Signal processing method, signal processing device, and signal processing program

Abstract

To provide a signal processing method that can generate a time waveform in which a signal component asynchronous with a target signal component is reduced without requiring a rotation pulse signal.SOLUTION: A signal processing method includes: a step of acquiring first to N-th time waveforms; a step of generating first to N-th frequency spectra; a trigger time series generation step of generating the time series of a trigger; a step of generating first to N-th transformed frequency spectra obtained by transforming the first to N-th frequency spectra; a step of, for each integer i of 1 or more and N or less, inverting the i-th transformed frequency spectrum in a time domain to generate an i-th inverted time waveform; and a step of, for each integer i, performing synchronous addition of the i-th inverted time waveform on the basis of the time series of the trigger, and thereby generating an i-th synchronous addition waveform. For each integer i, a time length of the i-th time waveform is equal to or more than a time corresponding to a product of a time interval of the time series of the trigger and the number of times of the synchronous addition.SELECTED DRAWING: Figure 1

Description

[[Technical Field]]

[0001] The present invention relates to a signal processing method, a signal processing apparatus, and a signal processing program. [[Background Art]]

[0002] Conventionally, when performing diagnosis using rotation phase information in vibration diagnosis of a rotating device, after acquiring a vibration time waveform and a rotation pulse signal from the rotating device, a vibration waveform component synchronized with the rotation pulse signal is extracted to perform the diagnosis. For example, Non-Patent Document 1 discloses a method and procedure for implementing a target diagnosis by obtaining a vibration time waveform or an orbit diagram with a rotation pulse as an absolute reference, and extracting a vibration waveform component synchronized with the rotation pulse to obtain a full spectrum. [[Prior Art Literature]] [[Non-Patent Literature]]

[0003] [[Non-Patent Document 1]] API Standard 670. Machinery Protection Systems. FIFTH EDITION | NOVEMBER 2014. [[Summary of the Invention]] [[Problem to be Solved by the Invention]]

[0004] When performing diagnosis using rotation phase information, in conventional methods, it is necessary to reduce vibration waveform components asynchronous to the target vibration waveform by performing preprocessing using a signal conditioner such as a PLL, tracking filter, or low-pass filter in combination in order to obtain a rotation pulse signal from the rotating device and extract the vibration waveform component synchronized with the rotation pulse signal. [[Means for Solving the Problem]]

[0005] One aspect of the signal processing method according to the present invention is A time waveform acquisition step in which, with N being a predetermined integer of 1 or more, for each integer i between 1 and N, the i-th sensor acquires an i-th time waveform relating to an i-th physical quantity caused by an external force, velocity, or displacement acting on an object, the i-th sensor acquiring the i-th time waveform relating to an i-th physical quantity that includes at least a periodic fluctuation having a first frequency component and a second frequency component different from the first frequency and a higher-order frequency of the first frequency. For each of the aforementioned integers i, a frequency spectrum generation step is performed to generate the ith frequency spectrum based on the i-th time waveform, A trigger time series generation step that generates a trigger time series having a time interval of a first period which is the reciprocal of the first frequency or a real multiple of the first period defined by the components of the object, where the period corresponds to the frequency of any of the multiple peaks included in at least one of the first to Nth frequency spectra, and the time interval is a first period which is the reciprocal of the first frequency or a period which is a real multiple of the first period defined by the components of the object, A frequency spectrum deformation step is performed to generate an i-th deformed frequency spectrum by multiplying a window function over at least one of a plurality of frequency intervals obtained by dividing the frequency spectrum of i such that each interval contains a plurality of distinct frequencies that are rational multiples of the frequency corresponding to the reciprocal of the time interval of the trigger time series, thereby deforming the i-th frequency spectrum. For each of the aforementioned integers i, the process includes an inverse transform time waveform generation step, which inversely transforms the modified frequency spectrum of i into the time domain to generate the ith inverse transform time waveform, A synchronous addition step is performed on each of the aforementioned integers i, based on the time series of the trigger, to generate the ith synchronous addition waveform by performing synchronous addition on the inverse time waveform of i, Includes, For each integer i, the duration of the time waveform of i is equal to or greater than the duration of the product of the time interval of the trigger time series and the number of synchronous additions.

[0006] One aspect of the signal processing apparatus according to the present invention is: A time waveform acquisition circuit that, for each integer i between 1 and N, where N is a predetermined integer of 1 or more, acquires an i-th time waveform from the i-th sensor relating to an i-th physical quantity caused by an external force, velocity, or displacement acting on an object, the i-th sensor having at least a periodic fluctuation having a first frequency component and a second frequency component different from the first frequency and a higher-order frequency of the first frequency. For each of the integers i, a frequency spectrum generation circuit generates the i-th frequency spectrum based on the time waveform of i, A trigger time series generation circuit generates a trigger time series whose time interval is a period corresponding to the frequency of any of the multiple peaks included in at least one of the first to Nth frequency spectra, and which is a first period that is the reciprocal of the first frequency or a period that is a real multiple of the first period defined by the components of the object, A frequency spectrum deformation circuit generates an i-th deformed frequency spectrum by multiplying the frequency spectrum of i by a window function over at least one of a plurality of frequency intervals obtained by dividing the frequency spectrum of i such that each interval contains a plurality of distinct frequencies that are rational multiples of the frequency corresponding to the reciprocal of the time interval of the trigger time series, For each of the aforementioned integers i, an inverse transform time waveform generation circuit generates an i-th inverse transform time waveform by inversely transforming the modified frequency spectrum of i into the time domain, A synchronous summing circuit generates an i-th synchronous sum waveform by performing synchronous summing on the inverse time waveform of i based on the time series of the trigger for each of the aforementioned integers i, Equipped with, For each integer i, the duration of the time waveform of i is equal to or greater than the duration of the product of the time interval of the trigger time series and the number of synchronous additions.

[0007] One aspect of the signal processing program according to the present invention is: A time waveform acquisition step in which, with N being a predetermined integer of 1 or more, for each integer i between 1 and N, the i-th sensor acquires an i-th time waveform relating to an i-th physical quantity caused by an external force, velocity, or displacement acting on an object, the i-th sensor acquiring the i-th time waveform relating to an i-th physical quantity that includes at least a periodic fluctuation having a first frequency component and a second frequency component different from the first frequency and a higher-order frequency of the first frequency. For each of the aforementioned integers i, a frequency spectrum generation step is performed to generate the ith frequency spectrum based on the i-th time waveform, A trigger time series generation step that generates a trigger time series having a time interval of a first period which is the reciprocal of the first frequency or a real multiple of the first period defined by the components of the object, where the period corresponds to the frequency of any of the multiple peaks included in at least one of the first to Nth frequency spectra, and the time interval is a first period which is the reciprocal of the first frequency or a period which is a real multiple of the first period defined by the components of the object, A frequency spectrum deformation step is performed to generate an i-th deformed frequency spectrum by multiplying a window function over at least one of a plurality of frequency intervals obtained by dividing the frequency spectrum of i such that each interval contains a plurality of distinct frequencies that are rational multiples of the frequency corresponding to the reciprocal of the time interval of the trigger time series, thereby deforming the i-th frequency spectrum. For each of the aforementioned integers i, the process includes an inverse transform time waveform generation step, which inversely transforms the modified frequency spectrum of i into the time domain to generate the ith inverse transform time waveform, A synchronous addition step is performed on each of the aforementioned integers i, based on the time series of the trigger, to generate the ith synchronous addition waveform by performing synchronous addition on the inverse time waveform of i, Have the computer run it, For each integer i, the duration of the time waveform of i is equal to or greater than the duration of the product of the time interval of the trigger time series and the number of synchronous additions. [Brief explanation of the drawing]

[0008] [Figure 1] A flowchart illustrating the procedure of the signal processing method according to the first embodiment. [Figure 2]Schematic perspective view showing the configuration of a vacuum pump. [Figure 3] Figure showing an example of a time waveform. [Figure 4] Figure showing an example of a frequency spectrum. [Figure 5] Figure showing an example of a modified frequency spectrum. [Figure 6] Figure showing an example of an inverse-transformed time waveform. [Figure 7] Figure showing a shifted inverse-transformed time waveform. [Figure 8] Figure showing an example of a synchronous addition waveform. [Figure 9] Figure showing a comparative example of a synchronous addition waveform. [Figure 10] Figure showing another example of a synchronous addition waveform. [Figure 11] Figure showing a configuration example of a signal processing device that implements the signal processing method according to the first embodiment. [Figure 12] Flowchart showing the procedure of the signal processing method according to the second embodiment. [Figure 13] Figure showing a comparative example of a synchronous addition waveform. [Figure 14] Figure showing a configuration example of a signal processing device that implements the signal processing method according to the second embodiment. [Figure 15] Flowchart showing the procedure of the signal processing method according to the third embodiment. [Figure 16] Figure showing an example of a time waveform obtained by integrating a synchronous addition waveform. [Figure 17] Figure showing a comparative example of a time waveform obtained by integrating a synchronous addition waveform. [Figure 18] Figure showing an example of a time waveform obtained by differentiating a synchronous addition waveform. [Figure 19] Figure showing a comparative example of a time waveform obtained by differentiating a synchronous addition waveform. [Figure 20] Figure showing a configuration example of a signal processing device that implements the signal processing method according to the third embodiment. [Figure 21] Flowchart showing the procedure of the signal processing method according to the fourth embodiment. [Figure 22] Figure showing an example of a Lissajous figure. [Figure 23]A diagram showing the first comparative example of Lissajous figures. [Figure 24] A diagram showing a second comparative example of Lissajous figures. [Figure 25] A diagram showing an example of how Lissajous figures change over time. [Figure 26] A diagram showing another example of how Lissajous figures change over time. [Figure 27] A diagram showing another example of how Lissajous figures change over time. [Figure 28] A diagram showing another example of how Lissajous figures change over time. [Figure 29] A diagram showing an example configuration of a signal processing device that performs the signal processing method of the fourth embodiment. [Modes for carrying out the invention]

[0009] Preferred embodiments of the present invention will be described in detail below with reference to the drawings. The embodiments described below are not intended to unduly limit the scope of the present invention as described in the claims. Furthermore, not all of the configurations described below are necessarily essential components of the present invention.

