Acoustic analysis of machine status
By providing individual acoustic signals to machine components and generating reference signals, the operating noise of each component can be analyzed independently, solving the problems of large data volume and computational complexity in existing technologies, and realizing efficient and automated machine condition analysis.
Patent Information
- Application Number
- CN201980097893.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-06-26
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2039-06-26
AI Technical Summary
Existing technologies require large amounts of data and high computing power for acoustic analysis of machine conditions, and the instruments are expensive and lack automation, limiting their application scope.
By providing individual acoustic signals to each component of the machine, generating reference signals using computing units, independently analyzing the operating noise of each component, reducing interference with the overall noise, using a small number of sound converters to record the machine's operating noise, and optimizing signal processing in conjunction with models and environmental conditions.
It reduces instrument consumption and data requirements, enables efficient machine status analysis, can be automated on-site, and reduces computational complexity and cost.
Smart Images

Figure CN114026403B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method for generating at least one reference signal for acoustically analyzing a machine state, a method for acoustically analyzing a machine state, wherein acoustic operating signals are recorded during operation of the machine, a corresponding analysis device, as well as a computer program and a computer-readable storage medium. Background Art
[0002] For example, in industrial manufacturing, data analysis methods are becoming increasingly common, which also aim to predictively plan maintenance work on machines, in order to be able to thus minimize downtimes or quality losses in the manufacturing process (“predictive maintenance”).
[0003] In known methods, state data of machines associated with process detection are analyzed using machine learning methods and an attempt is made to deduce meaningful trends. This may require a large amount of data in order to be able to draw reliable predictions, which in turn requires a correspondingly high computing power.
[0004] BE 1 018 513 A3 already discloses an operating method for early detection of machine faults by means of acoustic features or features recorded by means of a plurality of microphones.
[0005] Further disadvantages of the known methods are that they in some cases require a great deal of equipment effort, have a limited range of application or lack automation. Summary of the invention
[0006] Against this background, the object of the present invention is to specify an improved concept for the acoustic analysis of machine states, according to which the amount of data to be processed and the computational effort required for the processing can be reduced.
[0007] According to the invention, these objects are achieved by the subject matter of the independent claims. Advantageous developments and embodiments are the subject matter of the dependent claims.
[0008] The design of this improvement is based on the idea of providing each component of the machine with an acoustic individual signal and generating a reference signal based on this signal.
[0009] According to a first independent aspect of the improved design, a method for generating at least one reference signal for acoustic analysis of a machine state is provided. At least one acoustic individual signal is provided for each of at least two components of the machine. An acoustic overall signal to be expected is determined based on the individual signals by means of a calculation unit. A reference signal is generated based on the overall signal by means of the calculation unit.
[0010] Acoustic analysis of a machine can be understood in particular as an analysis of the operating noise of the machine. In particular, the acoustic analysis includes an analysis of the airborne sound waves generated by the operation of the machine.
[0011] The computing unit can include, for example, one or more subunits or processor units which can be arranged in a spatially distributed manner. In particular, different method steps can be performed by different subunits of the computing unit.
[0012] The component of the machine is in particular a noise-generating component of the machine. For example, the component can have a motor, a transmission device and / or other electrical components or mechanically movable parts.
[0013] Providing the acoustic individual signals in particular comprises generating the respective acoustic individual signals. In this case, the individual signals can be generated by detecting the respective individual operating noises of the respective components by means of a sound converter. Alternatively, supplementally or additionally, the generation of the individual signals can comprise simulating the acoustic individual signals, i.e., artificially generating the acoustic individual signals to a certain extent.
[0014] Here and below, a sound converter can be understood as a sound sensor or a sound receiver, ie a sensor unit or a receiving unit which is provided for converting airborne sound into an electrical signal. In particular, the sound converter can be designed as a microphone.
[0015] The sound converter or the computing unit can be part of an analysis device for acoustic analysis of the machine state, for example.
[0016] According to at least one embodiment, for each of the at least two components of the machine, at least one associated acoustic individual signal is generated as a function of the respective individual operating noise of the respective component, in particular by means of a sound transducer.
[0017] In this case, in a reference state or new state of the machine, acoustic individual signals are generated, in particular based on individual operating noises of the respective components.
