WO2020245970A1 - 分析装置 - Google Patents
分析装置 Download PDFInfo
- Publication number
- WO2020245970A1 WO2020245970A1 PCT/JP2019/022501 JP2019022501W WO2020245970A1 WO 2020245970 A1 WO2020245970 A1 WO 2020245970A1 JP 2019022501 W JP2019022501 W JP 2019022501W WO 2020245970 A1 WO2020245970 A1 WO 2020245970A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- spectrogram
- intensity
- frequency
- integrated
- acquired
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H1/00—Measuring characteristics of vibrations in solids by using direct conduction to the detector
- G01H1/12—Measuring characteristics of vibrations in solids by using direct conduction to the detector of longitudinal or not specified vibrations
- G01H1/14—Frequency
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H1/00—Measuring characteristics of vibrations in solids by using direct conduction to the detector
- G01H1/003—Measuring characteristics of vibrations in solids by using direct conduction to the detector of rotating machines
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H17/00—Measuring mechanical vibrations or ultrasonic, sonic or infrasonic waves, not provided for in the preceding groups
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M99/00—Subject matter not provided for in other groups of this subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M99/00—Subject matter not provided for in other groups of this subclass
- G01M99/008—Subject matter not provided for in other groups of this subclass by doing functionality tests
Definitions
- the present invention relates to an analyzer for analyzing sound or vibration waveform data.
- Patent Document 1 describes a device for determining the presence or absence of abnormal noise.
- the intensity time series is acquired by time-frequency analysis of sound waveform data.
- the intensity time series is acquired by time-frequency analysis of sound waveform data.
- the apparatus described in Patent Document 1 is susceptible to disturbance and may not be able to obtain sufficient analysis accuracy. Such problems can also occur when analyzing vibration waveform data.
- An object of the present invention is to provide an analyzer capable of improving the analysis accuracy of waveform data.
- the analyzer uses a sensor that detects sound or vibration, a first analysis means that acquires a spectrogram by time-frequency analysis of the sound or vibration waveform data detected by the sensor, and a first analysis means.
- the second analysis means that divides the acquired spectrogram into a plurality of frequency bands and acquires the band intensity time series for each of the plurality of frequency bands, and the time frequency for each of the band intensity time series acquired by the second analysis means.
- a third analytical means that performs analysis and obtains an intensity spectrogram corresponding to each of the bandwidth intensity time series, and an integrated means that obtains an integrated spectrogram by integrating a plurality of intensity spectrograms acquired by the third analytical means. , Equipped with.
- the integrating means extracts the values of the components having the same frequency and time from each of the plurality of intensity spectrograms acquired by the third analytical means, and determines the values of the corresponding components of the integrated spectrogram based on the extracted values.
- the second analytical means divides the spectrogram acquired by the first analytical means into a plurality of frequency bands, and acquires a bandwidth intensity time series for each of the plurality of frequency bands.
- the third analytical means acquires an intensity spectrogram corresponding to each of the bandwidth intensity time series.
- the integration means obtains an integrated spectrogram by integrating a plurality of intensity spectrograms acquired by the third analytical means.
- FIG. It is a figure which shows the example of the analyzer in Embodiment 1.
- FIG. It is a flowchart which shows the operation example of the analyzer in Embodiment 1. It is a figure which shows the example which applied the analyzer shown in FIG. 1 to an elevator apparatus. It is a figure which shows the example which displayed the integrated spectrogram two-dimensionally. It is a figure which shows the example of the hardware resource of the analyzer. It is a figure which shows another example of the hardware resource of an analyzer.
- FIG. 1 is a diagram showing an example of the analyzer 1 according to the first embodiment.
- the analyzer 1 shown in FIG. 1 is suitable for analyzing waveform data of sound or vibration having periodicity. Therefore, it is preferable that the analyzer 1 is used to analyze the waveform data of the sound or vibration generated by the rotating body.
- the analyzer 1 includes, for example, a sensor 2, a signal processing unit 3, an analysis unit 4, an analysis unit 5, an analysis unit 6, an integration unit 7, a display control unit 8, and a display device 11.
