JP6369171B2 - Noise measurement method - Google Patents

Description

  The present invention relates to a noise measurement method.

  In recent years, there has been an increasing demand for noise reduction for devices constituting vehicles such as automobiles. Examples of noise countermeasures include a method of reducing vibration of the apparatus and a method of suppressing resonance generated by surface materials such as a housing that accommodates components in the apparatus being vibrated. In order to implement such noise countermeasures, it is necessary to specify the noise source of the apparatus.

  As a method for identifying a noise source of the above-described apparatus, sound source search based on acoustic intensity is known. Hereinafter, the sound source search based on the sound intensity will be described with reference to FIG. In FIG. 7, a motor 200 is used as a measurement object.

  The measuring apparatus 100 sets a virtual surface HP on the surface of the motor 200 that measures noise. The measuring apparatus 100 divides the virtual surface HP by a large number of grids, and uses the grid points indicated by black dots in the figure as noise level measurement points. In a state where the motor 200 is driven, the measuring apparatus 100 moves the noise meter 110 to each measurement point by the driving unit 120 and measures the noise level of the motor 200. The measurer specifies the measurement point showing the highest noise level among the noise levels measured at each measurement point as the noise source. Note that Patent Document 1 discloses an example of a noise measurement method using acoustic intensity.

Japanese Patent Laid-Open No. 5-118965

  The sound source search based on the sound intensity needs to measure the noise level of the motor 200 at a large number of measurement points by the sound level meter 110. For this reason, the time for specifying the noise source of the motor 200 becomes long. On the other hand, when the number of measurement points is reduced in order to shorten the measurement time of the noise level of the motor 200, the interval between the measurement points becomes large, and the noise source of the motor 200 cannot be specified accurately. Such a problem is not limited to the motor 200, and similarly occurs if the device is operated as a measurement object by a drive source.

  An object of the present invention is to provide a noise measurement method capable of quickly specifying a noise source of a measurement object.

  [1] According to an independent form of the present noise measurement method, a frequency extraction step of extracting a specific frequency component from the frequency components of the noise of the measurement object in a state where the measurement object is operated, and an imaging device The imaging step of measuring the temperature of the measurement object by imaging the measurement object at a frame rate different from that of the specific frequency component, and calculating the temperature change amount of the image of the captured measurement object A temperature change calculating step for calculating a temperature change amount of the measurement object, and a noise source specifying step for specifying a noise source of the measurement object based on the temperature change amount of the measurement object.

  The noise of the measurement object is generated when the measurement object vibrates, and increases as the amplitude increases. For this reason, the place where the measurement object vibrates most is the place where the noise level is the highest. Further, the vibration of the measurement object is generated by repeatedly bending the measurement object by repeatedly pulling and compressing the measurement object. When the measurement object is repeatedly pulled and compressed, a slight temperature change occurs in the measurement object due to the thermoelastic effect. This temperature change increases as the amplitude of vibration of the measurement object increases. Therefore, the inventor of the present application thought that if a minute temperature change of the measurement object based on the thermoelastic effect can be acquired, the noise source of the measurement object can be specified from the temperature change amount of the measurement object.

  Based on this idea, this noise measurement method calculates the temperature change amount of the measurement object by calculating the temperature change amount of the captured image of the measurement object, and obtains the temperature change amount of the measurement object. Based on this, the noise source of the measurement object is specified. As a result, since the captured image of the measurement object includes a large number of measurement points in sound intensity sound source exploration, the temperature change amount of the measurement object corresponding to the large number of measurement points is calculated. . Therefore, it is not necessary to measure the noise level at a large number of measurement points in order to specify the noise source of the measurement object as in the sound source search of the sound intensity. Therefore, the noise source of the measurement object can be quickly identified.

  [2] According to one aspect dependent on the noise measurement method, in the imaging step, the imaging device acquires the measurement object whose temperature changes in synchronization with the specific frequency component as the image, In the temperature change calculation step, the temperature change amount of the image of the measurement object whose temperature changes in synchronization with the specific frequency component is calculated.