[0018] In particular, an associated acoustic individual signal is provided or generated for each component of the machine and for each individual operating state of the component, wherein each individual operating state of the component is selected from a predetermined set of individual operating states of the component.
[0019] The acoustic individual signal for a given component of the at least two components is in particular independent of the other components of the at least two components. This means that the other components of the at least two components do not contribute to the respective individual operating noise of the given component. In other words, the individual operating noise of each component is detected, in particular during periods when the other components of the at least two components are not in operation or when the other components are acoustically isolated from the given component.
[0020] Here and hereinafter, recording or receiving noises, in particular operating noises or individual operating noises, can be understood as detecting the respective operating noise, in particular detecting the respective operating noise by means of at least one sound transducer and subsequently storing the resulting signal on a computer-readable storage medium, for example on a computing unit.
[0021] The individual signals can in particular be provided as time-dependent signals or as signals in the frequency domain. Correspondingly, the total signal and / or the reference signal can each be generated as a time-dependent signal or as a signal in the frequency domain.
[0022] By generating or providing one or more individual associated acoustic individual signals for each of the at least two components and using the individual signals as a basis for generating the overall signal and the reference signal, the individual components of the machine are considered individually and independently of one another in order to generate the reference signal. As a result, additional influences that contribute to the overall signal or the reference signal, such as environmental influences or interfering influences, can be taken into account in a targeted manner, for example based on empirical data, simulation data or a theoretical or partially theoretical model.
[0023] By means of known acoustic individual signals, which can be interpreted as nominal signals or nominal noises of individual components in a reference state or a new state of the machine, it is only necessary to detect the total operating noise of the entire machine in the subsequent analysis of the machine state with reference to the reference signal. In particular, it is not necessary to also record the individual operating noises of the individual components within the scope of the analysis of the machine state, since the influence of the individual components on the total noise or the total operating noise of the machine can be separated or isolated. Accordingly, for the analysis of the machine state, the operating noise can be recorded relatively far from the components, without the need to record the noise in the vicinity of the components.
[0024] Compared with other acoustic analysis methods, the following advantages are obtained: According to the improved design, less instrumentation is required, in particular this relates to the number and positioning of one or more sound converters. In particular, according to the improved design, the sound converter can be sufficient to record the operating noise of the machine to analyze the state.
[0025] According to at least one embodiment, a respective first individual signal is recorded for each of the at least two components, while the respective component is in a respective first individual operating state. The overall signal is determined by means of a calculation unit from the first individual signals. In this case, a combination, in particular a predefined combination, of the first individual operating states of the at least two components, in particular all of the at least two components, defines a first operating state of the machine.
[0026] In other words, a first single state is defined for each of the at least two components. The combination of all first single states defines or forms a first operating state of the machine. In particular, each possible operating state of the machine can be defined as a predefined combination of the corresponding single operating states of the at least two components of the machine.
[0027] The individual operating states of at least one of the two components can include, for example, one or more predetermined operating parameters of the component and / or one or more environmental characteristics of the component. Depending on the type of machine and component, the operating parameters can include, for example, rotational speed, transmission ratio, feed speed, load, etc. The environmental characteristics of the component can be, for example, the temperature or other environmental conditions in the environment of the component.
[0028] The overall signal is generated by taking into account first individual signals corresponding to respective individual operating states of the components which together form a first operating state of the machine, for example by determining the overall signal from the first individual signals.
[0029] By adapting or changing individual operating states or operating states of the machine, the model used to generate the overall signal can be improved, thereby improving the accuracy of the reference signal and thus the accuracy of the analysis of the machine state.
[0030] According to at least one embodiment, the second operating state of the machine is given by a defined combination of the respective second individual operating states of the at least two components. A respective second individual signal is recorded for each of the at least two components while the respective component is in the respective second individual operating state. The overall signal is determined from the second individual signals.
[0031] In particular, the second operating state of the machine is different from the first operating state of the machine. Accordingly, at least one of the second individual operating states of the at least two components is different from the corresponding first individual operating state of the respective component. It is possible, but not absolutely necessary, that all first individual operating states are different from the corresponding second individual operating states.
[0032] According to at least one embodiment, the reference signal is equal to the total signal.
[0033] According to at least one specific embodiment, the reference signal is generated by determining a plurality of total signals for different operating states of the machine, as described above, and generating the reference signal from the plurality of total signals, for example by interpolation.
