Processing system, storage device, and processing method

Abstract

The invention relates to a processing system, a storage device and a processing method. Provided are a processing system and the like that can perform state monitoring, quality management, or predictive maintenance of an object on the basis of vibration analysis using multi-channel vibration information. A processing system (100) is provided with an acquisition unit (111), a calculation unit (112), and an output unit (114). The acquisition unit (111) acquires first vibration information of a first channel and second vibration information of a second channel regarding vibration of an object. The calculation unit (112) calculates a spectrogram including at least one of a cross-power spectrogram, a cross-phase spectrogram, and a coherence spectrogram of the first vibration information and the second vibration information. The output unit (114) outputs presentation information on at least one of state monitoring, quality management, and predictive maintenance for the object on the basis of the spectrogram.

Description

Technical Field

[0001] The present invention relates to a processing system, a program, a processing method, etc. Background Art

[0002] In Patent Document 1, a method for inspecting an automotive component having an operating part such as an electric motor as an object to be inspected is disclosed. In the inspection method, waveform data of the operating sound of the object to be inspected is acquired by a microphone disposed near the object to be inspected, the waveform data is subjected to a short-time Fourier transform to generate a complex spectrogram, a prescribed phase feature amount is calculated based on the complex spectrogram, a group delay is calculated by performing a differential operation on the phase feature amount in the frequency direction, a smoothing filter process is performed on the group delay, and the intensity level of the abnormal sound component is calculated based on the group delay after the smoothing filter process.

[0003] Patent Document 1: Japanese Unexamined Patent Application Publication No. 2022-154180

[0004] In Patent Document 1, waveform data of one channel acquired by a microphone is used for inspection. In Patent Document 1, state monitoring, quality management, or predictive maintenance of an object based on vibration analysis using multi-channel vibration information is not disclosed. Summary of the Invention

[0005] One aspect of the present disclosure relates to a processing system including: an acquisition unit that acquires first vibration information of a first channel and second vibration information of a second channel regarding vibration of an object; an operation unit that operates a spectrogram including at least one of a cross-power spectrogram, a cross-phase spectrogram, and a coherence spectrogram of the first vibration information and the second vibration information; and an output unit that outputs, based on the spectrogram, prompt information regarding at least one of state monitoring, quality management, and predictive maintenance of the object.

[0006] Furthermore, other aspects of the present disclosure relate to a program that causes a computer to function as an acquisition unit, an operation unit, and an output unit, the acquisition unit acquiring first vibration information of a first channel and second vibration information of a second channel regarding vibration of an object, the operation unit operating a spectrogram including at least one of a cross-power spectrogram, a cross-phase spectrogram, and a coherence spectrogram of the first vibration information and the second vibration information, and the output unit outputting, based on the spectrogram, prompt information regarding at least one of state monitoring, quality management, and predictive maintenance of the object.

[0007] In addition, another aspect of the present disclosure relates to a processing method that acquires first vibration information of a first channel and second vibration information of a second channel regarding vibrations of an object, calculates a spectrogram of at least one of a cross-power spectrogram, a cross-phase spectrogram, and a coherence spectrogram including the first vibration information and the second vibration information, and outputs a prompt message regarding at least one of state monitoring, quality management, and predictive maintenance for the object based on the spectrogram. Description of the Drawings

[0008] Figure 1 It is an explanatory diagram of a sensor that detects vibrations of an object.

[0009] Figure 2 It is an explanatory diagram of a sensor that detects vibrations of an object.

[0010] Figure 3 It is an explanatory diagram of a sensor that detects vibrations of an object.

[0011] Figure 4 It is a configuration example of a processing system.

[0012] Figure 5 It is a first flow example of the processing executed by the processing unit.

[0013] Figure 6 It is a second flow example of the processing executed by the processing unit.

[0014] Figure 7 It is a third flow example of the processing executed by the processing unit.

[0015] Figure 8 It is a configuration example of an experimental apparatus.

[0016] Figure 9 It is an example of a spectrogram obtained from vibration data output by a first vibration sensor.

[0017] Figure 10 It is an example of a spectrogram obtained from vibration data output by a first vibration sensor.

[0018] Figure 11 It is an example of a spectrogram obtained from vibration data output by a first vibration sensor.

[0019] Figure 12 They are spectrograms of the first vibration sensor immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later.

[0020] Figure 13 They are spectrograms of the first vibration sensor immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later.

[0021] Figure 14 It is the difference between the spectrogram of the first vibration sensor immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later and the reference spectrogram.

[0022] Figure 15 It is the difference between the spectrogram of the first vibration sensor immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later and the reference spectrogram.

[0023] Figure 16 It is an example of the spectrogram obtained from the vibration data output by the second vibration sensor.

[0024] Figure 17 It is an example of the spectrogram obtained from the vibration data output by the second vibration sensor.

[0025] Figure 18 It is an example of the spectrogram obtained from the vibration data output by the second vibration sensor.

[0026] Figure 19 It is the spectrogram of the second vibration sensor immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later.

[0027] Figure 20 It is the spectrogram of the second vibration sensor immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later.

[0028] Figure 21 It is the difference between the spectrogram of the first vibration sensor and the spectrogram of the second vibration sensor immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later.

[0029] Figure 22 It is the difference between the spectrogram of the first vibration sensor and the spectrogram of the second vibration sensor immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later.

[0030] Figure 23 It is the first example of the prompt information.

[0031] Figure 24 It is the second example of the prompt information.

[0032] Figure 25 It is the second example of the prompt information.

[0033] Explanation of Reference Numerals

[0034] 10…Object, 11…Vibration source, 100…Processing system, 110…Processing unit, 111…Acquisition unit, 112…Arithmetic unit, 114…Output unit, 120…Storage unit, 135…Program, 140…Sensor, 141…First vibration sensor, 142…Second vibration sensor, 150…Prompt unit, 210…Motor, 220…Controller, 230…Flask, 240…Blade of stirrer, 250…Glycerol, 260…Hot plate, CH1…First channel, CH2…Second channel. Detailed implementation mode

[0035] Hereinafter, preferred implementation modes of the present disclosure will be described in detail. It should be noted that the implementation modes described below do not unduly limit the content described in the claims, and not all of the components described in the implementation modes are necessarily essential components.

[0036] 1. Processing system

[0037] Figures 1 to 3 It is an explanatory diagram of a sensor that detects the vibration of an object. As Figure 1 shown, the object 10 includes, for example, a vibration source 11 that generates vibration through mechanical operation. The sensor 140 detects the vibration of the object 10 generated by the vibration source 11. As an example, the vibration source 11 is a motor, an engine, a turbine, or the like. In addition, it is also possible that the object 10 does not include the vibration source 11, vibration is applied from the outside of the object 10, and the sensor 140 detects this vibration.

[0038] The object 10 is, for example, the vibration source 11 itself, that is, a motor, an engine, a turbine, or the like. Or, the object 10 is a machine, device, or apparatus including the vibration source 11, such as a household device or an industrial device such as a printer, an air conditioner, a robot, a pump, a belt conveyor, or a processing device, a moving body such as an automobile or an airplane, or an industrial facility such as a generator or a manufacturing plant. Or, the object 10 is a structure such as a building, a road, or a bridge that vibrates due to an external force.

[0039] The sensor 140 detects acceleration, velocity, displacement, angular acceleration, angular velocity, or angle, and outputs a signal representing the detected physical quantity as vibration information. The sensor 140 may be a sensor that detects one physical quantity, or may be a sensor that detects multiple physical quantities. In addition, the sensor 140 may be a sensor that detects the physical quantity of one axis, or may be a sensor that detects the physical quantity of two or more axes. The sensor 140 outputs vibration information from one or more channels. One channel refers to a channel that outputs a signal of one physical quantity on one axis.

[0040] Figure 2A first configuration example of a sensor having two channels is shown. The sensor 140 includes a first vibration sensor 141, and the first vibration sensor 141 has a first channel CH1 and a second channel CH2. It should be noted that the first vibration sensor 141 may also have three or more channels.

