Method for operating a flow generation unit for at least one fan and control system

Abstract

The invention relates to a method (V1, V2) for operating a flow generation unit (11) and to a control system (10) comprising at least one flow generation unit (11). The flow generation unit (11) is, in particular, a system permanently installed in a building (12). The flow generation unit (11) has a fan controller (14) that controls a fan (13) to generate an air flow (L). The fan controller (14) is provided with a control model (CM) that specifies a relationship between a room air parameter (R), an ambient air parameter (U), and a fan operating parameter (B) for the fan (13).By measuring a room air value (Rm) for the room air parameter (R) and measuring an ambient air value (Um) for the ambient air parameter (U), a suitable fan operating parameter (B) for controlling the fan (13) can then be selected and set based on the control model (CM). The fan operating parameter (B) is selected such that the electrical energy required to operate the fan (13) is minimal.

Description

[0001] The invention relates to a method for operating a flow generation unit. The flow generation unit is configured to control at least one fan.

[0002] WO 2022 / 094281 A1 discloses a system and method for configuring a building management system using a cloud management platform. The cloud management platform is configured to generate digital data of physical spaces, equipment, people, or events, for example, entity graphs. Such entity graphs depict relationships between the individual physical elements. They can be optimized using machine learning to improve the building management system.

[0003] Based on the known system and method, it can be considered an object of the present invention to provide a method and a control system which is designed to control a room air parameter in a room of a building and which allows high efficiency.

[0004] The object is achieved by a method according to patent claim 1 and a control system according to patent claim 15.

[0005] The flow generation unit has a fan controller and at least one fan. The at least one fan is controlled by the fan controller. The flow generation unit is, in particular, a non-mobile system, for example, a ventilation system, a heating system, a cooling system, or an air conditioning system in a building or part of a building.

[0006] To control the at least one fan, the fan controller uses a control model that specifies a relationship between at least one room air parameter in a room of a building, optionally at least one ambient air parameter in the surroundings of the building, and at least one fan operating parameter. The control model specifies, in particular, the influence of the at least one fan operating parameter on the at least one room air parameter.

[0007] The control model is generated based on learning data by a modeling device configured for machine learning. The learning data provided to the modeling device includes at least the at least one room air parameter and the at least one fan operating parameter. Additionally, the learning data may also include the at least one ambient air parameter and / or other parameters.

[0008] The learning data contains historical data that indicate the at least one room air parameter—optionally also the at least one ambient air parameter—and the at least one fan operating parameter at a respective historical point in time. These can be continuous temporal profiles or discrete-time groups or tuples of the parameters contained in the learning data or in each learning data set at a respective historical point in time. In one embodiment, the learning data can consist exclusively of historical data.

[0009] To generate the control model, the modeling facility can, for example, proceed as follows: 1. Using the learning data, the relationship between the at least one fan operating parameter and the at least one room air parameter can be mapped in the control model. The at least one ambient air parameter can preferably also be taken into account in the control model. 2. If the learning data does not contain any fan operating parameters, for example, it only contains the at least one room air parameter and the at least one ambient air parameter, a parameter model, for example a temperature model, can first be generated in the modeling device. For example, the temperature model can describe the relationship between the at least one room air parameter and the at least one ambient air parameter. Based on this parameter model, the modeling device can perform a simulation with different fan operating parameters in order to determine the influence of the at least one fan operating parameter on the at least one room air parameter. Based on the simulation result, the control model for the fan control can then be generated.

[0010] In both cases, the modeling device can optimize the control model, for example, based on a cost function or another known optimization function or optimization algorithm. The control of the at least one fan operating parameter can be optimized such that a room air parameter—which represents a parameter to be controlled or controlled—remains within a predetermined tolerance range or parameter value range in a future time interval (prediction interval), and the energy required for fan operation within this prediction interval is minimal.

[0011] At least one current room air measurement value, at least for the room air parameter used as the controlled parameter, is provided to the fan controller. Furthermore, at least one current ambient air measurement value for the ambient air parameter or for one or more of the ambient air parameters can be provided to the fan controller. In particular, one, several, or all of the parameters used in the learning data and / or in the simulation and / or optimization can be measured and provided to the fan controller as a measured value.

