Acoustic control method for actively controlling acoustic emission from fluid machinery, acoustic control system including fluid machinery, and acoustic control device for executing said acoustic control method

Abstract

The invention relates to an acoustic control method for active control of the acoustic emissions of turbomachines, in particular turbomachines with electric motors, preferably ventilators or turbomachines, wherein an acoustic signal generated by superimposing at least one inverse acoustic signal on the acoustic emissions of the turbomachines is received by at least one receiver at at least one receiver location and transmitted to a control unit, the control unit having an artificial intelligence by means of which control signals for at least one actuator are generated taking into account the acoustic signal, the actuator generating an inverse acoustic signal which interacts with the acoustic emissions of the turbomachines, so that the acoustic load at least in the area of ​​the receiver location is reduced.The invention relates to an acoustic control system and an acoustic control device for carrying out the acoustic control method according to any one of claims 1 to 8.

Description

[Technical field]

[0001] The present invention relates to a method for actively controlling the acoustic emissions of a fluid machine, in particular a fluid machine having an electric motor, preferably a fan or turbomachine.

[0002] Furthermore, the invention relates to an acoustic control system comprising a fluid machine, in particular a fan or a turbomachine, for implementing the active acoustic control method according to any one of claims 1 to 8. [Background technology]

[0003] In particular, in fluid machines, it is always desirable to minimize the noise emission for a given power data or operating point, or to optimize the noise emission so that it is subjectively perceived as more comfortable by humans. The use of fluid machines, turbomachines or fans has been and will continue to be increasingly restricted due to the noise they generate. On the other hand, with the development of quieter devices, it appears that, at least for a given structural size and drive torque, noise is asymptotically approaching a natural lower limit.

[0004] Many optimization attempts can have very negative impacts on energy, material, or cost efficiency (e.g., passive sound damper technologies). For this reason, so-called "active noise canceling" (ANC) is highly anticipated these days. This involves either cancelling unpleasant sounds with phase-shifted inverse sound where necessary, thereby reducing the unpleasant sounds, or using a more complex approach ('acoustic design') to configure the sounds so that they are more comfortable for people. However, the complexity of acoustic events and their sensitivity to installation conditions, operating states of the fluid machinery, and receiver position make it difficult to design corresponding acoustic control systems in a robust and reliable manner. For this reason, ANC has not yet been used comprehensively in the field of turbomachinery, and it is likely to be some time before this becomes a reality.

[0005] Overall, the following procedures for controlling acoustic emissions are therefore known in the prior art: Use ANC for sound control systems with fixed source and receiver positions. Provide passive acoustic protection. Low noise devices are designed using reduced speed configurations, etc. Summary of the Invention [Problem to be solved by the invention]

[0006] The object of the present invention is to provide a method for active control of acoustic emissions in fluid machinery and to further develop it in order to achieve optimised active acoustic control of acoustic emissions with little complexity. It is a further object of the present invention to provide an acoustic control system having a fluid machine capable of optimizing active acoustic control of acoustic emission. It is a further object of the present invention to provide an acoustic control device which optimizes active acoustic control of acoustic emissions in fluid machinery. [Means for solving the problem]

[0007] According to the invention, the above mentioned object is achieved by means of the features of claim 1. As a result, an acoustic control method for actively controlling acoustic emissions of a fluid machinery, in particular a fluid machinery having an electric motor, preferably a fan or turbomachine, is claimed, in which an acoustic signal generated by superimposing the acoustic emission from the fluid machinery with at least one inverse acoustic signal is recorded by at least one receiver at at least one receiver location and transmitted to a control unit, the control unit having an artificial intelligence, and a control signal for at least one actuator is generated by the artificial intelligence taking into account the acoustic signal, whereby the actuator generates an inverse acoustic signal cooperating with the acoustic emission of the fluid machinery and the acoustic load at least in the area of ​​the receiver location is reduced or minimized, and at least two state values ​​of the fluid machinery are transmitted to the control unit, and the control signal is generated by the artificial intelligence taking into account the at least two state values.

[0008] With regard to the sound control system according to the invention, the above mentioned object is achieved by the features of claim 9. As a result, an acoustic control system is claimed comprising a fluid machine, in particular a fluid machine having an electric motor, preferably a fan or turbomachine, for performing an acoustic control method for actively controlling the acoustic emissions of a fluid machine as described in any one of claims 1 to 8, at least one receiver for detecting an acoustic signal at at least one receiver position, the acoustic signal being generated by superimposing the acoustic emission generated by the fluid machine and at least one inverse acoustic signal, a control unit and at least one actuator, the control unit having an artificial intelligence which controls the actuator taking into account the detected acoustic signal and at least two state values ​​of the fluid machine to generate an inverse acoustic signal which cooperates with the acoustic emission of the fluid machine, so that the acoustic load at least in the area of ​​the receiver position is reduced or minimized.