[0041] Figure 3 A second configuration example of a sensor having two channels is shown. The sensor 140 includes a first vibration sensor 141 and a second vibration sensor 142. The first vibration sensor 141 has a first channel CH1, and the second vibration sensor 142 has a second channel CH2. It should be noted that the sensor 140 may also include three or more vibration sensors, and each vibration sensor may also have two or more channels.

[0042] Hereinafter, the first vibration sensor 141 and the second vibration sensor 142 will also be simply referred to as vibration sensors respectively.

[0043] The vibration sensor is an acceleration sensor or a gyro sensor that uses a crystal oscillator as a detection element, or an acceleration sensor or a gyro sensor that uses MEMS as a detection element, etc. In addition, the vibration sensor may also be an IMU in which an acceleration sensor and a gyro sensor are combined and unitized. The vibration sensor may also detect speed or displacement by integrating the acceleration detected by the detection element, or may use a detection element that detects speed or the like. The vibration sensor may also detect angular acceleration or angle by differentiating or integrating the angular velocity detected by the detection element, or may use a detection element that detects angular acceleration or the like. An example of an acceleration sensor is a sensor that detects acceleration by measuring the vibration frequency, which changes according to the stress applied to the crystal oscillator. An example of a gyro sensor is a sensor that detects angular velocity by detecting the Coriolis force applied to the crystal oscillator. Other examples of the acceleration sensor or the gyro sensor are sensors described as follows: a mass portion and electrodes are formed of MEMS, and acceleration or angular velocity is detected by detecting the electrostatic capacitance between the electrodes that changes according to the inertial force applied to the mass portion.

[0044] It is assumed that the vibration sensor is installed in contact with the object 10, but it is not limited thereto, as long as the state is such that vibration is transmitted from the object 10 to the vibration sensor. The first vibration sensor 141 and the second vibration sensor 142 are installed at different positions on the object 10. One vibration sensor is a sensor unit installed at a certain position on the object 10. However, one vibration sensor may also be composed of a plurality of sensor units installed at substantially the same position on the object 10.

[0045] Figure 4This is a configuration example of a processing system. The processing system 100 infers the state of an object by analyzing the vibration information of the object. Detailed examples of the state will be described later. The processing system 100 includes a processing unit 110, a storage unit 120, a sensor 140, and a presentation unit 150. Note that the sensor 140 may also be provided outside the processing system 100 and connected to the processing system 100 via a cable or a network, etc.

[0046] The processing unit 110 includes an acquisition unit 111, an arithmetic unit 112, and an output unit 114.

[0047] The acquisition unit 111 acquires the vibration information of the object from the sensor 140 by receiving the signal of the physical quantity output by the sensor 140. The sensor 140 may output either an analog signal or digital data. The acquisition unit 111 may also include an A / D converter that A / D-converts the analog signal into digital data when receiving an analog signal. The acquisition unit 111 acquires at least the first vibration information of the first channel and the second vibration information of the second channel, and outputs them to the arithmetic unit 112. The vibration information is vibration data in the form of a time series representing acceleration, velocity, displacement, angular acceleration, angular velocity, or angle detected by the sensor 140. Note that the acquisition unit 111 may also acquire vibration information of three or more channels.

[0048] The arithmetic unit 112 calculates at least one of a cross power spectral density diagram, a cross phase spectral density diagram, and a coherence spectral density diagram based on the first vibration information and the second vibration information. In addition to the above, the arithmetic unit 112 may calculate a spectral density diagram of amplitude, power, or phase based on the first vibration information, and may also calculate a spectral density diagram of amplitude, power, or phase based on the second vibration information.

[0049] Regarding a detailed example of the spectral density diagram, it will be based on Figures 9 to 22 described later. In addition, the detailed method of calculating various spectral density diagrams is also described later. Here, an overview of the spectral density diagram will be explained. A spectral density diagram means continuously obtaining a spectrum while shifting the intercept range of the time series signal in the time direction, and showing the time change of the spectrum. Generally, the horizontal axis represents time, the vertical axis represents frequency, and the signal intensity is often represented by color or shading. A spectrum means decomposing a time series signal into components of each frequency according to an operation rule such as Fourier transform, and representing it as a distribution of a function of frequency.

[0050] More specifically, a spectrogram is two-dimensional data with data arranged at each point on the time axis and the frequency axis. A column of data in the frequency axis direction at a certain time shows the frequency spectrum of the vibration at that time. A spectrogram is generated by sequentially obtaining frequency spectra. For example, in an amplitude spectrogram, the data at each point is amplitude data, and in a phase spectrogram, the data at each point is phase data. It should be noted that when obtaining multiple spectrograms in the case of using a multi-axis sensor or the like, the multiple spectrograms can also be overlapped in the depth direction as three-dimensional data.

[0051] As a spectrogram regarding phase, there are a phase spectrogram, a cross-phase spectrogram, or a cross-power spectrogram. The phase spectrogram represents the frequency characteristics of the phase in the vibration data of one channel. The cross-phase spectrogram shows the frequency characteristics of the phase difference between channels of multiple channels as an angle. The cross-power spectrogram represents the frequency characteristics of the phase difference and magnitude (size) between channels of multiple channels. The cross-power spectrogram is obtained by calculating the product of the Fourier transform of the vibration data of one channel and the complex conjugate of the Fourier transform of the vibration data of the other channel.

[0052] As a spectrogram regarding amplitude, there are an amplitude spectrogram, a power spectrogram, or a cross-power spectrogram. The amplitude spectrogram represents the frequency characteristics of the amplitude in the vibration data of one channel. The power spectrogram represents the frequency characteristics of the power in the vibration data of one channel. The cross-power spectrogram is as described above.

[0053] The coherence spectrogram is a spectrogram obtained by normalizing the cross-power spectrogram by the product of the powers of each signal.

[0054] The operation unit 112 converts the spectrogram into image data using color mapping or the like, and outputs the image data to the output unit 114. In the spectrogram after being converted into image data, each point on the time axis and the frequency axis corresponds to a pixel. Color mapping is a mapping that makes the data values in the spectrogram correspond to color data.

[0055] The number of pixels in the frequency direction of the spectrogram can also be more than the number of pixels in the time direction. For example, the operation unit 112 can also generate a spectrogram with the same number of pixels in the frequency direction and the time direction, and perform compression processing on this spectrogram in the time direction. Or, the operation unit 112 can also generate a spectrogram with more pixels in the frequency direction than in the time direction according to the vibration data. It is assumed that there is more information on the vibration state in the frequency direction than in the time direction. Therefore, it is desirable that in the spectrogram, the resolution in the frequency direction is higher than the resolution in the time direction.

[0056] Note that the arithmetic unit 112 may output the spectrogram itself as image data to the output unit 114 without converting the spectrogram using a color map. In addition, the arithmetic unit 112 may convert the spectrogram regarding the phase into image data using a non-cyclic color map. In the image data after conversion based on the color map, each pixel generally has three elements. However, when the value of the spectrogram is complex, the values corresponding to the real part and the imaginary part may be regarded as image data having two elements for conversion. Alternatively, two grayscale images having values corresponding to the real part and the imaginary part respectively may be used. Since a complex number includes phase information, even if the phase θ is not explicitly calculated, the phase can be processed. Since the number of elements is reduced, the memory used is saved.

[0057] Based on the spectrogram from the arithmetic unit 112, the output unit 114 outputs prompt information for the user to the prompting unit 150. The output unit 114 analyzes the state of vibration according to the spectrogram and generates prompt information based on the analysis result. The state of vibration is the state of at least one of state monitoring, quality management, and predictive maintenance of the object. That is, the prompt information is information regarding at least one of state monitoring, quality management, and predictive maintenance of the object. Detailed examples of state monitoring, quality management, and predictive maintenance will be described later. The output unit 114, for example, classifies which of a plurality of vibration states the vibration state of the object belongs to, or detects an abnormality or a failure of the object. The classification result is the probability of each vibration state, or a flag indicating which vibration state it is, etc. The detection result is a flag indicating whether an abnormality or a failure has been detected, etc. The prompt information is information for prompting these to the user.