[0010] 1. First Embodiment 1-1. Signal Processing Method Figure 1 is a flowchart showing the procedure of the signal processing method of the first embodiment. As shown in Figure 1, the signal processing method of the first embodiment includes a time waveform acquisition step S10, a frequency spectrum generation step S20, a trigger time series generation step S30, a frequency spectrum deformation step S40, an inverse conversion time waveform generation step S50, and a synchronous addition step S60. Note that some of these steps may be omitted or modified, or other steps may be added to the signal processing method of the first embodiment. The signal processing method of the first embodiment is executed by, for example, a signal processing device 100. An example of the configuration of the signal processing device 100 that executes the signal processing method of the first embodiment will be described later.

[0011] As shown in Figure 1, first, in the time waveform acquisition step S10, the signal processing device 100 acquires the i-th time waveform relating to the i-th physical quantity from the i-th sensor for each integer i between 1 and N, where N is a predetermined integer of 1 or more.

[0012] The i-th physical quantity is a physical quantity that arises when an external force, velocity, or displacement acts on an object, and this external force, velocity, or displacement includes at least a periodic fluctuation having a first frequency F1 component and a second frequency F2 component that is different from the first frequency F1 and the higher-order frequencies of the first frequency F1.

[0013] The i-th time waveform may be time-series data of a digital signal output from the i-th sensor, or time-series data of a digital signal converted by an analog front-end from an analog signal output from the i-th sensor. In this embodiment, when the integer N is 2 or greater, the first to n-th time waveforms are synchronized with each other. The object is the object to be processed by the signal, and its type is not particularly limited. For example, it may be various devices such as motors with rotational or vibrational mechanisms, structures such as bridges or buildings that vibrate due to external forces, or electrical circuits that generate periodic signals.

[0014] The types of physical quantities from the 1st to the Nth are not particularly limited. For example, the 1st to the Nth physical quantities may be acceleration, angular velocity, speed, displacement, pressure, current, voltage, etc. If the integer N is 2 or greater, the 1st to the Nth physical quantities may be of the same type. That is, the 1st to the Nth sensors may be sensors that detect the same type of physical quantity. For example, for mutually orthogonal x, y, and z axes, the 1st sensor may detect velocity in the x-axis direction as the 1st physical quantity, the 2nd sensor may detect velocity in the y-axis direction as the 2nd physical quantity, and the 3rd sensor may detect velocity in the z-axis direction as the 3rd physical quantity. Alternatively, some of the 1st to Nth sensors may be sensors that detect different types of physical quantities than the others. For example, the 1st sensor may detect acceleration in the x-axis direction as the 1st physical quantity, and the 2nd sensor may detect angular velocity in the y-axis direction as the 2nd physical quantity. Furthermore, the 1st to Nth sensors may be, for example, sensors using MEMS or sensors using quartz crystal oscillators. MEMS stands for Micro Electro Mechanical Systems. The first to Nth sensors may be integrated into a single device, such as an IMU, or at least one of the first to Nth sensors may be physically separated from the others. IMU stands for Inertial Measurement Unit.

[0015] Figure 2 shows a vacuum pump 1, which is an example of the object. As shown in Figure 2, the vacuum pump 1 is installed on a base 20. The vacuum pump 1 has a roughly elongated oval cross-sectional shape. The longitudinal direction of the vacuum pump 1 is defined as the X direction. The direction of the long axis of the oval is defined as the Y direction, and the direction of the short axis of the oval is defined as the Z direction.

[0016] The vacuum pump 1 comprises a housing 3. The housing 3 includes a motor case 4, a connecting section 5, a pump case 6, and a gear case 7 arranged from the -X direction to the +X direction. The housing 3 includes a first side wall 8, which serves as a bearing casing, between the connecting section 5 and the pump case 6. The housing 3 includes a second side wall 9 between the pump case 6 and the gear case 7.

[0017] An intake pipe 11 is connected to the pump case 6 on the side facing the +Z direction. An exhaust pipe 12 is connected to the pump case 6 on the side facing the -Z direction.

[0018] The connecting section 5 is provided with a first leg 13 and a second leg on the base 20 side. The first leg 13 is positioned on the -Y direction side, and the second leg is positioned on the +Y direction side. The gear case 7 is provided with a third leg 14 and a fourth leg on the base 20 side. The third leg 14 is positioned on the -Y direction side, and the fourth leg is positioned on the +Y direction side. The first to fourth legs are fastened to the base 20 by a first bolt 15.

[0019] A sensor unit 17 is mounted on the housing 3. The sensor unit 17 is attached, for example, to the connection part 5. The sensor unit 17 contains first to Nth sensors (not shown) inside. For example, the first sensor may be a velocity sensor that detects velocity in the x-axis direction, the second sensor may be a velocity sensor that detects velocity in the y-axis direction, and the third sensor may be a velocity sensor that detects velocity in the z-axis direction. For example, the sensor unit 17 is mounted so that the x, y, and z axes coincide with the +X, +Y, and +Z directions, respectively.

[0020] Figure 3 shows an example of the first to nth time waveforms acquired by the signal processing device 100 in the time waveform acquisition process S10, with integer N set to 3. The three time waveforms shown in Figure 3 are time waveforms of the three-axis velocity based on the output signals of the three-axis velocity sensors attached to a rotating device such as the vacuum pump 1 shown in Figure 2. In the example in Figure 3, the integer N is 3, the time waveform of the x-axis velocity Vx corresponds to the first time waveform, the time waveform of the y-axis velocity Vy corresponds to the second time waveform, and the time waveform of the z-axis velocity Vz corresponds to the third time waveform. In Figure 3, the horizontal axis is time and the vertical axis is velocity. The x-axis velocity Vx is the velocity detected by the x-axis velocity sensor, which is the first sensor built into the three-axis velocity sensor. The y-axis velocity Vy is the velocity detected by the y-axis velocity sensor, which is the second sensor built into the three-axis velocity sensor. The z-axis velocity Vz is the velocity detected by the z-axis velocity sensor, which is the third sensor built into the three-axis velocity sensor. In Figure 3, only the 1-second time waveform from each acquired time waveform is shown, but in reality, time waveforms for the duration required for processing from step S20 onward, for example, 30 seconds or more, are acquired.

[0021] As shown in Figure 1, the signal processing device 100 then generates the i-th frequency spectrum for each integer i between 1 and N, based on the i-th time waveform acquired in the time waveform acquisition step S10, in the frequency spectrum generation step S20. For example, the signal processing device 100 may generate the first to N frequency spectra by performing a Fast Fourier Transform on each of the first to N time waveforms.

[0022] Figure 4 shows the frequency spectra of the time waveforms of the x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz, respectively, as shown in Figure 3. In Figure 4, the horizontal axis represents frequency and the vertical axis represents intensity. For example, in the frequency spectrum generation step S20, the signal processing device 100 generates the frequency spectra of the x-axis velocity Vx, the y-axis velocity Vy, and the z-axis velocity Vz.

[0023] As shown in Figure 1, the signal processing device 100 then generates a trigger time series in the trigger time series generation step S30, where the time interval ΔT is a first period T1 or a real number multiple of the first period T1, corresponding to the frequency of any of the multiple peaks included in at least one of the first to Nth frequency spectra generated in the frequency spectrum generation step S20. The first period T1 is the reciprocal of the first frequency F1. The real number multiple of the first period T1 is a period defined by the components of the object. For example, if the object is a rotating machine, the components are, for example, gears and bearings. The time interval ΔT of the trigger time series may also be a natural number multiple of the first period T1. For example, the time interval ΔT may be 2 or 3 times the first period T1. The trigger time series is a time series in which triggers occur at approximately ΔT intervals, and the average value of the trigger occurrence period is ΔT. For example, the time interval ΔT of the trigger time series is input by the user of the signal processing device 100.

[0024] In Figure 4 above, the frequency spectra of the x-axis velocity Vx, the y-axis velocity Vy, and the z-axis velocity Vz contain multiple peaks. Some of these peaks, indicated by black circles, are the fundamental wave component at approximately 84 Hz and its harmonic components. For example, if we want to focus on the vibration of the fundamental wave component at approximately 84 Hz, the signal processing device 100 generates a trigger time series in the trigger time series generation step S30, for example, by setting the first frequency F1 to approximately 84 Hz and the time interval ΔT to approximately 12 ms, which is the reciprocal of approximately 84 Hz. Alternatively, if we want to focus on the vibration of gears or bearings, the signal processing device 100 generates a trigger time series in the trigger time series generation step S30, where the vibration period of the gears or bearings is the time interval ΔT. In this case, if the first frequency F1 is approximately 84 Hz, the vibration frequency of the gears or bearings is a real multiple of the first frequency F1, and the reciprocal of this vibration frequency becomes the time interval ΔT of the trigger time series.

[0025] As shown in Figure 1, the signal processing device 100 then, in the frequency spectrum deformation step S40, multiplies the ith frequency spectrum generated in the frequency spectrum generation step S20 by a window function over at least one of several frequency intervals FS, each of which contains several distinct frequencies that are rational multiples of the frequency corresponding to the reciprocal of the time interval ΔT of the trigger time series generated in the trigger time series generation step S30. The type of window function is not particularly limited, and examples of window functions include the Hanning window function, rectangular window function, Gaussian window function, Hamming window function, Blackman window function, and Kaiser window function. Specifically, the signal processing device 100 divides the ith frequency spectrum such that each distinct frequency that is a rational multiple of the frequency corresponding to the reciprocal of the time interval ΔT is near the center of each frequency interval FS.

[0026] Figure 5 shows the modified frequency spectra of x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz, respectively, obtained by modifying the frequency spectra of x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz shown in Figure 4. In the example in Figure 5, each frequency spectrum is divided into multiple frequency intervals FS of the same frequency width such that the peak at approximately 84 Hz, which is the fundamental wave component, and the peaks of its harmonic components are located near the center of the multiple frequency intervals FS. Each modified frequency spectrum is obtained by multiplying all frequency intervals FS by a Hanning window function. Therefore, in each modified frequency spectrum, the intensity of the signal components near the ends of the multiple frequency intervals FS is greatly attenuated.

[0027] As shown in Figure 1, the signal processing device 100 then, in the inverse conversion time waveform generation step S50, inversely converts the ith modified frequency spectrum generated in the frequency spectrum deformation step S40 into the time domain for each integer i between 1 and N, thereby generating the ith inverse conversion time waveform.