[0009] With regard to the sound control device according to the invention, the above mentioned object is achieved by the features of claim 11. As a result, an acoustic control device is claimed having at least one receiver for detecting an acoustic signal at at least one receiver position for performing an acoustic control method for actively controlling the acoustic emissions of a fluid machinery as described in any one of claims 1 to 8, the acoustic signal being generated by superimposing the acoustic emission generated by the fluid machinery and at least one inverse acoustic signal, a control unit and at least one actuator, the control unit having an artificial intelligence which controls the actuator taking into account the detected acoustic signal and at least two state values ​​of the fluid machinery, such that the actuator generates an inverse acoustic signal which cooperates with the acoustic emission of the fluid machinery, and the acoustic load at least in the area of ​​the receiver position is reduced or minimized.

[0010] It should be noted that the features of the active sound control method according to the present invention may also have a similar nature for sound control devices. It is not only possible but advantageous to combine these features with features relating to the audio control system and / or with features relating to the audio control device.

[0011] According to the present invention it has first been recognised that by using AINC (Artificial Intelligence Noise Cancelling) methods it is possible to create a robust ANC system (e.g. active control of sound by one or more additional sound sources to generate phase shifted inverse sound) for fluid machinery or turbomachines or fans. This allows the inverse sound source to be dynamically controlled. In this case, the term "robust" means that the effectiveness of the acoustic control system is not very sensitive to changes in the operating conditions of the fluid machine (speed, volumetric flow rate, pressure rise, etc.) and / or is not very sensitive to the installation environment of the fluid machine and / or is not very sensitive to the position of the receiver. According to the invention, a control unit in which artificial intelligence is implemented is arranged to automatically find the optimal inverse acoustic signal for the respective configuration in a short time. The invention is therefore based on the physical principle that superimposing a first acoustic signal with a second acoustic signal that is 180° out of phase with the second signal and has the same frequency and amplitude results in a cancellation effect between the two signals, resulting in a reduced (and ideally cancelled) signal being generated at the receiver. In another active acoustic control method according to the present invention, an existing fluid machine can be retrofitted with an acoustic control device as recited in claim 11.

[0012] Advantageously, the time signal and / or the frequency range and / or the phase position of the inverse acoustic signal generated by the actuator may be controlled by the control signal. Furthermore, the artificial intelligence may be pre-trained, for example at the factory. Advantageously, the artificial intelligence may use reinforcement learning techniques ("reinforcement learning") to generate the control signals.

[0013] Specifically, a device-specific pre-trained "reinforcement learning" control unit can adaptively control the time signal of the inverse sound source so that the noise signal at a particular receiver position is minimized or optimized based on psychoacoustic considerations. This requires a control signal from a receiver at a given location. The flexibility of AINC allows the receiver location to be configured in an application-specific manner as well.

[0014] In the active acoustic control method according to the present invention, state values ​​of the at least two fluid machines are transmitted to the control unit, and the control signal is generated by the artificial intelligence taking into account the at least two state values. As a result, it is possible to react quickly and clearly to changing operating conditions. In this case, the expression "state values ​​of the fluid machinery" should be clearly understood to include all values ​​which represent or describe the current operating state of the fluid machinery and therefore all values ​​which represent or describe the acoustic emissions which occur. This may also include components associated with the fluid machinery, such as anemometer speed, hot wire anemometer signal, or differential pressure sensors.

[0015] The at least two state values ​​are: the motor speed and impeller anemometer speed of said fluid machinery; the motor speed and hot wire anemometer signals of said fluid machinery; a motor speed of the fluid machine and a motor current of the fluid machine; the motor current and impeller anemometer speed of said fluid machinery; a motor current of the fluid machine and a signal from a hot wire anemometer; The motor speed of the fluid machinery and the pressure difference between the upstream and downstream (in the flow direction) of the fluid machinery, Or the motor current of the fluid machine and the pressure difference between the upstream and downstream (in the flow direction) of the fluid machine, It is advantageous if The combination of state values ​​mentioned above has the advantage that it is particularly accurate in representing the first acoustic result to be corrected. In such cases, other pairs of state values ​​and / or combinations of three or more of the above state values ​​may be used.

[0016] A microphone may be used as the receiver. Furthermore, the actuator may be a speaker.

[0017] Advantageously, the actuator excites a component of the fluid machine to produce sound. For this purpose, for example, piezoelectric actuators may be used and / or the excitation current or voltage of an electric motor may be modulated with a suitable superimposed excitation signal. Thus, in particular, structure-borne sound may be emitted via a spectrally modulated excitation voltage, such as a drive motor.

[0018] Furthermore, the control unit may be in the form of a component that is integral to the fluid machine, or the control unit may be in the form of a separate control module.

[0019] Various methods are currently available for configuring and further developing the present invention in an active acoustic control manner. For this purpose, reference is made, on the one hand, to the claims dependent on claims 1 and 9 and, on the other hand, to the following description of embodiments of the invention with reference to the drawings, in which: In connection with the description of embodiments of the invention with reference to the drawings, general embodiments and further developments of the teachings are also described. [Brief description of the drawings]