[0058] Various methods for analyzing the state of vibration according to the spectrogram are envisioned. The output unit 114, for example, classifies or detects the state of vibration based on statistical information obtained from the spectrogram or its temporal change. The statistical information is an average value, a median value, a minimum value, a maximum value, a variance, a standard deviation, or a histogram, etc. These statistical information may also be calculated based on the data columns arranged along the time direction or the frequency direction in the spectrogram, or based on the data of the whole or a partial area of the spectrogram. The output unit 114, for example, classifies or detects the state of vibration by using threshold processing, pattern matching, or AI recognition, etc. on the statistical information or its temporal change.

[0059] Alternatively, the output unit 114 classifies or detects the vibration state based on the temporal change of the spectrogram within a specific frequency range. The specific frequency range refers to the frequency range that is prone to change under the influence of the vibration state in the frequency characteristics of the vibration. The specific frequency range includes the resonance frequency of the object or the frequency range nearby, the vibration frequency of the vibration source or the frequency range nearby, or the harmonic frequency of the vibration frequency of the vibration source or the frequency range nearby, etc., but is not limited to these. The temporal change of the spectrogram is the temporal change within a single spectrogram obtained within a certain time range. Alternatively, the temporal change of the spectrogram means that when the first to the nth spectrograms are obtained within the first to the nth time ranges in the time series, where n is an integer of 2 or more, the temporal change generated in the first to the nth spectrograms in the time series.

[0060] Alternatively, the output unit 114 may also input the spectrogram into the learned model, and the learned model outputs the above classification result or detection result through inference based on the spectrogram. In the learning stage, the learning system generates the learned model by pre-training the model using the teacher data. The learning system is a computer or a cloud system in which multiple computers are connected through a network or the like. The teacher data includes multiple spectrograms and the correct labels for each spectrogram. The correct label in the classifier is, for example, a flag indicating which of the multiple vibration states it conforms to. The correct label in the detector is, for example, a flag indicating whether the object is abnormal or faulty.

[0061] The learned model is a neural network for image recognition using deep learning. As an example, the neural network for image recognition is a CNN or a ViT. CNN is an abbreviation for Convolutional Neural Network. ViT is an abbreviation for Vision Transformer. Transformer is used in models in various fields, but the transformers used for image recognition are collectively called ViT.

[0062] The prompting unit 150 prompts the user with the prompting information from the output unit 114. The prompting unit 150 is, for example, a display, a speaker, a lamp, or a vibrator, etc. The prompting information shows the above classification result or detection result through numerical values, characters, colors, images, sounds, lights, or vibrations, etc. It should be noted that the output unit 114 may also save the classification result of the vibration state, the detection result of the vibration state, or the prompting information generated based on these in a memory or a storage. The memory or the storage may also be common with the following storage unit 120.

[0063] The storage unit 120 stores a program 135 that describes the functions of the respective units of the processing unit 110. The processing unit 110 implements the processing of the acquisition unit 111, the arithmetic unit 112, and the output unit 114 by executing the program 135 read from the storage unit 120. It should be noted that the program 135 may also include the learned model described above.

[0064] Various configurations can also be adopted as the hardware configuration of the processing system 100. The processing system 100 is, for example, a computer or a cloud system in which multiple computers are connected via a network or the like. The computer is not limited to a general-purpose computer such as a personal computer, and may also be a dedicated computer for performing the vibration analysis of the present embodiment or a computer incorporated in a specific device or the like.

[0065] As an example, the processing unit 110 is a processor. The processor includes, for example, one or more of a CPU, a GPU, a microcomputer, a DSP, an ASIC, or an FPGA. The CPU is an abbreviation for Central Processing Unit. The GPU is an abbreviation for Graphics Processing Unit. The DSP is an abbreviation for Digital Signal Processor. The ASIC is an abbreviation for Application Specific Integrated Circuit. The FPGA is an abbreviation for Field Programmable Gate Array. The storage unit 120 stores a program 135 that describes the functions of the respective units of the processing unit 110. The processor implements the functions of the respective units of the processing unit 110 as processing by executing the program 135 stored in the storage unit 120.

[0066] The processing unit 110 is not limited to the software processing as described above, and may also be a circuit in which the functions of the respective units are installed hardware-wise. In this case, the storage unit 120 may not store a program.

[0067] The storage unit 120 is a memory or a register. The memory is a volatile memory such as a RAM, or a non-volatile memory such as an OTP memory or an EEPROM. The RAM is an abbreviation for Random Access Memory. The OTP is an abbreviation for One Time Programmable. The EEPROM is an abbreviation for Electrically Erasable Programmable Read Only Memory.

[0068] Note that the non - transitory information storage medium as a computer - readable medium can also store the above - mentioned program 135. The information storage medium is, for example, an optical disc, a memory card, a hard disk drive, or a non - volatile semiconductor memory.

[0069] 2. Processing Flow

[0070] Figure 5 This is a first flow example of the processing executed by the processing unit. In steps S31 and S32, the acquisition unit 111 acquires the first vibration information of the first channel and the second vibration information of the second channel from the sensor 140. The first vibration information and the second vibration information are, for example, vibration information at the same time. However, this flow can also be applied equally to vibration information that is not at the same time.

[0071] In step S33, the arithmetic unit 112 calculates a cross - power spectral density diagram, a cross - phase spectral density diagram, or a coherence spectral density diagram based on the first vibration information and the second vibration information.

[0072] In step S34, the output unit 114 analyzes the spectral density diagram output by the arithmetic unit 112 and outputs a prompt message to the prompt unit 150 based on the result.

[0073] Note that the arithmetic unit 112 can also calculate the first to m - th time spectral density diagrams corresponding to the first to m - th times. The output unit 114 can also output a prompt message based on the first to m - th time spectral density diagrams. Or, the arithmetic unit 112 can calculate a reference spectral density diagram obtained by smoothing the first - time spectral density diagram in the time direction, and calculate the first to m - th difference spectral density diagrams obtained by subtracting the reference spectral density diagram from the first to m - th time spectral density diagrams. m is an integer of 1 or more. The output unit 114 can also output a prompt message based on the first to m - th difference spectral density diagrams.

[0074] Note that the acquisition unit 111 can also acquire vibration information of three or more channels. The arithmetic unit 112 can calculate a cross - power spectral density diagram, etc., based on the vibration information of any two channels among the three channels.

[0075] Figure 6 This is a second flow example of the processing executed by the processing unit. In steps S41 and S42, the acquisition unit 111 acquires the first vibration information of the first channel and the second vibration information of the second channel from the sensor 140.

[0076] In step S43, the arithmetic unit 112 calculates the first type of spectral density diagram among the cross - power spectral density diagram, the cross - phase spectral density diagram, or the coherence spectral density diagram based on the first vibration information and the second vibration information. Let this be the first spectral density diagram.

[0077] In step S44, the arithmetic unit 112 calculates a second spectrogram different from the first type among the cross-power spectrogram, the cross-phase spectrogram, or the coherence spectrogram based on the first vibration information and the second vibration information. This is set as the second spectrogram.

[0078] In step S45, the arithmetic unit 112 performs arithmetic operations on the first spectrogram and the second spectrogram. The arithmetic operations are addition, subtraction, multiplication, division, or a combination thereof of the first spectrogram and the second spectrogram at the same time. The arithmetic unit 112 performs arithmetic operations on the spectrogram before conversion based on color mapping, for example. That is, the arithmetic unit 112 performs the above arithmetic operations on the spectrogram value corresponding to the pixel of the first spectrogram and the spectrogram value corresponding to the pixel of the second spectrogram at the same position, and executes this arithmetic operation for each pixel. It should be noted that the arithmetic unit 112 may also perform arithmetic operations on the spectrogram after conversion based on color mapping.

[0079] In step S46, the output unit 114 analyzes the spectrogram after the arithmetic operation output by the arithmetic unit 112, and outputs a prompt message to the prompt unit 150 based on the result.

[0080] It should be noted that the acquisition unit 111 may also acquire vibration information of three or more channels. The arithmetic unit 112 may also calculate the first spectrogram and the second spectrogram based on the vibration information of any two of the three channels.

[0081] Figure 7 This is a third process example of the process executed by the processing unit. In steps S51 to S54, the acquisition unit 111 acquires the first vibration information of the first channel, the second vibration information of the second channel, the third vibration information of the third channel, and the fourth vibration information of the fourth channel from the sensor 140.