  According to this noise measurement method, since the imaging device acquires an image of a measurement object whose temperature changes in synchronization with a specific frequency component, the temperature change of the measurement object based on the specific frequency component is extracted, The temperature change of the measurement object based on the frequency other than the frequency component is excluded. For this reason, in the temperature change calculation process, the temperature change amount of the measurement object whose temperature changes in synchronization with a specific frequency component is calculated. For this reason, it becomes easy to specify the noise source of the measurement object based on a specific frequency component.

  [3] According to an embodiment dependent on the noise measurement method, in the temperature change calculation step, the image of the measurement target imaged over a plurality of periods of the specific frequency component is integrated and processed. A temperature change amount of the measurement object in one cycle of a specific frequency component is calculated.

  According to this noise measurement method, an image of a measurement object that changes in temperature in synchronization with a specific frequency component is subjected to integration processing, so that a temperature change that is synchronized with pulling and compression of the measurement object based on the specific frequency component. Are extracted with high accuracy. For this reason, since the maximum value and the minimum value of the temperature of the measurement object over one cycle of the specific frequency component can be obtained with high accuracy, the calculation accuracy of the temperature change amount of the measurement object based on the specific frequency component is improved. . Therefore, the noise source of the measurement object based on the specific frequency component can be specified with high accuracy.

  [4] According to one aspect dependent on the noise measurement method, an image processing step of reflecting the temperature change amount of the measurement object calculated by the temperature change calculation step on the imaged image of the measurement object. In the noise source specifying step, the noise source of the measurement object is specified based on the image processed image.

  The noise measurement method can quickly identify the noise source of the measurement object.

The block diagram which shows the structure of the electric power steering apparatus as a measuring object. The lineblock diagram showing the composition of the noise measuring device of one embodiment. The flowchart which shows the procedure of the identification method of the noise source of one Embodiment. (A) The graph which shows transition of the temperature change in a predetermined pixel, (b) The graph which shows transition of the signal which shows imaging timing. The graph which shows transition of the temperature change in 1 period after an integration process. The schematic diagram which shows the temperature image data of an assist motor and its periphery. The schematic diagram of the apparatus of the sound source search based on the conventional sound intensity.

With reference to FIG. 1, the structure of the electric power steering apparatus used as the measuring object of this noise measuring apparatus is demonstrated.
The electric power steering device 50 includes a steering mechanism 51 that steers the left and right steered wheels 70 and 70 based on the steering operation of the driver, and a steering assist mechanism 60 that assists the steering operation of the driver.

  The steering mechanism 51 includes a steering wheel 52 that is operated by a driver, a steering shaft 53 that rotates integrally with the steering wheel 52, and a rack shaft 54 that steers the steered wheels 70 and 70. The steering shaft 53 includes a column shaft 53A connected to the center of the steering wheel 52, an intermediate shaft 53B connected to the lower end portion of the column shaft 53A, and a pinion shaft 53C connected to the lower end portion of the intermediate shaft 53B. Become. A lower portion of the pinion shaft 53C is meshed with a rack shaft 54 (more precisely, a portion 54A in which rack teeth are formed) extending in a direction intersecting with the pinion shaft 53C. Therefore, the rotational motion of the steering shaft 53 is converted into the reciprocating linear motion of the rack shaft 54 by the rack and pinion mechanism 55 including the pinion shaft 53C and the rack shaft 54. The reciprocating linear motion is transmitted to the left and right steered wheels 70 and 70 through tie rods 56 respectively connected to both ends of the rack shaft 54, whereby the steered angles of the steered wheels 70 and 70 are changed. .

  The steering assist mechanism 60 includes an assist motor 61 that is a generation source of a steering assist force (assist force). The assist motor 61 employs a brushless motor and is connected to the column shaft 53 </ b> A via a speed reduction mechanism 63. The deceleration mechanism 63 decelerates the rotation of the assist motor 61, and transmits the rotational force in the decelerated state to the column shaft 53A. That is, the steering operation of the driver is assisted by applying the torque of the assist motor 61 to the steering shaft 53 as an assist force.