[0012] In the fan control system, the control model is then used to control the at least one fan of the flow generation unit. For this purpose, as explained, at least one of the room air parameters is used as the parameter to be controlled or as the controlled parameter. In particular, the at least one controlled parameter includes the room air temperature. The at least one controlled parameter can be controlled in an open control loop or regulated in a closed control loop. In the case of a closed control loop, the controlled parameter represents the controlled variable.

[0013] Based on the at least one current room air measurement value and preferably also the at least one current ambient air measurement value, the fan operating parameter results from the control model or several or all adjustable fan operating parameters result from the control model such that the controlled parameter (e.g. room air temperature) remains within the parameter value range in the preceding forecast interval and the energy expenditure required for the operation of the fan within this time interval is minimal.

[0014] In other words, the control of the controlled parameter is carried out under the boundary condition that the fan power is as small as possible at any time or for a time interval in order to achieve the control objective, namely to keep the controlled parameter within the specified parameter value range.

[0015] According to the invention, the modeling device is configured for machine learning. For this purpose, it may comprise and / or use known machine learning components and / or known artificial intelligence components (AI components). The term "components" refers here to devices and / or methods and / or procedures. The components may therefore be hardware and / or software.

[0016] Machine learning can be any known form of machine learning, in particular supervised machine learning, unsupervised machine learning, reinforcement machine learning, etc. Within the framework of machine learning, for example, methods for pattern recognition, pattern analysis or pattern prediction can be used.

[0017] An AI component can be any known implementation of artificial intelligence (AI), such as an artificial neural network (ANN) or a support vector machine (SVM).

[0018] Optionally, in addition to the modeling device, the fan control can also be set up for machine learning.

[0019] According to the invention, it is sufficient to create the control model once and make it available to the fan controller. The fan controller can then operate autonomously, independently of the modeling device. Optionally, the control model used by the fan controller can be updated, for example, at regular intervals or when an update condition is met. An update condition can be met, for example, if continued machine learning in the modeling device results in a more current control model that differs from the control model currently used in the fan controller in such a way that a predetermined condition is met. Such a predetermined condition can, for example, be a predetermined reduction in the electrical energy required to achieve the control objective.

[0020] Using the method according to the invention, the fan control can be implemented using simple standard components. It can, but does not have to, be configured for machine learning. For example, the fan control can have a conventional microprocessor and optionally a conventional data storage (RAM and / or non-volatile memory). The more complex calculations for determining the control model can, for example, be performed independently of the fan control in the modeling device. If the fan control has sufficient computing and / or storage capacity, it can also be configured for machine learning, for example to determine the control model.

[0021] The modeling device can be located remotely from the fan controller. For example, the modeling device can be an internet service (cloud service) that provides the necessary computing power. If a communication connection exists between the fan controller and such an internet service, the control model can be transmitted to the fan controller by the internet service and optionally updated, for example, via an internet connection. During operation of the flow generation unit, the fan controller can operate independently of the modeling device and control or regulate the at least one fan.

[0022] A communication connection between the modeling device and the fan controller may be any wireless and / or wired communication connection, preferably an Internet connection.

[0023] A room air parameter can be one of the following parameters, or several parameters can be used in any combination: a room air temperature, a room air humidity, a room air pressure, or a room air component that describes a proportion or amount of a gas component of the room air, such as a carbon dioxide content. One or more of these room air parameters can be used as a controlled parameter (e.g., as a control variable).

[0024] The ambient air parameter can be one of the following parameters or several parameters can be used in any combination: an ambient air temperature, an ambient air humidity, an ambient air pressure or an ambient air component that describes a proportion or amount of a gas component of the ambient air.

[0025] The at least one fan operating parameter can be one of the following parameters, or several of the following parameters can be used in any combination: a fan speed, a fan torque, a motor voltage of an electric motor of the fan, a motor current of an electric motor of the fan, or a mechanical and / or electrical power of the fan. For example, to influence the controlled parameter, the speed of the fan can be controlled or regulated. For this purpose, for example, an electrical variable of the electric motor of the fan that influences the speed is controlled or regulated, such as the motor voltage.

