanded view of building 1 in Figure 1. It shows that the temperature control system 5 for implementing the operating method according to the invention can have a central control unit S. This central control unit S can also be referred to as an energy management system (EMS). Steps a) to c) can be carried out within this central control unit S. For this purpose, this control unit can in particular have an integrated computing unit R or at least be connected to such a computing unit R. In order to enable the separate controllability of the individual temperature control devices 11 and 21, these can each be connected to associated control devices s11 and s21, respectively. These associated control devices s11 and s21 can also be integrated into the respective temperature control devices 11 and 21, respectively, or they can be arranged locally in their vicinity.The dashed lines and boxes therefore only outline the hierarchy of the control flow and not the spatial arrangement of the elements s11, s21, and S. The higher-level control unit S can also be located within building 1. This higher-level control unit S can transmit the corresponding optimized partial operating parameter set to each local control unit s11 or s21, which is then implemented in the local control of the corresponding temperature control unit 11 or 21. List of reference symbols.

[0062] 1 building

[0063] 2 photovoltaic systems

[0064] 3 Power grid

[0065] 5 T emperation system

[0066] 7 Surroundings

[0067] 8 passive area

[0068] 10 first temperature-controlled zone

[0069] 11 First zone heater

[0070] 12 first zone equipment

[0071] 13 internal heat source(s)

[0072] 14 windows or exterior doors

[0073] 15 Exterior wall

[0074] 16 Radiation source

[0075] 20 second temperature-controlled zone

[0076] 21 Heating device of the second zone a Room air of the considered zone m Thermal mass of the considered zone

[0077] M thermal model P total operating parameter set T1-T11 heat transfer paths

[0078] R arithmetic unit

[0079] S central control unit s11 ,s21 control units z adjacent zone

[0080] Z objective function

Claims

Patent claims 1. A method for operating a temperature control system (5) for a building (1), wherein the temperature control system (1) comprises a plurality of independently controllable temperature control devices (11, 21), the method comprising the following steps: a) providing a thermal model (M) of the building (1), wherein the building (1) is divided in the model (M) into a plurality of zones (10, 20), each of which is assigned at least one temperature control device (11, 21), wherein the thermal model (M) is suitable for predicting a time-dependent room temperature profile for the respective zone (10, 20) as a function of a time-dependent operating parameter set (P) of the temperature control system (5), wherein the building parts belonging to the respective zone (10, 20) are each represented in the thermal model (M) by at least one separate thermal mass (m) with a heat capacity (C m), b) providing a target function (Z) for an optimization method, wherein the target function (Z) contains at least one optimization criterion (c2) which depends directly or indirectly on temporal profiles of power consumption of the temperature control devices (11, 21), c) extremalizing the target function (Z) on the basis of computer-aided determined prediction values ​​on the basis of the thermal model (M), whereby an optimized time-dependent operating parameter set (P) is obtained, and d) operating the temperature control devices (11, 21) of the temperature control system (5) according to the optimized time-dependent operating parameter set (P).

2. Method according to claim 1, wherein the extremalization in step c) is carried out under the boundary condition that a predetermined time-dependent room temperature profile is maintained within predetermined limits for the respective zone (10, 20).

3. Method according to one of claims 1 or 2, wherein the temperature control system (5) comprises an associated control device (s11, s21) for each temperature control device (11, 21) and further comprises a higher-level control system (S) for controlling these control devices (s11, s21), wherein the optimized time-dependent operating parameter set (P) of the temperature control system comprises a plurality of time-dependent partial operating parameter sets for the individual temperature control devices (11, 21), which are transmitted from the higher-level control system (S) to the associated control devices (s11, s21) of the individual temperature control devices (11, 21).

4. Method according to one of the preceding claims, in which the objective function (Z) contains an optimization criterion (c1) which represents an energy efficiency in achieving predetermined room temperature profiles for the respective zones (10, 20), and in which the objective function (Z) contains at least one additional optimization criterion (c2) which depends directly or indirectly on temporal shifts in power consumption of the temperature control devices (11, 21).