  The configuration of the noise measuring device 1 will be described with reference to FIGS. 1 and 2. The noise measurement device 1 according to the present embodiment specifies a noise source of the housing 62 of the assist motor 61 in the electric power steering device 50.

  As shown in FIG. 1, the noise measuring device 1 includes a noise meter 10 that measures a noise level, and an FFT device (Fast Fourier Transform) that takes in the noise level measured by the noise meter 10 and frequency-decomposes the noise level. 20. The sound level meter 10 is a microphone and is disposed on the side of the housing 62 of the assist motor 61. The sound level meter 10 detects the noise level of the housing 62 of the assist motor 61. The FFT device 20 performs frequency decomposition on the noise level measured by the sound level meter 10 with the audible range as the upper limit (20 kHz), and displays the result on the display unit 21.

  Further, as shown in FIG. 2, the noise measurement device 1 takes in the infrared thermography 30 that is an imaging device that images the housing 62 of the assist motor 61, and temperature image data captured by the infrared thermography 30, and the temperature image thereof. And an image processing device 40 that performs image processing of data.

  The infrared thermography 30 includes an infrared sensor (not shown) that outputs an electrical signal corresponding to the intensity of infrared rays emitted from the surface of an object. The infrared thermography 30 converts an electrical signal of the infrared sensor into an image signal and outputs the image signal to the image processing device 40.

  The image processing device 40 includes an image processing unit 41 that performs image processing based on an image signal input from the infrared thermography 30, and a display unit 42 that displays temperature image data image-processed by the image processing unit 41. . The image processing unit 41 calculates a temperature change calculation unit 41A that calculates the temperature change of the housing 62 based on the image signal, and each pixel of the temperature image data based on the temperature change of the housing 62 calculated by the temperature change calculation unit 41A. And a hue setting unit 41B for setting the hue.

  With reference to FIGS. 3-6, the noise measuring method of the electric power steering apparatus 50 by the noise measuring apparatus 1 is demonstrated. Hereinafter, as a noise measurement method, a method for specifying a noise source of the housing 62 of the assist motor 61 in the electric power steering apparatus 50 will be described. In the following description with reference to FIGS. 3 to 6, the components of the noise measuring device 1 and the electric power steering device 50 to which reference numerals are attached are the noise measuring device 1 and the electric motor described in FIGS. 1 and 2. The components of the power steering apparatus 50 are shown.

  The outline of the noise source specifying method is based on the amount of change in the temperature of the housing 62 indicated by the temperature image data obtained by imaging the housing 62 at an arbitrary frame rate by the lock-in infrared thermography 30 and image-processing by the image processing device 40. Thus, the noise source of the housing 62 is specified. Hereinafter, a detailed procedure of the noise source specifying method will be described.

  The procedure of the noise source specifying method is shown by the flowchart of FIG. In the flowchart, steps S1 and S2 indicate a frequency extraction process, steps S3 and S4 indicate an imaging process, step S5 indicates a temperature change calculation process, step S6 indicates an image processing process, and step S7 indicates a noise source identification. The process is shown.

  In step S1, the noise level of the housing 62 and its surroundings is measured by the sound level meter 10 disposed 30 cm away from the side of the housing 62 while the assist motor 61 is rotated at 2000 rpm. In the present embodiment, the assist motor 61 measures the noise level of the housing 62 of the assist motor 61 and its surroundings while the steering wheel 52 rotates from the neutral position to the limit position in the predetermined turning direction. The noise level indicates a sound pressure level measured by the A characteristic which is a weighting characteristic of a frequency close to human hearing (good sensitivity in the frequency band of 3 to 4 kHz) by the sound level meter 10.

  Further, the noise measured by the sound level meter 10 is frequency-resolved by the FFT device 20. The FFT apparatus 20 displays on the display unit 21 a graph indicating the relationship between the noise level and the frequency component, and a table indicating the peak value of the noise level and the frequency component corresponding to the peak value. Note that the graph showing the relationship between the noise level and the frequency component has a plurality of peak values.

  In step S <b> 2, a specific frequency component fs is extracted based on the frequency component of the noise level that is frequency-resolved by the FFT device 20. The specific frequency component fs of this embodiment is a frequency component corresponding to the maximum peak value among a plurality of peak values of the noise level.