[0082] In step S56, the arithmetic unit 112 calculates a first type of spectrogram among the cross-power spectrogram, the cross-phase spectrogram, or the coherence spectrogram based on the first vibration information and the second vibration information. This is set as the first spectrogram.

[0083] In step S57, the arithmetic unit 112 calculates a spectrogram of the same type as the first type among the cross-power spectrogram, the cross-phase spectrogram, or the coherence spectrogram based on the third vibration information and the fourth vibration information. This is set as the second spectrogram.

[0084] In step S58, the arithmetic unit 112 performs arithmetic operations on the first spectrogram and the second spectrogram.

[0085] In step S59, the output unit 114 analyzes the spectrogram after the arithmetic operation output by the arithmetic unit 112, and outputs a prompt message to the prompt unit 150 based on the result.

[0086] 3. Vibration Analysis Example

[0087] Next, a specific example of vibration analysis using the above method will be described.

[0088] In conventional vibration analysis, there are few detection methods that focus on phase changes on the premise that the amplitude intensity or frequency peak changes. In addition, even when focusing on phase changes, as in Patent Document 1 described above, a sensor with one channel is used.

[0089] In the present embodiment, a multi-channel sensor is used to acquire time-series data of vibration. In addition, based on the time-series data of vibration, not only an amplitude spectrogram is obtained, but also a cross spectrogram, a cross-phase spectrogram, a coherence spectrogram, or a spectrogram obtained by arithmetic operations on them is obtained and analyzed. Since sensor data needs to be synchronized with each other between multiple axes or at different measurement points, a sensor with excellent synchronization performance is used. For example, an integrated sensor such as an IMU is used, but it is not limited thereto.

[0090] The processing system 100 of the present embodiment uses multi-channel vibration data, for example, multi-channel vibration data at different azimuths at the same position of the object. The processing system 100 calculates a spectrogram regarding at least one of amplitude, power, cross, cross-phase, and coherence based on the vibration data. The processing system 100 uses the coherence of the vibration data to weight information. The processing system 100 uses the spectrogram for state monitoring, quality management, or predictive maintenance of the object. The processing system 100 sets statistical information such as the average value or variance value obtained from the spectrogram as an index, or sets the change in cross-phase or coherence in a specific frequency range as an index. According to the present embodiment, since the vibration of the device is directly used, it is possible to comprehensively grasp minute changes in the mass, spring coefficient, or damping coefficient of the object without missing any, without sweeping the vibration frequency.

[0091] The processing system 100 of the present embodiment may also use a spectrogram obtained by arithmetic operations on two or more spectrograms obtained at different positions of the object regarding the spectrograms of amplitude, cross, cross-phase, and coherence. Thereby, even if the data is not synchronized with each other in measurements at different positions, comparison can be performed.

[0092] In addition, the processing system 100 of the present embodiment may also use multi-channel vibration data at different azimuths at the same position of the object. Thereby, even if the vibration intensity does not change, a clear change can also be regarded as a situation where the phase relationship between the vibration directions or azimuths changes.

[0093] In addition, the processing system 100 of the present embodiment can also use the average value or variance value of the spectrogram as an index to detect the vibration state. Thereby, it is possible to detect minute state changes with high precision.

[0094] Next, an example of vibration analysis is shown. Figure 8 This is a configuration example of the experimental apparatus. A flask 230 containing 250 of glycerol is placed on a hot plate 260, and the glycerol 250 is heated by the hot plate 260. A stirring rod is installed on a motor 210. The blade 240 of the stirring rod is inserted into the glycerol 250, and the glycerol 250 is stirred by the rotation of the motor 210. A controller 220 is installed on the motor 210 and controls the rotation of the motor 210.

[0095] A first vibration sensor SENA is installed on the wall surface of the flask 230, and a second vibration sensor SENB is installed on the wall surface of the controller 220. Each vibration sensor is a three-axis velocity sensor. xa, ya, and za are the detection axes of the first vibration sensor SENA, corresponding to the x-axis, y-axis, and z-axis respectively, and are orthogonal to each other. xb, yb, and zb are the detection axes of the second vibration sensor SENB, corresponding to the x-axis, y-axis, and z-axis respectively, and are orthogonal to each other.

[0096] The viscosity of glycerol 250 changes according to the temperature. It is known that its viscosity is approximately 100 mPa·s at 40 degrees Celsius and approximately 10 mPa·s at 70 degrees Celsius.

[0097] The measurement sequence is as follows. First, insert the stirring rod into the flask 230. Put 200 ml of glycerol into the flask 230. The flask 230 is a separatory funnel, and its upper and lower parts are joined together and fixed by a binding clamp. Set the motor 210 etc. as the main body of the stirrer on the flask 230, and place the flask 230 on the hot plate 260. Fix the first vibration sensor SENA and the second vibration sensor SENB to the apparatus with double-sided tape.

[0098] Set the hot plate 260 to 40 degrees Celsius, and heat the glycerol 250 by the hot plate 260. Set the stirring speed to 256 rpm and start stirring. Gradually raise the temperature of the hot plate 260 to 70 degrees Celsius over one hour, and measure the vibration accompanied by stirring during the temperature rise by the first vibration sensor SENA and the second vibration sensor SENB. Temporarily stop the stirrer at the 45-minute mark in the middle to measure the temperature. The stirring speed during the re-stirring is 250 rpm, which is slightly shifted to the lower speed side compared to before the stop.

[0099] It should be noted that when visually observing, it was also confirmed from the situation of the liquid surface during stirring that: as the temperature rises, the viscosity of glycerol 250 decreases significantly.

[0100] The first vibration sensor SENA and the second vibration sensor SENB are three-axis digital output crystal vibration sensors, and the vibration frequency band is 10 to 1000 Hz. Double-sided tape that has been confirmed to have no effect on the measurement frequency band is used for fixing the vibration sensors. The measurement sampling rate is 3000 samples per second.

[0101] In the forced vibration frequency response model of a degree-of-freedom viscous damping system, it is known that if the elastic coefficient or the damping coefficient changes, the vibration amplitude or the phase changes. By sensing the vibration that is the response of the forced vibration of the agitator and confirming the change over time of the relative phase in the time series data, it is expected that the change in viscoelasticity can be detected. Although it is difficult to measure the absolute phase of the vibration, as long as it is the change in the relative phase between two vibration data, it can be measured. It is considered that the relative phase includes information corresponding to the state of the rotating equipment, and the relative phase itself can be an index indicating the state of the analysis object. However, the object to which the vibration analysis method of the present embodiment can be applied is not limited to a degree-of-freedom viscous damping system. For example, it can also be a device composed only of solids, etc.

[0102] Next, an example of a spectrogram obtained by Figure 8 the device is shown. Since the vibration energy is concentrated in a relatively low frequency band, a spectrogram in the frequency range of 0 to 200 Hz is shown.

[0103] Figures 9 to 11 is an example of a spectrogram obtained from the vibration data output by the first vibration sensor. The "x-y axis" shows the relative phase between the vibration data using the x-axis and the y-axis vibration data, etc. The same applies to the "y-z axis" and the "z-x axis".

[0104] The amplitude spectrogram is a spectrogram obtained as follows: Using a rectangular window function with a width of 4 seconds, while overlapping each window function by 2 seconds, it is shifted in the time direction. The relative phase spectrogram is a spectrogram as follows: Regarding the phase of the complex number when calculating the amplitude spectrogram, the phase difference between the object channels is plotted as the relative phase. The coherence spectrogram is a spectrogram obtained by preparing time series data with a width of 4 seconds intercepted by a rectangular window function for two channels, calculating their coherence, and repeating this by overlapping the window function by 2 seconds. The color mapping is shown on the right side of each spectrogram. The amplitude is in decibel values. The cross phase is shown numerically from -π to +π radians. The coherence is a value in the range of 0 to 1.

[0105] Figures 9 to 11 shows a spectrogram calculated from 5 minutes of time series data intercepted from the vibration data of the first vibration sensor SENA. By comparing the amplitude spectrogram, the cross phase spectrogram, and the coherence spectrogram, it is possible to see which vibration frequency components have a large correlation.