[0026] The learning data may additionally contain at least one further state parameter and / or at least one geographical parameter, for example one or more of the following parameters: - weather data, such as the amount and / or type of precipitation and / or the degree of cloud cover and / or the intensity of solar radiation; - a state parameter indicating the duration of the day and / or night. - Building data of a room or part of a building to which the flow generation device is assigned, such as the size of the at least one room and / or the size of one or more of the existing windows and / or the orientation of at least one existing window; - the geographical position of the building or part of the building to which the flow generation unit is assigned, for example a geographical longitude, a geographical latitude, an altitude above sea level;

[0027] In one embodiment, the at least one fan of the flow generation unit is controlled such that the fan speed is as low as possible in order to keep the at least one room air parameter within the specified parameter value range at least in a preceding time interval. For example, the fan speed can be selected as low as possible so that a room air temperature remains within a specified temperature value range for the preceding time interval (forecast interval). The speed to be set results from the control model, to which the at least one current room air measured value and preferably also the at least one current ambient air measured value are provided as input variables.

[0028] The prediction interval can be in the range of seconds (e.g. up to 60 seconds) or minutes (e.g. at least 5 minutes or at least 10 minutes or at least 15 minutes and for example a maximum of 15 minutes or 30 minutes or 45 minutes) or this time interval can also cover one or more hours (e.g. 1 to 3 or 4 hours).

[0029] It is also advantageous if the control model depends on the fan type. The learning data can therefore, in particular, be from the same or comparable fan types. Additionally or alternatively, the learning data can contain data that characterizes the respective fan type, such as a fan characteristic curve or characteristic data that describe the generated air flow (e.g., pressure and / or flow velocity and / or volumetric flow and / or mass flow) depending on one or more fan operating parameters that can be adjusted by the fan control.

[0030] The control system according to the invention comprises the modeling device and the flow generation unit with the at least one fan and the fan controller, as described above. The control system can be configured to carry out any embodiment of the method explained above.

[0031] The flow generation unit also has at least one room sensor configured to determine a current room air measurement value, for example, for the or each room air parameter used as a controlled parameter. The flow generation unit optionally also has at least one environmental sensor configured to determine at least one current ambient air measurement value for the ambient air parameter or for several of the ambient air parameters. As explained, the fan controller can use the at least one room air measurement value, preferably additionally the at least one current ambient air measurement value, and the control model to control the at least one fan of the flow generation unit.

[0032] Advantageous embodiments of the invention will become apparent from the dependent claims, the description, and the drawings. Preferred embodiments of the invention are explained in detail below with reference to the accompanying drawings. The drawings show: Fig. 1 a block diagram of an embodiment of a control system comprising a flow generation unit with a fan control and at least one fan, Fig. 2 and Fig. 3 each show a flow diagram of an embodiment of a method for operating the flow generation unit from Fig. 1, Fig. 4 an exemplary representation of a real and a predicted course of a room air parameter based on learning data, for example a room air temperature, Fig. 5 an exemplary time-dependent course of a room air parameter, for example the room air temperature, depending on the setting of a fan operating parameter for controlling a fan of a flow generation unit and Fig. 6 an exemplary time-dependent course of a room air parameter and a fan operating parameter for controlling the fan of the flow generation unit based on a determined control model.

[0033] In Fig. 1 shows an embodiment of a control system 10 in the form of a simplified block diagram. The control system 10 has a flow generation unit 11. In a modification of the illustrated embodiment, a control system 10 can also have multiple flow generation units 11.

[0034] In the exemplary embodiment, the flow generation unit 11 is a non-mobile flow generation unit and, for example, is a component of a system installed in a building 12, such as a heating system, a cooling system, a ventilation system, or an air conditioning system. The flow generation unit 11 is configured to generate a fluid flow. In the exemplary embodiment, the flow generation unit 11 has at least one fan 13 for generating an air flow L. In addition to or alternatively to the at least one fan 13, another flow generation unit for generating a fluid flow may also be present, for example a pump for generating a liquid flow.

[0035] The flow generation unit 11 also has a controller, in the present case a fan controller 14 for controlling the at least one fan 13, in particular for controlling an electric motor 15 of the fan 13. The fan controller 14 can set at least one fan operating parameter B, for example a motor voltage for the electric motor 15 and / or a motor current for the electric motor 15. By means of the at least one set fan operating parameter B, at least one further fan operating parameter B can be influenced, such as a speed of the fan 13 and / or a mechanical or electrical power and / or a torque of the fan 13.