  In step S3, the frame rate FR of the infrared thermography 30 is set as a frequency different from the specific frequency component fs. In step S4, the infrared thermography 30 images the housing 62 and its periphery. The infrared thermography 30 captures an image based on the frame rate FR, and outputs the captured temperature image data to the image processing device 40 as an image signal. The infrared thermography 30 receives a lock-in signal that is a signal synchronized with a specific frequency component fs from the image processing device 40. The infrared thermography 30 acquires temperature information of the housing 62 and its surroundings as temperature image data in response to only the lock-in signal. That is, the infrared thermography 30 can separate temperature information of frequency components other than the specific frequency component fs.

The frame rate FR is preferably a frequency that is smaller than the specific frequency component fs and that cannot divide the specific frequency component fs. The reason for this is as follows.
It is considered that the housing 62 vibrating at the specific frequency component fs changes in temperature at the specific frequency component fs due to the thermoelastic effect. For this reason, in order to calculate the temperature change of the housing 62 based on the specific frequency component fs, it is necessary to acquire the temperature of the housing 62 over one cycle of the specific frequency component fs.

  However, when the frame rate FR is the same frequency as the specific frequency component fs, the imaging timing of the infrared thermography 30 and the cycle of the specific frequency component fs are synchronized, so one of the one cycle of the specific frequency component fs. The temperature of the housing 62 at the time of the point is acquired in each cycle of the specific frequency component fs. For this reason, since the temperature of the housing 62 over one period of the specific frequency component fs cannot be acquired, the maximum value and the minimum value of the temperature of the housing 62 in one period of the specific frequency component fs cannot be acquired. Therefore, the temperature change of the housing 62 at the specific frequency component fs cannot be obtained with high accuracy. In addition, when the frame rate FR is a frequency that can divide the specific frequency component fs, for example, when the frame rate FR is a half frequency of the specific frequency component fs, the imaging timing of the infrared thermography 30 is 1 of the specific frequency component fs. The temperature of the housing 62 at two points of the cycle start point and the cycle center point in the cycle is acquired in each cycle. For this reason, as in the case where the frame rate FR has the same frequency as the specific frequency component fs, the temperature change of the housing 62 in one cycle of the specific frequency component fs cannot be obtained with high accuracy.

  On the other hand, in the case of a frequency that cannot divide the specific frequency component fs, the imaging timing of the infrared thermography 30 is different for each cycle of the specific frequency component fs. Therefore, the temperature of the housing 62 is acquired over many cycles, and the result will be described later. By performing the integration process as in step S5, the temperature of the housing 62 over one cycle of the specific frequency component fs can be acquired. Therefore, since the maximum value and the minimum value of the temperature of the housing 62 in one cycle of the specific frequency component fs can be acquired, the temperature change amount of the housing 62 based on the specific frequency component fs can be acquired with high accuracy.

Next, in step S5, the image processor 40 calculates the temperature change of the housing 62 and its surroundings as follows.
The temperature change calculation unit 41A performs integration processing for integrating image signals indicated by black dots at the frame rate FR shown in FIG. 4 in each cycle of the specific frequency component fs. Since the frame rate FR is set as described above for these black dots, the image signals (black dots) acquired over a plurality of periods are superimposed on one period of a specific frequency component fs as shown in FIG. A graph showing the transition of the temperature of the housing 62 in one cycle of the specific frequency component fs is created. Then, the temperature change calculation unit 41A extracts the maximum value T1 and the minimum value T2 of the temperature in the graph of FIG. 5, and calculates the temperature change amount ΔT (= T1-T2) that is the difference between them. Then, the temperature change calculation unit 41A outputs the temperature change amount ΔT to the hue setting unit 41B.

  Next, in step S <b> 6, the temperature image data is subjected to image processing by the image processing device 40. The image processing device 40 reflects the temperature change amount ΔT in the housing 62 and its surroundings to each pixel of the temperature image data by the hue setting unit 41B as image processing of the temperature image data.