[0020] [Figure 1] 1 is a schematic diagram of an embodiment of an active system according to the invention, on the basis of which the active sound control method according to the invention and the sound control device according to the invention are also explained. [Diagram 2] FIG. 1 is a three-dimensional bar graph showing a schematic representation of the fan pressure-side acoustic power Lw6 (dB) as a function of both motor speed and impeller anemometer speed. [Diagram 3] FIG. 1 is a view of a radial fan having a housing that is particularly suitable for use with active acoustic control methods, the view being taken from the rotor axial direction and a cross-section taken in a plane transverse to the rotor axis. [Figure 4] FIG. 1 shows perspective and cross-sectional views of an embodiment of a fan taken in a plane through the axis of rotation of the rotor, with an impeller anemometer generating anemometer speed as an input sensor variable for an active acoustic control method. [Diagram 5] FIG. 1 shows a perspective view from the inlet side of an embodiment of a fan having a support module with support struts in the form of vectored vanes that are particularly suitable for the use of active acoustic control methods. [Figure 6]FIG. 1 shows a side view and a cross section in a plane through the axis of a fan with a conveying deflection unit (carrier module) with two different types of carrier struts and an intermediate ring in the form of deflecting vanes, which are particularly suitable for the use of active acoustic control methods. [Figure 7] FIG. 1 is a perspective view from the outflow side of an assembly of four fans connected in parallel, which is particularly suitable for use with active acoustic control methods. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] FIG. 1 shows a schematic diagram of an artificial intelligence based active acoustic control method for fluid machinery, more particularly for turbomachines, more particularly for fans.

[0022] FIG. 1 shows an outline of an acoustic control system 1, which in this embodiment is an AINC system (AINC is an abbreviation for Active Intelligent Noise Control).

[0023] The electric fluid machine 2 operates to perform energy transfer (or power transfer) between a fluid and an electrical connection. In an embodiment, the fluid machine 2 is a turbomachine or a fan driven by an electric motor 3 . As a result of its operation, the fluid machine 2 converts electrical energy into fluid energy (specifically, a total pressure increase of the transported volume flow rate is caused). The proposed technology is also relevant to fluid machines operated by a generator to transfer power from a fluid to the generator (eg, a wind turbine). Experience has shown that in such cases, unpleasant noise emissions (=first acoustic signals) occur periodically. The occurrence of this first acoustic signal is unavoidable, and in fluid machines developed at the same time according to the prior art, reducing this first acoustic signal on a more extensive scale would be technically and / or developmentally very complicated.

[0024] The active acoustic control method outlined above is based physically on the fact that superimposing a first acoustic signal with a second acoustic signal that is 180 degrees out of phase and of identical frequency and amplitude creates a cancellation effect between the two signals, resulting in a reduction (or, in ideal cases, cancellation) of the signal at the receiver. The active acoustic control system 1 has an actuator 5 for physically generating one or more second acoustic signals (=second acoustic signals) that are used to cancel or reduce the total acoustic signal received by the receiver 4. This actuator 5 then has the function of transmitting an inverse acoustic signal, which is flexibly controllable in terms of time signal and / or frequency range and / or phase position, to the surrounding fluid medium, typically the conveying medium of the fluid machine 2 . A typical actuator 5 is a speaker, although other actuators 5 are suitable and contemplated. Specifically, components of the fluid machine 2 itself may be excited to produce sound, for example by a piezoelectric actuator or by modulating the excitation current or excitation voltage of the drive device 3 (e.g. an electric motor) with an appropriate superimposed excitation signal.

[0025] In general, this method of cancelling a given sound source by means of so-called inverse acoustics is well known under the term ANC (Active Noise Cancelling). In this case, one of the most important technical challenges is always to determine and generate one or more suitable inverse acoustic signals. This is because in fluid machinery we are interested in the exact structure of the first acoustic signal at the receiver location. The structure is unpredictable, very difficult to predict, or unknown. For example, the turbulent events that produce sound often do not have a deterministically predictable frequency, phase position or amplitude. Furthermore, acoustic events may vary significantly depending on the receiver position. For various reasons, sound production can be highly dependent on the turbomachine installation and cannot be determined in advance by laboratory operation in a representative manner. For example, the inflow turbulence that significantly affects the first acoustic signal is significantly affected by the installation conditions on the inflow side. The manner in which the first acoustic signal is transmitted to the observer is also greatly affected by the installation conditions on the inflow side or outflow side (which differ depending on the position of the observer).

[0026] The proposed active sound control method further uses one (or more) receivers 4 (eg microphones), as also shown in FIG. The total acoustic signal recorded in real time at the location of the receiver 4 means the "acoustic at the receiver side", where the total acoustic is minimized as much as possible and / or optimized as much as possible for good subjective perception of the object. All recorded acoustic signals are transferred to a control unit 6 which contains an artificial intelligence (artificial intelligence module). There, first an assessment of the "quality" of this total acoustic signal needs to be carried out. The evaluation results in one or more quantitatively determinable characteristic variables, the goal of which is to optimize (minimize or maximize) these characteristic variables.

[0027] A simple example of a characteristic variable is the A-weighted sound pressure level. Also other characteristic variables can be used, for example psychoacoustic characteristic variables such as sharpness, harshness, timbre, loudness, etc. The control unit 6 advantageously comprises an interface 7 . Via this interface 7, the user or a higher-level system can specifically control this evaluation and, if necessary, flexibly control the weighting of the various evaluation factors. This provides the sound control system 1 with more flexibility.