[0028] Figure 6 shows the inverse transformed time waveforms of the x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz, respectively, obtained by inversely transforming the deformed frequency spectra of the x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz shown in Figure 5 above.

[0029] As shown in Figure 1, the signal processing device 100 then performs a synchronous addition in the synchronous addition step S60 on the ith inverse converted time waveform generated in the inverse converted time waveform generation step S50, based on the trigger time series generated in the trigger time series generation step S30, for each integer i between 1 and N, thereby generating the ith synchronous addition waveform. Specifically, in the synchronous addition step S60, the signal processing device 100 generates multiple time waveforms for each integer i between 1 and N, by shifting the ith inverse converted time waveform by the timing of each trigger included in the trigger time series, i.e., M-1 time waveforms obtained by shifting the ith inverse converted time waveform by ΔT, and then adds all M time waveforms consisting of the ith inverse converted time waveform and the M-1 time waveforms to generate the ith synchronous addition waveform. The integer M corresponds to the number of synchronous additions. In this embodiment, for each integer i between 1 and N, the duration of the i-th time waveform and the duration of the i-th inverse time waveform are equal to or greater than the time equivalent to the product of the time interval ΔT of the trigger time series and the integer M which is the number of synchronous additions. Under this condition, the first to Nth synchronous addition waveforms with durations equal to or greater than the time interval ΔT of the trigger time series are obtained.

[0030] Figure 7 shows the inverse time waveform of the x-axis velocity Vx in Figure 6, the time waveform of the x-axis velocity Vx' obtained by shifting the inverse time waveform of the x-axis velocity Vx by ΔT, and the time waveform of the x-axis velocity Vx'' obtained by shifting the inverse time waveform of the x-axis velocity Vx by ΔT × 2. ΔT is approximately 12 ms. In the synchronous addition process S60, the signal processing device 100 generates M-1 time waveforms for each integer j between 1 and M, obtained by shifting the time waveform of the x-axis velocity Vx by ΔT × j. M is an integer of 2 or more. The signal processing device 100 then adds the inverse time waveform of the x-axis velocity Vx and the M-1 time waveforms together to generate a synchronous addition waveform of the x-axis velocity Vx as the first synchronous addition waveform. Similarly, in the synchronous addition step S60, the signal processing device 100 generates M-1 time waveforms by shifting the inverse time waveform of the y-axis velocity Vy by ΔT × j, and adds the inverse time waveform of the y-axis velocity Vy and the M-1 time waveforms together to generate a second synchronous addition waveform of the y-axis velocity Vy. Similarly, in the synchronous addition step S60, the signal processing device 100 generates M-1 time waveforms by shifting the inverse time waveform of the z-axis velocity Vz by ΔT × j, and adds the inverse time waveform of the z-axis velocity Vz and the M-1 time waveforms together to generate a third synchronous addition waveform of the z-axis velocity Vz.

[0031] Figure 8 shows examples of synchronous summation waveforms based on the inverse time waveform of x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz. Figure 9 shows, as a comparative example, synchronous summation waveforms based on the time waveform of x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz. In both the example in Figure 8 and the example in Figure 9, the number of synchronous summations is 216. Comparing Figure 8 and Figure 9, it can be seen that by performing synchronous summation on the inverse time waveform of velocity, the influence of the signal component at approximately 84 Hz and the asynchronous signal component is reduced, resulting in smaller fluctuations. Figure 10 shows another example of synchronous summation waveforms based on the inverse time waveform of x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz. In the example in Figure 10, the number of synchronous summations is 1728. Comparing Figure 8 and Figure 10, it can be seen that the more synchronous summation cycles there are, the smaller the amplitude fluctuation of the signal component at approximately 84 Hz, and the asynchronous low-frequency components at approximately 84 Hz are also reduced.

[0032] As shown in Figure 1, the signal processing device 100 repeatedly performs steps S10 to S60 until the signal processing is completed (step N of step S100).

[0033] 1-2. Signal Processing Device Figure 11 shows an example configuration of a signal processing device 100 that performs the signal processing method of the first embodiment. As shown in Figure 11, the signal processing device 100 includes first to Nth sensors 200-1 to 200-N, N analog front ends 210-1 to 210-N, a processing circuit 110, a memory circuit 120, an operation unit 130, a display unit 140, a sound output unit 150, and a communication unit 160. Note that the signal processing device 100 may have a configuration in which some of the components in Figure 11 are omitted or changed, or other components are added. For example, the first to Nth sensors 200-1 to 200-N and the analog front ends 210-1 to 210-N do not have to be components of the signal processing device 100.

[0034] For each integer i between 1 and N, the i-th sensor 200-i detects the i-th physical quantity resulting from an external force, velocity, or displacement acting on the object, and outputs a signal of a magnitude corresponding to the detected i-th physical quantity. The output signals of the first to Nth sensors 200-1 to 200-N are input to the analog front ends 210-1 to 210-N, respectively.

[0035] Each of the analog front-ends 210-1 to 210-N performs amplification and A / D conversion processing on the output signals of the first to Nth sensors 200-1 to 200-N, and outputs a digital time-series signal.

[0036] The processing circuit 110 acquires N digital time-series signals output from the analog front-ends 210-1 to 210-N as first to ninth time waveforms and performs signal processing. Specifically, the processing circuit 110 executes the signal processing program 121 stored in the memory circuit 120 and performs various calculations on the first to ninth time waveforms. In addition, the processing circuit 110 performs various processes in response to operation signals from the operation unit 130, processes to transmit display signals to display various information to the display unit 140, processes to transmit sound signals to generate various sounds to the sound output unit 150, and processes to control the communication unit 160 for data communication with external devices (not shown). The processing circuit 110 can be implemented by, for example, a CPU or a DSP. CPU stands for Central Processing Unit, and DSP stands for Digital Signal Processor.

[0037] The processing circuit 110 functions as a time waveform acquisition circuit 111, a frequency spectrum generation circuit 112, a trigger time series generation circuit 113, a frequency spectrum deformation circuit 114, an inverse time waveform generation circuit 115, and a synchronous summing circuit 116 by executing the signal processing program 121. In other words, the signal processing device 100 includes a time waveform acquisition circuit 111, a frequency spectrum generation circuit 112, a trigger time series generation circuit 113, a frequency spectrum deformation circuit 114, an inverse time waveform generation circuit 115, and a synchronous summing circuit 116.

[0038] The time waveform acquisition circuit 111 acquires the i-th time waveform relating to the i-th physical quantity from the i-th sensor 200-i for each integer i between 1 and N, where N is a predetermined integer of 1 or greater. That is, the time waveform acquisition circuit 111 performs the time waveform acquisition process S10 shown in Figure 1. The first to N time waveforms acquired by the time waveform acquisition circuit 111 are stored in the memory circuit 120.

[0039] The frequency spectrum generation circuit 112 generates the ith frequency spectrum for each integer i between 1 and N, based on the ith time waveform acquired by the time waveform acquisition circuit 111. That is, the frequency spectrum generation circuit 112 performs the frequency spectrum generation process S20 shown in Figure 1. The first to Nth frequency spectra generated by the frequency spectrum generation circuit 112 are stored in the memory circuit 120.

[0040] The trigger time series generation circuit 113 generates a trigger time series with a time interval ΔT that corresponds to the frequency of any of the multiple peaks included in at least one of the first to Nth frequency spectra generated by the frequency spectrum generation circuit 112, and has a first period T1 or a real multiple of the first period T1. That is, the trigger time series generation circuit 113 performs the trigger time series generation step S30 shown in Figure 1. The trigger time series generated by the trigger time series generation circuit 113 is stored in the memory circuit 120.

[0041] The frequency spectrum deformation circuit 114, for each integer i between 1 and N, multiplies the ith frequency spectrum generated by the frequency spectrum generation circuit 112 by a window function over at least one of several frequency intervals FS, each containing multiple distinct frequencies that are rational multiples of the frequency corresponding to the reciprocal of the time interval ΔT of the trigger time series generated by the trigger time series generation circuit 113. In other words, the frequency spectrum deformation circuit 114 performs the frequency spectrum deformation process S40 shown in Figure 1. The first to Nth deformation frequency spectra generated by the frequency spectrum deformation circuit 114 are stored in the memory circuit 120.

[0042] The inverse time waveform generation circuit 115 generates the ith inverse time waveform by inversely transforming the ith deformed frequency spectrum generated by the frequency spectrum deformation circuit 114 into the time domain for each integer i between 1 and N. That is, the inverse time waveform generation circuit 115 performs the inverse time waveform generation process S50 shown in Figure 1. The first to Nth inverse time waveforms generated by the inverse time waveform generation circuit 115 are stored in the memory circuit 120.

[0043] The synchronous adder circuit 116 performs synchronous addition on the ith inverse transform time waveform generated by the inverse transform time waveform generation circuit 115, based on the trigger time series generated by the trigger time series generation circuit 113, for each integer i between 1 and N, thereby generating the ith synchronous adder waveform. That is, the synchronous adder circuit 116 executes the synchronous adder process S60 shown in Figure 1. The first to Nth synchronous adder waveforms generated by the synchronous adder circuit 116 are stored in the memory circuit 120.

[0044] Thus, the signal processing program 121 is a program that causes the computer-based processing circuit 110 to execute the following steps: time waveform acquisition step S10, frequency spectrum generation step S20, trigger time series generation step S30, frequency spectrum deformation step S40, inverse conversion time waveform generation step S50, and synchronous addition step S60.

[0045] The memory circuit 120 includes a ROM and RAM (not shown). ROM stands for Read Only Memory, and RAM stands for Random Access Memory. The ROM stores various programs such as the signal processing program 121 and predetermined data, while the RAM stores data generated by the processing circuit 110. The RAM is also used as a workspace for the processing circuit 110, and stores programs and data read from the ROM, data input from the operation unit 130, and data temporarily generated by the processing circuit 110.

[0046] The operation unit 130 is an input device consisting of operation keys, button switches, etc., and outputs operation signals to the processing circuit 110 in response to user operations.

[0047] The display unit 140 is a display device composed of an LCD or the like, and displays various information based on the display signal output from the processing circuit 110. LCD stands for Liquid Crystal Display. The display unit 140 may also be provided with a touch panel that functions as an operation unit 130. For example, the display unit 140 may display a screen that includes at least a portion of the various data stored in the memory circuit 120 based on the display signal output from the processing circuit 110.