  The hue setting unit 41B stores a conversion table between the temperature change amount ΔT and the hue. The hue setting unit 41B sets the hue from the temperature change amount ΔT using the conversion table. The hue setting unit 41B sets the hue of each pixel so that the hue changes from blue to red as the temperature change amount ΔT increases.

  The hue setting unit 41B sets the hue for all the pixels of the temperature image data, thereby creating the temperature image data shown in FIG. In FIG. 6, each grid-like square region indicates a pixel. A pixel to which dots are not attached shows blue, and a pixel to which dots are attached most finely shows red. Each pixel has a hue close to red as the dots become finer. FIG. 6 shows the temperature image data with 50 × 68 pixels for convenience, but the actual temperature image data is composed of finer pixels.

  In step S7, the noise source of the housing 62 is specified based on the temperature image data subjected to the image processing. Specifically, as shown in FIG. 6, the measurer sets an image region R corresponding to the finest pixel (red) surrounded by a thick line in the temperature image data reflecting the temperature change in the housing 62. Identify noise sources.

Further, in the above-described method for specifying the noise source of the housing 62, the noise source of the housing 62 is specified as follows with respect to the frequency caused by the assist motor 61 or the speed reduction mechanism 63.
In the noise measurement at step S1, the plurality of peak values of the noise level include, for example, the switching frequency of the assist motor 61 and the meshing frequency of the speed reduction mechanism 63. The switching frequency and the meshing frequency can be calculated based on the number of magnetic poles of the assist motor 61, the rotation speed of the assist motor 61, and the number of teeth of the speed reduction mechanism 63. For this reason, the measurer can grasp whether the frequency of the peak value of the measured noise level is caused by the switching frequency or the meshing frequency.

  Then, the frequencies of the plurality of peak values of the noise level are extracted as specific frequency components fs, respectively, and the noise source of the housing 62 corresponding to each specific frequency component fs is specified sequentially in the procedure of steps S2 to S7. Thereby, the noise source of the housing 62 caused by the switching frequency and the noise source of the housing 62 caused by the meshing frequency are specified.

  Further, when the noise source of the housing 62 is specified, a weight (not shown) is attached to the noise source of the housing 62 as a measure for reducing the noise of the housing 62. Then, the measurer measures the temperature change at the noise source of the housing 62 again in steps S2 to S6 in order to verify whether or not the noise level of the noise source of the housing 62 has been reduced. At this time, when the hue of the image region R in the noise source of the housing 62 changes from red to blue, it can be seen that the vibration of the noise source of the housing 62 is reduced by this weight. In order to reduce the noise of the housing 62, a design change such as providing a rib (not shown) on the housing 62 can be performed. In addition, when a weight is attached to the housing 62, a new noise source may be generated in another part of the housing 62. Even in this case, a new noise source can be specified by acquiring the temperature change of the housing 62 in the procedure of steps S2 to S6.

Below, the effect of the noise measuring method by the noise measuring apparatus 1 is demonstrated.
(1) The noise of the housing 62 is generated when vibrations of the rotor and the stator (both not shown) of the assist motor 61 propagate to the housing 62. The noise of the housing 62 is caused by the vibration of the housing 62 and increases as the amplitude increases. For this reason, the portion where the housing 62 vibrates most is the portion where the noise is the largest in the housing 62.

  Further, the vibration of the housing 62 is caused by the housing 62 being repeatedly bent due to repeated pulling and compression of the housing 62. When the housing 62 is bent, a minute temperature change occurs in the housing 62 due to the thermoelastic effect resulting from the bending.

  Therefore, the present inventor considered that if a minute temperature change occurring in the housing 62 is known, the portion where the temperature change is the largest is the portion where the amplitude of the housing 62 is the largest, and that portion can be specified as the noise source. The inventor of the present application considered that a slight temperature change of the housing 62 can be acquired by imaging the housing 62 using the infrared thermography 30, and the noise source of the housing 62 can be specified from the acquired temperature change.