- FIG. 2 is a flowchart showing an operation example of the analyzer 1 according to the first embodiment.
- FIG. 3 is a diagram showing an example in which the analyzer 1 shown in FIG. 1 is applied to an elevator device.
- FIG. 3 shows an example in which the sensor 2 is provided on the elevator car 12.
- the elevator car 12 moves up and down the hoistway 19.
- Guide rails 13 and 14 are provided on the hoistway 19.
- the car 12 is arranged between the guide rails 13 and 14.
- the basket 12 is provided with guide rollers 15 to 18.
- the guide roller 15 is arranged above the guide roller 16.
- the guide roller 15 rotates while contacting the guide rail 13.
- the guide roller 16 rotates while contacting the guide rail 13.
- the guide roller 17 is arranged above the guide roller 18.
- the guide roller 17 rotates while contacting the guide rail 14.
- the guide roller 18 rotates while contacting the guide rail 14. The movement of the car 12 is guided by the guide rails 13 and 14 and the guide rollers 15 to 18.
- the sensor 2 is arranged between the guide rollers 15 and 17.
- the sensor 2 is a sensor that detects sound
- the sound emitted by the guide rollers 15 and 17 is detected by the sensor 2.
- the sensor 2 is a sensor that detects vibration
- the vibration generated by the guide rollers 15 and 17 is detected by the sensor 2.
- the sensor 2 may be provided under the car 12.
- the sensor 2 is arranged between the guide rollers 16 and 18.
- the sensor 2 is a sensor that detects sound
- the sound emitted by the guide rollers 16 and 18 is detected by the sensor 2.
- the sensor 2 is a sensor that detects vibration
- the vibration generated by the guide rollers 16 and 18 is detected by the sensor 2.
- the sensor 2 may be provided both above the car 12 and below the car 12. In the following, an example in which the sensor 2 detects sound will be described.
- the signal representing the sound detected by the sensor 2 is input to the signal processing unit 3.
- the signal processing unit 3 acquires the waveform data wave [t] of the sound detected by the sensor 2 (S102). For example, the signal processing unit 3 samples a waveform representing the sound detected by the sensor 2 at the sampling frequency fs and the word length ws, and converts it into a digital signal. As a result, the signal processing unit 3 acquires the waveform data wave [t] including the digital signal.
- wave [t] represents the amplitude of the waveform.
- the sampling frequency fs is 48 kHz.
- the word length ws is 24 bits.
- the analysis unit 4 acquires the spectrogram X [i, j] by time-frequency analysis of the waveform data wave [t] acquired by the signal processing unit 3 (S103). For example, the analysis unit 4 extracts a frame having a specific length at a specific time interval from the waveform data wave [t] acquired by the signal processing unit 3. The analysis unit 4 calculates the power spectral density by multiplying the extracted frame by a time window and performing frequency analysis based on the fast Fourier transform operation. As a result, the analysis unit 4 acquires the spectrogram X [i, j]. In the following, the fast Fourier transform will be referred to as FFT.
- i the index of the frame.
- j the frequency index in the FFT.
- NFFT is the size of the FFT, that is, the number of points.
- X [i, j] represents the intensity of the time frequency component at the indexes i and j.
- the start time time [i] of the time window corresponding to the index i of the frame is calculated by the following equation.
- NS is the frame interval, that is, the number of waveform samples.
- the frame length NW is 480 points, which is 10 ms in terms of time.
- the frame interval NS is 48 points, which is 1 ms in terms of time.
- the size of the FFT is 512 points.
- a humming window can be used as the time window.
- the analysis unit 5 divides the spectrogram X [i, j] acquired by the analysis unit 4 into a plurality of frequency bands, and acquires a bandwidth intensity time series Y [n, i] for each of the plurality of frequency bands.
- a bandwidth intensity time series Y [n, i] for each of the plurality of frequency bands.
- the analysis unit 5 acquires the bandwidth intensity time series Y [n, i] of the frequency band n from the spectrogram X [i, j] acquired by the analysis unit 4 (S105).
- the bandwidth intensity time series Y [n, i] of the nth frequency band n out of the N frequency bands is calculated as shown in (1) to (3) below.
