JP2003214943A - Device and method for evaluating machine operating noise - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an effective technology for appropriately evaluating a machine operating noise. <P>SOLUTION: An operating noise evaluating device 100 is provided with a controller 130 and waveform data on the operating noise detected by an acceleration sensor 112 are processed with an attenuation filter by the controller 130. On the process with the attenuation filter, the waveform data are attenuated based on a rotation period detected by a revolution sensor 122 in accordance with a prescribed exponential function. More specifically, a fundamental frequency is derived from the rotation period, a most appropriate time constant corresponding to the prescribed exponential function is derived, and the process with the attenuation filter is carried out using the time constant. By using the most appropriate time constant, a judgment on an abnormal noise after an FFT processing following the process with the attenuation filter can be properly carried out. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a technique for evaluating operating noise generated during operation of a vehicle engine in a completion inspection process for a machine such as a vehicle engine.

[0002]

2. Description of the Related Art For example, an operation noise generated from a vehicle engine is evaluated in a completion inspection process before the product is shipped from the vehicle engine. Most of such operation sounds include impact sounds (tapping sounds, interference sounds), and in particular, in a vehicle engine, various impact sounds generated from a valve operating system, a combustion chamber, and the like are combined. Therefore, there is a demand for a technique for correctly analyzing and determining only abnormal noise (abnormal noise) from the combined operating noise. Therefore, conventionally, a technique using frequency analysis such as FFT (Fast Fourier Transform) calculation is known as a technique for evaluating the mechanical operation noise of this type.
In this technique, for example, sound or vibration during machine operation is sampled at a predetermined frequency, the sampled waveform data is subjected to frequency analysis to obtain an intensity level according to the frequency, and this intensity level is set to a preset threshold value. By comparing with, it is determined whether or not the mechanical operation sound includes abnormal noise.

[0003]

By the way, in the above-mentioned conventional technique for evaluating the mechanical operation noise, when the frequency analysis is performed, the data of the sound or the vibration are sampled several times and averaged to obtain the average of the data. It is designed to smooth out minute fluctuations. However, averaging the sound or vibration data smoothes the intensity level due to abnormal noise, and a significant difference may not be seen in the result of frequency analysis of waveform data that does not include abnormal noise. Therefore, the conventional technique for evaluating the mechanical operation sound has a limit in accurately determining the abnormal noise included in the mechanical operation sound. Therefore, the present invention has been made in view of the above points, and an object of the present invention is to provide an effective technique for appropriately evaluating mechanical operation noise.

[0004]

In order to solve the above-mentioned problems, an apparatus for evaluating mechanical operating noise according to the present invention is configured as described in claims 1 and 2. Further, the machine operation sound evaluation method of the present invention is configured as described in claims 3 and 4.
Note that the invention according to these claims attenuates the harmonic components other than the fundamental wave component in the waveform data related to the mechanical operation sound based on the rotation cycle data during the mechanical operation, thereby performing the subsequent frequency analysis well. It is a technology that enables the appropriate evaluation of machine operation noise.

