JP2656307B2 - Sound source locator - Google Patents

Description

[Detailed Description of the Invention] [Object of the Invention] (Industrial application field) The present invention relates to a sound source detection device suitable for use in, for example, abnormality diagnosis of equipment constituting a plant.

(Prior Art) Conventionally, a sound source search device that searches for a noise source using an acoustic technique has been considered. This sound source locator uses sound intensity, basically uses two microphones, and utilizes the phase difference between sounds captured by these microphones. That is, when the sound source is on the line connecting the two microphones, the phase difference is maximum and the sound intensity value is maximum. When the line connecting the two microphones is orthogonal to the sound source direction, the phase difference is zero, and the sound intensity value is zero. Using this principle, in a conventional sound source locating device, two microphones are rotated around one point, and the direction in which the sound intensity value is maximum is set as the sound source direction, and this direction is displayed on the display device. Like that.

By the way, humans make extremely accurate abnormality diagnosis based on the five senses, especially the ears and eyes. As described above, when the abnormality diagnosis is performed by combining the results of the acoustic processing and the image processing, the accuracy of the diagnosis can be expected to be improved. The characteristic of the sound heard by the ear is that it can be heard in all directions, that is, in any direction. On the other hand, the feature of an image viewed by the eye is that it is specified within the viewing angle in the direction in which the image is facing. Therefore, a method is considered in which the direction of the abnormality is determined by sound, the direction of the sound source is determined, and the abnormality is determined by an image.

(Problems to be Solved by the Invention) The present invention utilizes a method of searching for a sound source using sound intensity, obtains clear image information in a sound source direction, and uses both sound information and image information. It is an object of the present invention to provide a sound source detection device capable of comprehensively determining the presence or absence of an abnormality.

[Structure of the Invention] (Means for Solving the Problems) In order to achieve the above object, the present invention relates to a sound source detecting apparatus used for diagnosing abnormalities of equipment constituting a plant. A comparison is made between a sound receiver provided with a plurality of microphones arranged in a fixed relationship in the yard and the atmosphere in the installation yard of the equipment at the present time and in a steady state. Means for transmitting an output to the sound receiver, and operation is started by the output of the means, and the sound intensity in each direction centering on the position of the sound receiver is determined from the output signal of each microphone constituting the sound receiver. Means for determining the direction of the sound source with reference to the position of the sound receiver from the sound intensity determined by the means; a camera disposed near the sound receiver; And means for directing the sound source direction of the optical axis of the camera that is, is characterized by comprising; and a receiving apparatus Projects an image captured by at least the camera.

(Operation) If the current atmosphere of the equipment installation site is different from the normal condition, for example, if abnormal sound is detected,
The detection signal is used as a trigger signal to start the operation of the means for obtaining the sound intensity in each direction centering on the position of the sound receiver from the output signals of the microphones constituting the sound receiver. Then, when the direction of the sound source is determined based on the sound intensity in each direction, the optical axis of the camera is automatically directed in that direction. It is generally easy for a camera to have an autofocus function. Therefore, a clear image of the device serving as the sound source can be obtained, and abnormality diagnosis can be performed by processing the image.
Of course, at this time, the sound of the place can be output from the speaker of the monitor device, and the analysis result of the acoustic processing can be displayed. Therefore, assuming that the image receiving device is installed in the central control room, it is not only possible to confirm the normal state of the equipment that is the sound source in the central control room as a clear image, The processing can automatically determine the abnormality of the device. That is,
It is possible to comprehensively determine whether or not the device is abnormal based on the audio information and the image information.

(Example) Hereinafter, an example is described with reference to drawings. FIG. 1 shows the appearance of a sound source detection device according to one embodiment.

This sound source locating device is roughly divided into a detection unit 1 arranged in a room where various devices are installed in a plant or the like, and the detection unit 1 is connected via a cable 2 and arranged in a central control room or the like. And a signal processing unit 3.