- N the number of frequency bands.
- f 0 is the center frequency of the first frequency band 1.
- BW is the frequency bandwidth (in octaves). As an example, the center frequency f 0 of the frequency band 1 is 250 Hz.
- the frequency bandwidth BW is 1/3.
- the analysis unit 5 acquires the bandwidth intensity time series Y [n, i] of the frequency band n.
- Y [n, i] mean_ ⁇ j' ⁇ X [i, j] i is a time-wise index of the spectrogram.
- mean_ ⁇ j' ⁇ represents an operation for finding an average value for j'.
- the index i in Y [n, i] corresponds to a time advance of 1 ms for each increment. Therefore, when n is a fixed value, Y [n, i] can be regarded as a time-series signal having a sampling frequency of 1000 Hz.
- the analysis unit 6 analyzes each of the plurality of bandwidth intensity time series Y [n, i] acquired by the analysis unit 5 as a time series signal with i as the time, and performs time frequency analysis to perform the intensity spectrogram Z [n , K, l] is acquired (S106).
- the intensity spectrogram Z [1, k, l] is calculated from the band intensity time series Y [1, i] of the frequency band 1.
- the intensity spectrogram Z [2, k, l] is calculated from the band intensity time series Y [2, i] of the frequency band 2.
- the intensity spectrogram Z [N, k, l] is calculated from the band intensity time series Y [N, i] of the frequency band N.
- the analysis unit 6 extracts a frame having a specific frame length MW at a specific frame interval MS from the bandwidth intensity time series Y [n, i] acquired by the analysis unit 5.
- the analysis unit 6 calculates the power spectral density by multiplying the extracted frame by a time window and performing frequency analysis based on the FFT. As a result, the analysis unit 6 acquires the intensity spectrogram Z [n, k, l].
- the frame length MW is 500 points, which is 500 ms in terms of time.
- the frame interval MS is 50 points, which is 50 ms in terms of time.
- the integration unit 7 acquires the integrated spectrogram U [k, l] by integrating the plurality of intensity spectrograms Z [n, k, l] acquired by the analysis unit 6 (S109).
- the integration unit 7 extracts the values of the components having the same frequency and time from each of the N intensity spectrograms Z [n, k, l] acquired by the analysis unit 6. Then, the integration unit 7 determines the values of the corresponding components of the integrated spectrogram based on the extracted N values. The integration unit 7 acquires the integrated spectrogram U [k, l] by performing the same calculation for all the components.
- the integration unit 7 fixes the indexes k and l of the intensity spectrogram Z [n, k, l] and performs a specific operation on n. Then, the integration unit 7 calculates the integration spectrogram U [k, l] by performing the same calculation for all combinations of the indexes k and l.
- U [k, l] f_ ⁇ n' ⁇ Z [n', k, l] f_ ⁇ n' ⁇ represents an operation related to n'.
- the operation related to n' the maximum value, the average, the quantile, and the like can be used.
- the integration unit 7 extracts the values of the components having the same frequency and time from each of the N intensity spectrograms Z [n, k, l] acquired by the analysis unit 6. Then, the integration unit 7 determines the value of the corresponding component of the integrated spectrogram by using a specific number of values from the larger of the extracted N values. For example, the integration unit 7 selects three large values from the extracted N values and calculates the average value of the selected three values. The integration unit 7 determines the calculated average value as the value of the corresponding component of the integration spectrogram.
- the display control unit 8 displays the integrated spectrogram U [k, l] acquired by the integrated unit 7 on the display 11 (S110).
- FIG. 4 is a diagram showing an example in which the integrated spectrogram is displayed two-dimensionally.
- FIG. 4 shows an example of the integrated spectrogram U [k, l] obtained when the sensor 2 is provided on the car 12 and the car 12 is driven from the bottom floor to the top floor of the building.
- the horizontal axis shown in FIG. 4 is the index k of the frame, which corresponds to time.
- the vertical axis is the frequency index l.
- the intensity U [k, l] is represented by the size of the dots.
- the intensity U [k, l] may be expressed by different colors.