A machine operation sound evaluation device according to a first aspect of the present invention includes operation sound detection means, rotation cycle detection means, and processing means. When the machine operates, a machine operation sound is generated due to sound or vibration. The operating sound detecting means detects the waveform data of the mechanical operating sound. This operation sound detecting means is configured by appropriately using, for example, an acceleration sensor, an amplifier and the like. The rotation cycle detection means detects data relating to the rotation cycle during machine operation. This rotation cycle detection means is, for example, a rotation speed sensor,
It is configured by appropriately using an amplifier or the like. The processing means of the present invention analyzes the frequency of the waveform data processed by the attenuation filter, and evaluates the mechanical operating noise based on the frequency analysis result. In frequency analysis, the intensity level (peak value) of waveform data is calculated according to each frequency. In this frequency analysis, for example, FFT (Fast Fourier Transform) calculation is preferably used. In the evaluation of the mechanical operation sound, for example, the intensity level (peak value) obtained by the frequency analysis is compared with a preset threshold value, and when the intensity level exceeds the threshold value, it is determined as abnormal noise. The attenuation filter of the present invention has a configuration in which harmonic components other than the fundamental wave component in the waveform data are attenuated in a predetermined attenuation mode. The attenuation filter exponentially attenuates the harmonic components. In the waveform data of the mechanical operating sound, a harmonic component is usually combined with a fundamental wave component. Therefore, it becomes possible to relatively emphasize the fundamental wave component by attenuating the harmonic component via the attenuation filter. The harmonic referred to here is a so-called n-th harmonic, which is n times the fundamental frequency of the fundamental component (n
= 2,3, ...). When the fundamental wave component of the waveform data contains an abnormal sound, the harmonic component is attenuated and the fundamental wave component is emphasized, so that the abnormal sound in the mechanical operation sound after the frequency analysis can be easily determined. Moreover, the attenuation filter of the present invention is configured to perform the attenuation processing of the harmonic component based on the rotation cycle data detected by the rotation cycle detection means. The rotation cycle during operation of the machine varies even when the machine is apparently in a steady rotation state, and correct frequency analysis cannot be performed unless damping processing is performed based on an appropriate rotation cycle. For example, when a harmonic component is attenuated according to a predetermined exponential function, it is necessary to determine a time constant (attenuation characteristic) corresponding to the exponential function according to the rotation cycle. In the present invention, the data regarding the rotation cycle detected via the rotation cycle detection means is
Since it can be reflected in the attenuation processing of harmonic components,
It is possible to perform an appropriate damping process corresponding to a minute fluctuation during the machine operation. This makes it possible to more accurately determine the abnormal noise in the mechanical operation noise. As described above, according to the machine operation sound evaluation device of the first aspect, it is possible to appropriately evaluate the machine operation sound. In particular, it is possible to more accurately determine the abnormal noise in the mechanical operating noise. The evaluation device of the present invention is intended for operating sounds generated during the operation of various machines, and is preferably used when inspecting a vehicle engine. This makes it possible to detect the following defects of the vehicle engine. (1) Engagement sound Crankshaft gear and oil pump gear (2) Valve operating system tapping sound (tappet sound) Poor camshaft machining Hydroricklash adjuster (HLA) improper assembly valve clearance (3) Incorrect intake / exhaust valve Scratches on the crankshaft dent on the journal of the crankshaft Poor assembly of the piston ring Poor machining of the cylinder bore Interference between the crankshaft and oil pan parts

Here, it is preferable that the attenuation filter described in claim 1 has a structure as described in claim 2. That is, the attenuation filter according to the second aspect uses an attenuation index determined corresponding to the data regarding the rotation period. As the attenuation index, an attenuation function for attenuating the harmonic components of the waveform data in a predetermined manner, especially an exponential function, is preferably used. As described above, according to the machine operation sound evaluation device of the second aspect, it is possible to obtain the optimum damping characteristic corresponding to the rotation cycle during the machine operation.

According to the method for evaluating the mechanical operation sound of the third aspect, it is possible to appropriately evaluate the mechanical operation sound.
In particular, it is possible to more accurately determine the abnormal noise in the mechanical operating noise.

Further, according to the method for evaluating the mechanical operation sound of the fourth aspect, it is possible to obtain the optimum damping characteristic corresponding to the rotation cycle during the mechanical operation.

[0009]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the present invention will be described below with reference to the drawings. Here, FIG. 1 is a block diagram showing a configuration of an operation sound evaluation apparatus 100 according to an embodiment of the present invention. FIG. 2 is a conceptual diagram showing the processing performed by the controller 130 in FIG. In the present embodiment, a case will be described where the present invention is applied to a technique for determining an abnormal sound (abnormal sound) included in an operating sound generated from the vehicle engine in a vehicle engine inspection line.

As shown in FIG. 1, the operation sound evaluation apparatus 100 of the present embodiment is roughly classified into a first data detection means 11
0, second data detecting means 120, controller 13
0, the display device 160. First data detecting means 110 and second data detecting means 120
Are connected to the controller 130, and the data detected by each data detecting means is transmitted to the controller 130. In addition, the controller 130
Is connected to the display device 160, and the controller 13
The data processed by 0 is output to the display device 160.

The first data detecting means 110 is mainly composed of an acceleration sensor 112, an amplifier 114, an LPF (low pass filter) 116, and an HPF (high pass filter) 118. The acceleration sensor 112 detects the operating sound or vibration of the vehicle engine. Amplifier 11
4 amplifies the detection signal detected by the acceleration sensor 112 and transmits the amplified signal to the LPF 116 and HPF 118. The LPF 116 and the HPF 118 constitute a so-called bandpass filter, and remove unnecessary low frequency components and high frequency components of the signal amplified by the amplifier 114. The signals processed by the LPF 116 and the HPF 118 are sent to the A / D converter 1 of the controller 130 described later.
32 is transmitted. The first data detecting means 110 corresponds to the operating sound detecting means in the present invention.