The detection unit 1 includes a power supply box 11 and the power supply box 11.
Fixed table 13 fixed horizontally via a support 12 on the
And a rotary table 15 rotatably supported on the fixed table 13 via a bearing (not shown) in a direction indicated by a solid arrow 14 in the figure, and a rotation angle of the rotary table 15 fixed to the lower surface of the fixed table 13. Motors such as step motors to be controlled
A rotation angle detector that detects the rotation angle of the rotary table 16 and 16
17, an autofocus camera 18 fixed on the rotary table 15, and a sound receiver 20 fixed on the camera 18 via a support 19.

The sound receiver 20 rotates the sound receiving portions of the omnidirectional microphones 21 and 22 formed in a bar shape on a line 24 that is parallel to the optical axis of the camera 18 and intersects the rotation center axis 23 of the turntable 15 at right angles. They are arranged symmetrically about the central axis 23.

On the other hand, the signal processing unit 3 converts the signals obtained by the microphones 21 and 22 into analog-to-digital (A / D) converters, a circuit that processes angle signals obtained by an angle detector, and a control signal to the motor 16. It comprises an input / output device 31 having a built-in sending circuit, a processing device 32 mainly composed of a computer, and a display device 33. The processing device 32 constantly monitors the output signal of one of the microphones 21 and sends a signal for controlling the motor 16 when the level of the output signal exceeds a certain value, and sends the signal to the rotary table.
15 is intermittently rotated, the sound intensity value in each direction is calculated from the outputs of the microphones 21 and 22 a total of 6 times every 30 degrees in the range of 0 to 150 degrees, and finally, A function for sending a signal for positioning the rotary table 15 at the rotation angle at which the largest sound intensity value is obtained, that is, a sound source direction detection function, a function for processing the obtained sound signal and image signal, and presence / absence of abnormality And a function for determining Then, the sound source direction detected by the processing device 32, the result of the acoustic processing / image processing, the presence / absence of an abnormality, and the image captured by the camera 18 are displayed on the image receiving device 33.

As described above, the sound source direction is detected by the processing device 32. Here, the theory of sound intensity and the flow of signal processing at the time of sound source direction detection will be described.

Sound intensity value in the r direction at a given point
Ir is given assuming that the sound pressure at that point is P (t) and the r-direction component of the particle velocity is Vr (t). Becomes Assuming that the air density is ρ, the following relationship exists between the particle velocity Vr and the sound pressure P.

(∂Vr) / (∂t) = − 1 / ρ ₀ · (∂P) / (∂r) (2) Therefore, Vr = −1 / ρ ₀ ∫ (∂P) / (∂r) dt ... (3)

Now, if two microphones are arranged in the r direction at intervals of △ r and the sound pressures P1 (t) and P2 (t) are measured, when △ r is small, (∂P) / (∂r) ｛P2 ( t) −P1 (t)｝ / Δr (4) If the sound pressure P is an average value of P1 (t) and P2 (t), P = {P1 (t) + P2 (t)} / 2 (5) Therefore, the sound intensity value Ir is Becomes There are the following three methods for obtaining Ir in equation (6).

Configure the analog circuit and find it.

Determined using a digital filter.

Determined from the cross spectrum using a Fourier analyzer.

In this embodiment, the following method is used. That is,
Now, considering the sound of a single frequency, the average power of one cycle is given by Percival's theorem. Becomes Here, P (f) and Vr (f) are complex Fourier coefficients of P (t) and Vr (t), and Vr (f) ^* is a conjugate complex number of Vr (t).

(3) - (5) below, Vr (f) = {P1 (f) -P2 (f)} / jωρ 0 △ r ... (8) P (f) = {P1 (f) -P2 (f) ｝ / 2 (9) is obtained. Than this, Ir = 2Re [[{P1 (f) + P2 ( f)} / 2 ] · ^{[{P1 (f) * + P2} (f) *} -jωρ 0 Δr ]] = Re [{| P1 | 2 - ^{| P2 | 2 + P1 * P2} -P1P2 *} / - jωρ 0 △ r] = Im {P1 · P2 *} / πfρ 0 △ r ... is (10). Im ｛P1 · P2 ^* ｝ means the imaginary part of the cross spectrum of P1 · P2.