[0034] It is thus not necessary to determine a separate acoustic total signal for each relevant operating state of the machine (which should, if necessary, be acoustically analyzed) by combining the separately recorded individual signals. Rather, it is sufficient to select a representative operating state of the machine as an interpolation point to a certain extent and to generate the total signal for this interpolation point as described above. The reference signal of the representative operating state then corresponds to the respective total signal, while the reference signal of the operating state which does not belong to the representative operating state is generated from the total signals of two or more representative operating states, for example by interpolation or combination.
[0035] According to at least one embodiment, an overall signal is generated from the individual signals based on a model.
[0036] In the simplest case, the model can consist in adding the individual individual signals or adding them in a weighted manner, wherein the model includes corresponding weighting factors. In more complex embodiments, in addition to the individual signals, the model can also take into account other influences, such as interference noise or other disturbances, the geometry of the machine, the ambient conditions of the machine and / or the material-specific noise characteristics of the machine, in order to generate an overall signal.
[0037] According to at least one embodiment, an acoustic total signal is determined from the position of one of the at least two components or from two or more positions of two or more of the at least two components by means of a computing unit.
[0038] The pose of a component can be understood here and below as a combination of the position and orientation of the respective component in three-dimensional space. In this case, the position and orientation of the component can be defined relative to one or more other components of the at least two components, relative to other components of the machine or relative to objects in the machine environment, such as a building floor, a building wall or another machine.
[0039] The pose of the component can influence how the individual signals are weighted to form the total signal to be expected. In addition, the influence of external influences can also be related to the pose. Therefore, by taking the pose into account, an improved accuracy of the reference signal and ultimately an improved accuracy of the machine state analysis is achieved.
[0040] According to at least one embodiment, the pose of one of the at least two components or the poses of all of the at least two components are stored on a storage medium, for example on a computing unit. The computing unit can access one or more stored poses to generate the overall signal.
[0041] According to at least one specific embodiment, the acoustic overall signal is determined by means of a computing unit as a function of nominal ambient conditions of the machine.
[0042] The nominal environmental conditions are in particular not measured values, but one or more model parameters for generating or determining the total signal from the individual signals. For example, depending on the environmental conditions, different individual signals can be reflected differently in the total signal and / or external influences on the total signal can be related to the environmental conditions.
[0043] Nominal ambient conditions may be, for example, temperature or air humidity in the machine environment, secondary or interfering noise, the spatial size or spatial properties of the space in which the machine is arranged and / or the type of machine or other object and its position or orientation in space.
[0044] By taking the nominal ambient conditions into account in the model used to determine the total signal, the accuracy in generating the total signal can be improved.
[0045] According to at least one embodiment, the current environmental conditions of the machine are determined, in particular measured, by means of a sensor device, in particular an evaluation device. A reference signal is determined by means of a calculation unit as a function of the current environmental conditions, in particular as a function of the difference between the current environmental conditions and nominal environmental conditions.
[0046] By taking into account actual ambient conditions or deviations of actual ambient conditions from nominal ambient conditions, the determination of the total signal can be further improved.
[0047] According to a further independent aspect of the improved design, a method for acoustically analyzing a machine state is provided, wherein an acoustic operating signal is recorded during operation of the machine, in particular based on the acoustic operating noise of the machine, in particular by means of at least one sound transducer. Based on the method according to the improved design for generating at least one reference signal for acoustic analysis of the machine state, a reference signal is provided. The machine state is analyzed based on deviations of the operating signal from the reference signal.
[0048] The reference signal is provided in particular by generating the reference signal by means of the method for generating a reference signal according to an improved embodiment.
[0049] As already mentioned above, a single acoustic transducer or a few acoustic transducers are sufficient to be able to record operating signals and nevertheless carry out a component-specific analysis of the machine state.
[0050] Compared to analysis methods based on structure-borne sound measurements or thermographic measurements, the method according to the improved design reduces the instrumentation expenditure.
[0051] Furthermore, by means of the improved design, the condition analysis can be carried out automatically, since no manual checks are necessary, for example for wear residues in the engine oil, as is the case with ferrography.
[0052] A further advantage of the improved design is that the method can be used universally for almost any machine having noise-emitting components.