[0048] The sound output unit 150 is composed of a speaker or the like and generates various sounds based on the sound signals output from the processing circuit 110. For example, the sound output unit 150 may generate sounds to indicate the start or end of signal processing based on the sound signals output from the processing circuit 110.

[0049] The communication unit 160 performs various controls to establish data communication between the processing circuit 110 and the external device. For example, the communication unit 160 may transmit at least a portion of the information of various data stored in the memory circuit 120 to the external device, and the external device may display the received information on a display unit (not shown).

[0050] Furthermore, at least a portion of the time waveform acquisition circuit 111, frequency spectrum generation circuit 112, trigger time series generation circuit 113, frequency spectrum deformation circuit 114, inverse conversion time waveform generation circuit 115, and synchronous summing circuit 116 may be implemented in dedicated hardware. Also, the signal processing device 100 may be a single device or may be composed of multiple devices. For example, the first to Nth sensors 200-1 to 200-N and analog front-ends 210-1 to 210-N may be included in the first device, while the processing circuit 110, memory circuit 120, operation unit 130, display unit 140, sound output unit 150, and communication unit 160 may be included in a second device separate from the first device. Alternatively, for example, the processing circuit 110 and the memory circuit 120 may be implemented by a device such as a cloud server, which generates the first to Nth synchronous summation waveforms and transmits the generated first to Nth synchronous summation waveforms to a terminal including an operation unit 130, a display unit 140, a sound output unit 150, and a communication unit 160 via a communication line.

[0051] 1-3. Effects According to the signal processing method of the first embodiment, the signal processing device 100 generates a trigger time series from the first to the Nth time waveforms, so a rotating pulse signal is not required. Furthermore, by performing synchronous summation on the first to the Nth time waveforms, the signal component of the target synchronized with the first period T1, which is the reciprocal of the first frequency F1, or a real multiple of the first period T1, is emphasized, and the signal component of the second frequency F2 that is asynchronous with the target signal component is reduced, resulting in the first to the Nth synchronous summation waveforms, so there is no need to use a PLL, tracking filter, low-pass filter, etc. in combination. In addition, when the target signal component is a low-intensity signal component having a real multiple of the first period T1 defined by the components of the object, the time interval ΔT of the trigger time series can be matched with the period of the target signal component using the first period T1, which is the period of the high-intensity signal component, so that the first to the Nth synchronous summation waveforms in which the target signal component is emphasized can be obtained. Furthermore, since the time lengths of the first to nth time waveforms are each equal to or greater than the time corresponding to the product of the time interval ΔT of the trigger time series and the number of synchronous additions, the first to nth synchronous addition waveforms with a time length equal to or greater than the time interval ΔT can be obtained. In this way, according to the signal processing method of the first embodiment, the signal processing device 100 can generate the first to nth synchronous addition waveforms that reduce the signal components asynchronous with the target signal components without requiring a rotation pulse signal.

[0052] Furthermore, according to the signal processing method of the first embodiment, the frequency spectrum deformation step S40 and the inverse conversion time waveform generation step S50 can reduce the signal components that are asynchronous with the target signal component in the step prior to the synchronous addition step S60, so that the number of synchronous additions can be reduced compared to when the frequency spectrum deformation step S40 and the inverse conversion time waveform generation step S50 are not performed.

[0053] Furthermore, according to the signal processing method of the first embodiment, when the integer N is 2 or greater, synchronous addition can be performed on multiple inverse-converted time waveforms based on the time series of a common trigger. Moreover, according to the signal processing method of the first embodiment, when the integer N is 2 or greater, the first to Nth time waveforms are synchronized with each other, so the accuracy of synchronous addition is improved, and the signal-to-noise ratio of the first to Nth synchronously added waveforms is improved.

[0054] Furthermore, according to the signal processing method of the first embodiment, when the time interval ΔT of the trigger time series is a natural number multiple of the first period T1, first to N synchronous summation waveforms are obtained in which signal components with frequencies below the first frequency F1 that are synchronized with the signal component of the first frequency F1 are emphasized.

[0055] 2. Second Embodiment In the following description of the second embodiment, the same reference numerals are used for components similar to those in the first embodiment, and explanations that overlap with those in the first embodiment will be omitted or simplified. The main points to be described will be those that differ from the first embodiment.

[0056] Figure 12 is a flowchart showing the procedure of the signal processing method of the second embodiment. As shown in Figure 12, the signal processing method of the second embodiment includes a time waveform acquisition step S10, a time waveform synchronization step S12, a frequency spectrum generation step S20, a trigger time series generation step S30, a frequency spectrum deformation step S40, an inverse conversion time waveform generation step S50, and a synchronous addition step S60. Note that in the signal processing method of the second embodiment, some of these steps may be omitted or modified, or other steps may be added. The signal processing method of the second embodiment is executed by, for example, a signal processing device 100. An example of the configuration of the signal processing device 100 that executes the signal processing method of the second embodiment will be described later.

[0057] As shown in Figure 12, first, the signal processing device 100 performs the same time waveform acquisition process S10 as in Figure 1.

[0058] In the second embodiment, the integer N is 2 or greater, and two or more of the first to nth time waveforms acquired by the signal processing device 100 in the time waveform acquisition step S10 are asynchronous with respect to each other. Therefore, the signal processing device 100 then synchronizes the first to nth time waveforms acquired in the time waveform acquisition step S10 with respect to each other in the time waveform synchronization step S12, before the frequency spectrum generation step S20. For example, the signal processing device 100 may synchronize the first to nth time waveforms by resampling them at a predetermined sampling rate.

[0059] Next, in the frequency spectrum generation step S20, the signal processing device 100 generates the i-th frequency spectrum for each integer i between 1 and N, based on the first to N time waveforms synchronized with each other in the time waveform synchronization step S12. The details of the processing by the signal processing device 100 in this frequency spectrum generation step S20 are the same as in Figure 1.

[0060] Next, the signal processing device 100 performs a trigger time series generation process S30 similar to that in Figure 1. Next, the signal processing device 100 performs a frequency spectrum deformation process S40 similar to that in Figure 1. Next, the signal processing device 100 performs an inverse conversion time waveform generation process S50 similar to that in Figure 1.

[0061] Next, in the synchronous addition step S60, the signal processing device 100 performs synchronous addition on each integer i between 1 and N, based on the trigger time series generated in the trigger time series generation step S30, to the ith inverse transform time waveform generated in the inverse transform time waveform generation step S50, thereby generating the ith synchronous add waveform. The details of the processing by the signal processing device 100 in this synchronous addition step S60 are the same as in Figure 1.

[0062] Then, until the signal processing is completed (N in step S100), the signal processing device 100 repeatedly performs steps S10 to S60.

[0063] In the example shown in Figure 3, we assume that the sampling frequencies of the x-axis velocity sensor and the z-axis velocity sensor are equal, and the sampling frequency of the y-axis velocity sensor is 100 ppm higher. Figure 13 shows an example of the synchronous sum waveforms of the x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz generated in the synchronous sum step S60 when the signal processing device 100 does not perform the time waveform synchronization step S12. In the example in Figure 13, approximately 84 Hz, which corresponds to one of the multiple peaks in the frequency spectrum of the x-axis velocity Vx, is used as the first frequency F1, and the time interval ΔT of the trigger time series is set to the reciprocal of the first frequency F1, and 1728 synchronous sums are performed. On the other hand, an example of the synchronous sum waveforms of the x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz when the signal processing device 100 performs the time waveform synchronization step S12 is the same as in Figure 10. Comparing Figure 10 and Figure 13, the synchronous summation waveforms for x-axis velocity Vx and z-axis velocity Vz in Figure 10 are the same as those for x-axis velocity Vx and z-axis velocity Vz in Figure 13, respectively. However, the periodic peak of approximately 84 Hz included in the synchronous summation waveform for y-axis velocity Vy is sharper in Figure 10 than in Figure 13. This indicates that the signal component of approximately 84 Hz is further emphasized when the signal processing device 100 performs the time waveform synchronization process S12.

[0064] Figure 14 shows an example configuration of a signal processing device 100 that performs the signal processing method of the second embodiment. As shown in Figure 14, the signal processing device 100 includes first to nth sensors 200-1 to 200-N, N analog front ends 210-1 to 210-N, a processing circuit 110, a memory circuit 120, an operation unit 130, a display unit 140, an audio output unit 150, and a communication unit 160. Note that the signal processing device 100 may have a configuration in which some of the components in Figure 14 are omitted or changed, or other components are added. For example, the first to nth sensors 200-1 to 200-N and the analog front ends 210-1 to 210-N do not have to be components of the signal processing device 100.

[0065] The configuration and functions of the first to nth sensors 200-1 to 200-N, the analog front end 210-1 to 210-N, the memory circuit 120, the operation unit 130, the display unit 140, the sound output unit 150, and the communication unit 160 are the same as in the first embodiment, so their description is omitted.

[0066] The processing circuit 110 functions as a time waveform acquisition circuit 111, a frequency spectrum generation circuit 112, a trigger time series generation circuit 113, a frequency spectrum deformation circuit 114, an inverse time waveform generation circuit 115, a synchronous summing circuit 116, and a time waveform synchronization circuit 117 by executing the signal processing program 121 stored in the memory circuit 120. In other words, the signal processing device 100 includes a time waveform acquisition circuit 111, a frequency spectrum generation circuit 112, a trigger time series generation circuit 113, a frequency spectrum deformation circuit 114, an inverse time waveform generation circuit 115, a synchronous summing circuit 116, and a time waveform synchronization circuit 117.

[0067] The functions of the time waveform acquisition circuit 111, the trigger time series generation circuit 113, the frequency spectrum deformation circuit 114, and the inverse transform time waveform generation circuit 115 are the same as in the first embodiment, so their description is omitted.

[0068] The time waveform synchronization circuit 117 synchronizes the first to the Nth time waveforms acquired by the time waveform acquisition circuit 111 with each other. That is, the time waveform synchronization circuit 117 performs the time waveform synchronization process S12 shown in Figure 12. The first to the Nth time waveforms synchronized by the time waveform synchronization circuit 117 are stored in the memory circuit 120.