The second data detecting means 120 is mainly composed of a rotation speed sensor 122 (rotation cycle sensor) and an amplifier 124. The rotation speed sensor 122 outputs a periodic pulse according to the rotation of the vehicle engine. The amplifier 124 amplifies the detection signal detected by the rotation speed sensor 122, and outputs the amplified signal to a controller 130 described later.
To the A / D converter 134. The second data detection means 120 corresponds to the rotation cycle detection means in the present invention.

The controller 130 mainly comprises A / D (analog / digital) converters 132 and 134, an I / F (interface) 136, a memory 138, and a CPU 140. A / D converters 132 and 134
Through the I / F 136, the memory 138, the CPU 14
0, connected to the display device 160. A / D converter 1
Reference numeral 32 converts the sound or vibration signal of the vehicle engine processed by the first data detection means 110 from an analog signal to a digital signal. The digital signal output from the A / D converter 132 is transferred to the memory 13 via the I / F 136.
8, the CPU 140, and the display device 160. A /
The D converter 134 converts the signal of the rotation cycle of the vehicle engine detected by the second data detection means 120 from an analog signal to a digital signal. The digital signal output from the A / D converter 134 is transmitted to the memory 138, the CPU 140, and the display device 160 via the I / F 136. The data output from the A / D converters 132 and 134 is temporarily stored in the memory 138, and various processes are performed according to the instruction of the CPU 140. Display device 160
Is configured using at least one of display means such as a display and a printer, and displays and outputs various data in accordance with a command from the CPU 140. This controller 130 corresponds to the processing means in the present invention.

The controller 130 has a processing configuration as shown in FIG. The first FFT circuit 151 performs FFT (Fast Fourier Transform) arithmetic processing of the data detected by the second data detection means 120. By this FFT calculation processing, the fundamental frequency of the vehicle engine (first-order component frequency)
Calculate f ₁ . Even when the vehicle engine is apparently in a steady rotation state, variations in the number of rotations occur, so that the fundamental frequency f ₁ is usually in a constantly changing state. Therefore, it is preferable that the second data detecting means 120 continuously detects the rotation cycle T _R , and the fundamental frequency f ₁ is calculated and updated each time.

The fundamental wave emphasizing section 152 performs processing for emphasizing the fundamental wave in the waveform data detected by the first data detecting means 110. The fundamental wave emphasizing unit 152 includes a squaring processing unit 153 and an attenuation filter 154. The squaring processing unit 153 obtains a squared waveform of the waveform data. The attenuating filter 154 performs an attenuating filter process for exponentially attenuating the nth harmonic component other than the fundamental wave component in the waveform data according to a predetermined exponential function G (attenuation function). By this attenuation filter process, the fundamental wave component of the waveform data is emphasized by selectively attenuating the harmonic components from the waveform data detected by the first data detection means 110.

In the second FFT circuit 155, the waveform data on which the fundamental wave is emphasized is subjected to frequency analysis by FFT (Fast Fourier Transform) calculation processing, and the intensity level corresponding to the frequency is calculated. Further, in the abnormal sound determination unit 156,
The result of the frequency analysis by the second FFT circuit 155 is used to determine whether the operating noise of the vehicle engine includes an abnormal noise. The abnormal sound determination unit 156 includes a peak value calculation unit 157, a logarithmic compression unit 158, and a comparison unit 159. The peak value calculation unit 157 calculates the peak value of the data frequency-analyzed by the second FFT circuit 155. The logarithmic compression unit 158 performs a process of logarithmically compressing the peak value calculated by the peak value calculation unit 157. The comparison unit 159 compares the data logarithmically compressed by the logarithmic compression unit 158 with a preset threshold value, and determines that the data exceeding the threshold value is abnormal noise. Then, the abnormal sound determination result by the abnormal sound determination unit 156 is output to the display device 160.

Next, a method of evaluating the operation sound of the vehicle engine using the operation sound evaluation apparatus 100 will be described in detail with reference to FIGS. For example, it is possible to target abnormal noise generated from a portion where the crankshaft gear of the vehicle engine and the oil pump gear mesh with each other. Here, FIG. 3 is a flowchart showing the abnormal sound determination processing. FIG. 4 is a diagram showing an example of the waveform data detected by the acceleration sensor 112 in step S10. FIG. 5 shows step S2
It is a figure which shows an example of the waveform data after the band-pass filter process and the square process were performed in 0 and S30. FIG. 6 is a diagram showing an example of the waveform data after the attenuation filter processing is performed in step S60. FIG. 7 shows step S
FIG. 10 is a diagram showing an example of data after being subjected to FFT calculation processing at 70.