[0028] As a central basic algorithm on the basis of which the control unit 6 establishes the second acoustic signal (the inverse acoustic signal) a "reinforcement learning" algorithm can be used, although the use of other algorithms is also conceivable. The algorithm itself is known and is characterized by an adaptive behavior based on a reward principle. Briefly, a trial-based algorithm optimizes the second audio signal (the inverse audio signal) until the overall audio signal at the receiver or microphone 4 is optimized according to a criterion.

[0029] "Reinforcement learning" requires a "learning period" during which one or more optimal second acoustic signals are eventually established by a trial and error strategy. In order to minimize the learning time during practical operation, it is therefore proposed to carry out an initial preliminary learning phase specific to the particular type of fluid machine during laboratory operation and to configure the control unit 6 accordingly in advance. At least two status values ​​that are as representative as possible of the current operating state of the fluid machine 2 are transmitted to the control unit 6 so that the control unit 6 can adapt the second acoustic signal quickly and in real time to changes in the operating state of the fluid machine 2. For example, speed, motor current, impeller anemometer speed, hot wire anemometer signal, differential pressure, or vibration sensor signals can be used. The control unit 6 then directly uses these state or measurement values ​​to establish the second acoustic signal. In this way, the acoustic control system 1 is able to adapt the second acoustic signal in a highly dynamic manner to changing operating conditions of the fluid machine 2, resulting in changes to the first acoustic signal. The operating conditions of the fluid machinery 2 may change more dynamically, for example when wind affects the load of the fluid machinery 2 (wind turbine or turbomachine).

[0030] Figure 2 shows the pressure-side acoustic power L of a fan similar to that in Figure 3. w6 (dB) is shown diagrammatically in a three-dimensional bar graph as a function of both motor speed and impeller anemometer speed, which may be used to determine the conveyed volumetric flow rate, for example, in a configuration similar to that shown in FIG. In this case, combinations for which no bar graph is shown represent combinations that were not measured. The representation of sound power is only one way of representing sound events with individual characteristic values; many other characteristic values ​​can be derived for each sound event, e.g. sound power in only one freely selectable frequency range, psychoacoustic characteristic variables or, e.g., the tonal proportions of the sound event. In either case, in order to capture the entire acoustic event as quickly and optimally as possible with the receiver microphone and actively control the acoustic emissions of the fluid machine 2, the acoustic event is characterized as well as possible by transmitting state or sensor values ​​to the control unit 6 during operation, and when the state of the acoustic event changes rapidly (highly dynamic state changes), the active acoustic control method according to the present invention is executed to highly dynamically adapt the inverse acoustic signal (second acoustic signal) to the first acoustic event, which depends on the flow state of the fluid machine 2.

[0031] The acoustic phenomena of the fluid machinery 2, in particular of the fan, are then substantially predefined, since they are strongly linked to the flow or operating conditions of the fluid machinery 2 for given installation conditions and a certain conveying medium. This also makes it possible to derive appropriate conclusions regarding the first acoustic event generated from the state or sensor values, thereby making it possible to derive appropriate and unambiguous conclusions regarding the flow state of the fluid machine 2 as far as possible. Thus, during operation, the active acoustic control method according to the present invention transmits to the control unit 6 state or sensor values ​​from which a conclusion as appropriate and unambiguous as possible regarding the flow state of the fluid machine 2 can be derived, and, if applicable, a conclusion from which the active acoustic control method according to the present invention can unambiguously derive a conclusion regarding the current first acoustic event.

[0032] Typically, in a fluid machine 2, and in particular a fan, for certain installation conditions and a certain conveying medium, the flow conditions (and therefore also the first acoustic event generated) depend on two parameters. In particular, it is not sufficient to use only one characteristic variable or one sensor variable in characterizing the first acoustic event. For example, this can be easily seen in the diagram shown in FIG. FIG. 2 shows the pressure-side acoustic power L of a fan 2 similar to the fan 2 shown in FIG. w 6, motor speed n Mot and an impeller anemometer n mounted on the suction side of the fan similar to the impeller anemometer shown in Figure 4. Ane The figure shows the relative velocity of the Sound power L w It can be seen that 6 varies depending on the operating state of fan 2. Motor speed n Mot Even if we know the sound power L w It is not possible to draw any appropriate and unequivocal conclusions regarding 6 and, as a consequence, the first acoustic event. Because the motor speed n Mot Even if the anemometer speed n Ane Depending on the acoustic output L w 6 is because there is variety. Constant motor speed n Mot The behavior of is given by Mot At different operating points of the fluid machine 2 on the characteristic line, the first acoustic event varies. But the motor speed n Mot In addition, the anemometer speed n Ane is transmitted to a control unit 6 of the active acoustic control method according to the invention, an optimal second acoustic signal can be generated appropriately and highly dynamically from the two measured variables with respect to a changing first acoustic signal (if applicable, without having to be explicitly calculated). Conversely, the anemometer speed n Ane Knowing only this, no good, unequivocal conclusions can be drawn about the first acoustic event.