[0069] The frequency spectrum generation circuit 112 generates the i-th frequency spectrum for each integer i between 1 and N, based on the first to N time waveforms synchronized by the time waveform synchronization circuit 117. That is, the frequency spectrum generation circuit 112 performs the frequency spectrum generation process S20 shown in Figure 12. The frequency spectrum generated by the frequency spectrum generation circuit 112 is stored in the memory circuit 120.

[0070] The synchronous adder circuit 116 performs synchronous addition on the ith inverse transform time waveform generated by the inverse transform time waveform generation circuit 115, based on the trigger time series generated by the trigger time series generation circuit 113, for each integer i between 1 and N, thereby generating the ith synchronous adder waveform. That is, the synchronous adder circuit 116 executes the synchronous adder step S60 shown in Figure 12. The first to Nth synchronous adder waveforms generated by the synchronous adder circuit 116 are stored in the memory circuit 120.

[0071] Thus, the signal processing program 121 is a program that causes the computer-based processing circuit 110 to execute the following steps: time waveform acquisition step S10, time waveform synchronization step S12, frequency spectrum generation step S20, trigger time series generation step S30, frequency spectrum deformation step S40, inverse conversion time waveform generation step S50, and synchronous addition step S60.

[0072] Furthermore, at least a portion of the time waveform acquisition circuit 111, frequency spectrum generation circuit 112, trigger time series generation circuit 113, frequency spectrum deformation circuit 114, inverse conversion time waveform generation circuit 115, synchronous summing circuit 116, and time waveform synchronization circuit 117 may be implemented using dedicated hardware.

[0073] The signal processing method of the second embodiment described above provides the same effects as the signal processing method of the first embodiment. Furthermore, according to the signal processing method of the second embodiment, even if two or more of the first to Nth time waveforms are asynchronous with respect to each other, the signal processing device 100 synchronizes the first to Nth time waveforms with respect to each other before performing synchronous addition, thereby improving the accuracy of synchronous addition and improving the signal-to-noise ratio of the first to Nth synchronously added waveforms.

[0074] 3. Third Embodiment In the following description of the third embodiment, the same reference numerals are used for components similar to those in the first or second embodiment. Descriptions that overlap with those of the first or second embodiment will be omitted or simplified, and the differences from the first and second embodiments will be described primarily.

[0075] Figure 15 is a flowchart showing the procedure of the signal processing method of the third embodiment. As shown in Figure 15, the signal processing method of the third embodiment includes a time waveform acquisition step S10, a time waveform synchronization step S12, a frequency spectrum generation step S20, a trigger time series generation step S30, a frequency spectrum deformation step S40, an inverse conversion time waveform generation step S50, and a synchronous addition step S60. Furthermore, the signal processing method of the third embodiment includes at least one of an integration step S70 and a differentiation step S80. Note that in the signal processing method of the third embodiment, some of these steps may be omitted or modified, or other steps may be added. The signal processing method of the third embodiment is executed, for example, by a signal processing device 100. An example of the configuration of the signal processing device 100 that executes the signal processing method of the third embodiment will be described later.

[0076] As shown in Figure 15, first, the signal processing device 100 performs a time waveform acquisition process S10 similar to that in Figure 1 or Figure 12.

[0077] Next, if the integer N is 2 or greater and two or more of the first to nth time waveforms acquired in the time waveform acquisition step S10 are asynchronous with respect to each other, the signal processing device 100 performs a time waveform synchronization step S12 similar to that shown in Figure 12. Note that if the first to nth time waveforms acquired in the time waveform acquisition step S10 are synchronized with each other, the signal processing device 100 does not need to perform the time waveform synchronization step S12.

[0078] Next, the signal processing device 100 performs a frequency spectrum generation step S20 similar to that in Figure 1 or Figure 12. Next, the signal processing device 100 performs a trigger time series generation step S30 similar to that in Figure 1 or Figure 12. Next, the signal processing device 100 performs a frequency spectrum deformation step S40 similar to that in Figure 1 or Figure 12. Next, the signal processing device 100 performs an inverse conversion time waveform generation step S50 similar to that in Figure 1 or Figure 12. Next, the signal processing device 100 performs a synchronous addition step S60 similar to that in Figure 1 or Figure 12.

[0079] Next, in the integration step S70, the signal processing device 100 performs integration on the i-th synchronous sum waveform generated in the synchronous sum step S60 for each integer i between 1 and N.

[0080] Figure 16 shows the time waveforms of x-axis displacement Dx, y-axis displacement Dy, and z-axis displacement Dz, obtained by integrating the synchronous summation waveforms of x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz, respectively, as shown in Figure 8. As a comparative example, Figure 17 shows the time waveforms of x-axis displacement Dx, y-axis displacement Dy, and z-axis displacement Dz, obtained by integrating the time waveforms obtained by synchronously summing the time waveforms of x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz, respectively, as shown in Figure 3. In the examples in Figure 16 and Figure 17, the number of synchronous summations is 216 in both cases. Comparing Figure 16 and Figure 17, it can be seen that the signal processing device 100 performs integration processing on the inverse conversion time waveform of velocity after synchronous summation, which reduces the influence of the signal component of approximately 84 Hz and the asynchronous signal component in the time waveform of the integrated displacement, resulting in smaller fluctuations.

[0081] Next, in the differentiation step S80, the signal processing device 100 performs differentiation on the i-th synchronous sum waveform generated in the synchronous sum step S60 for each integer i between 1 and N (inclusive).

[0082] Figure 18 shows the time waveforms of x-axis acceleration Ax, y-axis acceleration Ay, and z-axis acceleration Az, obtained by differentiating the synchronous summation waveforms of x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz, respectively, as shown in Figure 8. As a comparative example, Figure 19 shows the time waveforms of x-axis acceleration Ax, y-axis acceleration Ay, and z-axis acceleration Az, obtained by differentiating the time waveforms obtained by synchronously summing the time waveforms of x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz, respectively, as shown in Figure 3. Comparing Figure 18 and Figure 19, it can be seen that the signal processing device 100 performs differentiation after synchronous summation on the inverse conversion time waveform of velocity, which reduces the influence of the signal component at approximately 84 Hz and the asynchronous signal component in the time waveform of the acceleration after differentiation, resulting in smaller fluctuations.

[0083] Then, until the signal processing is completed (N in process S100), the signal processing device 100 repeatedly performs processes S10 to S80. Note that the signal processing device 100 may perform the differentiation process S80 before the integration process S70, or it may omit either the integration process S70 or the differentiation process S80.

[0084] Figure 20 shows an example configuration of a signal processing device 100 that performs the signal processing method of the third embodiment. As shown in Figure 20, the signal processing device 100 includes first to nth sensors 200-1 to 200-N, N analog front ends 210-1 to 210-N, a processing circuit 110, a memory circuit 120, an operation unit 130, a display unit 140, an audio output unit 150, and a communication unit 160. Note that the signal processing device 100 may have a configuration in which some of the components in Figure 20 are omitted or changed, or other components are added. For example, the first to nth sensors 200-1 to 200-N and the analog front ends 210-1 to 210-N do not have to be components of the signal processing device 100.

[0085] The configuration and functions of the first to nth sensors 200-1 to 200-N, the analog front ends 210-1 to 210-N, the memory circuit 120, the operation unit 130, the display unit 140, the sound output unit 150, and the communication unit 160 are the same as in the first or second embodiment, so their description is omitted.

[0086] The processing circuit 110 functions as a time waveform acquisition circuit 111, a frequency spectrum generation circuit 112, a trigger time series generation circuit 113, a frequency spectrum deformation circuit 114, an inverse time waveform generation circuit 115, a synchronous summing circuit 116, a time waveform synchronization circuit 117, an integrating circuit 118, and a differentiating circuit 119 by executing the signal processing program 121 stored in the memory circuit 120. In other words, the signal processing device 100 includes a time waveform acquisition circuit 111, a frequency spectrum generation circuit 112, a trigger time series generation circuit 113, a frequency spectrum deformation circuit 114, an inverse time waveform generation circuit 115, a synchronous summing circuit 116, a time waveform synchronization circuit 117, an integrating circuit 118, and a differentiating circuit 119.

[0087] The functions of the time waveform acquisition circuit 111, frequency spectrum generation circuit 112, trigger time series generation circuit 113, frequency spectrum deformation circuit 114, inverse conversion time waveform generation circuit 115, synchronous summing circuit 116, and time waveform synchronization circuit 117 are the same as in the first or second embodiment, so their description is omitted.

[0088] The integrating circuit 118 performs integration on the i-th synchronous sum waveform generated by the synchronous summing circuit 116 for each integer i between 1 and N (inclusive). That is, the integrating circuit 118 performs the integration process S70 shown in Figure 15. The N time waveforms obtained by the integration process of the integrating circuit 118 are stored in the memory circuit 120.

[0089] The differentiating circuit 119 performs differentiation on the i-th synchronous sum waveform generated by the synchronous summing circuit 116 for each integer i between 1 and N (inclusive). That is, the differentiating circuit 119 executes the differentiation process S80 shown in Figure 15. The N time waveforms obtained by the differentiation process of the differentiating circuit 119 are stored in the memory circuit 120.

[0090] Thus, the signal processing program 121 is a program that causes the computer-based processing circuit 110 to execute the following steps: time waveform acquisition step S10, time waveform synchronization step S12, frequency spectrum generation step S20, trigger time series generation step S30, frequency spectrum deformation step S40, inverse conversion time waveform generation step S50, synchronous addition step S60, integration step S70, and differentiation step S80.

[0091] Furthermore, at least a portion of the time waveform acquisition circuit 111, frequency spectrum generation circuit 112, trigger time series generation circuit 113, frequency spectrum deformation circuit 114, inverse transform time waveform generation circuit 115, synchronous summing circuit 116, time waveform synchronization circuit 117, integration circuit 118, and differentiation circuit 119 may be implemented using dedicated hardware.

[0092] According to the third embodiment described above, the same effects as those of the signal processing method of the first or second embodiment can be obtained. Furthermore, according to the signal processing method of the third embodiment, the signal processing device 100 performs integration and differentiation on the first to Nth synchronous summation waveforms in which the target signal component is emphasized, thereby obtaining a time waveform of a physical quantity different from the physical quantity corresponding to the target signal component.