[0053] By means of the analysis according to the improved design, in particular by means of the analysis using the reference signal generated according to the improved design, the amount of data that must be examined for the analysis is relatively small. Therefore, complex data analysis methods, such as those used in deep learning or machine learning, can be dispensed with. This also results in a reduction in the computational effort for the analysis. Accordingly, it is particularly not necessary to process the data to be analyzed in a central backend server or cloud server. Rather, the analysis can be performed on site (“on premises”).
[0054] According to at least one embodiment of the method for acoustically analyzing a machine state according to an improved embodiment, the current operating state of the machine is determined by means of a computing unit. A reference signal is provided by means of the computing unit as a function of the current operating state.
[0055] In order to determine the current operating state, for example, the current operating parameters can be read from the control unit of the machine, in particular from the control program for controlling the machine, by means of the computing unit. For this purpose, the computing unit can be coupled in particular to the control unit of the machine.
[0056] For example, the current operating state of the machine is determined deterministically at each point in time by a control program for the machine. Thus, by providing reference signals accordingly, the expected nominal acoustic behavior of the machine can be predicted within a very narrow time frame.
[0057] In order to provide a reference signal as a function of the current operating state, a total signal generated for a representative operating state can be selected as a reference signal, or a reference signal can be generated as a function of the total signals of two or more representative operating states, as described above. In this case, after determining the current operating state, the respective total signal can be generated with the aid of a model or can have been generated and stored in advance.
[0058] According to at least one embodiment, a differential signal is generated by means of a computing unit based on a deviation of the operating signal from a reference signal. The state is evaluated by means of the computing unit based on the differential signal.
[0059] This makes it possible to reduce the influence of noise and also reduce the complexity of the evaluation.
[0060] According to at least one embodiment, a frequency signal is generated based on the difference signal by means of a computing unit. The state is analyzed based on the frequency signal by means of a computing unit.
[0061] In order to generate the frequency signal, the difference signal is in particular subjected to a Fourier transformation.
[0062] In this embodiment, the difference signal and in particular the reference signal and the operating signal are present as time-dependent signals.
[0063] For example, in the frequency domain, the noise contribution can be masked or removed, thereby making it possible to more clearly identify abnormal signal features or signal features attributable to equipment wear.
[0064] According to at least one specific embodiment, a check is carried out using a computer unit based on a state analysis to determine whether anomalies are present in the operation of the machine.
[0065] An anomaly is in particular a deviation of the current operating state of a machine or of a current individual operating state of a component from an expected state. For example, an anomaly can be attributed to an error or a defect.
[0066] According to at least one embodiment, the computing unit identifies, based on the analysis, which of the at least two components exhibits an abnormality.
[0067] This is possible in particular due to the possible separation of the individual noise characteristics of the components.
[0068] According to at least one specific embodiment, the degree of wear of the machine is determined by means of a computing unit based on an analysis of the state.
[0069] According to at least one specific embodiment, the individual degree of wear of one or more of the at least two components is determined by means of the computing unit based on the evaluation of the state.
[0070] In particular, it is possible to distinguish between possible anomalies and signs of wear based on typical features in the differential signal or the frequency signal.
[0071] According to an improved embodiment, further embodiments of the method for generating a reference signal follow directly from the various embodiments of the method for analyzing a machine state according to an improved embodiment, and vice versa.
[0072] According to another independent aspect of the improved design, an analysis device for acoustic analysis of a machine state is provided, wherein the analysis device has at least one sound converter and a computing unit. The at least one sound converter is configured to generate a respective acoustic individual signal from respective individual operating noises of at least two components of the machine. The computing unit is configured to determine an acoustic total signal to be expected based on the individual signals. The computing unit is configured to generate a reference signal from the total signal.
[0073] For example, at least one sound converter can be constituted by just one sound converter. In this embodiment, the sound converter can be positioned at different positions, for example, to record respective single operating noises and / or the sound converter can be used to record respective operating noises of different components in sequence.
[0074] Alternatively, the at least one sound transducer can include two or more sound transducers, wherein one or more of the sound transducers are used to receive each of the individual operating noises.
[0075] The acoustic individual signals are stored in particular on a storage medium, in particular by means of a computing unit.
[0076] In other words, at least two acoustic individual signals are generated, wherein each of the at least two acoustic individual signals belongs to an associated component of the at least two components.