[0033] It is easily conceivable to transmit other pairs of state or sensor variables to the control unit 6 of the active acoustic control method, as long as this allows for drawing as relevant and unambiguous a conclusion as possible regarding the first acoustic event of the fluid machine 2 or fan in the respective operating environment. According to the invention, it is advantageous if there are at least two state values ​​(per fluid machine 2). A particularly possible and advantageous pair, also using suitable sensors, is: a.) Motor speed and impeller anemometer speed b.) Motor speed and hot wire anemometer signals c.) Motor Speed ​​and Motor Current d.) Motor current and impeller anemometer speed e.) Motor current and hot wire anemometer signals f.) Motor Speed ​​and Differential Pressure g.) Motor current and differential pressure or alternative pairs of identical information items that are derived directly from such pairs.

[0034] In the active acoustic control method according to the present invention, if several fluid machines 2 operating simultaneously are connected in parallel in series (see example in Figure 6), for each operating fluid machine 2 at least two sensor variables characterizing the flow state of the respective fluid machine 2 are transmitted to the control unit 6 and processed to generate a second acoustic signal, at least insofar as these sensor variables are independent of each other.

[0035] FIG. 3 is a view seen from the direction of the impeller shaft, and is a cross-sectional view taken along a plane intersecting the impeller shaft, showing a fluid machine 2 (fan) having a housing 10. As shown in FIG. This cross-sectional plane is perpendicular to the fan axis and is axially located at the center of the flow path. In addition to a housing 10 , the fan 2 in particular further comprises a drive device 3 , which is shown only diagrammatically in cross section, and a rotor 9 or impeller 9 which has blades 8 . The rotor 9 is driven by a drive device 3, for example an electric motor, advantageously an external rotor motor, and rotates in operation in a clockwise direction in the figure. This is therefore a backward curved impeller 9, i.e. an impeller 9 with backward curved blades 8. Specifically, when the range of the blade 8 is viewed from a radially inner position (from the front end) toward the radially outer direction (toward the rear end), the blade 8 is curved in the direction opposite to the direction of rotation.

[0036] During operation of the fan, entrained air is discharged radially outwardly from the rotor 9 into a flow passage in the housing 10 which extends generally circumferentially about the impeller axis. This flow passage widens in the circumferential direction from its narrowest point in the region of the tongue 11 to the outlet 12 of the fluid machine 2 or of the helical housing 10 in order to accommodate an air flow that increases in the circumferential direction. As a result of the interaction between the blades 8 and the tongue 11 or scraper 11, during operation of the fluid machine 2, the rotating blades 8 of the rotor 9 interact with the rear end of the rotor 9 relatively close to the tongue 11 or scraper 11, which may generate rotational noise that accounts for a large portion of the first acoustic signal. This rolling noise can sound loud, harsh and unpleasant. Such a fan 2 or such a fluid machine 2 is particularly suitable for use with the active acoustic control method according to the invention, since the frequencies are very discrete and at very low frequencies. For example, the rotor 9 can be used as an actuator 5 and excited in a suitable manner, for example via the drive 3 . The housing 10 or its walls may be used as an actuator 5 in combination with a vibration generating element or a separate actuator 5 may be fixed within the housing 10 . The rotational noise generated as the first acoustic part is determined by two sensor parameters of the fluid machine 2, e.g. the rotor speed n Mot For example, the anemometer speed n of an impeller anemometer (not shown) installed upstream of the inlet of the fluid machine 2 AneIt depends on the pairing with . In this case, the rotor speed n Mot However, this determines the frequency of the rotational noise and also has a significant impact on the intensity of the rotational noise. anemometer speed n Ane has a significant effect on the intensity of rotational noise.

[0037] In this embodiment, and also in other embodiments involving the interaction of rotating and stationary parts, it may be advantageous to use the current rotational angular position of the rotor 9 as additional input information in the control unit 6 . This gives information about the current phase position of the rotational noise generated by this interaction, which is dependent on the relative rotational positions of the rotating and fixed parts. Usually, for this purpose, a signal (trigger, pulse) is always required, indicating when the rotor 9 passes a particular position. This can easily be achieved, for example, with a Hall sensor.

[0038] FIG. 4 shows an embodiment of a fluid machine 2 (in this case a fan 2) in perspective and cross-sectional view in a plane passing through the axis of rotation of the rotor 9, with a rotatable impeller anemometer wheel 13 attached on the inlet side. The impeller anemometer wheel 13 is essentially formed of a hub and blades 15 fixed to the hub. FIG. 4 clearly shows the impeller anemometer wheel 13 and its bearing relative to the inlet structure (in this case the inlet grid 14). An inlet grid 14 is attached to the rotor 9 or to the inlet side of the inlet nozzle 16 , through which the incoming conveying medium can enter the rotor 9 . The inlet grid 14 homogenizes the inlet and thereby improves the measurement accuracy of the impeller anemometer wheel 13 . During operation of the fluid machine 2, the speed of the impeller anemometer wheel 13, n Ane is constantly measured by a suitable sensor (e.g. a Hall sensor). And this speed n of this impeller anemometer wheel 13 Ane can be used particularly advantageously as an input variable for active sound control methods. For example, the motor speed n of the drive 3 or the rotor 9 Mot Together, the operating state of the fluid machine 2, and therefore the first acoustic signal generated, can be very well characterized in a particular operating environment. The control unit 6 can use this sensor signal in a highly dynamic manner to generate the most optimal possible second acoustic signal.