- the elevator maintenance staff can confirm that a pattern similar to the traveling speed pattern of the car 12 exists by looking at the display on the display 11. For example, FIG. 4 shows that the sensor 2 detected a sound in which the signal level peaked every 1 / 9.2 sec while the car 12 was traveling at the rated speed. If the rated speed of the car 12 is 105 m / min and the diameter of the guide roller 15 is 6 cm, the rotation frequency of the guide roller 15 is 9.28 Hz. Therefore, it is highly possible that the sound whose signal level peaks at a period of 1 / 9.2 sec, which can be recognized from the display of the display 11, is generated from the guide roller 15 or 17. By using the analyzer 1, the elevator maintenance staff can easily identify the source of the abnormal noise.
- the spectrogram X [i, j] is divided into N frequency bands, and the bandwidth intensity time series Y [n, i] is acquired for each of the N frequency bands. .. Then, the integrated spectrogram U [k, l] is acquired based on the intensity spectrogram Z [n, k, l] corresponding to each of the N bandwidth intensity time series Y [n, i]. Therefore, the analyzer 1 shown in the present embodiment can improve the analysis accuracy of the waveform data.
- the analyzer 1 includes the display control unit 8 and the display 11 .
- the analyzer 1 may have only a function of outputting the information of the integrated spectrogram U [k, l] acquired by the integration unit 7.
- the analyzer 1 may have a function of remotely transmitting the information of the integrated spectrogram U [k, l] acquired by the integration unit 7 using a network such as the Internet.
- the analyzer 1 may further include a calculation unit 9.
- the calculation unit 9 calculates the frequency of the intensity peak of the integrated spectrogram U [k, l] acquired by the integration unit 7.
- the calculation unit 9 outputs "9.2 Hz" as the calculation result of the frequency of the intensity peak of the integrated spectrogram U [k, l].
- FIG. 4 shows an example in which the display control unit 8 displays the result of calculation by the calculation unit 9 on the display 11.
- the analyzer 1 may further include a determination unit 10.
- the determination unit 10 determines the presence or absence of an abnormality based on the frequency calculated by the calculation unit 9.
- the sensor 2 detects a sound emitted by a specific rotating body.
- the determination unit 10 determines whether or not there is an abnormality in the rotating body based on the frequency of rotation of the rotating body and the frequency calculated by the calculation unit 9.
- the sensor 2 detects the sound emitted by the guide rollers 15 and 17.
- Information on the frequency at which the guide roller 15 rotates when the car 12 moves at the rated speed is stored in advance in the analyzer 1.
- the determination unit 10 detects that the guide roller 15 or 17 is emitting an abnormal sound if the frequency calculated by the calculation unit 9 is within a specific range based on the frequency stored in advance.
- the analyzer 1 is applied to the elevator device.
- the analyzer 1 may be applied to a device other than the elevator device.
- the analyzer 1 is preferably used for analyzing the waveform data of the sound or vibration generated by the rotating body. Therefore, it is preferable that the device to which the analyzer 1 is applied includes a rotating body.
- FIG. 5 is a diagram showing an example of hardware resources of the analyzer 1.
- the analyzer 1 includes a processing circuit 20 including, for example, a processor 21 and a memory 22 as hardware resources.
- the memory 22 is, for example, a semiconductor memory.
- the memory 22 does not have to be a semiconductor memory.
- the analyzer 1 realizes the functions of the respective parts shown by reference numerals 3 to 10 by executing the program stored in the memory 22 by the processor 21.
- FIG. 6 is a diagram showing another example of the hardware resource of the analyzer 1.
- the analyzer 1 includes, for example, a processing circuit 20 including a processor 21, a memory 22, and dedicated hardware 23.
- FIG. 6 shows an example in which a part of the functions of the analyzer 1 is realized by the dedicated hardware 23. All the functions of the analyzer 1 may be realized by the dedicated hardware 23.
- the dedicated hardware 23 a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, or a combination thereof can be adopted.