[0093] 4. Fourth Embodiment In the following description of the fourth embodiment, the same reference numerals are used for components similar to those in the first, second, or third embodiments. Descriptions that overlap with those of the first, second, or third embodiments will be omitted or simplified, and the differences from the first, second, and third embodiments will be described primarily.

[0094] Figure 21 is a flowchart showing the procedure of the signal processing method of the fourth embodiment. As shown in Figure 21, the signal processing method of the fourth embodiment includes a time waveform acquisition step S10, a time waveform synchronization step S12, a frequency spectrum generation step S20, a trigger time series generation step S30, a frequency spectrum deformation step S40, an inverse conversion time waveform generation step S50, a synchronization addition step S60, an integration step S70, a differentiation step S80, a state index calculation step S90, and a Lissajous figure generation step S92. Note that in the signal processing method of the fourth embodiment, some of these steps may be omitted or modified, or other steps may be added. The signal processing method of the fourth embodiment is executed by, for example, a signal processing device 100. An example of the configuration of the signal processing device 100 that executes the signal processing method of the fourth embodiment will be described later.

[0095] As shown in Figure 21, first the signal processing device 100 performs a time waveform acquisition process S10 similar to that in Figures 1, 12, or 15.

[0096] Next, if the integer N is 2 or greater and two or more of the first to nth time waveforms acquired in the time waveform acquisition step S10 are asynchronous with respect to each other, the signal processing device 100 performs a time waveform synchronization step S12 similar to that shown in Figure 12. Note that if the first to nth time waveforms acquired in the time waveform acquisition step S10 are synchronized with each other, the signal processing device 100 does not need to perform the time waveform synchronization step S12.

[0097] Next, the signal processing device 100 performs a frequency spectrum generation step S20 similar to that in Figure 1, Figure 12, or Figure 15. Next, the signal processing device 100 performs a trigger time series generation step S30 similar to that in Figure 1, Figure 12, or Figure 15. Next, the signal processing device 100 performs a frequency spectrum deformation step S40 similar to that in Figure 1, Figure 12, or Figure 15. Next, the signal processing device 100 performs an inverse conversion time waveform generation step S50 similar to that in Figure 1, Figure 12, or Figure 15. Next, the signal processing device 100 performs a synchronous addition step S60 similar to that in Figure 1, Figure 12, or Figure 15. Next, the signal processing device 100 performs an integration step S70 similar to that in Figure 1, Figure 12, or Figure 15. Next, the signal processing device 100 performs a differentiation step S80 similar to that in Figure 1, Figure 12, or Figure 15. Note that the signal processing device 100 may perform the differentiation step S80 before the integration step S70, or it may omit either the integration step S70 or the differentiation step S80.

[0098] Next, in the state index calculation step S90, the signal processing device 100 calculates an index vector representing the state of the object based on the first to nth synchronous summation waveforms generated in the synchronous summation step S60. The signal processing device 100 may calculate an N-dimensional synchronous summation vector whose elements are the values ​​at each time point of the first to nth synchronous summation waveforms as an index vector representing the state of the object, or it may calculate an N-dimensional tangent vector by differentiating the N-dimensional synchronous summation vector, or it may calculate an N-dimensional principal normal vector by further differentiating the N-dimensional tangent vector. For example, if the first to nth synchronous summation waveforms are time waveforms of displacement, the synchronous summation vector is the displacement vector, the tangent vector is the velocity vector, and the principal normal vector is the acceleration vector. The signal processing device 100 may also calculate a binormal vector, which is the cross product of the tangent vector and the principal normal vector, as an index vector representing the state of the object, or it may further calculate a vibration surface normal vector by converting the binormal vector into a unit vector.

[0099] Next, in the Lissajous figure generation step S92, the signal processing device 100 generates a Lissajous figure based on the vector calculated in the state index calculation step S90. The signal processing device 100 may also generate a Lissajous figure representing the trajectory of an N-dimensional synchronous sum vector, tangent vector, principal normal vector, binormal vector, or vibration surface normal vector.

[0100] Then, until the signal processing is completed (N in process S100), the signal processing device 100 repeatedly performs processes S10 to S92. Note that the signal processing device 100 may perform the differentiation process S80 before the integration process S70, or it may omit either the integration process S70 or the differentiation process S80.

[0101] Figure 22 shows an example of a Lissajous figure generated in the Lissajous figure generation process S92. In Figure 22, A1 is a Lissajous figure representing the trajectory of a three-dimensional displacement vector as a synchronous sum vector calculated from the time waveforms of the x-axis displacement Dx, y-axis displacement Dy, and z-axis displacement Dz shown in Figure 16. B1 is a Lissajous figure representing the trajectory of a three-dimensional velocity vector as a tangent vector obtained by differentiating the displacement vector of A1. C1 is a Lissajous figure representing the trajectory of a three-dimensional acceleration vector as a principal normal vector obtained by further differentiating the velocity vector of B1. D1 is a Lissajous figure representing the trajectory of a vibration surface normal vector obtained by converting the binormal vector, which is the cross product of the velocity vector of B1 and the acceleration vector of C1, into a unit vector. As a first comparative example, Figure 23 shows an example of a Lissajous figure when the signal processing device 100 does not perform the synchronous summing process S60. In Figure 23, A2 is a Lissajous figure representing the trajectory of the displacement vector calculated from the time waveforms of the x-axis displacement Dx, y-axis displacement Dy, and z-axis displacement Dz, respectively, obtained by integrating the inverse time waveforms of the x-axis velocity Vx, y-axis velocity Vy, and z-axis velocity Vz shown in Figure 6 without synchronous addition. B2 is a Lissajous figure representing the trajectory of the 3D velocity vector as a tangent vector obtained by differentiating the displacement vector in A2. C2 is a Lissajous figure representing the trajectory of the 3D acceleration vector as a principal normal vector obtained by further differentiating the velocity vector in B2. D2 is a Lissajous figure representing the trajectory of the vibration surface normal vector obtained by converting the binormal vector, which is the cross product of the velocity vector in B2 and the acceleration vector in C2, into a unit vector. Comparing Figure 22 and Figure 23, it can be seen that when synchronous addition is not performed, the trajectories of all Lissajous figures are unclear in terms of regularity, whereas when synchronous addition is performed, the trajectories of all Lissajous figures exhibit regularity.

[0102] Furthermore, as a second comparative example, Figure 24 shows an example of a Lissajous figure when the signal processing device 100 does not perform the frequency spectrum deformation process S40 and the inverse conversion time waveform generation process S50. In Figure 24, A3 is a Lissajous figure representing the trajectory of the displacement vector calculated from the time waveform of the x-axis displacement Dx, the time waveform of the y-axis displacement Dy, and the z-axis displacement Dz, which are obtained by integrating the synchronous summation waveform based on the time waveform of the x-axis velocity Vx, the synchronous summation waveform based on the time waveform of the y-axis velocity Vy, and the synchronous summation waveform based on the z-axis velocity Vz, respectively, as shown in Figure 9 above. B3 is a Lissajous figure representing the trajectory of the three-dimensional velocity vector as a tangent vector obtained by differentiating the displacement vector of A3. C3 is a Lissajous figure representing the trajectory of the three-dimensional acceleration vector as a principal normal vector obtained by further differentiating the velocity vector of B3. D3 is a Lissajous figure representing the trajectory of the vibration surface normal vector obtained by converting the binormal vector, which is the cross product of the velocity vector of B3 and the acceleration vector of C3, into a unit vector. Comparing Figure 22 and Figure 24, it can be visually confirmed that the signal processing device 100 reduces fluctuations caused by the approximately 84 Hz signal component and asynchronous signal components by performing synchronous addition on the inverse conversion time waveform of the speed.

[0103] The user can monitor the Lissajous figure generated in the Lissajous figure generation process S92, and if a change in the regularity of the vector trajectory is observed over time, it can infer that the state of the object has changed. In Figure 25, the upper panel shows an example of the change over time of the displacement vector trajectory shown in A1 of Figure 22, and the lower panel shows an example of the change over time of the displacement vector trajectory shown in A3 of Figure 24. Similarly, in Figure 26, the upper panel shows an example of the change over time of the velocity vector trajectory shown in B1 of Figure 22, and the lower panel shows an example of the change over time of the displacement vector trajectory shown in B3 of Figure 24. Furthermore, in Figure 27, the upper panel shows an example of the change over time of the acceleration vector trajectory shown in C1 of Figure 22, and the lower panel shows an example of the change over time of the displacement vector trajectory shown in C3 of Figure 24. Furthermore, in Figure 28, the upper panel shows an example of the time-dependent change in the trajectory of the vibration surface normal vector shown in D1 of Figure 22, and the lower panel shows an example of the time-dependent change in the trajectory of the displacement vector shown in D3 of Figure 24. In all of Figures 25 to 28, comparing the upper and lower panels, it can be visually confirmed that the signal processing device 100 reduces fluctuations caused by the signal component of approximately 84 Hz and the asynchronous signal component by performing synchronous summation on the inverse conversion time waveform of the velocity. Moreover, the regularity of the trajectories in all of Figures 25 to 28 changes after 6 months, suggesting that the state of the object changed during that 6-month period.

[0104] Figure 29 shows an example configuration of a signal processing device 100 that performs the signal processing method of the fourth embodiment. As shown in Figure 29, the signal processing device 100 includes first to nth sensors 200-1 to 200-N, N analog front ends 210-1 to 210-N, a processing circuit 110, a memory circuit 120, an operation unit 130, a display unit 140, a sound output unit 150, and a communication unit 160. Note that the signal processing device 100 may have a configuration in which some of the components in Figure 29 are omitted or changed, or other components are added. For example, the first to nth sensors 200-1 to 200-N and the analog front ends 210-1 to 210-N do not have to be components of the signal processing device 100.

[0105] The configuration and functions of the first to nth sensors 200-1 to 200-N, the analog front ends 210-1 to 210-N, the memory circuit 120, the operation unit 130, the display unit 140, the sound output unit 150, and the communication unit 160 are the same as in the first, second, or third embodiment, so their description is omitted.