[0039] The rotor 9 / impeller 9 of the fan 2 is fixed to a drive unit 3 / motor 3 . In operation, the rotor 9 rotates with its blades 8 and transports the transport medium sequentially through the inlet grid 14 , over the anemometer wheel 13 , through the inlet nozzles 16 and radially outwardly within the rotor 9 . This may result in the generation of a first acoustic signal having a plurality of acoustic components (timbre components). The plurality of tone components may be, for example, tone components generated by the interaction of the webs of the inlet grid 14 with the impeller anemometer wheel 13 or the rotor 9 or the blades 8 of the rotor 9, or tone components generated by the flow of the conveying medium with a velocity n Ane The tonal components may be generated by the interaction between the freely rotating impeller anemometer wheel 13 and the rotor 9 or between the blades 8 and the rotor 9. In order to reduce the acoustic discomfort of such a first acoustic signal at the receiver location, the active acoustic control method according to the invention generates in the control unit 6 a second acoustic signal which is superimposed on the first acoustic signal to make the acoustics at the receiver location lower and / or more pleasant. In order to be able to react highly dynamically to changes in the first acoustic signal, the control unit 6 advantageously also processes, in addition to at least one signal from the receiver microphone, at least two sensor variables which are constantly measured during operation and accurately characterise the operating state of the fluid machine 2. In the control unit 6, inter alia, a reinforcement learning algorithm is used.

[0040] The impeller anemometer wheel 13 is typically mounted in an inlet grid 14 or housing 10 of a fan, for example at the inlet or outlet side of the rotor 9 of the fluid machinery 2 .

[0041] FIG. 5 shows a fluid machine 2 (fan 2) having a support module with support struts 17 in the form of deflecting vanes 17, in a perspective view seen from the inlet side. In this embodiment, a rotor 9 / impeller 9 can be seen having blades 8 with a radial or diagonal flow configuration therein, which in operation, in this embodiment, is driven by an external rotor motor, here a drive device not shown. Furthermore, on the inlet side, the inlet nozzles 16 can be seen which are attached to a nozzle plate 19 and which draw the conveying medium into the rotor 9 during operation of the fluid machine 2 . In addition to the nozzle plate 19 , the support module comprises a base plate 18 and eight lateral support struts 17 radially outside the air outlet (outlet side) of the impeller / rotor 9 . The support struts 17 in the form of deflection vanes 17 have both an aerodynamic function, by virtue of their presence increasing the efficiency of the fluid machine 2, and a supporting function, by connecting the nozzle plate 19 to the base plate 18 in a supporting manner, so as to ultimately hold the rotor 9 on the nozzle plate 19.

[0042] During operation of the fan / fluid machine 2 a first acoustic signal is generated. This first acoustic signal may include multiple components, for example components generated as a result of the interaction of the blades 8 of the rotor 9 with the diverting vanes 17 in the form of tonal and / or broadband components. In order to reduce the acoustic discomfort of such a first acoustic signal at the receiver location, the active acoustic control method according to the invention generates in the control unit 6 a second acoustic signal which is superimposed on the first acoustic signal to make the acoustics at the receiver location lower and / or more pleasant. In addition to the rotor 9, for example the carrier module with the deflection carrying posts 17, the nozzle plate 19 and the base plate 18 can also be used as components of the actuator 5 for generating the second acoustic signal. For example, a piezoelectric actuator can be arranged to excite vibrations that generate the second acoustic signal.

[0043] FIG. 6 shows a side view and a cross-section in a plane through the axis of a fan 2 (fluid machine 2) having a conveying direction-diverting unit (support module) with two different types of support struts 17 in the form of deflection vanes 17 and an intermediate ring 22, which are particularly suitable for the use of active acoustic control methods. The support module in particular has two different types of turning vanes 17 (radially inner and radially outer turning vanes 17), an intermediate ring 22 connecting the radially inner and radially outer turning vanes 17 to each other, a hub ring 23 to which the drive 3 of the rotor 9 is fixed, and an outer housing 10. The housing 10 here comprises, in one piece and integrally, an inlet nozzle 16 which draws the conveying medium towards the rotor 9 during operation of the fluid machine 2, an operating area 21 (advantageously having the shape of a substantially cylindrical cover) inside which the rotor 9 operates together with the blades 8, and a radial expansion section 20 in which the outer deflection blades 17 are fixed and whose outlet end forms the outlet 12 from the fluid machine 2. The diverting wheel has outer and inner diverting vanes 17 . The intermediate ring 22 has both an aerodynamic function to increase the efficiency of the fluid machine 2, and a transport function to connect the drive unit 3 (in this case, an external rotor type electric motor) to the external housing 10, which connects the fluid machine 2 to a higher unit in a supporting manner. During operation of the fan 2 / fluid machinery 2, a first acoustic signal is generated which includes multiple components, e.g. tonal and / or broadband components, resulting from the interaction of the blades 8 of the rotor 9 with the diverting vanes 17 or the intermediate ring 22. In order to reduce the acoustic discomfort of such a first acoustic signal at the receiver location, the active acoustic control method according to the invention generates in the control unit 6 a second acoustic signal which is superimposed on the first acoustic signal to make the acoustics at the receiver location lower and / or more pleasant. In addition to the rotor 9, for example a carrier module having a deflection vane 17 or a housing 10 having an inlet nozzle 16, a rotor region 21 or a radial expansion 20 can also be used as components of the actuator 5 for generating the second acoustic signal. For example, a piezoelectric actuator can be arranged to excite vibrations that generate the second acoustic signal. In such an embodiment, the speed n of the drive 3 or the rotor 9 in combination with the hot wire anemometer signal is Mot can also be advantageously used as a pair of sensor signals to accurately characterize the operating condition of the fluid machine 2 in the operating environment and serve as current inputs to the control unit 6 during use of the active acoustic control method.