- the analyzer according to the present invention is suitable for analyzing waveform data of sound or vibration having periodicity.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
Description
図1は、実施の形態1における分析装置1の例を示す図である。図1に示す分析装置1は、周期性がある音或いは振動の波形データの分析に好適である。このため、回転体が発する音或いは振動の波形データを分析するために分析装置1が用いられることが好ましい。分析装置1は、例えばセンサ2、信号処理部3、分析部4、分析部5、分析部6、統合部7、表示制御部8、及び表示器11を備える。
tは波形サンプルのインデックスである。wave[t]は波形の振幅を表す。一例として、サンプリング周波数fsは48kHzである。ワード長wsは24ビットである。
iはフレームのインデックスである。jはFFTにおける周波数のインデックスである。NFFTは、FFTのサイズ、即ちポイント数である。X[i,j]は、インデックスi及びjにおける時間周波数成分の強度を表す。
time[i]=(NS/fs)×i (i=0,1,2,…,T)
NSは、フレーム間隔、即ち波形のサンプル数である。
freq[j]=(fs/NFFT)×j (j=0,1,2,…,NFFT/2)
一例として、フレーム長NWは480点であり、時間換算で10msである。フレーム間隔NSは48点であり、時間換算で1msである。FFTのサイズNFFTは512点である。また、時間窓としてハミング窓を用いることができる。
(1)先ず、周波数帯域nの下限周波数fL[n]と上限周波数fH[n]を求める。
fL[n]=f0×2(n-1-BW/2) (n=1,2,…,N)
fH[n]=f0×2(n-1+BW/2) (n=1,2,…,N)
Nは、周波数帯域の数である。f0は、1番目の周波数帯域1の中心周波数である。BWは、周波数帯域幅(オクターブ単位)である。一例として、周波数帯域1の中心周波数f0は250Hzである。周波数帯域幅BWは1/3である。
{j´}={j´|fL[n]≦freq[j´]<fH[n]}
Y[n,i]=mean_{j´} X[i,j]
iは、スペクトログラムの時間方向のインデックスである。mean_{j´}は、j´に関する平均値を求める演算を表す。上述した具体例では、Y[n,i]におけるインデックスiは、1増えるごとに1msの時間の進みに対応する。このため、nを固定値とした場合、Y[n,i]は、サンプリング周波数1000Hzの時系列信号とみなすことができる。
kは、フレームのインデックスである。但し、インデックスkは、スペクトログラムX[i,j]のフレームのインデックスiとは異なる。lは、周波数のインデックスである。但し、インデックスlは、スペクトログラムX[i,j]の周波数のインデックスjとは異なる。一例として、フレーム長MWは500点であり、時間換算で500msである。フレーム間隔MSは50点であり、時間換算で50msである。
U[k,l]=f_{n´} Z[n´,k,l]
f_{n´}は、n´に関する演算を表す。n´に関する演算としては、最大値、平均、及び分位数などを用いることができる。
U[k,l]=max_{n´} Z[n´,k,l]
Claims (9)
- 音又は振動を検出するセンサと、
前記センサが検出した音又は振動の波形データを時間周波数分析することにより、スペクトログラムを取得する第1分析手段と、
前記第1分析手段によって取得されたスペクトログラムを複数の周波数帯域に分割し、前記複数の周波数帯域のそれぞれについて帯域強度時系列を取得する第2分析手段と、
前記第2分析手段によって取得された帯域強度時系列のそれぞれについて時間周波数分析を行い、帯域強度時系列のそれぞれに対応する強度スペクトログラムを取得する第3分析手段と、
前記第3分析手段によって取得された複数の強度スペクトログラムを統合することにより、統合スペクトログラムを取得する統合手段と、
を備え、
前記統合手段は、前記第3分析手段によって取得された複数の強度スペクトログラムのそれぞれから周波数及び時間が同じ成分の値を抽出し、抽出した値に基づいて統合スペクトログラムの対応する成分の値を決定する分析装置。 - 表示器と、
前記統合手段によって取得された統合スペクトログラムを前記表示器に2次元的に表示する表示制御手段と、
を更に備えた請求項1に記載の分析装置。 - 前記統合手段によって取得された統合スペクトログラムの強度ピークの周波数を演算する第2演算手段を更に備えた請求項1又は請求項2に記載の分析装置。
- 判定手段を更に備え、
前記センサは、特定の回転体が発する音又は振動を検出し、
前記判定手段は、前記回転体の回転の周波数と前記第2演算手段によって演算された周波数とに基づいて、前記回転体の異常の有無を判定する請求項3に記載の分析装置。 - 前記統合手段は、前記第3分析手段によって取得された複数の強度スペクトログラムのそれぞれから周波数及び時間が同じ成分の値を抽出し、抽出した値の中で最も大きな値を、統合スペクトログラムの対応する成分の値として決定する請求項1から請求項4の何れか一項に記載の分析装置。
- 前記統合手段は、前記第3分析手段によって取得された複数の強度スペクトログラムのそれぞれから周波数及び時間が同じ成分の値を抽出し、抽出した値のうち大きい方から特定の数の値を用いて、統合スペクトログラムの対応する成分の値を決定する請求項1から請求項4の何れか一項に記載の分析装置。
- 前記統合手段は、前記第3分析手段によって取得された複数の強度スペクトログラムのそれぞれから周波数及び時間が同じ成分の値を抽出し、抽出した値の中で大きい方から3つの値を選択し、選択した3つの値の平均値を、統合スペクトログラムの対応する成分の値として決定する請求項1から請求項4の何れか一項に記載の分析装置。