When several periodic signals are included in a certain frequency band, the sound intensity value of that band is the sum of the individual frequency components. Therefore, considering the case where the signal is a random signal, the sound intensity value Ir in the frequency band of f ₁ ≦ f ≦ f ₂ is Becomes Here, P1 and P2 are cross spectrum densities. As can be seen from the above description, the sound intensity value Ir can be finally obtained by the equation (11).

In this embodiment, the output signal level of the microphone 21 is constantly monitored, and when this level exceeds a certain value, it is determined that an abnormality has occurred, and the output signals of the microphones 21 and 22 are fetched. Signal processing is performed according to the flow shown in the figure. That is, when the output signal level of the microphone 21 exceeds a certain value, a start command is transmitted. As a result, the A / D converter in the input / output device 31 starts operating and converts the output signals of the microphones 21 and 22 into digital signals, respectively (step S1). Each converted signal
D21 and D22 are separated into respective frequency components by a frequency analyzer (steps S2 and S3). Next, the microphone 21
Complex conjugate processing (step
After performing S4), both signals are introduced into a multiplier and multiplied for each frequency component (step S5). The imaginary part is extracted from the signal (step S6), and the imaginary part signal is divided by a constant (step S7). That is, in step S6, the signal component serving as the numerator in equation (10) is obtained, and in step S7, the signal component is divided by the constant component forming the denominator in equation (10). Next, addition in a frequency band (step S8) is performed to calculate a sound intensity value Ir in that direction, which is represented by Expression (11). After the acoustic intensity value Ir ₁ in that direction that this manner was calculated has been stored in the memory (step S9), and the rotary table 15
Send a signal to rotate 30 degrees. The rotation angle of the turntable 15 is detected by a rotation angle detector 17. Therefore, the rotation of the turntable 15 is accurately controlled by 30 degrees. When the rotary table 15 is rotated 30 degrees, again performs processing from step S1 to step S9, calculates the sound intensity Ir ₂ in that direction, and stores it in memory.
Thus, to calculate the acoustic intensity value Ir ₁ ~Ir ₆ in each direction over the total of six times to 150 degrees.

Then, select the maximum value from among six stored in the memory of the sound intensity values Ir ₁ ~ ₆ (step S10), and
The angle value at which the maximum value is obtained is provided to the input / output device 31 and to the image receiving device 33 (step S11). When a sound source direction angle value is given, the input / output device 31 controls the rotation angle of the turntable 15 so that the direction of the line 24 becomes the angle direction. Therefore, the optical axis of the camera 18 is automatically directed toward the sound source. In addition, the image receiving device 33 displays the direction of the sound source as a vector on a part of the CRT display surface (step S1).
2). The rest of the display surface of the image receiving device 33 includes a camera 1
The image captured in step 8, ie, the image in the direction of the sound source, the result of the image processing, and the presence / absence of an abnormality are displayed. The sound obtained by the microphone 21 is also output. Therefore, the sound source direction can be recognized, and the current situation of the device serving as the sound source can be recognized from the image information.

In this case, when the sound source direction is obtained, the optical axis of the camera 18 can be automatically directed in that direction. This is abnormal because, for example, a camera incorporating a wide-angle lens is mounted in a fixed state, and image information with a higher resolution can be obtained as compared with a case where the entire part where the equipment is installed is photographed with this camera. It is possible to facilitate the process of determining whether or not the determination is made.

In the embodiment described above, the optical axis direction of the camera 18 is moved by 30 degrees. However, if the camera 18 is moved at a finer angle and the sound intensity value is calculated for each angle, the sound source direction can be changed. Exploration accuracy can be improved.

In the above embodiment, the direction of the optical axis of the camera 18 is moved by a certain angle, and the sound intensity value is calculated for each angle, and the angle direction at which the maximum sound intensity value is obtained from these is determined as the sound source direction. However, the sound intensity direction calculated at each angle may be vector-wise combined to determine the sound source direction.