[0077] According to at least one embodiment, the evaluation device has a sensor device which is configured to determine current environmental conditions of the machine. The calculation unit is configured to determine the overall signal as a function of the current environmental conditions.
[0078] According to at least one embodiment, at least one sound converter is configured to generate an acoustic operating signal as a function of the operating noise of the machine. The computing unit is configured to evaluate the state based on a deviation of the operating signal from a reference signal.
[0079] The operating noise of the machine is, in particular, the total operating noise of the machine, ie, the operating noise jointly generated by a plurality of components of the at least two components.
[0080] Other embodiments of the analysis device are derived directly from the described embodiments of the various methods according to the improved design, and vice versa. In particular, the analysis device according to the improved design can be configured or programmed to: perform the method according to the improved design, or the analysis device performs the method according to the improved design.
[0081] According to another independent aspect of the improved design, a computer program with instructions is provided. If the computer program is executed by a computer system, in particular by an analysis device according to the improved design, such as a computing unit of the analysis device, the instructions cause the computer system, in particular the analysis device, such as a computing unit, to perform the method according to the improved design for generating at least one reference signal for acoustic analysis of a machine state.
[0082] According to another independent aspect of the improved design, a further computer program with further instructions is provided. If the further computer program is executed by the analysis device according to the improved design, in particular by the computing unit of the analysis device, the further instructions cause the analysis device, in particular the computing unit, to execute the method according to the improved design for acoustic analysis of machine states.
[0083] According to a further independent aspect of the improved design, a computer-readable storage medium is proposed, on which the computer program according to the improved design and / or a further computer program is stored.
[0084] The features and feature combinations mentioned above in the description and the features and feature combinations mentioned below in the description of the figures and / or shown individually in the figures can be used not only in the respectively specified combination but also in other combinations without departing from the scope of the invention. Embodiments and feature combinations are also regarded as disclosed which do not have all the features of the originally formulated independent claims and / or which exceed or deviate from the feature combinations stated in the reference to the claims. BRIEF DESCRIPTION OF THE DRAWINGS
[0085] The present invention will be explained in more detail below based on specific embodiments and the associated schematic diagrams. Identical or functionally identical elements may be provided with the same reference numerals in the drawings. The description of identical or functionally identical elements may not necessarily be repeated in different drawings if necessary.
[0086] In the accompanying drawings:
[0087] Figure 1 A schematic diagram showing an exemplary embodiment of an analysis device according to an improved design;
[0088] Figure 2 a flow chart showing an exemplary embodiment of a method according to an improved design, in particular a method for analyzing a machine state according to an improved design; and
[0089] Figure 3 Show according to Figure 2 A signal in the context of a method. DETAILED DESCRIPTION
[0090] exist Figure 1 8 is shown, which can be designed, for example, as an industrial robot or as a machine for additive or subtractive production of components.
[0091] The machine 8 has, for example, a plurality of components. Figure 1 In the schematic example of the present invention, there are three components 14, 15, 16. The components 14, 15, 16 can have, for example, one or more corresponding motors or other noise-emitting components. Here, noise-emitting components are only described as such when, for example, noise is generated during the operation of the machine 8, in particular in a certain operating state of the machine 8.
[0092] The machine 8 also has a control unit 18 which is coupled to at least one of the components 14 , 15 , 16 in order primarily to control it.
[0093] exist Figure 1 2 also shows a schematic diagram of an exemplary embodiment of an analysis device 9 for acoustically analyzing the state of a machine 8 .
[0094] The analysis device 9 has a computing unit 13, which is configured, for example, as a computer system having one or more processor units. For example, the computing unit 13 is coupled to a control unit 18. In addition, the analysis device 9 has at least one sound converter. Figure 1 In the non-limiting example of , there are for example three microphones 10 , 11 , 12 . The first microphone 10 can be arranged in the environment of the first component 14 , the second microphone 11 can be arranged in the environment of the second component 15 , and the third microphone 12 can be arranged in the environment of the third component 16 , for example.
[0095] The described arrangement of microphones 10, 11, 12 can be provided, for example, as shown, for carrying out a method for generating at least one reference signal according to an improved design. In particular, the described arrangement of microphones 10, 11, 12 can be used to record the respective acoustic individual signals of components 14, 15, 16 in order to generate a reference signal.