- 前記センサは、エレベーターの昇降路を移動するかごの上又は前記かごの下に設けられた請求項1から請求項7の何れか一項に記載の分析装置。
- 前記センサは、エレベーターのかごの移動を案内するためのガイドローラが発する音又は振動を検出する請求項1から請求項7の何れか一項に記載の分析装置。
Priority Applications (5)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2019/022501 WO2020245970A1 (ja) | 2019-06-06 | 2019-06-06 | 分析装置 |
CN201980095986.6A CN113767267A (zh) | 2019-06-06 | 2019-06-06 | 分析装置 |
JP2021524592A JP7001200B2 (ja) | 2019-06-06 | 2019-06-06 | 分析装置 |
KR1020217028591A KR102658693B1 (ko) | 2019-06-06 | 2019-06-06 | 분석 장치 |
TW108142493A TW202046036A (zh) | 2019-06-06 | 2019-11-22 | 分析裝置 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/JP2019/022501 WO2020245970A1 (ja) | 2019-06-06 | 2019-06-06 | 分析装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020245970A1 true WO2020245970A1 (ja) | 2020-12-10 |
Family
ID=73653098
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2019/022501 WO2020245970A1 (ja) | 2019-06-06 | 2019-06-06 | 分析装置 |
Country Status (5)
Country | Link |
---|---|
JP (1) | JP7001200B2 (ja) |
KR (1) | KR102658693B1 (ja) |
CN (1) | CN113767267A (ja) |
TW (1) | TW202046036A (ja) |
WO (1) | WO2020245970A1 (ja) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2023226572A1 (zh) * | 2022-05-25 | 2023-11-30 | 腾讯科技(深圳)有限公司 | 特征表示的提取方法、装置、设备、介质及程序产品 |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
TWI839948B (zh) * | 2022-11-14 | 2024-04-21 | 財團法人工業技術研究院 | 促進加工缺陷肇因分析的方法 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH11108806A (ja) * | 1997-10-06 | 1999-04-23 | Oki Electric Ind Co Ltd | 機械運動特徴抽出装置 |
JP2000275098A (ja) * | 1999-03-26 | 2000-10-06 | Toshiba Corp | 波形信号解析装置 |
JP2005164378A (ja) * | 2003-12-02 | 2005-06-23 | Hitachi Ltd | 移動体データ解析装置および移動体データ解析システム |
JP2006047265A (ja) * | 2004-02-09 | 2006-02-16 | National Institute Of Advanced Industrial & Technology | 非定常波解析システム |
JP2019039859A (ja) * | 2017-08-28 | 2019-03-14 | 沖電気工業株式会社 | 分析装置、分析装置の表示方法および表示プログラム |
Family Cites Families (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP3211615B2 (ja) * | 1995-04-06 | 2001-09-25 | 日産自動車株式会社 | 機械作動音の異常判定方法 |
JP4785563B2 (ja) * | 2006-03-03 | 2011-10-05 | グローリー株式会社 | 音声処理装置および音声処理方法 |
JP4974345B2 (ja) * | 2006-06-30 | 2012-07-11 | 一般財団法人電力中央研究所 | 信号検出装置、信号検出方法および信号検出プログラム |
EP2372134B1 (en) * | 2009-02-06 | 2015-09-16 | Honda Motor Co., Ltd. | Frequency component analyzing device |
JP5783808B2 (ja) * | 2011-06-02 | 2015-09-24 | 三菱電機株式会社 | 異常音診断装置 |
CN102984634A (zh) * | 2011-11-22 | 2013-03-20 | 南京工程学院 | 一种数字助听器非等宽子带自动增益控制方法 |
JP5930789B2 (ja) * | 2012-03-23 | 2016-06-08 | 三菱電機株式会社 | 異常音診断装置 |