[0106] A dark frequency in the coherence spectrogram means that the correlation between channels at that frequency is small, while a bright frequency in the coherence spectrogram means that the correlation between channels at that frequency is large. For example, between 150 and 200 Hz, the correlation between channels is small. At frequencies where the correlation between channels is small, the relationship of the relative phase of the vibration components is unstable. Therefore, in the cross-phase spectrogram, the values are random, and thus the colors are random. When changes occur in the vibration situation, the hue or distribution of the spectrogram changes. Since the stripe pattern in the horizontal axis direction is stable, in this specific example, it can be known that the vibration is stable for about 5 minutes.

[0107] Figure 12 and Figure 13 Spectrograms of the first vibration sensor are shown immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later. Here, the y-axis amplitude spectrogram, the y-x axis cross-phase spectrogram, and the y-x coherence spectrogram are shown. Similar studies can also be carried out on the x-axis and z-axis as follows.

[0108] 15 minutes later: In the amplitude spectrogram, a change in the distribution of the peak intensity can be confirmed. In the relative phase spectrogram and the coherence spectrogram, the changes can be clearly captured. For example, in the cross-phase spectrogram, the stripe pattern between 100 and 150 Hz has changed. In addition, in the coherence spectrogram, the bright stripe pattern around 75 Hz has become coarser and brighter.

[0109] 30 minutes later: In the amplitude spectrogram, there is an impression that the overall intensity value has increased. The impressions of the cross-phase spectrogram and the coherence spectrogram remain unchanged. Therefore, from the perspective of viscoelasticity, it can be known that the changes are small.

[0110] 60 minutes later: Clear changes can be confirmed in the amplitude spectrogram, the cross-phase spectrogram, and the coherence spectrogram respectively. For example, in the cross-phase spectrogram and the coherence spectrogram, the frequency or concentration of the stripe pattern has changed. From this, it can be known that the viscosity of glycerol 250 has decreased and the vibration situation has changed.

[0111] As described above, as a whole, compared with the amplitude spectrogram, the changes in the cross-phase spectrogram and the coherence spectrogram are greater, and the sensitivity to changes in the vibration state is higher. In addition, it is known that, for example, changes occur in the amplitude spectrogram 60 minutes later, and by combining the amplitude spectrogram, the sensitivity is further improved.

[0112] Figure 14 and Figure 15 The differences between the spectrograms of the first vibration sensor immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later and the reference spectrogram are shown.

[0113] The reference spectrogram of the amplitude spectrogram is the spectrogram obtained by averaging the amplitude spectrogram immediately after the start of measurement in the time axis direction. The same applies to the reference spectrogram of the cross-phase spectrogram and the reference spectrogram of the coherence spectrogram. By subtracting the reference spectrogram from the spectrogram at each time, it becomes easier to capture the changes from the reference spectrogram.

[0114] Immediately after the start of measurement: Since the reference spectrogram is subtracted, the value approaches 0 in the frequency band where the value is stable. In the case where the value fluctuates, it becomes a distribution with an average value of 0, giving a rough impression. The greater the fluctuation, the darker the tone becomes.

[0115] After 15 minutes: In the amplitude spectrogram, an increase in the distribution of the peak intensity can be confirmed. In the cross-phase spectrogram and the coherence spectrogram, changes in the vibration state are captured. In particular, the change in the coherence spectrogram is large.

[0116] After 30 minutes: An overall increase in the intensity of the amplitude spectrogram can be confirmed. The impression of the cross-phase spectrogram remains unchanged. The shade of the coherence spectrogram changes. Specifically, in the coherence spectrogram, the vibrations with low correlation become even less correlated, and the vibrations with high correlation become even more correlated. From the perspective of viscoelasticity, the change is small.

[0117] After 60 minutes: Clear changes can be confirmed in all three spectrograms. However, this change is also due to a slight change in the stirring speed, and the frequency shift is also enhanced. From the perspective of capturing the change in viscosity reduction, it is desirable to separate the change due to the slight change in the stirring speed.

[0118] Figures 16 to 18 This is an example of the spectrogram obtained from the vibration data output by the second vibration sensor. The production method of each spectrogram is Figures 9 to 11 the same. The installation positions of the first vibration sensor SENA and the second vibration sensor SENB are different. If the Figures 16 to 18 spectrogram is compared with the Figures 9 to 11 spectrogram, it can be seen that the vibration situation is different due to the different measurement positions.

[0119] Figure 19 and Figure 20 show the spectrograms of the second vibration sensor immediately after the start of measurement, after 15 minutes, after 30 minutes, and after 60 minutes. Here, the y-axis amplitude spectrogram, the y-x axis cross-phase spectrogram, and the y-x coherence spectrogram are shown. The same study as below can also be carried out on the x-axis and z-axis.

[0120] After 15 minutes: The impression of the three spectrograms does not change significantly immediately after the start of measurement.

[0121] After 30 minutes: There is an impression that the intensity values in the amplitude spectrum diagram increase as a whole. In the cross-phase spectrum diagram and the coherence spectrum diagram, the change in the vibration state is clearly captured.

[0122] After 60 minutes: A viscosity decrease is detected, and clear changes can be confirmed in each of the three spectrum diagrams.

[0123] Figure 21 and Figure 22 Shows the differences between the spectrum diagrams of the first vibration sensor and the second vibration sensor immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later. Immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later, the spectrum diagrams of the first vibration sensor are averaged in the time axis direction, and the differences between the average spectrum diagrams and the spectrum diagrams of the second vibration sensor are shown.

[0124] Immediately after the start of measurement: In each spectrum diagram, in the frequency band where the values are stable, the values are close to 0. When the values change, it becomes a distribution with an average value of 0, giving a rough impression. The greater the change, the darker the hue becomes.

[0125] After 15 minutes: The impressions of the three spectrum diagrams do not change significantly immediately after the start of measurement. However, in the coherence spectrum diagram, changes can be confirmed in the region below 70 Hz.

[0126] After 30 minutes: The intensity values in the amplitude spectrum diagram increase as a whole. In the cross-phase spectrum diagram and the coherence spectrum diagram, the change in the vibration state is clearly captured.

[0127] After 60 minutes: A viscosity decrease is detected, and clear changes can be confirmed in each of the three spectrum diagrams. 45 minutes after the start of measurement, the rotation frequency of the mixer changes from 256 rpm to 250 rpm. By taking the differences between the spectrum diagrams between the two vibration sensors, the influence of the rotation frequency of the mixer can be eliminated. Thus, the change in viscosity decrease can be divided into the change in which the stirring speed slightly changes, and the vibration state can be detected with higher precision.

[0128] It should be noted that in the above, the average value of the spectrum diagrams is obtained in the time axis direction for each frequency and used as the reference spectrum. However, the method for producing the reference spectrum diagram is not limited to this. For example, instead of taking the average value, any spectrum diagram itself can be used as the reference to calculate the differences between the spectrum diagrams between the vibration sensors.

[0129] 4. Examples of methods for producing hint information from spectrum diagrams

[0130] The output unit 114 extracts characteristic frequencies from the entire spectrogram. The characteristic frequencies are, for example, the peak frequency or the frequency with a stable intensity distribution, etc. The output unit 114 represents the change in the amount of object information in the extracted frequencies as a function of time through a graph or the like. The amount of object information is the intensity value, relative phase, etc. The function only needs to represent the temporal change, and it can be a mathematical formula or the time-series data of the amount of object information, etc. The output unit 114 uses the above function to obtain the statistic of the amount of object information in the steady state. The statistic is the average value, variance, etc. of the above function. The output unit 114 sets a threshold value with respect to the statistic, and outputs a warning from the prompting unit 150 when the statistic exceeds the threshold value.

[0131] Figure 23 This is the first example of the prompt information. Regarding the viscosity change of glycerol caused by temperature rise, the vibration generated during stirring is measured to obtain an amplitude spectrogram, a cross-phase spectrogram, and a coherence spectrogram.