[0096] After recording the acoustic individual signals, one or more or all of the microphones 10, 11, 12 can be arranged offset, in particular in order to record the operating noise of the machine 8, in particular the overall operating noise of the machine 8, in a further step of the method for acoustic analysis according to the improved design. For this purpose, in various embodiments, all microphones 10, 11, 12 can be used. In alternative embodiments, not all microphones 10, 11, 12 are used, for example only one of the microphones 10, 11, 12 is used to record the operating noise of the machine 8.
[0097] The evaluation device 9 optionally has a sensor device 17 which is connected or coupled to the computing unit 13 , in particular in order to determine current ambient conditions of the machine 8 , for example the ambient temperature.
[0098] The computing unit 13 optionally has a computer-readable storage medium 19 , in particular a computer-readable storage medium according to an improved embodiment, or the computing unit 13 is coupled to such a computer-readable storage medium 19 .
[0099] For example, the computing unit 13 can execute a computer program according to the improved embodiment, which is stored on a computer-readable storage medium 19 , in order to cause the evaluation device 9 to carry out the method according to the improved embodiment.
[0100] Figure 1The mode of operation of the analysis device 9 is explained below with reference to an exemplary embodiment of a method for acoustic analysis of a machine state according to an improved design, in particular with reference to Figure 2 , Figure 3 The method steps 1 - 7 described here do not necessarily have to be performed in the order suggested by their numbering.
[0101] exist Figure 2 Detailed description of the invention A flow chart of an exemplary embodiment of a method for acoustically analyzing the state of a machine 8 according to an improved design is shown in FIG.
[0102] In step 1 of the method, for example, acoustic individual signals of the individual components 14, 15, 16 are recorded by means of the corresponding microphones 10, 11, 12. The individual signals of the respective components 14, 15, 16 are recorded in particular isolated from the remaining components. For example, if the individual signals of the first component 14 are recorded, any operating noises of the other components 15, 16 are shielded by the first microphone 10. If possible, during the recording of the individual signals of the first component 14, individual components of the remaining components 15, 16 can also be stopped. For example, individual signals are generated for different individual operating states of the components 14, 15, 16.
[0103] Instead of using a dedicated microphone 10 , 11 , 12 for each of the components 14 , 15 , 16 , it is also possible to record the individual signals of all components 14 , 15 , 16 with a small number of microphones, for example only one microphone, for example successively.
[0104] In optional step 2 of the method, the environmental conditions of the machine 8 are detected, for example, by means of the sensor device 17. Alternatively or additionally, in optional step 2, information about the arrangement and / or orientation (e.g. relative to one another), i.e. the position and / or orientation, of the components 14, 15, 16 can be loaded by means of the computing unit 13, for example from a storage medium 19 or from the control unit 18.
[0105] In step 3 of the method, the current operating state of machine 8 is determined. For this purpose, computer unit 13 can access control unit 18 , for example a control program, which is executed by control unit 18 to determine current operating parameters or state parameters of machine 8 .
[0106] In step 4, information from steps 1 and 3, and optionally from step 2, is processed to provide a reference signal.
[0107] In particular, the calculation unit 13 determines, depending on the current operating state of the machine 8, which of the individual signals are combined with each other and, if necessary, with other information, such as the information from step 2 or other information about interfering influences or environmental influences, to generate a reference signal. Alternatively, different reference signals can also be generated in advance, and the calculation unit 13 selects one of them depending on the current operating state.
[0108] In step 5 of the method, operating noises are detected during the current operating state of machine 8 by means of at least one of microphones 10 , 11 , 12 or by means of one of the further microphones (not shown) and recorded as an operating signal.
[0109] In step 6 of the method, the computing unit 13 generates a difference signal 22 , for example based on a deviation of the operating signal 21 from a reference signal.
[0110] Likewise, in step 6 , the difference signal 22 can be converted into a frequency signal, for example, in particular by means of a Fourier transformation of the difference signal 22 .
[0111] In step 7 , the computing unit 13 can, for example, analyze the difference signal 22 or the frequency signal in order to analyze the state of the machine 8 .
[0112] For example, the computing unit 13 can determine from the difference signal or the frequency signal whether an anomaly is present in the operation of the machine 8 and, if necessary, which component 14 , 15 , 16 caused the anomaly.