[0036] An operating measurement value Bm for at least one fan operating parameter B can be detected by means of an operating sensor 16, such as a rotational speed n of the fan 13. The operating sensor 16 is optional. A closed control loop can be implemented for the fan 13 by means of the operating sensor 16. The fan controller 14 is configured to directly or indirectly control or regulate at least one fan operating parameter B. For example, a rotational speed n of the fan 13 to be controlled or regulated can be set indirectly via a motor voltage of the electric motor 15.

[0037] The flow generation unit 11 also has at least one room sensor 17, which is arranged in a room of the building 12 and detects a room air measurement value Rm of a room air parameter R there and provides it to the fan control 14. For example, a room air temperature RT, a room air humidity, a room air pressure, or a room air component, or any combination of the aforementioned room air parameters R, can be used as the room air parameter R. The room air component can indicate a proportion or amount of a gas component of the room air, such as a carbon dioxide proportion.

[0038] According to the example, at least one environmental sensor 18 is arranged outside the building 12. The environmental sensor 18 is configured to detect an ambient air measurement value Um of an ambient air parameter U and to provide the ambient air measurement value Um to the fan control 14. The ambient air parameter U can be an ambient air temperature, an ambient air pressure, an ambient air humidity or an ambient air component or any combination of these ambient air parameters. The ambient air component can describe a proportion or a quantity of a gas component of the ambient air. Additionally or alternatively, the at least one ambient air parameter U can also be retrieved from available weather data (e.g. via weather data from the Internet) or determined based thereon.

[0039] In the exemplary embodiment described here, at least the room temperature RT is measured as the room air parameter R by means of the room sensor 17, and at least the ambient air temperature in the vicinity of the building 12 is measured as the ambient air parameter U. In the exemplary embodiment, the room temperature RT is the room air parameter R, which is used as the parameter Rc to be controlled or controlled. The aim of the fan control 14 is to keep the controlled parameter Rc within a predetermined parameter value range W. In the exemplary embodiment, the parameter value range W is a temperature range between a minimum room temperature RT min and a maximum room temperature RT max . The minimum room temperature RT min and / or the maximum room temperature RT maxcan be constant or vary depending on the time of day. For example, a room temperature RT for a room used by people in a building 12 in a given time of day D, for example between 6:00 a.m. and 6:00 p.m., can be defined differently than outside this time of day D. For example, the maximum room temperature value RT max and / or the minimum room temperature value RT min be specified in a stepwise manner.

[0040] In addition to the flow generation unit 11, the control system 10 has a modeling device 22 configured for machine learning. The modeling device 22 serves to generate a control model CM. The control model CM is provided to the fan controller 14 for controlling the at least one fan 13. For this purpose, a communication connection, for example an internet connection, can optionally exist between the modeling device 22 and the fan controller 14. The modeling device 22 can be implemented remotely from the fan controller 14 and provided, for example, in the form of an internet service or cloud service. However, the communication connection is not mandatory.The control model CM can also be determined by the modeling device 22 and stored in a memory of the fan control 14 as part of the production and configuration of the fan control 14.

[0041] The model building device 22 may include any known component that enables machine learning based on the learning data LD and, for example, may include any known artificial intelligence (AI) component, such as a neural network (NN). A support vector machine (SVM) or other known components may also be used.

[0042] The control model CM defines a relationship between the controlled parameter Rc, the current room air measured value Rm, the current ambient air measured value Um, and at least one fan operating parameter B, for example, the fan speed n. The current room air measured value Rm and the current ambient air measured value Um can be provided to the fan controller 14. Using the control model CM, the fan operating parameter B of the assigned fan 13 can be set such that the controlled parameter Rc (here: room air temperature RT) remains within the specified parameter value range W during a preceding time interval (prediction interval).The control of the fan 13 is such that a forecast for the course of the controlled parameter Rc is made by means of the control model CM in the forecast interval and the fan operating parameter B is controlled such that the parameter value range W is maintained using the lowest possible electrical power or electrical energy within the forecast interval.

[0043] To determine the control model CM, learning data LD is provided to the modeling device 22. The learning data LD can be fixed data and / or historical data. The learning data LD comprises at least historical data of the at least one ambient air parameter U and the at least one room air parameter R, in particular the room air parameter R used as the controlled parameter Rc. The learning data LD can also include historical data of the at least one fan operating parameter B. The learning data LD can optionally also include at least one state parameter Z and / or at least one geographical parameter G.The at least one state parameter Z and / or the at least one geographical parameter G and / or the at least one fan operating parameter B do not have to be part of the learning data LD and can optionally be taken into account elsewhere in the modeling device 22, for example in the context of a simulation or optimization when determining the control model CM.