[0132] As the characteristic frequency, 60.5 Hz is focused on. In Figure 23 the upper part shows the amplitude value of 60.5 Hz of the y-axis vibration data plotted in time series. In Figure 23 the middle part shows the cross-phase of 60.5 Hz of the vibration data of the y-axis and z-axis plotted in time series. Regarding the cross-phase diagram, for easy observation, the 60-second moving average is plotted with thick black dots. In Figure 23 the lower part shows the coherence of 60.5 Hz of the vibration data of the y-axis and z-axis plotted in time series.

[0133] Although no clear temporal change can be confirmed from the amplitude diagram and the coherence diagram, a certain phase change related to the viscosity change can be extracted from the cross-phase diagram. By using this as an index, it can be flexibly applied to quality control in the material manufacturing process, for example.

[0134] Figure 24 and Figure 25 This is the second example of the prompt information. In a measurement different from Figure 23 the temperature was raised to 80 degrees Celsius to measure the viscosity change of glycerol. Regarding the viscosity change of glycerol caused by temperature rise, the vibration generated during stirring is measured to obtain an amplitude spectrogram, a cross-phase spectrogram, and a coherence spectrogram.

[0135] As the characteristic frequency, 108.5 Hz, which is one of the peaks of the amplitude spectrum, is focused on. In Figure 24 the upper part shows the amplitude value of 108.5 Hz of the z-axis vibration data plotted in time series. In Figure 24 the middle part shows the cross-phase of 108.5 Hz of the vibration data of the z-axis and x-axis plotted in time series. In Figure 24The lower part of shows the coherence of 108.5 Hz obtained by plotting the vibration data of the z-axis and the x-axis in time series. Note that the jumps in the values that occasionally occur at 700 seconds, 1600 seconds, 2800 seconds, etc. are vibration noises accompanying the equipment operation.

[0136] Figure 25 Shows the time-series change of the variance in the cross-phase diagram. Values are extracted from the cross-phase diagram using a rectangular window function of 60 seconds, the variance of the extracted values is calculated, and the variance is continuously calculated while shifting the start point of the rectangular window function by 1 second in sequence, and the time-series data thereof is plotted.

[0137] In the second example, the values obtained by summing and averaging the values from 107.0 Hz to 110.0 Hz centered on 108.5 Hz at each time are used. Thereby, the influence caused by the fluctuation of the vibration, that is, the influence caused by the peak change, can be mitigated.

[0138] As Figure 24 As shown in the middle part of , regarding the viscosity change caused by the temperature rise, if the viscosity decreases, the cross-phase changes.

[0139] As Figure 25 As shown, in the steady state, the variance of the cross-phase is less than 0.1. In addition, the noise accompanying the equipment operation does not exceed 0.3. Therefore, 0.4 is set as the threshold value, and when it exceeds this threshold value, it is determined as abnormal.

[0140] As Figure 25 As shown, at the time point of 6400 seconds, a change considered to be a certain phase change occurs. Since the variance of the cross-phase exceeds the threshold value of 0.4, the system senses an abnormal state and outputs an alarm. The operator noticed the alarm and ended the experiment 30 minutes later.

[0141] Note that the method for detecting system abnormalities is not limited to the above method. For example, it is also possible to calculate the moving average of the coherence, and based on its inclination, predict the time when the coherence is lower than 0.5, so as to estimate the time until the system abnormality.

[0142] As Figure 24 As shown, no sign of viscosity change clearly appears in the amplitude diagram. Therefore, it is difficult to detect the sign of viscosity change from the amplitude diagram. In contrast, signs of viscosity change clearly appear in the cross-phase diagram and the coherence diagram, and by using their plots, the sensitivity for detecting the sign of viscosity change can be improved.

[0143] The processing system 100 of the present embodiment described above includes an acquisition unit 111, a calculation unit 112, and an output unit 114. The acquisition unit 111 acquires first vibration information of a first channel and second vibration information of a second channel regarding the vibration of an object. The calculation unit 112 calculates at least one spectrogram including a cross-power spectrogram, a cross-phase spectrogram, and a coherence spectrogram of the first vibration information and the second vibration information. The output unit 114 outputs a prompt message regarding at least one of state monitoring, quality management, and predictive maintenance for the object based on the spectrogram.

[0144] According to the present embodiment, it is possible to perform state monitoring, quality management, or predictive maintenance of an object based on vibration analysis using vibration information of multiple channels. Even a small change in the vibration state that is difficult to appear in the amplitude spectrogram of a single channel can be detected by using the relative spectrogram of multiple channels. For example, even when the intensity value in the amplitude spectrogram changes little, there may be a change in the cross-phase or coherence between multiple channels. By using these spectrograms, the detection sensitivity of the vibration state can be improved.

[0145] In addition, in the present embodiment, the first vibration information may also be vibration information of a first axis detected by a first vibration sensor. The second vibration information may also be vibration information of a second axis detected by the first vibration sensor.

[0146] For example, in Figure 9 , the first vibration information is vibration information of the x-axis detected by the first vibration sensor SENA, and the second vibration information is vibration information of the y-axis detected by the first vibration sensor SENA.

[0147] According to the present embodiment, it is possible to calculate at least one spectrogram including a cross-power spectrogram, a cross-phase spectrogram, and a coherence spectrogram based on the vibration information of the first axis and the vibration information of the second axis at the same position of the object. Thus, by evaluating the correlation of vibrations between multiple axes at the same position, it is possible to detect changes in the vibration state. For example, in linear vibration and elliptical vibration, or in the case where the axis of elliptical vibration stops and rotates, the phase correlation changes and changes occur in each spectrogram. By detecting these situations, it is possible to detect small changes in the vibration state.

[0148] In addition, in the present embodiment, the first vibration information may also be vibration information of a first axis detected by a first vibration sensor. The second vibration information may also be vibration information of the first axis or the second axis detected by a second vibration sensor arranged at a position different from the first vibration sensor.

[0149] According to the present embodiment, it is possible to calculate a spectrogram including at least one of a cross power spectrogram, a cross phase spectrogram, and a coherence spectrogram based on the vibration information of the first axis at different positions of the object and the vibration information of the first axis or the second axis. The path for transmitting vibration from the vibration source to the first vibration sensor is different from the path for transmitting vibration from the vibration source to the second vibration sensor. By evaluating the correlation of vibrations at such different positions, it is possible to detect a change in the vibration state.

[0150] In addition, in the present embodiment, the first vibration information may also be a first physical quantity that is the acceleration, velocity, displacement, angular acceleration, angular velocity, or angle of the first axis among the x-axis, y-axis, and z-axis. The second vibration information may also be the first physical quantity of the second axis different from the first axis among the x-axis, y-axis, and z-axis.

[0151] For example, in Figure 9 , the first physical quantity is velocity, the first vibration information is the velocity of the x-axis, and the second vibration information is the velocity of the y-axis.

[0152] In addition, in the present embodiment, the first vibration information may also be a first physical quantity that is the acceleration, velocity, displacement, angular acceleration, angular velocity, or angle of the first axis among the x-axis, y-axis, and z-axis. The second vibration information is a second physical quantity different from the first physical quantity among the acceleration, velocity, displacement, angular acceleration, angular velocity, and angle of the first axis.

[0153] When the object vibrates, the acceleration, velocity, displacement, angular acceleration, angular velocity, or angle changes. That is, a spectrogram is created based on these physical quantities, and by analyzing the spectrogram, the vibration state can be estimated.

[0154] In addition, in the present embodiment, the output unit 114 may also output a prompt message based on the statistical information about the spectrogram.

[0155] When the vibration state changes, the statistical information of the spectrogram changes. In the Figure 25 example, due to the change in the viscosity of glycerol, the variance of the cross phase changes. This situation can be utilized to estimate the vibration state based on the statistical information of the spectrogram and output a prompt message about the vibration state.

[0156] In addition, in the present embodiment, the output unit 114 may also output a prompt message based on the time change of the spectrogram in a specific frequency or specific frequency range.

[0157] In Figure 23 the example, the specific frequency is 60.5 Hz. In Figure 24 and Figure 25In the example, the specific frequency range is the range from 107.0 Hz to 110.0 Hz centered at 108.5 Hz.

[0158] According to the present embodiment, in the spectrogram, there is a specific frequency or a specific frequency range that is liable to be affected by changes in the vibration state. By analyzing the temporal change of the spectrogram in the specific frequency or the specific frequency range, the vibration state can be estimated, and prompt information regarding the vibration state can be output.