When the operation sound evaluation apparatus 100 having the above-described configuration is activated, steps S10 to S1 shown in FIG. 3 are performed.
The abnormal noise determination process can be performed according to 10. In step S10, the acceleration sensor 112 is used to detect the sound or vibration of the vehicle engine. In this case, the acceleration sensor 112 is installed in a portion of the vehicle engine corresponding to the portion where the crankshaft gear and the oil pump gear mesh with each other. This makes it possible to detect the waveform data corresponding to the position where the crankshaft gear and the oil pump gear mesh with each other. For example, waveform data as shown in FIG. 4 is detected. Next, step S20 and step S40.
I do. In step S20, the waveform data detected by the acceleration sensor 112 is band-pass filtered through the LPF 116 and HPF 118. As a result, unnecessary low frequency components and high frequency components of the waveform data are removed. Further, in step S30, the squaring processing unit 153 squares the waveform data subjected to the band pass filter processing.
As a result, for example, waveform data as shown in FIG. 5 is output. When step S30 ends, the process proceeds to step S60.

On the other hand, in step S40, the rotation speed sensor
The rotation signal of the vehicle engine is detected by 122. Su
In step S50, the rotation signal detected in step S40
Is processed by the first FFT circuit 151.
In this FFT calculation process, the rotation at the time of detecting the rotation signal is performed.
Cycle T_R, Fundamental frequency f₁Is calculated. This rotation cycle
T _R, Fundamental frequency f₁Changes the vehicle engine speed
It changes depending on the situation. The rotation cycle T_RAnd fundamental frequency
Number f₁And f₁= 1 / T_RIt is shown by the relationship of.

In step S60, the attenuation filter 154
Performs the attenuation filter processing of the waveform data processed in step S30. This step S60 corresponds to the attenuation filter processing step in the present invention. In this attenuation filter processing, the fundamental frequency f ₁ calculated in step S50 is used. When the waveform data is attenuated according to a predetermined exponential function G, when the fundamental frequency f ₁ is determined, the optimum time constant τ of the exponential function G is determined. Fundamental frequency f
₁ and the time constant τ are represented by the relationship of τ = f ₁ × n (n: order). Appropriate attenuation filter processing can be performed by attenuating the waveform data based on the time constant τ. Thereby, for example, the waveform data shown by the solid line in FIG. 6 is output. Next, step S
At 70, the waveform data that has been subjected to the attenuation filter processing at step S60 is subjected to F by the second FFT circuit 155.
FT operation processing is performed. As a result, for example, as shown in FIG. 7, the intensity level (peak value) for each frequency is output.

Here, in step S60, the attenuation filter processing is performed based on the optimum time constant τ (example) and the attenuation filter processing is performed based on a time constant different from the time constant τ (comparison). The difference in the processing results obtained in (Example) will be described with reference to FIGS. 6 and 7.
In the comparative example, a time constant of 1/2 of the optimum time constant τ of the example, that is, (1/2) × τ was used. When the attenuation filter processing is performed using this time constant, the waveform data shown by the broken line in FIG. 6 was obtained. Further, when the waveform data indicated by the broken line is subjected to frequency analysis in step S70, the intensity level (peak value indicated by ◯ plot in FIG. 7) lower than the intensity level (peak value indicated by ● plot in FIG. 7) of the example. Data were obtained. As described above, when attenuation filtering and frequency analysis are performed using a time constant different from the original optimum time constant τ, it is often the case that an accurate intensity level cannot be obtained during frequency analysis.
Then, in such a case, it becomes difficult to determine the abnormal sound thereafter.

In step S80, the peak value calculator 157 calculates the intensity level (peak value) for each frequency output by the FFT calculation process. In step S90, the peak value calculated in step S80 is set to the logarithmic compression unit 1
Logarithmic compression processing is performed by 58. In step S100, the comparison unit 159 compares the logarithmically compressed data processed in step S90 with a preset threshold value, and determines that the data exceeding the threshold value is abnormal noise. Further, in step S110, the comparison unit 159 outputs the determination result in step S100 to the display device 160. By sequentially performing the above steps, it is possible to evaluate whether or not abnormal noise is generated from a position where the crankshaft gear and the oil pump gear mesh with each other. Since the evaluation result by such abnormal noise evaluation matches the evaluation result by the sensory test well, it is effective for accurate and rational evaluation of abnormal noise.