[0113] Alternatively or additionally, the computing unit 13 can infer a degree of wear of the machine 8 and / or a respective degree of wear of one or more of the components 14 , 15 , 16 based on the difference signal 22 or the frequency signal.
[0114] exist Figure 3 , the described signals are schematically shown. Figure 3 The respective signal amplitudes of the operating signal 21 , the inverted reference signal 20 and the difference signal 22 are shown as a function of time. The difference signal results as the sum of the signals 21 , 20 .
[0115] According to the improved design, a possibility for analyzing the machine status or a method for generating corresponding reference signals is proposed, which method does not require a large amount of data and correspondingly does not require high computing power, and thus the method allows predictive planning of maintenance work in order to minimize downtimes or quality losses in manufacturing processes carried out with the aid of the machine.
[0116] According to the improved design, an acoustic method for early identification of wear or anomalies of mechanical machine components is proposed, in which the deviation between the operating noise of the machine recorded by one or more microphones accompanying the process and the expected nominal noise is taken into account as a measure of the wear state or the presence of anomalies. Here, the nominal noise is predicted as a reference signal. For example, the modeling of the reference signal or the generation of the reference signal can be regarded as a description of the digital twin of the machine to be analyzed. With the help of the digital twin, the prediction of the development of the machine noise uses the concept of determinism about the nominal noise of different operating states of the machine, so that the association with different wear levels and a smaller database already provide more reliable conclusions compared to conventional solutions.
[0117] Reference numerals list
[0118] 1 Methods and Steps
[0119] 2 Methods and Steps
[0120] 3 Methods and Steps
[0121] 4 Methods and Steps
[0122] 5 Methods and Steps
[0123] 6 Methods and Steps
[0124] 7 Methods and Steps
[0125] 8 Machines
[0126] 9 Analysis device
[0127] 10 Microphone
[0128] 11 Microphone
[0129] 12 Microphone
[0130] 13 Computational Unit
[0131] 14 Components
[0132] 15 Components
[0133] 16 Components
[0134] 17 Sensor device
[0135] 18 Control Unit
[0136] 19 Storage Media
[0137] 20 Reverse reference signal
[0138] 21 Operation signal
[0139] 22 differential signals.
Claims
1. A method for generating at least one reference acoustic signal (20) for acoustically analyzing a state of a machine (8), The method comprises: recording acoustic individual signals by means of at least one sound converter, the acoustic individual signals comprising at least one acoustic individual signal for each of at least two components (14, 15, 16) of the machine (8); Recognizing, by means of a computing unit (13), a pose of one of the at least two components from a storage medium, the pose being a combination of a position and an orientation of the respective component in three-dimensional space; determining an expected acoustic total signal based on the acoustic individual signals and the posture of the component by means of a calculation unit (13), wherein the determination of the expected acoustic total signal includes weighting at least one acoustic individual signal of the component based on the recognized posture; and The at least one reference acoustic signal (20) is generated by means of the computing unit (13) as a function of the acoustic total signal to be expected, wherein anomalies in the operation of the machine can be detected based on the at least one reference acoustic signal.
2. The method according to claim 1, in, recording a respective first individual signal of the acoustic individual signal for each of the at least two components (14, 15, 16), while the respective component (14, 15, 16) is in a respective first individual operating state, wherein a combination of first individual operating states of at least two components (14, 15, 16) defines an operating state of the machine (8); and Therein, determining the acoustic total signal to be expected comprises determining the acoustic total signal to be expected from the first individual signals by means of the calculation unit (13).
3. The method according to claim 1, in, The acoustic total signal to be expected is determined by means of the computing unit (13) as a function of nominal ambient conditions of the machine (8).
4. The method according to claim 1, Also includes: determining current environmental conditions of the machine (8) by means of a sensor device (17); and The at least one reference acoustic signal (20) is determined by means of the computing unit (13) as a function of the current ambient conditions.