[0044] The at least one further state parameter Z can, for example, comprise weather data and / or building data. Weather data can, for example, be a degree of cloud cover and / or solar radiation intensity and / or precipitation data (type and / or amount of precipitation).

[0045] Building data can, for example, describe the building 12 or the part of the building 12 to which the flow generation unit 11 is assigned and can, for example, include the size of a room, the size of at least one window, the orientation of at least one window (compass direction).

[0046] The at least one geographical parameter G can, for example, indicate a geographical longitude and / or a geographical latitude of the installation location of the flow generation unit 11. The daytime and / or nighttime duration, which vary depending on the season, can also be considered as a state parameter. The at least one further state parameter Z can be variable (e.g., weather data) or fixed (e.g., building data).

[0047] In the Fig. 2 and Fig. 3, embodiments of a method are illustrated in the form of a flow chart, wherein this method can be carried out with any embodiment of the control system 10 described above. Fig. 2 shows an example of a first method V1 and Fig. 3 shows an example of a second method V2.

[0048] In the first method V1, a parameter model is generated in a first step V11 based on the learning data LD. The parameter model specifies a relationship between the at least one ambient air parameter U and the at least one room air parameter R, for example, between a room temperature RT and the at least one ambient air parameter U, which in particular comprises the ambient air temperature.

[0049] In a second step V12 of this first method V1, based on the parameter model, it is simulated how a controlled parameter Rc develops in a preceding time interval (prediction interval) under different operating conditions of the at least one fan 13.

[0050] In a subsequent third method step V13, the control model CM is generated from the simulation results. This control model CM represents the relationship between the at least one ambient air parameter U, the controlled parameter Rc, and optionally at least one further parameter R, Z, G, on the one hand, and the fan operating parameter B (output variable of the control model CM), on the other hand. This control model CM is provided to the fan controller 14 in a fourth step V14 of the first method V1, which can then use the control model CM and the current measured values ​​BM, RM, UM to control the at least one fan 13 (fifth step V15 of this first method V1). Ü

[0051] By means of the control model CM, the development of the controlled parameter Rc (for example, the room air temperature RT) can be predicted in a preceding prediction interval, and the operation of the at least one fan 13 can be adjusted such that the controlled parameter Rc remains within the parameter value range W. The available parameter value range W can be fully utilized to minimize the electrical power or electrical energy required to operate the at least one fan 13 within the prediction interval. The controlled parameter Rc (for example, room temperature RT) is thus not adjusted to a target value as precisely as possible, but fluctuations within the parameter value range W are permitted in order to optimize the energy efficiency of the flow generation unit 11.

[0052] The forecast interval can, for example, have a duration of a few seconds to a few hours, e.g., at least 5 or 10 seconds, at least 5 to 30 minutes, or at least 60 minutes. Additionally or alternatively, the forecast interval can have a duration of a maximum of 24 hours, a maximum of 12 hours, a maximum of 8 hours, or a maximum of 4 hours.

[0053] In the Fig. In the second method V2 illustrated by way of example in Figure 3, the learning data LD comprise the at least one fan operating parameter B, so that the relationship between the at least one fan operating parameter B, the at least one ambient air parameter U, and the at least one room temperature parameter R can be established by the learning process, and the control model CM can be provided based thereon (first step V21 of the second method V2). In comparison to the first method V1, the second step V12 and / or the third step V13 of the first method V1 can thus be omitted. The second step V22 and the third step V23 of the second method V2 correspond to the fourth step V14 and the fifth step V15 of the first method V1. In this respect, reference can be made to the above explanation.

[0054] In a modification of the illustrated embodiment, the control model CM learned in the first step V21 can be optimized in a subsequent optional step by a simulation and / or an optimization method to minimize the electrical energy required for the at least one fan 13 in the prediction interval.

[0055] Based on the Fig. 4 to 6, the principles of methods V1 and V2 are illustrated schematically and merely by way of example.

[0056] In Fig. 4 schematically illustrates the temporal course of a room temperature RT, as it can be predicted based on a parameter model or the control model CM at least always for a previous time period (prediction interval) using current measurement data, in particular a current room measurement value (e.g. current room temperature RT) and a current ambient air measurement value Um (e.g. current ambient air temperature).