[0106] The processing circuit 110 functions as a time waveform acquisition circuit 111, a frequency spectrum generation circuit 112, a trigger time series generation circuit 113, a frequency spectrum deformation circuit 114, an inverse conversion time waveform generation circuit 115, a synchronous summing circuit 116, a time waveform synchronization circuit 117, an integrating circuit 118, a differentiating circuit 119, a state index calculation circuit 171, and a Lissajous figure generation circuit 172 by executing the signal processing program 121 stored in the memory circuit 120. In other words, the signal processing device 100 includes a time waveform acquisition circuit 111, a frequency spectrum generation circuit 112, a trigger time series generation circuit 113, a frequency spectrum deformation circuit 114, an inverse conversion time waveform generation circuit 115, a synchronous summing circuit 116, a time waveform synchronization circuit 117, an integrating circuit 118, a differentiating circuit 119, a state index calculation circuit 171, and a Lissajous figure generation circuit 172.

[0107] The functions of the time waveform acquisition circuit 111, frequency spectrum generation circuit 112, trigger time series generation circuit 113, frequency spectrum deformation circuit 114, inverse transform time waveform generation circuit 115, synchronous summing circuit 116, time waveform synchronization circuit 117, integrating circuit 118, and differentiating circuit 119 are the same as in the first, second, or third embodiment, so their description is omitted.

[0108] The state index calculation circuit 171 calculates an index vector representing the state of the object based on the first to the Nth synchronous summation waveforms generated by the synchronous summation circuit 116. That is, the state index calculation circuit 171 performs the state index calculation step S90 shown in Figure 21. The vector calculated by the state index calculation circuit 171 is stored in the memory circuit 120.

[0109] The Lissajous figure generation circuit 172 generates a Lissajous figure based on the vector calculated by the state index calculation circuit 171. That is, the Lissajous figure generation circuit 172 performs the Lissajous figure generation process S92 shown in Figure 21. The Lissajous figure generated by the Lissajous figure generation circuit 172 is displayed on the display unit 140.

[0110] Thus, the signal processing program 121 is a program that causes the computer-based processing circuit 110 to execute the following steps: time waveform acquisition step S10, time waveform synchronization step S12, frequency spectrum generation step S20, trigger time series generation step S30, frequency spectrum deformation step S40, inverse conversion time waveform generation step S50, synchronous addition step S60, integration step S70, differentiation step S80, state index calculation step S90, and Lissajous figure generation step S92.

[0111] Furthermore, at least a portion of the time waveform acquisition circuit 111, frequency spectrum generation circuit 112, trigger time series generation circuit 113, frequency spectrum deformation circuit 114, inverse transform time waveform generation circuit 115, synchronous summing circuit 116, time waveform synchronization circuit 117, integration circuit 118, differentiation circuit 119, state index calculation circuit 171, and Lissajous figure generation circuit 172 may be implemented using dedicated hardware.

[0112] According to the fourth embodiment described above, the same effects as those of the signal processing methods of the first, second, or third embodiment can be obtained. Furthermore, according to the signal processing method of the fourth embodiment, the signal processing device 100 calculates a vector that serves as an indicator representing the state of the object, so that information reflecting the state of the object can be obtained with a relatively small amount of data. In addition, according to the signal processing method of the fourth embodiment, the user can visually understand the state of the object using Lissajous figures.

[0113] The embodiments and variations described above are examples only and are not limited thereto. For example, each embodiment and each variation can be combined as appropriate.

[0114] The present invention includes configurations that are substantially identical to those described in the embodiments (for example, configurations with the same function, method, and result, or configurations with the same purpose and effect). Furthermore, the present invention includes configurations in which non-essential parts of the configurations described in the embodiments are replaced. Furthermore, the present invention includes configurations that produce the same effects or achieve the same purpose as those described in the embodiments. Furthermore, the present invention includes configurations that add known technology to the configurations described in the embodiments.

[0115] The following can be derived from the embodiments and modifications described above.

[0116] One aspect of the signal processing method is: A time waveform acquisition step in which, with N being a predetermined integer of 1 or more, for each integer i between 1 and N, the i-th sensor acquires an i-th time waveform relating to an i-th physical quantity caused by an external force, velocity, or displacement acting on an object, the i-th sensor acquiring the i-th time waveform relating to an i-th physical quantity that includes at least a periodic fluctuation having a first frequency component and a second frequency component different from the first frequency and a higher-order frequency of the first frequency. For each of the aforementioned integers i, a frequency spectrum generation step is performed to generate the ith frequency spectrum based on the i-th time waveform, A trigger time series generation step that generates a trigger time series having a time interval of a first period which is the reciprocal of the first frequency or a real multiple of the first period defined by the components of the object, where the period corresponds to the frequency of any of the multiple peaks included in at least one of the first to Nth frequency spectra, and the time interval is a first period which is the reciprocal of the first frequency or a period which is a real multiple of the first period defined by the components of the object, A frequency spectrum deformation step is performed to generate an i-th deformed frequency spectrum by multiplying a window function over at least one of a plurality of frequency intervals obtained by dividing the frequency spectrum of i such that each interval contains a plurality of distinct frequencies that are rational multiples of the frequency corresponding to the reciprocal of the time interval of the trigger time series, thereby deforming the i-th frequency spectrum. For each of the aforementioned integers i, the process includes an inverse transform time waveform generation step, which inversely transforms the modified frequency spectrum of i into the time domain to generate the ith inverse transform time waveform, A synchronous addition step is performed on each of the aforementioned integers i, based on the time series of the trigger, to generate the ith synchronous addition waveform by performing synchronous addition on the inverse time waveform of i, Includes, For each integer i, the duration of the time waveform of i is equal to or greater than the duration of the product of the time interval of the trigger time series and the number of synchronous additions.

[0117] This signal processing method generates a trigger time series from the first to the Nth time waveforms, thus eliminating the need for a rotating pulse signal. Furthermore, by performing synchronous summation on each of the first to the Nth time waveforms, the target signal component synchronized with the first period (the reciprocal of the first frequency) or a real multiple of the first period is emphasized, while the second frequency signal component, which is asynchronous with the target signal component, is reduced. This results in the first to the Nth synchronous summation waveforms, eliminating the need for PLLs, tracking filters, low-pass filters, etc. Additionally, if the target signal component is a low-intensity signal component with a real multiple of the first period defined by the components of the object, the time interval of the trigger time series can be adjusted to match the period of the target signal component, using the first period (the period of the high-intensity signal component) as a reference. This results in the first to Nth synchronous summation waveforms in which the target signal component is emphasized. Furthermore, since the time lengths of the first to Nth time waveforms are each equal to or greater than the time corresponding to the product of the time interval of the trigger time series and the number of synchronous additions, the first to Nth synchronous addition waveforms with time lengths equal to or greater than the time interval of the trigger time series are obtained. In this way, this signal processing method makes it possible to generate time waveforms that reduce asynchronous signal components with respect to the target signal component without requiring a rotating pulse signal. Moreover, with this signal processing program, the frequency spectrum deformation process and the inverse conversion time waveform generation process reduce asynchronous signal components with respect to the target signal component in a stage prior to the synchronous addition process, thus reducing the number of synchronous additions compared to when the frequency spectrum deformation process and the inverse conversion time waveform generation process are not performed.

[0118] In one embodiment of the signal processing method, The time interval of the trigger time series may be a natural number multiple of the first period.

[0119] According to this signal processing method, first to Nth synchronous summation waveforms are obtained in which signal components with frequencies below the first frequency, synchronized with the signal component at the first frequency, are emphasized.

[0120] One embodiment of the signal processing method is: For each of the integers i, the system may include at least one of an integration step that performs integration on the synchronous sum waveform of i and a differentiation step that performs differentiation on the synchronous sum waveform of i.

[0121] According to this signal processing method, by performing integration and differentiation on the first to Nth synchronous summation waveforms, each of which emphasizes the target signal component, a time waveform of a physical quantity different from the physical quantity corresponding to the target signal component can be obtained.

[0122] In one embodiment of the signal processing method, The integer N may be 2 or greater.

[0123] This signal processing method allows for synchronous addition of multiple inverse time waveforms based on a common trigger time series.

[0124] In one embodiment of the signal processing method, The first to Nth time waveforms may be synchronized with each other.

[0125] This signal processing method ensures the synchronization accuracy of the first to Nth time waveforms, thereby improving the accuracy of synchronous summation and increasing the signal-to-noise ratio of the first to Nth synchronous summation waveforms.

[0126] One embodiment of the signal processing method is: Two or more of the first to Nth time waveforms are asynchronous with respect to each other. Prior to the frequency spectrum generation step, a time waveform synchronization step may be included in which the first to Nth time waveforms are synchronized with one another.

[0127] This signal processing method ensures the synchronization accuracy of the first to Nth time waveforms, thereby improving the accuracy of synchronous summation and increasing the signal-to-noise ratio of the first to Nth synchronous summation waveforms.

[0128] One embodiment of the signal processing method is: The process may include a state index calculation step, in which a vector representing the state of the object is calculated based on the first to Nth synchronous summation waveforms.

[0129] This signal processing method allows for obtaining information that reflects the state of the object, even with a relatively small amount of data.

[0130] One embodiment of the signal processing method is: The process may include a step of generating a Lissajous figure that represents the trajectory of the aforementioned vector.

[0131] This signal processing method allows users to visually understand the state of an object through Lissajous figures.

[0132] One aspect of a signal processing device is: A time waveform acquisition circuit that, for each integer i between 1 and N, where N is a predetermined integer of 1 or more, acquires an i-th time waveform from the i-th sensor relating to an i-th physical quantity caused by an external force, velocity, or displacement acting on an object, the i-th sensor having at least a periodic fluctuation having a first frequency component and a second frequency component different from the first frequency and a higher-order frequency of the first frequency. For each of the integers i, a frequency spectrum generation circuit generates the i-th frequency spectrum based on the time waveform of i, A trigger time series generation circuit generates a trigger time series whose time interval is a period corresponding to the frequency of any of the multiple peaks included in at least one of the first to Nth frequency spectra, and which is a first period that is the reciprocal of the first frequency or a period that is a real multiple of the first period defined by the components of the object, A frequency spectrum deformation circuit generates an i-th deformed frequency spectrum by multiplying the frequency spectrum of i by a window function over at least one of a plurality of frequency intervals obtained by dividing the frequency spectrum of i such that each interval contains a plurality of distinct frequencies that are rational multiples of the frequency corresponding to the reciprocal of the time interval of the trigger time series, For each of the aforementioned integers i, an inverse transform time waveform generation circuit generates an i-th inverse transform time waveform by inversely transforming the modified frequency spectrum of i into the time domain, A synchronous summing circuit generates an i-th synchronous sum waveform by performing synchronous summing on the inverse time waveform of i based on the time series of the trigger for each of the aforementioned integers i, Equipped with, For each integer i, the duration of the time waveform of i is equal to or greater than the duration of the product of the time interval of the trigger time series and the number of synchronous additions.