[0044] FIG. 7 shows a perspective view from the outlet side of an assembly of four fluid machines 2 / fans 2 connected in parallel, which is highly suitable for the use of active acoustic control methods. These are fluid machines 2 / fans 2 without the housing 10 and are arranged side by side and connected in parallel to each other. Each of the fans 2 has a nozzle plate 19 in which inlet nozzles 16 are fixed, through which the conveying medium is drawn towards the rotor 9 during operation. The drive 3 with the rotor 9 is fixed to the nozzle plate 19 via a support post 25 . The rotor 9 has blades 8 . In this embodiment, the fan 2 is further fitted with a backflow prevention member 26 which prevents undesirable backflow downstream of the fan 2 in the area near the hub, thereby increasing the efficiency of the fan assembly 24.

[0045] The fluid mechanical assembly 24 is highly suitable for use with the active acoustic control method according to the present invention. However, in such a fluid machinery assembly 24, the function is related to the entire fluid machinery 2. This is because the receiving microphone receives all the inseparable sounds associated with all the fluid machines 2 . This means that there is an active acoustic control method coupled to each fluid machinery assembly 24 . As with the individual fluid machines 2 , signals from one or more microphones at the receiver locations are used as input signals to the control unit 6 . With regard to the sensor signals characterizing the flow conditions of the fluid machinery 2, as explained with reference to Figure 2, typically at least two sensor signals must be transmitted to the control unit 6 for each fluid machinery 2 in order to be able to detect the flow conditions of each fluid machinery 2. However, it may be advantageous to connect different fluid machines 2 with respect to one or more sensor variables. For example, all of the fluid machines 2 can be operated at the same speed. Depending on the arrangement, the fluid machines 2 can be connected to each other in relation to their operating state (pressure rise or volumetric flow rate). As a result, depending on the situation, all of the fluid machines 2 may always operate in the same operating state. If this is possible, depending on the placement, it may be possible to reduce the number of sensor signals sent. However, a pair of sensor variables that accurately and unambiguously characterises the flow conditions (and therefore the first acoustic event) in the current operating environment must be able to be derived directly for each fluid machine 2 itself from the sensor signals used and the connection state of the fluid machine 2.

[0046] Regarding the actuator 5, different approaches are also possible. It is also possible to distribute the actuators 5 symmetrically among all the fluid machines 2 or to reduce the number of actuators 5 . Typically, one or more actuators 5 can be used per fluid machine 2 .

[0047] In an embodiment, the backflow prevention member 26 and / or the nozzle plate 19 can be used as an effective actuator 5 in combination with a vibration generating device.

[0048] In an embodiment, as many components of the acoustic control system 1 as possible can be integrated into the fluid machine 2 . In particular, a fluid machine 2 having electronic speed control (e.g. via an electronically controlled frequency converter) will in any case already have a powerful electronic system built in that can be relatively easily expanded by a control unit 6, thereby advantageously providing an AINC control unit that is fully integrated in the electric motor or its electronic control unit.

[0049] In the embodiment shown in FIG. 1, the fluid machine 2 comprises an electric motor 3 incorporating an electronic control unit 6 . The electric motor 3 is preferably an external rotor motor in order to achieve a compact construction. The electric motor is an EC motor incorporating an electronic control unit 6, but may also be an AC motor. Alternatively, the control unit 6 may be integrated in the area of ​​the fluid machine 2 as an insulating module. As already explained, the actuator 5 can also be incorporated near the fluid machine 2 . The actuators 5 are connected in the most compact way possible with minimal hardware complexity, for example via available conductors in the electrical equipment with signals controllable by the control unit 6, to excite the available components, in particular the blades 8 of the fluid machine 2. It is also conceivable to excite components within the structure via actuators 5, for example piezoelectric actuators. It is also conceivable to use a speaker integrated close to the fluid machine 2, for example in or on the housing 10. An important advantage of incorporating the actuator 5 close to the fluid machine 2, and therefore close to the sound source for the first acoustic signal, is that the desired effect of the second acoustic signal emitted via the actuator 5 becomes very low directional dependency.

[0050] Also envisaged are embodiments in which one or more microphone signals recorded close to the sound source, i.e. close to the fluid machine 2 and representative of the first acoustic signal, are used as input to the control unit 6 .

[0051] Depending on the embodiment, it is also possible for the described active acoustic control functionality to be retrofitted to already developed or manufactured or operational fluid machinery, for example as an optional product feature extension or as an add-on, as claimed in independent claim 11. The necessary hardware components (such as microphones and actuators) need to be additionally connected or attached to the current interface. Software components, where applicable, can be installed on available hardware.

[0052] Regarding further embodiments of the sound control method according to the invention and of the sound control device according to the invention, reference is made to the general part of the specification and to the appended claims in order to avoid repetition.