[0057] Based on the predicted temperature change, the setting of one or more fan operating parameters B, for example the fan speed n, can then be investigated within the framework of a simulation and it can be determined within the framework of the simulation how different settings for the at least one fan operating parameter B affect the development of the controlled parameter Rc - in this case the room temperature RT.

[0058] In the Fig. In the example illustrated in Figure 5, the at least one fan 13 is switched on during several time phases and switched off during several time phases. In a modification of the illustrated embodiment, the rotational speed n can also vary between several speed values ​​in the phases in which the at least one fan 13 is switched on. The rotational speed n can be changed in stages or continuously during the switch-on phases.

[0059] In Fig. Figure 6 illustrates, by way of example, a control of the at least one fan 13 as it results from the implementation of the invention. During a time period D between a first time t1 and a second time t2, the maximum room temperature value RT max smaller than outside the daytime range D and the minimum room temperature value RT minis greater within the daytime range D than outside this daytime range D. The parameter value range W defined in this way can be used, for example, for rooms in buildings 12 that are used by people during the daytime range D. During the night hours and in the early morning hours, greater temperature fluctuations can be permitted than during the daytime range D.

[0060] The room temperature RT to be controlled or regulated (dashed line in Fig. 6) is at the beginning of the exemplary time range outside the parameter value range W, which is desired during the daytime range D and, for example, above the maximum room temperature value RT max, which is specified for the time of day D. In order to reduce the room temperature RT before the start of the time of day D, the at least one fan 13 is operated at a high speed n before the first time t1 at a switch-on time te, whereby the room temperature RT decreases. As shown by way of example, the fan speed n can be reduced stepwise before the first time t1. At the start of the time of day D at the first time t1, the room temperature RT has a value that lies within the permissible parameter value range W. The at least one fan 13 is operated at a speed n that is as low as possible. The required speed n is selected based on the control model CM such that, based on a prediction in a respective preceding time interval or prediction interval, the room temperature RT lies within the respective permissible parameter value range W, which here is defined by the maximum room temperature value RTmax and the minimum room temperature value RT min is defined.

[0061] Based on Fig. 6 also shows that the room temperature RT is initially significantly reduced before reaching the second time t2 (end of the daytime range D), making it possible to switch off the at least one fan 13 before the second time t2 at a switch-off time ta. Subsequently, the room temperature RT rises. Based on the control model CM, the switch-off time ta is selected before the second time t2 such that the room temperature remains within the permissible parameter value range W until the end of the daytime range D.

[0062] The invention relates to a method V1, V2 for operating a flow generation unit 11 and to a control system 10 comprising at least one flow generation unit 11. The flow generation unit 11 is, in particular, a system permanently installed in a building 12. The flow generation unit 11 has a fan controller 14 that controls a fan 13 to generate an air flow L. The fan controller 14 is provided with a control model CM that specifies a relationship between a room air parameter R, an ambient air parameter U, and a fan operating parameter B for the fan 13. By measuring a room air measured value Rm for the room air parameter R and measuring an ambient air measured value Um for the ambient air parameter U, a suitable fan operating parameter B for controlling the fan 13 can then be selected and set based on the control model CM.The fan operating parameter B is selected such that the electrical energy required to operate the fan 13 is minimal. List of reference symbols: 10 Control system 11 Flow generation unit 12 buildings 13 Fan 14 Fan control 15 Electric motor 16 Operating sensor 17 Room sensor 18 Environmental sensor 22 Modeling facility B Fan operating parameters Bm Operating measured value for the fan operating parameter CM control model D Daytime range G geographical parameter L Air flow LD learning data n Fan speed NN artificial neural network R Indoor air parameters Rc controlled parameter Rm indoor air measurement value RT room temperature RT maxmaximum room temperature value RT min minimum room temperature value t time t1 first time point t2 second time point ta switch-off time th switch-on time U Ambient air parameters To ambient air measurement V1 first procedure V11 first step of the first procedure V12 second step of the first procedure V13 third step of the first procedure V14 fourth step of the first procedure V15 fifth step of the first procedure V2 second procedure V21 first step of the second procedure V22 second step of the second procedure V23 third step of the second procedure W Parameter value range Z state parameters QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] WO 2022 / 094281 A1

[0002]