FIG. 3 shows the flow of signal processing in an example in which the sound intensity directions are determined by vectorwise combining the sound intensity values. That is, the steps up to step S9 are the same as those in the above embodiment. In step S10, the obtained 6
Vector synthesis of two sound intensity values. Now, consider the orthogonal coordinates drawn on the plane orthogonal to the rotation center axis 23 of the rotary table 15 with the rotation center axis 23 as the origin,
When it is on the axis, it is defined as 0 degree. The combined value of the X direction component of the turntable 15 the six obtained by rotating by 30 degrees counterclockwise sound intensity values Ir ₁ ~Ir ₆ and Ix, when the Iy the combined value of the Y direction component, Ix = Ir ₁ cos θ ₁ + Ir ₂ cos θ ₂ +... Ir ₆ cos θ ₆ Iy = Ir ₁ sin θ ₁ + Ir ₂ sin θ ₂ +... Ir ₆ sin θ ₆ . However, θ ₁ = 0 °, θ ₂ = 30 °,..., Θ ₆ =
150 ゜. The direction of the vector I obtained by combining Ix and Iy is the direction of the sound source as shown in FIG. The angle θ ₀ between the vector I and the X axis is calculated, and this angle information is sent to the input / output device 31 and also sent to the image receiving device 33 (step
S11).

In this way, in the method of obtaining the sound source direction by vector synthesis of the sound intensity values calculated at each angle,
Data in each direction can be averaged compared to the method in which the maximum value direction is set as the sound source direction, and the disturbance component can be removed, so that the accuracy of the sound source search can be improved.

In each of the embodiments described above, a sound receiver is constructed by arranging two microphones on the same axis, and the sound receiver is rotated about one point to search for a two-dimensional sound source. Although an image in that direction is obtained, a three-dimensional sound source search and image information in that direction can also be obtained by changing the arrangement of the microphones, the mechanism for moving the position of the camera, and the signal processing means.

FIG. 5 shows an embodiment of a sound source searching apparatus capable of searching for a sound source in three dimensions and obtaining image information in the direction. In this figure, the same parts as those in FIG. 1 are indicated by the same reference numerals.

In the sound source detection device according to this embodiment, the configuration of the detection unit 1a is greatly different. That is, a rectangular coordinate composed of the X axis, the Y axis, and the Z axis is drawn on the power supply box 11, and the four microphones 41 are arranged at equal intervals on the X axis, the Y axis, and the Z axis, respectively. , 42, 43, and 44 are arranged to constitute the sound receiver 20a. Note that the sound receiver 20a is fixed on the power supply box 11 via the support 45. On the other hand, an autofocus camera 18 is arranged between the sound receiver 20a and the power supply box 11, and the camera 18 is mounted on the rotary table 15 on the X axis and the Y axis.
An axis that is parallel to the plane connecting the axes and that is arranged rotatably
Supported by 46. On the turntable 15, there are provided an alternator 47 for selectively applying rotational power to the shaft 46 and a rotational angle detector 48 for detecting the rotational angle of the shaft 46. The rotary table 15 is fixed to an upper end of a shaft 49 parallel to the Z axis, and the shaft 49 is rotatably supported by a bearing 50 provided on the power supply box 11. On the power supply box 11, a motor 51 for selectively applying rotational power to the shaft 49 and a rotation angle detector 52 for detecting the rotation angle of the shaft 49 are arranged. Accordingly, by controlling the drive of the motor 51, the rotary table 15 can be rotated in the direction indicated by the solid arrow 53 in the figure, that is, around the Z axis. By controlling the drive of the motor 47, the camera 18 can be rotated by the solid arrow in the figure. It can be rotated in the direction indicated by 54. Then, the microphones 41 to 44, the camera 18, the motors 47, 51,
The rotation angle detectors 48, 52 are connected to an input / output device 56, a processing device 57, and an image receiving device 58 constituting the signal processing unit 3a via a cable 55 as in the above-described embodiment.