JP6103899B2 (ja) * | 2012-11-28 | 2017-03-29 | 三菱電機株式会社 | 故障個所推定装置 |
CN105705928B (zh) * | 2013-11-08 | 2018-07-24 | 三菱电机株式会社 | 异常声音诊断装置 |
KR101627107B1 (ko) * | 2014-06-26 | 2016-06-07 | 고려대학교 산학협력단 | 음향 센서를 이용하여 분리막 모듈의 완결성을 판단하는 장치 및 방법 |
KR101608275B1 (ko) * | 2014-08-21 | 2016-04-04 | 호원대학교 산학협력단 | 전력설비 음향 진단 시스템 |
JP6315810B2 (ja) * | 2014-08-29 | 2018-04-25 | 東日本旅客鉄道株式会社 | 鉄道橋の支承部の異常検出方法 |
JP6190343B2 (ja) * | 2014-09-12 | 2017-08-30 | 株式会社神戸製鋼所 | 回転機の異常診断装置、回転機の異常診断方法、及び、回転機 |
KR101962558B1 (ko) * | 2015-02-03 | 2019-03-26 | 미쓰비시덴키 가부시키가이샤 | 이상음 진단 장치, 이상음 진단 시스템, 이상음 진단 방법 및 이상음 진단 프로그램 |
WO2016146265A1 (en) * | 2015-03-17 | 2016-09-22 | Zynaptiq Gmbh | Methods for extending frequency transforms to resolve features in the spatio-temporal domain |
CN108291837B (zh) * | 2015-12-09 | 2020-02-14 | 三菱电机株式会社 | 劣化部位估计装置、劣化部位估计方法以及移动体的诊断系统 |
KR101696105B1 (ko) * | 2015-12-10 | 2017-01-24 | 주식회사 포스코 | 결함원인 분석장치 및 방법 |
JP6540962B2 (ja) * | 2016-02-02 | 2019-07-10 | 富士ゼロックス株式会社 | 診断装置、診断システムおよびプログラム |
KR20180086628A (ko) * | 2017-01-23 | 2018-08-01 | 현대엘리베이터주식회사 | 전동기 전류 및 속도 실시간 주파수 분석을 통한 엘리베이터 원격 진단 방법 |
CN108647667B (zh) * | 2017-09-19 | 2019-09-27 | 长安大学 | 一种基于信号时频分解的频域幅值谱峭度图的实现方法 |
CN107808114B (zh) * | 2017-09-19 | 2021-11-05 | 长安大学 | 一种基于信号时频分解的幅值谱峭度图的实现方法 |
CN108549105A (zh) * | 2018-07-02 | 2018-09-18 | 西北工业大学 | 一种自适应随机共振的地震波特征提取方法 |
CN109459131A (zh) * | 2018-11-29 | 2019-03-12 | 郑州工程技术学院 | 一种旋转机械多通道振动信号的时频特征提取方法及装置 |
-
2019
- 2019-06-06 WO PCT/JP2019/022501 patent/WO2020245970A1/ja active Application Filing
- 2019-06-06 KR KR1020217028591A patent/KR102658693B1/ko active IP Right Grant
- 2019-06-06 JP JP2021524592A patent/JP7001200B2/ja active Active
- 2019-06-06 CN CN201980095986.6A patent/CN113767267A/zh active Pending
- 2019-11-22 TW TW108142493A patent/TW202046036A/zh unknown
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH11108806A (ja) * | 1997-10-06 | 1999-04-23 | Oki Electric Ind Co Ltd | 機械運動特徴抽出装置 |
JP2000275098A (ja) * | 1999-03-26 | 2000-10-06 | Toshiba Corp | 波形信号解析装置 |
JP2005164378A (ja) * | 2003-12-02 | 2005-06-23 | Hitachi Ltd | 移動体データ解析装置および移動体データ解析システム |
JP2006047265A (ja) * | 2004-02-09 | 2006-02-16 | National Institute Of Advanced Industrial & Technology | 非定常波解析システム |
JP2019039859A (ja) * | 2017-08-28 | 2019-03-14 | 沖電気工業株式会社 | 分析装置、分析装置の表示方法および表示プログラム |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2023226572A1 (zh) * | 2022-05-25 | 2023-11-30 | 腾讯科技(深圳)有限公司 | 特征表示的提取方法、装置、设备、介质及程序产品 |
Also Published As
Publication number | Publication date |
---|---|
JP7001200B2 (ja) | 2022-01-19 |