Claims

[1] Method (V1, V2) for operating a flow generation unit (11) having a fan control (14) which is designed to control at least one fan (13), the method (V1, V2) comprising: - Providing learning data (LD) for a modeling device (22) which is configured for machine learning, wherein the learning data (LD) comprise at least one room air parameter (R) and at least one fan operating parameter (B), - determining a control model (CM) which represents the relationship between the at least one room air parameter (R) and the at least one fan operating parameter (B) by means of the modeling device (22), - Determining a current room air measurement value (Rm) for each room air parameter (R) that is used as a controlled parameter (Rc), - Controlling the at least one fan (13) using the control model (CM) and the at least one current room air measurement value (Rm) in such a way that a room air parameter (R) used as a controlled parameter (Rc) remains within a predetermined parameter value range (W) and the electrical energy required for operating the fan (13) is minimal. [2] Method according to claim 1, wherein the learning data (LD) further comprises at least one ambient air parameter (U), wherein the control model (CM) is determined such that it describes a relationship between the at least one room air parameter (R), the at least one ambient air parameter (U) and the at least one fan operating parameter (B), wherein at least one current ambient air measured value (Um) is determined for the one or more of the ambient air parameters (U), and wherein the at least one fan (13) is controlled using the control model (CM), the at least one current room air measured value (Rm) and the at least one current ambient air measured value (Um). [3] Method according to claim 1 or 2, wherein the fan controller (14) controls the at least one fan (13) based on the control model (CM) independently of the modeling device (22). [4] Method according to one of the preceding claims, wherein the modeling device (22) is implemented separately from the fan control (14). [5] Method according to one of the preceding claims, wherein the modeling device (22) and the fan control (14) are communicatively connected. [6] Method according to one of the preceding claims, wherein the at least one room air parameter (R) comprises one or more of the following parameters: a room air temperature (RT), a room air humidity, a room air pressure, a room air component which describes a proportion or a quantity of a gas component of the room air. [7] Method according to one of the preceding claims, wherein the at least one ambient air parameter (U) comprises one or more of the following parameters: an ambient air temperature, an ambient air humidity, an ambient air pressure, an ambient air component which describes a proportion or an amount of a gas component of the ambient air. [8] Method according to one of the preceding claims, wherein the at least one fan operating parameter (B) comprises one or more of the following parameters: a fan speed (n), a fan torque, a motor voltage of an electric motor (15) of the fan (13), a motor current of an electric motor (15) of the fan (13), a mechanical and / or electrical power of the fan (13). [9] Method according to one of the preceding claims, wherein the learning data (LD) additionally contain at least one further state parameter (Z) and / or at least one geographical parameter (G). [10] Method according to one of the preceding claims, wherein the learning data (LD) further comprises the at least one fan operating parameter (B), wherein the modeling device (22) generates the control model (CM) by learning the learning data (LD). [11] Method according to one of claims 1 to 9, wherein the modeling device (22) carries out a simulation based on the learning data (LD) at different parameter values ​​of the at least one fan operating parameter (B) in order to generate the control model (CM). [12] Method according to one of the preceding claims, wherein the at least one fan (13) is controlled such that the speed of the fan (13) is as low as possible in order to keep the at least one room air parameter (R) in the predetermined parameter value range (W). [13] Method according to one of the preceding claims, wherein the controlled parameter (Rc) is a room air temperature (RT). [14] Method according to one of the preceding claims, wherein the control model (CM) is dependent on the fan type. [15] Control system (10) comprising a modeling device (22), a flow generation unit (11) with at least one fan (13), with at least one room sensor (17) and with a fan controller (14) which is configured to control the at least one fan (13), wherein the modeling device (22) is configured for machine learning and is configured to determine a control model (CM) for the fan controller (14) based on provided learning data (LD), wherein the learning data (LD) comprise at least one room air parameter (R) and at least one fan operating parameter (B), and wherein the control model (CM) describes a relationship between the at least one room air parameter (R) and the at least one fan operating parameter (B), wherein the at least one room sensor (17) is configured to determine a current room air measurement value (Rm) for each room air parameter (R) that is used as a controlled parameter (Rc), and wherein the fan control (14) is configured to control the at least one fan (13) using the control model (CM) and the at least one current room air measurement value (Rm) in such a way that a room air parameter (R) used as a controlled parameter (Rc) remains within a predetermined parameter value range (W) and the electrical energy required for operating the fan (13) is minimal.

Images