[0133] This signal processing device can generate a time waveform that reduces asynchronous signal components from the target signal component without requiring a rotational pulse signal. Furthermore, this signal processing device, through the frequency spectrum deformation circuit and the inverse time waveform generation circuit, can reduce asynchronous signal components from the target signal component in the stage prior to synchronous addition, thus reducing the number of synchronous addition steps compared to when the frequency spectrum deformation step and inverse time waveform generation step are not performed.

[0134] One aspect of a signal processing program is: A time waveform acquisition step in which, with N being a predetermined integer of 1 or more, for each integer i between 1 and N, the i-th sensor acquires an i-th time waveform relating to an i-th physical quantity caused by an external force, velocity, or displacement acting on an object, the i-th sensor acquiring the i-th time waveform relating to an i-th physical quantity that includes at least a periodic fluctuation having a first frequency component and a second frequency component different from the first frequency and a higher-order frequency of the first frequency. For each of the aforementioned integers i, a frequency spectrum generation step is performed to generate the ith frequency spectrum based on the i-th time waveform, A trigger time series generation step that generates a trigger time series having a time interval of a first period which is the reciprocal of the first frequency or a real multiple of the first period defined by the components of the object, where the period corresponds to the frequency of any of the multiple peaks included in at least one of the first to Nth frequency spectra, and the time interval is a first period which is the reciprocal of the first frequency or a period which is a real multiple of the first period defined by the components of the object, A frequency spectrum deformation step is performed to generate an i-th deformed frequency spectrum by multiplying a window function over at least one of a plurality of frequency intervals obtained by dividing the frequency spectrum of i such that each interval contains a plurality of distinct frequencies that are rational multiples of the frequency corresponding to the reciprocal of the time interval of the trigger time series, thereby deforming the i-th frequency spectrum. For each of the aforementioned integers i, the process includes an inverse transform time waveform generation step, which inversely transforms the modified frequency spectrum of i into the time domain to generate the ith inverse transform time waveform, A synchronous addition step is performed on each of the aforementioned integers i, based on the time series of the trigger, to generate the ith synchronous addition waveform by performing synchronous addition on the inverse time waveform of i, Have the computer run it, For each integer i, the duration of the time waveform of i is equal to or greater than the duration of the product of the time interval of the trigger time series and the number of synchronous additions.

[0135] This signal processing program can generate a time waveform that reduces asynchronous signal components from the target signal component without requiring a rotational pulse signal. Furthermore, this signal processing program reduces asynchronous signal components from the target signal component in the step prior to the synchronous addition step through the frequency spectrum deformation step and the inverse conversion time waveform generation step, thus reducing the number of synchronous addition steps compared to when the frequency spectrum deformation step and the inverse conversion time waveform generation step are not performed. [Explanation of Symbols]

[0136] 1...Vacuum pump, 3...Housing, 4...Motor case, 5...Connection part, 6...Pump case, 7...Gear case, 8...First side wall, 9...Second side wall, 11...Intake pipe, 12...Exhaust pipe, 13...First leg, 14...Third leg, 15...First bolt, 17...Sensor unit, 20...Base, 100...Signal processing unit, 110...Processing circuit, 111...Time waveform acquisition circuit, 112...Frequency spectrum generation circuit, 113...Trigger time series generation circuit, 114...Frequency spectrum Lissajous figure generation circuit, 115...Inverse transform time waveform generation circuit, 116...Synchronous summing circuit, 117...Time waveform synchronization circuit, 118...Integration circuit, 119...Differentiation circuit, 120...Memory circuit, 121...Signal processing program, 130...Operation unit, 140...Display unit, 150...Sound output unit, 160...Communication unit, 171...State index calculation circuit, 172...Lissajous figure generation circuit, 200-1~200-N...1st to Nth sensors, 210-1~210-N...Analog front end

Claims

1. A time waveform acquisition step in which, with N being a predetermined integer of 1 or more, for each integer i between 1 and N, the i-th sensor acquires an i-th time waveform relating to an i-th physical quantity caused by an external force, velocity, or displacement acting on an object, the i-th sensor acquiring the i-th time waveform relating to an i-th physical quantity that occurs when an external force, velocity, or displacement acts on an object, the i-th sensor having at least a periodic fluctuation having a first frequency component and a second frequency component different from the first frequency and a higher-order frequency of the first frequency. For each of the integers i, a frequency spectrum generation step is performed to generate the i-th frequency spectrum based on the time waveform of i, A trigger time series generation step that generates a trigger time series having a time interval of a first period which is the reciprocal of the first frequency or a real multiple of the first period defined by the components of the object, which corresponds to the frequency of any of the multiple peaks included in at least one of the first to Nth frequency spectra, A frequency spectrum deformation step is performed to generate an i-th deformed frequency spectrum by multiplying a window function over at least one of a plurality of frequency intervals obtained by dividing the frequency spectrum of i such that each interval contains a plurality of distinct frequencies that are rational multiples of the frequency corresponding to the reciprocal of the time interval of the trigger time series, for each integer i; For each of the aforementioned integers i, the process includes an inverse transform time waveform generation step, which inversely transforms the modified frequency spectrum of i into the time domain to generate the ith inverse transform time waveform, A synchronous addition step is performed on each of the aforementioned integers i, based on the time series of the trigger, to generate the ith synchronous addition waveform by performing synchronous addition on the inverse time waveform of i, Includes, A signal processing method wherein, for each integer i, the time length of the time waveform of i is equal to or greater than the time corresponding to the product of the time interval of the trigger time series and the number of synchronous additions.

2. In claim 1, A signal processing method wherein the time interval of the trigger time series is a natural number multiple of the first period.

3. In claim 1, A signal processing method comprising, for each integer i, an integration step of performing an integral operation on the synchronous sum waveform of i, and a differentiation step of performing a differential operation on the synchronous sum waveform of i.

4. In claim 1, A signal processing method wherein the integer N is 2 or greater.

5. In claim 4, A signal processing method in which the first to Nth time waveforms are synchronized with each other.

6. In claim 4, Two or more of the first to Nth time waveforms are asynchronous with respect to each other. A signal processing method comprising a time waveform synchronization step of synchronizing the first to N time waveforms with respect to each other, prior to the frequency spectrum generation step.

7. In claim 4, A signal processing method including a state index calculation step of calculating a vector that serves as an index representing the state of the object based on the first to Nth synchronous summation waveforms.

8. In claim 7, A signal processing method including a step of generating a Lissajous figure that generates a Lissajous figure representing the trajectory of the aforementioned vector.

9. A time waveform acquisition circuit that, for each integer i between 1 and N, where N is a predetermined integer of 1 or more, acquires an i-th time waveform from the i-th sensor relating to an i-th physical quantity caused by an external force, velocity, or displacement acting on an object, which includes at least a periodic fluctuation having a first frequency component and a second frequency component different from the first frequency and a higher-order frequency of the first frequency. For each of the integers i, a frequency spectrum generation circuit generates the i-th frequency spectrum based on the time waveform of i, A trigger time series generation circuit generates a trigger time series whose time interval is a period corresponding to the frequency of any of the multiple peaks included in at least one of the first to Nth frequency spectra, and which is a first period that is the reciprocal of the first frequency, or a period that is a real multiple of the first period defined by the components of the object, A frequency spectrum deformation circuit generates an i-th deformed frequency spectrum by multiplying the frequency spectrum of i by a window function over at least one of a plurality of frequency intervals obtained by dividing the frequency spectrum of i such that each interval contains a plurality of distinct frequencies that are rational multiples of the frequency corresponding to the reciprocal of the time interval of the trigger time series, For each of the aforementioned integers i, an inverse transform time waveform generation circuit generates an i-th inverse transform time waveform by inversely transforming the modified frequency spectrum of i into the time domain, A synchronous summing circuit generates an i-th synchronous sum waveform by performing synchronous summing on the inverse time waveform of i based on the time series of the trigger for each of the aforementioned integers i, Equipped with, A signal processing device wherein, for each integer i, the time length of the time waveform of i is equal to or greater than the time corresponding to the product of the time interval of the trigger time series and the number of synchronous additions.

10. A time waveform acquisition step in which, with N being a predetermined integer of 1 or more, for each integer i between 1 and N, the i-th sensor acquires an i-th time waveform relating to an i-th physical quantity caused by an external force, velocity, or displacement acting on an object, the i-th sensor acquiring the i-th time waveform relating to an i-th physical quantity that occurs when an external force, velocity, or displacement acts on an object, the i-th sensor having at least a periodic fluctuation having a first frequency component and a second frequency component different from the first frequency and a higher-order frequency of the first frequency. For each of the integers i, a frequency spectrum generation step is performed to generate the i-th frequency spectrum based on the time waveform of i, A trigger time series generation step that generates a trigger time series having a time interval of a first period which is the reciprocal of the first frequency or a real multiple of the first period defined by the components of the object, which corresponds to the frequency of any of the multiple peaks included in at least one of the first to Nth frequency spectra, A frequency spectrum deformation step is performed to generate an i-th deformed frequency spectrum by multiplying a window function over at least one of a plurality of frequency intervals obtained by dividing the frequency spectrum of i such that each interval contains a plurality of distinct frequencies that are rational multiples of the frequency corresponding to the reciprocal of the time interval of the trigger time series, for each integer i; For each of the aforementioned integers i, the process includes an inverse transform time waveform generation step, which inversely transforms the modified frequency spectrum of i into the time domain to generate the ith inverse transform time waveform, A synchronous addition step is performed on each of the aforementioned integers i, based on the time series of the trigger, to generate the ith synchronous addition waveform by performing synchronous addition on the inverse time waveform of i, Have the computer run it, A signal processing program in which, for each integer i, the time length of the time waveform of i is equal to or greater than the time equivalent to the product of the time interval of the trigger time series and the number of synchronous additions.

Images