[0053] Finally, the embodiments of the sound control method according to the present invention and the sound control device according to the present invention are intended to illustrate the claimed teachings and are not intended to limit the claimed teachings to the embodiments. [Explanation of symbols]

[0054] 1. System 2. Fluid machinery (fans) 3. Driving device (of fluid machinery) 4. Receiver 5. Actuator 6. Control unit 7. Interface 8 Rotor blades 9 Rotor 10. Housing 11 Tongue, scraper 12 Outlet from fluid machinery 13 Impeller anemometer wheel 14...Inflow grid 15... Impeller anemometer blades 16 Inlet nozzle 17. Directional blade 18 Support plate 19 Nozzle plate 20... Diameter expansion section 21 Rotor operating area 22 Middle ring of swerving wheel 23 - Steering wheel hub ring 24 Fan assembly, fluid machine assembly 25 Support column 26 Backflow prevention member

Claims

1. An acoustic control method for actively controlling the acoustic emission of a fluid machine, particularly a fluid machine having an electric motor, preferably a fan or turbomachine, The acoustic signal generated by superimposing the acoustic emission from the fluid machine with at least one inverse acoustic signal is recorded by at least one receiver at at least one receiver position and transmitted to a control unit. The control unit has artificial intelligence, A control signal for at least one actuator is generated by the artificial intelligence taking into account the acoustic signal, so that the actuator generates the inverse acoustic signal that cooperates with the acoustic emission of the fluid machine, thereby reducing the acoustic load in at least the region of the receiver position. At least two state values ​​of the fluid machine are transmitted to the control unit. An acoustic control method for actively controlling the acoustic emission of a fluid machine, wherein the control signal is generated by the artificial intelligence taking into consideration the at least two state values.

2. An acoustic control method for actively controlling the acoustic emission of a fluid machine according to claim 1, characterized in that the time signal and / or frequency range and / or phase position of the inverse acoustic signal generated by the actuator are controlled by the control signal.

3. A method for actively controlling the acoustic emission of a fluid machine according to claim 1, characterized in that the artificial intelligence uses a reinforcement learning method to generate the control signal.

4. An acoustic control method for actively controlling the acoustic emission of a fluid machine according to claim 1, characterized in that the artificial intelligence is pre-trained in a factory.

5. An acoustic control method for actively controlling the acoustic emission of a fluid machine according to claim 1, characterized in that one or more of the at least two state values ​​are measured values ​​from the corresponding sensors.

6. The aforementioned at least two state values ​​are The motor speed of the fluid machine and the speed of the impeller anemometer, The motor speed of the fluid machine and the signal from the hot-wire anemometer, The motor speed of the fluid machine and the motor current of the fluid machine, The motor current of the fluid machine and the speed of the impeller anemometer, The motor current of the aforementioned fluid machine and the signal from the hot-wire anemometer, The motor speed of the fluid machine and the differential pressure between the upstream and downstream of the fluid machine, or the motor current of the fluid machine and the differential pressure between the upstream and downstream of the fluid machine, An acoustic control method for actively controlling the acoustic emission of a fluid machine as described in claim 1, characterized in that...

7. An acoustic control method for actively controlling the acoustic emission of a fluid machine according to claim 1, characterized in that a microphone is used as the receiver and / or a speaker is used as the actuator.

8. Acoustic control method for actively controlling the acoustic emission of a fluid machine according to claim 1, wherein the actuator excites the components of the fluid machine to emit sound by modulating the excitation current or excitation voltage of an electric motor using a piezoelectric actuator and / or an appropriate superposition excitation signal.

9. An acoustic control system for performing an acoustic control method for actively controlling the acoustic emission of a fluid machine as described in any one of claims 1 to 8, comprising: a fluid machine, in particular a fluid machine having an electric motor, preferably a fan or turbomachine; at least one receiver for detecting an acoustic signal at at least one receiver position, wherein the acoustic signal is generated by superimposing an acoustic emission generated by the fluid machine and at least one inverse acoustic signal; a control unit; and at least one actuator, the system comprising: The control unit has artificial intelligence, An acoustic control system in which artificial intelligence controls the actuator taking into account the detected acoustic signal and at least two state values ​​of the fluid machine to generate an inverse acoustic signal that cooperates with the acoustic emission of the fluid machine, thereby reducing the acoustic load in at least the region of the receiver position.

10. The acoustic control system according to claim 9, characterized in that the control unit is a component integrated with the fluid machine, or the control unit is a separate control module.

11. An acoustic control device comprising: at least one receiver for detecting an acoustic signal at at least one receiver position, for performing an acoustic control method for actively controlling the acoustic emission of a fluid machine as described in any one of claims 1 to 8, wherein the acoustic signal is generated by superimposing an acoustic emission generated by the fluid machine and at least one inverse acoustic signal; a control unit; and at least one actuator; The control unit has artificial intelligence, An acoustic control device in which artificial intelligence controls the actuator considering the detected acoustic signal and at least two state values ​​of the fluid machine, so that the actuator generates an inverse acoustic signal that cooperates with the acoustic emission of the fluid machine, thereby reducing the acoustic load in at least the region of the receiver position.

Images