  Therefore, according to the noise measurement method using the infrared thermography 30, the temperature change amount ΔT of the temperature image data of the imaged housing 62 is calculated, and the temperature change amount ΔT of the housing 62 is calculated based on the calculated temperature change amount ΔT of the housing 62. Noise sources are identified. Since the imaged temperature image data of the housing 62 includes a large number of measurement points in the sound intensity sound source search, the temperature of the housing 62 corresponding to a large number of measurement points based on the imaged temperature image data of the housing 62. The change will be calculated. For this reason, it is not necessary to measure the noise level at a large number of measurement points in the housing 62 in order to specify the noise source of the housing 62 as in the sound source search of the sound intensity. Therefore, the noise source of the housing 62 can be quickly identified.

  In particular, in the case of sound intensity sound source exploration, in order to reduce noise, when a weight is attached to the noise source of the housing 62, it is also determined whether or not a new noise source has occurred. The level needs to be measured again. For this reason, in the case of sound intensity sound source exploration, it takes time to measure the noise level. Further, when a plurality of types of weights are used, the noise level is measured at a large number of measurement points for each weight. The time for measuring is significantly increased.

  On the other hand, in the noise measurement method of the present embodiment, the degree of noise level reduction of the housing 62 and the presence or absence of a new noise source can be known from the temperature image data of the housing 62. For this reason, compared with the sound source search of the sound intensity, the time for the determination of the noise level reduction degree of the housing 62 and the presence or absence of a new noise source is shortened.

  (2) The infrared thermography 30 acquires temperature image data of the housing 62 whose temperature changes in synchronization with the specific frequency component fs. Accordingly, the temperature change calculation unit 41A can calculate the temperature change amount ΔT of the housing 62 according to the specific frequency component fs. For this reason, it becomes easy to specify the noise source of the housing 62 based on the specific frequency component fs.

  (3) The temperature change calculation unit 41A integrates the image signal of the image of the housing 62 whose temperature changes in synchronization with the extracted specific frequency component fs over a plurality of cycles of the specific frequency component fs. Thereby, the temperature change synchronized with the pulling and compression of the housing 62 based on the specific frequency component fs can be accurately extracted, and the maximum value and the minimum value of the temperature of the housing 62 in one cycle of the specific frequency component fs are accurately determined. You can get well. For this reason, the calculation accuracy of the temperature change amount ΔT of the housing 62 based on the specific frequency component fs is improved. As a result, temperature image data in which the temperature change amount ΔT of the housing 62 based on the specific frequency component fs is accurately reflected is created, so that the noise source of the housing 62 based on the specific frequency component fs can be accurately identified.

  In addition, the specific form which this noise measuring method can take is not limited to the content shown by the said embodiment. This noise measurement method can take the form of the modification of the said embodiment shown below, for example.

  The image processing unit 41 may include a calibration unit. The calibration unit calibrates the temperature change amount ΔT of the housing 62 and the noise level as follows. For example, the magnitude of the noise level of the noise source of the housing 62 and the magnitude of the temperature change amount ΔT of the noise source of the housing 62 are input to the calibration unit. Then, the calibration unit uses the relationship of the ratio of the noise level of the noise source of the housing 62 to the temperature change amount ΔT of the noise source of the housing 62 as a reference, and this relationship also applies to the temperature change amount ΔT of the portion other than the noise source of the housing 62. Based on this, the noise level is calculated.

  Further, the vibration of the noise source of the housing 62 may be measured by an acceleration pickup. The calibration unit calibrates the temperature change amount ΔT and vibration of the housing 62 as follows. For example, the magnitude of vibration of the noise source of the housing 62 and the temperature change ΔT of the noise source of the housing 62 are input to the calibration unit. The calibration unit uses the relationship of the ratio of the vibration of the noise source of the housing 62 to the temperature change amount ΔT of the noise source of the housing 62 as a reference, and the temperature change amount ΔT of the portion other than the noise source of the housing 62 is also based on this relationship. To calculate vibration.

  The specific frequency component fs may be set as a frequency having a maximum peak value in a frequency band that is harsh in the audible range of 20 Hz to 20 kHz. An example of the frequency band causing the annoyance is set as a frequency band excluding a high frequency band of 16 kHz or more, for example, as a frequency band in which sound is hard to hear in the audible range.