TW202046036A (zh) | 2020-12-16 |
KR20210122839A (ko) | 2021-10-12 |
CN113767267A (zh) | 2021-12-07 |
JPWO2020245970A1 (ja) | 2021-10-14 |
KR102658693B1 (ko) | 2024-04-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9177559B2 (en) | Method and apparatus for analyzing animal vocalizations, extracting identification characteristics, and using databases of these characteristics for identifying the species of vocalizing animals | |
TWI451070B (zh) | 異常音診斷裝置 | |
JP7001200B2 (ja) | 分析装置 | |
US8428945B2 (en) | Acoustic signal classification system | |
JP4547042B2 (ja) | 音判定装置、音検知装置及び音判定方法 | |
US20090154726A1 (en) | System and Method for Noise Activity Detection | |
KR880004426A (ko) | 언어 엔코딩 처리시스템 및 음성 합성방법 | |
KR910015109A (ko) | 신호처리장치 | |
JP2014512022A (ja) | スペクトル挙動の変換を実行する音響信号処理システム及び方法 | |
US6718302B1 (en) | Method for utilizing validity constraints in a speech endpoint detector | |
US20230317095A1 (en) | Systems and methods for pre-filtering audio content based on prominence of frequency content | |
US20170135649A1 (en) | Breath sound analyzing apparatus, breath sound analyzing method, computer program, and recording medium | |
US20050278173A1 (en) | Determination of the common origin of two harmonic signals | |
CN113109808A (zh) | 基于存在探测的多普勒信号处理方法和装置 | |
CN1771533A (zh) | 音频编码 | |
Gockel et al. | Pitch perception at very high frequencies: On psychometric functions and integration of frequency information | |
JP4654621B2 (ja) | 音声処理装置およびプログラム | |
US20150348536A1 (en) | Method and device for recognizing speech | |
US8554546B2 (en) | Apparatus and method for calculating a fundamental frequency change | |
Grassi et al. | Why are damped sounds perceived as shorter than ramped sounds? | |
EP3457395A1 (en) | Music structure analysis device, method for analyzing music structure, and music structure analysis program | |
Oetjen et al. | Spectro-temporal modulation masking patterns reveal frequency selectivity | |
WO2019123633A1 (ja) | 音響計測システム及びパラメータ生成装置 | |
Korvel et al. | Investigating Noise Interference on Speech Towards Applying the Lombard Effect Automatically | |
JP6501917B2 (ja) | 生体音解析装置及び生体音解析方法、並びにコンピュータプログラム及び記録媒体 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19931906 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2021524592 Country of ref document: JP Kind code of ref document: A |
|
ENP | Entry into the national phase |
Ref document number: 20217028591 Country of ref document: KR Kind code of ref document: A |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19931906 Country of ref document: EP Kind code of ref document: A1 |