5. Method according to one of the preceding claims, in which the optimization criterion (c2) represents the optimization of a match of the time-dependent power consumption of the energy system (5) with a predetermined and / or at least approximately known target power profile.

6. Method according to one of the preceding claims, in which the building (1) is connected to a power grid (3), wherein the optimization criterion (c2) represents the optimization of a provision of system services for the power grid (3).

7. Method according to one of the preceding claims, in which the building (1) is connected to a power grid (3), wherein the optimization criterion (c2) comprises a minimization of a purchase price for the purchase of required electrical energy.

8. Method according to one of the preceding claims, in which the building (1) is connected to a local generator (2) for electrical energy, wherein the optimization criterion (c2) comprises a maximization of the self-consumption of locally generated electrical energy and / or wherein the optimization criterion (c2) comprises a price for the sale of locally generated electrical energy.

9. Method according to one of the preceding claims, in which, within the thermal model (M), a balance of the thermal energy balance for the thermal mass (m) of the respective zone (10) is assumed, wherein in particular the following contributions are taken into account in this thermal energy balance: - heat transfers (T1) to the corresponding thermal masses (m) of adjacent zones (z), - heat transfer (T2) by conduction to the environment (7) of the building (1), - a heat transfer (T3) to an internal air (a) of the respective zone (10), - a proportional heat input (T4) by the at least one associated temperature control device (11), - a proportional heat input (T5) by thermal radiation from an external radiation source (16), - a proportional heat input (T6) from other heat sources (13) within the respective zone (10) and - a change in the thermal energy contained in the thermal mass (m).

10. Method according to one of the preceding claims, in which, within the thermal model (M), a balance of the thermal energy balance for the indoor air (a) of the respective zone (10) is assumed, wherein in particular the following contributions are taken into account in this thermal energy balance: - heat transfer (T7) to the outside air (7) by air exchange, - a heat transfer (T3) from the thermal mass (m) of the respective zone (10), - a proportional heat input (T8) by the at least one associated temperature control device (11), - a proportional heat input (T9) by thermal radiation from an external radiation source (16), - a proportional heat input (T10) from additional heat sources (13) within the respective zone (10).

11. Method according to one of the preceding claims, wherein the provision of the thermal model (M) includes the specification of values ​​for one or more of the following variables for the respective zone (10): - Coefficients (W tr z )) for the heat transfer between the thermal mass (m) of the respective zone (10) and the corresponding thermal masses of adjacent zones (z) and / or the environment (7), - a coefficient (H ma ) for the heat transfer from the thermal mass (m) of the respective zone (10) to the indoor air (a) of the respective zone (10), - a relative share for the heat input by the at least one associated temperature control device (11) into the thermal mass (m) of the respective zone (10), - a relative proportion (f m , S0() for the heat input by thermal radiation from an external radiation source (16) into the thermal mass (m) of the respective zone (10), - a relative proportion (^ m ,m t ) for the heat input by further heat sources (13) within the respective zone (10) into the thermal mass (m) of the respective zone (10), - an effective heat capacity (C m ) for the thermal mass (m) of the respective zone (10), - a coefficient (H ve ) for heat transfer to the outside air through air exchange, - a relative proportion (^ h ) for the heat input by the at least one associated temperature control device (11) into the internal air (a) of the respective zone (10), - a relative proportion (^ soi ) for the heat input by thermal radiation from an external radiation source (16) into the interior air (a) of the respective zone (10) and - a relative share for the heat input by further heat sources (13) within the respective zone (10) into the interior air (a) of the respective zone (10).

12. The method according to claim 11, wherein the temperature control system (5) has a plurality of sensors and wherein the values ​​for said variables are specified on the basis of measured values ​​of these sensors.

13. Method according to one of the preceding claims, in which the determination of the predicted values ​​is based on a temporal discretization of the temporal behavior of the heat transfers.

14. Method according to one of the preceding claims, in which at least a subset of the temperature control devices (11, 21) are electrical temperature control devices, in particular electrical direct heating devices.

15. Temperature control system (5) for a building (1), which comprises a plurality of independently controllable temperature control devices (11, 21), and which comprises a central control unit (S) designed to carry out the method according to one of the preceding claims.

Images