  In step S1 of the noise source specifying method, the noise level near the assist motor 61 may be measured by the sound level meter 10 in a continuous steering state in which the steering wheel 52 repeats a right turn and a left turn.

  In step S <b> 1 of the noise source specifying method, the sound level meter 10 may be arranged at a position other than a position 30 cm away from the housing 62. The rotation speed of the assist motor 61 may be a rotation speed other than 2000 rpm.

  In the noise source identification method, for example, the noise source of the housing 62 at a plurality of rotation speeds of the assist motor 61 such as 1500 rpm, 2000 rpm, and 3000 rpm may be identified. Thus, when the rotation speed of the assist motor 61 is changed, resonance may occur in the housing 62. Therefore, by specifying the noise source of the housing 62 using the resonance frequency of the housing 62 as the specific frequency component fs, the portion that resonates in the housing 62 is specified.

  Step S1 may be omitted from the noise source identification method. In this case, for example, the switching frequency of the assist motor 61 may be set as the specific frequency component fs. In short, the specific frequency component fs is not limited to being extracted from the result of actually measuring the noise of the measurement object, and may be set to a predetermined frequency.

  -You may abbreviate | omit step S6 which is an image processing process from the identification method of a noise source. In this case, the temperature change ΔT of the housing 62 calculated in step S5 may be displayed in a graph with each pixel as the horizontal axis and the temperature change ΔT of each pixel as the vertical axis. In step S <b> 7, the pixel or the image region where the temperature change amount ΔT is the maximum value in the graph is specified as the noise source of the housing 62.

In step S7 of the noise source specifying method, in the temperature image data indicating the temperature change amount ΔT, an image region showing the temperature change amount ΔT greater than or equal to a predetermined value may be specified as the noise source.
The noise level of the electric power steering device 50 in the driver's seat may be measured by the sound level meter 10. In this case, the specific frequency component fs is set to a frequency that is annoying in the driver's seat.

  As a measure for reducing the noise of the noise source of the housing 62, other parts of the housing 62 may be cut instead of attaching a weight to the noise source of the housing 62. Thereby, since the rigidity of the other part of the housing 62 is lowered, the bending state of the entire housing 62 is changed, and the vibration of the noise source of the housing 62 is likely to be reduced.

  The measurement target may be another steering device such as a rack assist type electric power steering device or an electrohydraulic type power steering device instead of the column assist type electric power steering device 50.

  The measurement object may be a mechanism in which the assist motor 61 is detached from the column shaft 53A and the assist motor 61 and the speed reduction mechanism 63 are combined. The measurement object may be the assist motor 61 alone. Further, the measurement object may be a part other than the assist motor 61 such as the rack and pinion mechanism 55 of the electric power steering apparatus 50. In addition, the measurement object is not limited to the electric power steering device 50, and may be any device that is driven by a predetermined drive source.

  30 ... Infrared thermography (imaging device), 50 ... Electric power steering device (measurement object), ΔT ... Temperature change amount, FR ... Frame rate, fs ... Specific frequency component.

Claims (3)

A frequency extraction step of extracting a specific frequency component of the frequency components of the noise of the measurement object in a state where the measurement object is activated;
Unlike the specific frequency component by the imaging device, by imaging the measurement object at a frame rate which is set to a frequency which the indivisible a specific frequency component, an imaging step of measuring the temperature of the object to be measured When,
A temperature change calculation step of calculating a temperature change amount of the measurement object in one cycle of the specific frequency component by calculating a temperature change amount of the image of the measured measurement object;
A noise measurement method comprising: a noise source identification step of identifying a noise source of the measurement object based on a temperature change amount of the measurement object. In the temperature change calculation step, the temperature change amount of the measurement object in one period of the specific frequency component is obtained by integrating the images of the measurement object captured over a plurality of periods of the specific frequency component. Is computed
The noise measurement method according to claim 1 . An image processing step of reflecting the amount of temperature change of the measurement object calculated by the temperature change calculation step on the image of the measured measurement object;
In the noise source specifying step, the noise source of the measurement object is specified based on the image processed image
The noise measurement method according to claim 1 or 2 .