As described above, according to the present embodiment, since the detection data of the rotation period T _R is used when performing the attenuation filter processing via the attenuation filter 154, the optimum time constant τ for the attenuation of the harmonic component is obtained. Appropriate attenuation filter processing used can be performed. By performing such an appropriate attenuation filter process, it is possible to prevent the intensity level (peak value) for each frequency after the FFT calculation process from decreasing, and it is possible to accurately evaluate abnormal noise.

The present invention is not limited to the above-mentioned embodiments, and various applications and modifications can be considered. For example, each of the following modes to which the above-described embodiment is applied can be implemented.

(A) In the above embodiment, the case where the operating noise of the vehicle engine is evaluated has been described, but the present invention can also be applied to a technique for evaluating mechanical operating noise other than that of the vehicle engine.

(B) In the above embodiment, the case where the harmonic component of the waveform data is exponentially attenuated by using the exponential function G has been described. However, the attenuation function related to this attenuation characteristic is an exponential function. However, various damping functions can be used as needed.

[0027]

As described above, according to the present invention,
It is possible to realize an effective technique for appropriately evaluating the mechanical operating noise.

[Brief description of drawings]

FIG. 1 is an operation sound evaluation apparatus 100 according to an embodiment of the present invention.
3 is a block diagram showing the configuration of FIG.

FIG. 2 is a conceptual diagram showing a process performed by a controller 130 in FIG.

FIG. 3 is a flowchart showing an abnormal sound determination process.

FIG. 4 is a diagram showing an example of waveform data detected by an acceleration sensor 112 in step S10.

FIG. 5 is a diagram showing an example of waveform data after band-pass filtering processing and squaring processing have been performed in steps S20 and S30.

FIG. 6 is a diagram showing an example of waveform data after being subjected to an attenuation filter process in step S60.

FIG. 7 is a diagram showing an example of data after the FFT calculation processing is performed in step S70.

[Explanation of symbols]

100 ... Operating sound evaluation device 110 ... First data detecting means (operating sound detecting means) 112 ... Acceleration sensor 120 ... Second data detection means (rotation period detection means) 122 ... Revolution sensor 130 ... Controller (control means) 151 ... First FFT circuit 152 ... Fundamental wave emphasis section 153 ... Square processing unit 154 ... Attenuation filter 155 ... Second FFT circuit 156 ... Abnormal sound determination unit 157 ... Peak value calculation unit 158 ... Logarithmic compression unit 159 ... Comparison section 160 ... Display device

   ─────────────────────────────────────────────────── ─── Continued front page    (72) Inventor Kenji Akiyama             2-1, Toyota-cho, Kariya City, Aichi Stock Association             Inside Toyota Toyota Industries F term (reference) 2G064 AA15 AB13 BA02 CC02 CC41

Claims (4)

[Claims] 1. An operating sound detecting means and a processing means, wherein the operating sound detecting means detects waveform data relating to a mechanical operating sound during machine operation, and the processing means is detected by the operating sound detecting means. A method for evaluating a machine operation sound based on a result of frequency analysis of the waveform data and a result of the frequency analysis, further comprising: a rotation cycle detection for detecting data on a rotation cycle during machine operation. The processing means has an attenuation filter, and the attenuation filter attenuates a harmonic component other than the fundamental wave component in the waveform data based on the data regarding the rotation period before the frequency analysis. An apparatus for evaluating mechanical operating noise, which is configured as described above. 2. The machine operation noise evaluation device according to claim 1, wherein the attenuation filter attenuates the harmonic component using an attenuation index determined corresponding to the data regarding the rotation cycle. A mechanical operation sound evaluation device characterized by being configured. 3. A method for evaluating a machine operating sound, which comprises detecting waveform data relating to a machine operating sound during machine operation, frequency-analyzing the waveform data, and evaluating the machine operating sound based on the frequency analysis result. Then, before the frequency analysis, a step of detecting data regarding a rotation cycle during machine operation, and an attenuation filter process for attenuating a harmonic component other than a fundamental wave component of the waveform data based on the data regarding the rotation cycle. A method for evaluating a mechanical operating noise, comprising: 4. The method for evaluating mechanical operating noise according to claim 3, wherein in the attenuation filter processing step, the harmonic component is attenuated using an attenuation index determined corresponding to the data regarding the rotation cycle. A method for evaluating machine operation noise, which is characterized by the following.