[0159] In addition, in the present embodiment, the arithmetic unit 112 may also obtain a reference spectrogram that smooths the spectrogram in the time direction, and obtain a differential spectrogram obtained by subtracting the reference spectrogram from the spectrogram. The output unit 114 may also output prompt information based on the differential spectrogram.

[0160] In Figure 14 and Figure 15 In the example, the reference spectrogram is a spectrogram that smooths each spectrogram immediately after the start of measurement in the time direction. The differential spectrogram is a spectrogram obtained by subtracting the reference spectrogram from each of the spectrograms immediately after the start of measurement, 15 minutes later, 30 minutes later, and 60 minutes later.

[0161] By taking the difference from the reference spectrogram, the part without the change from the reference spectrogram is eliminated. Therefore, the change from the reference spectrogram is enhanced. By using such a differential spectrogram, the sensitivity for detecting changes in the vibration state is improved.

[0162] In addition, according to the present embodiment, the arithmetic unit 112 may also perform arithmetic operations on a first spectrogram of a first type among the three types of cross-power spectrogram, cross-phase spectrogram, and coherence spectrogram of the first vibration information and the second vibration information, and a second spectrogram of a second type different from the first type among the three types. The output unit 114 may also output prompt information based on the result of the arithmetic operation.

[0163] In addition, in the present embodiment, the acquisition unit 111 may also acquire third vibration information of a third channel and fourth vibration information of a fourth channel regarding the vibration of the object. The arithmetic unit 112 may also perform arithmetic operations on a first spectrogram and a second spectrogram. The first spectrogram is a spectrogram of a first type among the three types of cross-power spectrogram, cross-phase spectrogram, and coherence spectrogram of the first vibration information and the second vibration information. The second spectrogram is a spectrogram of a first type or a second type different from the first type among the three types of cross-power spectrogram, cross-phase spectrogram, and coherence spectrogram of the third vibration information and the fourth vibration information. The output unit 114 may also output prompt information based on the result of the arithmetic operation.

[0164] According to the present embodiment, by performing arithmetic operations on two spectrograms, the spectrograms can be processed in a manner that easily exhibits changes in the vibration state. Thereby, the sensitivity for detecting changes in the vibration state is improved. An example of the arithmetic operation is the multiplication operation of the cross-phase spectrogram and the coherence spectrogram. In the coherence spectrogram, the coherence value of the frequency band with low correlation is small. When the coherence spectrogram is multiplied by the cross-phase spectrogram, the frequency band portion where the value of the cross-phase spectrogram is random is multiplied by the frequency band portion where the value of the coherence spectrogram is small. Thereby, in the cross-phase spectrogram, the frequency band having the characteristic that the cross-phase is not random is enhanced. By using such an enhanced spectrogram, the sensitivity for detecting changes in the vibration state is improved. It should be noted that, as Figure 6 described in etc., the object of the arithmetic operation or the operation content can also be various.

[0165] In addition, the present embodiment can also be implemented as a computer-readable program. The program causes a computer to function as an acquisition unit 111, an arithmetic unit 112, and an output unit 114. The acquisition unit 111 acquires first vibration information of a first channel and second vibration information of a second channel regarding the vibration of an object. The arithmetic unit 112 calculates at least one spectrogram of a cross-power spectrogram, a cross-phase spectrogram, and a coherence spectrogram including the first vibration information and the second vibration information. The output unit 114 outputs a prompt message regarding at least one of state monitoring, quality management, and predictive maintenance for the object based on the spectrogram.

[0166] In addition, the present embodiment can also be implemented as a processing method. The processing method acquires first vibration information of a first channel and second vibration information of a second channel regarding the vibration of an object. The processing method calculates at least one spectrogram of a cross-power spectrogram, a cross-phase spectrogram, and a coherence spectrogram including the first vibration information and the second vibration information. The processing method outputs a prompt message regarding at least one of state monitoring, quality management, and predictive maintenance for the object based on the spectrogram. The processing method, for example, causes a computer to perform these steps.

[0167] 5. Examples of Vibration Analysis for State Monitoring, Quality Management, and Predictive Maintenance

[0168] Next, as an example of the processing system 100 to which the present embodiment can be applied, examples of vibration analysis for state monitoring, quality management, and predictive maintenance are shown.

[0169] Condition Monitoring refers to the technology of continuously or periodically monitoring the current operating state or performance of machinery or equipment. The main purpose of condition monitoring is to monitor specific parameters such as vibration, sound, temperature, or pressure, and evaluate the soundness of machinery or equipment. As systems for conducting condition monitoring, for example, there are the following (a) to (f).

[0170] (a) A system that installs sensors in industrial motors and issues an alarm when abnormal vibration or temperature rise is detected.

[0171] (b) A system that installs vibration sensors in various parts of a building and monitors the soundness of the building's structure by monitoring vibrations during earthquakes or strong winds.

[0172] (c) A system that installs vibration sensors to monitor the performance of a gas turbine and warns the operator when abnormal vibration is detected.

[0173] (d) A system that installs vibration sensors to monitor the soundness of heavy machinery in mines or quarries to assist in the early detection of wear or faults.

[0174] (e) A system that installs vibration sensors to monitor the soundness of each piece of machinery on a factory production line and issues an alarm when abnormal vibration is detected.

[0175] (f) A system that uses vibration sensors installed on the blades or gearboxes of turbines in wind power generation to detect whether abnormal wear or damage has occurred.

[0176] Quality Control refers to the process of confirming whether a product or service meets predetermined quality standards or required specifications. As systems for conducting quality control, for example, there are the following (g) to (k).

[0177] (g) A system that uses cameras or sensors on a production line to confirm the quality of a product such as its dimensions, color, or shape in real time and automatically rejects products that deviate from the standard.

[0178] (h) A system that, after the assembly process of a product that combines multiple components, uses a vibration sensor to confirm the vibration mode when the product is operating. If abnormal vibration is detected, it is suspected that there are assembly defects or component malfunctions.

[0179] (i) A system that operates a manufactured motor or generator and measures the vibration during this operation using a vibration sensor. If vibration exceeding a specific standard is detected, it is considered that there is internal imbalance or damage.

[0180] (j) A system: To verify whether a newly manufactured electronic device has the set vibration resistance, the electronic device is set on a vibration test bench, and a vibration sensor is used to monitor the impact of vibration.

[0181] (k) A system: To confirm whether a new type of railway vehicle or aircraft has the operating performance conforming to the design, the vibration during operation is monitored by a vibration sensor. In the case where abnormal vibration is detected, the cause is determined to assist in quality improvement.

[0182] Predictive Maintenance refers to a method of collecting and analyzing the operation data or status data of equipment or machinery. The purpose of predictive maintenance is to predict the risk of future failures or performance degradation. As a system for performing predictive maintenance, for example, there are the following (l) to (q).

[0183] (l) A system: A sensor is installed on an industrial robot, and the sensor data during the operation of the industrial robot and the sensor data when a failure occurred in the past are analyzed. By detecting a marker that is a precursor to the failure of a specific component, the replacement of the component is scheduled predictively.

[0184] (m) A system: A sensor installed on the wheel of a railway vehicle is used to monitor the degree of wear, and the replacement time of the wheel is predicted based on the data indicating excessive wear.

[0185] (n) A system: The state of the piping or valves in an oil refinery is monitored by sensors to predict the possibility of future leaks.

[0186] (o) A system: The sensor data during the operation of an elevator or escalator is analyzed to predict the risk of failure before the components need to be replaced.

[0187] (p) A system: Used to monitor the efficiency of an industrial cooling device or air conditioner based on sensor data and predict component wear or failure.

[0188] (q) A system: The sensor data during the operation of agricultural machinery is collected and analyzed to predict component wear or failure and support the scheduling of appropriate maintenance activities.

[0189] 6. Operation method of spectrogram

[0190] As the first method, an operation method of various spectrograms using Fourier transform is shown. Below, a method of calculating the spectrum based on the vibration data within a window is shown. The spectrum is obtained in time series while moving the window in the time direction to form a spectrogram. For example, the Short-Time Fourier Transform (STFT) using a window with a fixed width is utilized.