5. A method for acoustically analyzing the state of a machine (8), the method comprising: recording an acoustic operating signal (21) by means of at least one sound transducer during operation of the machine (8); Providing a reference acoustic signal, wherein providing the reference acoustic signal comprises generating the reference acoustic signal, wherein generating the reference acoustic signal comprises: - recording at least one acoustic individual signal for each of at least two components (14, 15, 16) of the machine (8) by means of at least one sound transducer; - identifying, by means of a computing unit (13), a pose of one of the at least two components from a storage medium, said pose being a combination of the position and orientation of the respective component in three-dimensional space; - determining, by means of a calculation unit (13), an acoustic total signal to be expected based on at least one of the acoustic individual signals for each of the at least two components (14, 15, 16) of the machine (8) and the posture of the components, the determination of the acoustic total signal to be expected comprising weighting the at least one acoustic individual signal of the component based on the recognized posture; and - generating the reference acoustic signal (20) by means of the calculation unit (13) as a function of the acoustic total signal to be expected; analyzing a state of the machine based on a deviation of the acoustic operating signal and the reference acoustic signal; and Based on the analysis, anomalies in the operation of the machine are detected with the aid of the computing unit.
6. The method according to claim 5, Also includes: determining the current operating state of the machine (8) by means of a computing unit (13); and The reference acoustic signal (20) is provided according to the current operating state.
7. The method according to claim 5, Also includes: generating a differential signal (22) based on a deviation of the acoustic operating signal (21) from the reference acoustic signal (20) by means of the computing unit (13); and The state of the machine is evaluated based on the difference signal (22) by means of the computing unit (13).
8. The method according to claim 7, Also includes: generating a frequency signal based on the difference signal (22) by means of the calculation unit (13); and The state of the machine is analyzed based on the frequency signal by means of the computing unit (13).
9. The method according to claim 5, Also includes: By means of the computing unit (13), based on an analysis of the state of the machine, it is checked whether an anomaly is present in the operation of the machine (8); and / or The degree of wear of the machine (8) is determined by means of the computing unit (13) based on an analysis of the state of the machine.
10. An analysis device for acoustically analyzing the state of a machine (8), the analysis device (9) comprising: at least one sound converter (10, 11, 12), which is configured to generate a respective acoustic individual signal as a function of a respective individual operating noise of at least two components (14, 15, 16) of the machine (8); A storage medium configured to store data representing a posture of one of at least two components, wherein the posture is a combination of a position and an orientation of the respective components in three-dimensional space, a sensor device configured to determine current environmental conditions of the machine, A computing unit configured for - recognize the pose of the component from a storage medium, - determining an acoustic total signal to be expected based on the acoustic individual signals, the posture of the component and current environmental conditions of the machine, the calculation unit being configured to determine the acoustic total signal to be expected comprising the calculation unit being configured to weight at least one acoustic individual signal of the component based on the recognized posture, - generating a reference acoustic signal (20) based on the acoustic total signal to be expected, and - Based on the reference acoustic signal, identifying anomalies in the operation of the machine.
11. The analysis device according to claim 10, in, The at least one sound converter (10, 11, 12) is configured to generate an acoustic operating signal (21) based on operating noise of the machine (8); and The computing unit (13) is configured to analyze the state of the machine based on a deviation of the acoustic operating signal (21) from the reference acoustic signal (20).
12. A computer-readable storage medium storing instructions executable by one or more processors for generating at least one reference acoustic signal for acoustic analysis of a state of a machine, the instructions comprising: recording at least one acoustic individual signal for each of at least two components (14, 15, 16) of the machine (8) by means of at least one sound transducer; Recognizing, by means of a computing unit (13), a pose of one of the at least two components from a storage medium, the pose being a combination of a position and an orientation of the respective component in three-dimensional space; determining an expected acoustic total signal based on the acoustic individual signals and the posture of the component by means of a calculation unit (13), wherein the determination of the expected acoustic total signal includes weighting at least one acoustic individual signal of the component based on the recognized posture; and generating the at least one reference acoustic signal (20) by means of the computing unit (13) as a function of the acoustic total signal to be expected, Therein, based on the at least one reference acoustic signal, anomalies in the operation of the machine can be identified.
Citation Information
Patent Citations
PROCEDURE FOR THE EARLY DETECTION OF DAMAGE TO A VEHICLE, MACHINE, OR PART THEREOF, AND A DETECTION SYSTEM APPLIED IN THIS MANNER.
BE1018513A3
Method for monitoring a vicinity using several acoustic sensors
CN102473338A
Method, system and computer program for the acoustic analysis of a machine
CN105659064A
Scraper conveyor chain fluttering fault diagnosis method
CN106404388A
Detection of anomalies in rail wheelsets
CN106662483A