[0191] The following equation (1) shows the power spectrum PWxy(f). t is time, and x(t) is a signal as vibration information. f is frequency, and X(f) is the Fourier transform of x(t).

[0192] [Equation 1]

[0193] PWx(f) = |X(f)| 2 ···(1)

[0194] The following equation (2) shows the cross-power spectrum CRxy(f). x(t) is a signal as the first vibration information, and y(t) is a signal as the second vibration information. X(f) is the Fourier transform of x(t), and Y(f) is the Fourier transform of y(t). * represents complex conjugate.

[0195] [Equation 2]

[0196] CRxy(f) = X(f)*Y(f) ···(2)

[0197] The above equation (2) can be rewritten as the following equation (3). The phase angle θ(f) of CRxy(f) represents the cross-phase spectrum. θ(f) is the phase difference between the signals x(t) and y(t) at frequency f. j is the imaginary unit.

[0198] [Equation 3]

[0199] CRxy(f) = |CRxy(f)|e jθ(f) ···(3)

[0200] The following equation (4) shows the coherence spectrum CHxy(f). <> represents an appropriate smoothing operation, such as averaging in the time direction, frequency direction, or both time and frequency directions.

[0201] [Equation 4]

[0202]

[0203] As a second method, an operation method of various spectrograms using wavelet transform is shown.

[0204] The following equation (5) shows the wavelet transform of the signal x(t). Wx(a, b) represents the wavelet transform of the signal x(t) at scale a and time point b. Ψ(t) represents the Morlet wavelet function. Ψ* represents the complex conjugate of the wavelet function. a is the scale parameter or dilation parameter, which controls the stretching and shrinking of the wavelet and corresponds to frequency. b is the translation position parameter or translation parameter, which controls the position of the wavelet and corresponds to time.

[0205] [Equation 5]

[0206]

[0207] The following equation (6) shows the Morlet wavelet function Ψ(t). ω0 specifies the center frequency. As an example, ω0 = 6, but it is not limited thereto. It should be noted that various functions such as the Haar wavelet function or the Daubechies wavelet function can also be used as the wavelet function.

[0208] [Equation 6]

[0209]

[0210] The following equation (7) shows the wavelet power spectrum WPW corresponding to the "power spectrum diagram". x The operation expressions of (a, b).

[0211] [Equation 7]

[0212] WPWx(a, b) = |Wx(a, b)| 2 ···(7)

[0213] The following equation (8) shows the operation expression of the cross-wavelet spectrum WCRxy(a, b) corresponding to the "cross power spectrum diagram". Wy(a, b) represents the wavelet transform of the signal y(t) at the scale a and the time point b.

[0214] [Equation 8]

[0215] WCRxy(a, b) = Wx(a, b) * Wy(a, b) ···(8)

[0216] The above equation (8) can be rewritten as the following equation (9). The phase angle θ(a, b) of WCRxy(a, b) represents the cross-phase spectrum. θ(a, b) is the phase difference between the signals x(t), y (t).

[0217] [Equation 9]

[0218] WCRxy(a, b) = |WCRxy(a, b)|e jθ(a，b) ···(9)

[0219] The following equation (10) shows the operation expression of the wavelet coherence spectrum WCH(a, b) corresponding to the "coherence spectrum diagram".

[0220] [Equation 10]

[0221]

[0222] It should be noted that, as described above, the present embodiment has been described in detail. However, those skilled in the art should easily understand that many modifications can be made without substantially departing from the new matters and effects of the present disclosure. Therefore, such modified examples are all included within the scope of the present disclosure. For example, in the specification or the drawings, a term that is described at least once together with a different term that is more general or synonymous can be replaced with that different term anywhere in the specification or the drawings. In addition, all combinations of the present embodiment and the modified examples are also included within the scope of the present disclosure. Furthermore, the configurations and operations of the processing unit, storage unit, learned model, sensor, processing system, object, vibration information, spectrogram, etc. are not limited to the content described in the present embodiment, and various modifications can be made.

Claims

1. A processing system, characterized in that: include: An acquisition unit that acquires first vibration information of a first channel and second vibration information of a second channel regarding vibration of an object; A calculation unit, which calculates a spectrum diagram including at least one of a cross-power spectrum diagram, a cross-phase spectrum diagram, and a coherence spectrum diagram of the first vibration information and the second vibration information; as well as The output unit outputs presentation information on at least one of state monitoring, quality management, and predictive maintenance for the object based on the spectrogram.

2. The processing system according to claim 1, characterized in that The first vibration information is vibration information of the first axis detected by the first vibration sensor. The second vibration information is vibration information of a second axis detected by the first vibration sensor.

3. The processing system according to claim 1, characterized in that The first vibration information is vibration information of the first axis detected by the first vibration sensor. The second vibration information is vibration information about the first axis or the second axis detected by a second vibration sensor disposed at a position different from that of the first vibration sensor.

4. The processing system according to claim 1, characterized in that The first vibration information is a first physical quantity that is acceleration, velocity, displacement, angular acceleration, angular velocity, or angle of a first axis among the x-axis, y-axis, and z-axis. The second vibration information is the first physical quantity of a second axis different from the first axis among the x-axis, the y-axis, and the z-axis.

5. The processing system according to claim 1, characterized in that The first vibration information is a first physical quantity that is acceleration, velocity, displacement, angular acceleration, angular velocity, or angle of a first axis among the x-axis, y-axis, and z-axis. The second vibration information is a second physical quantity different from the first physical quantity among acceleration, velocity, displacement, angular acceleration, angular velocity, and angle of the first axis.

6. The processing system according to claim 1, characterized in that The output unit outputs the prompt information based on statistical information about the spectrogram.

7. The processing system according to claim 1, characterized in that The output unit outputs the prompt information based on a time change of the spectrogram at a specific frequency or a specific frequency range.

8. The processing system according to claim 1, characterized in that The calculation unit obtains a reference spectrum diagram obtained by smoothing the spectrum diagram in the time direction, and obtains a differential spectrum diagram obtained by subtracting the reference spectrum diagram from the spectrum diagram. The output unit outputs the prompt information based on the differential spectrum graph.

9. The processing system according to claim 1, characterized in that The operation unit performs arithmetic operations on a first spectrum diagram of the first type among the three types of the cross-power spectrum diagram, the cross-phase spectrum diagram and the coherence spectrum diagram of the first vibration information and the second vibration information and a second spectrum diagram of the second type among the three types that is different from the first type. The output unit outputs the prompt information based on a result of the arithmetic operation.

10. The processing system according to claim 1, characterized in that The acquisition unit acquires third vibration information of a third channel and fourth vibration information of a fourth channel regarding the vibration of the object. The operation unit performs arithmetic operations on a first spectrum diagram of the first of the three types of the cross-power spectrum diagram, the cross-phase spectrum diagram and the coherence spectrum diagram of the first vibration information and the second vibration information and a second spectrum diagram of the first of the three types of the cross-power spectrum diagram, the cross-phase spectrum diagram and the coherence spectrum diagram of the third vibration information and the fourth vibration information, or a second spectrum diagram of a second type different from the first type. The output unit outputs the prompt information based on a result of the arithmetic operation.

11. A storage device, characterized in that: A program is stored, the program causing the computer to function as an acquisition unit, a calculation unit, and an output unit, The acquisition unit acquires first vibration information of a first channel and second vibration information of a second channel regarding vibration of the object. The operation unit operates a spectrum diagram including at least one of a cross-power spectrum diagram, a cross-phase spectrum diagram, and a coherence spectrum diagram of the first vibration information and the second vibration information, The output unit outputs presentation information regarding at least one of state monitoring, quality management, and predictive maintenance for the object based on the spectrogram.

12. A processing method, characterized in that: acquiring first vibration information of a first channel and second vibration information of a second channel regarding vibration of the object, Calculating a spectrum diagram including at least one of a cross-power spectrum diagram, a cross-phase spectrum diagram, and a coherence spectrum diagram of the first vibration information and the second vibration information, Based on the spectrogram, presentation information on at least one of state monitoring, quality management, and predictive maintenance for the object is output.

Images