The timing for transmitting a start command in the processing device 57 to detect the sound source direction and the flow of signal processing basically employ the method shown in FIG. However,
In the case of this embodiment, two microphones are provided on the X-axis, Y-axis, and Z-axis respectively. Specifically, as shown in FIG. 6, two microphones arranged on the X-axis are provided. The sound intensity value Irx in the X direction is calculated using the outputs of the microphones 41 and 42, and the sound intensity value Iry in the Y direction is calculated using the outputs of the two microphones 41 and 43 arranged on the Y axis. ,
Also, the sound intensity value Irz in the Z direction is calculated using the outputs of the two microphones 41 and 44 arranged on the Z axis. The three sound intensity values obtained in this manner are combined vectorwise, and the direction of the combined vector is calculated (step S13). Then, the obtained angle value in the vector direction is given to the input / output device 56 and the image receiving device 58
(Step S15). The input / output device 56 controls the motor 51 based on the given angle signal to
, And further controls the motor 47 to control the elevation angle of the camera 18. Therefore, the optical axis of the camera 18 automatically points toward the sound source.

In this way, the sound intensity values in three directions are simultaneously obtained, and the sound source direction is obtained by vector-wise synthesis of these values. In addition to the need to rotate the sound receiver, the sound source direction can be determined with high accuracy. It is possible to search, and even in the case of a single sound, the direction of the sound can be specified.

In addition, the normal video signal in each direction captured by the camera is stored, and the video signal obtained when the optical axis of the camera is directed to the sound source is compared with the stored video signal in that direction. Alternatively, an alarm may be generated when a difference of a certain level or more occurs.

[Effect of the Invention] In the sound source locating device according to the present invention, it is possible to specify the sound source direction and obtain clear image information of the sound source direction at the same time. A determination can be made. In the sound source detection device according to the present invention, the current atmosphere of the equipment installation site is compared with the atmosphere in a steady state, and when it is different from the steady state, an operation for obtaining the intensity in each direction is started. Therefore, when comparing the atmosphere at the present time with the atmosphere in the steady state, it is possible to set a margin and perform the comparison. Therefore, it is possible to easily cope with equipment installation places where the steady atmosphere fluctuates within a certain range, and eventually, it is possible to perform abnormality diagnosis of equipment in all equipment installation places.

[Brief description of the drawings]

FIG. 1 is an external view of a sound source detection apparatus according to one embodiment, FIG. 2 is a view for explaining the flow of signal processing in a processing apparatus in the apparatus, and FIG. 3 is another example of signal processing. FIG. 4 is a diagram for explaining the signal processing, and FIG.
FIG. 5 is an external view of a sound source detection device according to another embodiment, and FIG.
The figure is a diagram for explaining the flow of signal processing in the processing device in the same device. 1,1a Detector, 3,3a Signal processor, 20,20a Sound receiver, 11 Power supply box, 15 Rotary table, 18 Camera, 21,22,41,42 , 43,44 …… Microphone, 16,47,51…
… Motor, 31, 56 …… I / O device, 32,57 …… Processor, 33,
58 ... Image receiving device.

Claims (5)

(57) [Claims] 1. A sound receiver, which is used for abnormality diagnosis of equipment constituting a plant, wherein a plurality of microphones are arranged in a fixed relationship in a place where the equipment is installed, and A means for comparing the current atmosphere in the installation space of the equipment with the atmosphere in a steady state, and sending out an output when the atmosphere is different from the steady state; Means for determining the sound intensity in each direction centering on the position of the sound receiver from the output signals of the respective microphones constituting the microphone; and determining the sound intensity based on the position of the sound receiver from the sound intensity obtained by the means. Means for determining the direction of the sound source obtained, a camera disposed in the vicinity of the sound receiver, means for directing the optical axis of the camera in the direction of the sound source determined by the means, and at least the camera captured by the camera. Sound source search apparatus characterized by comprising; and a Projects receiving device image. 2. The sound source search according to claim 1, wherein the image receiving device simultaneously displays an image captured by the camera and a signal obtained by converting an acoustic signal obtained by the sound receiver into a displayable state. apparatus. And a driving device for changing the direction of the sound receiver, wherein the drive is stopped every time the direction of the sound receiver changes by a predetermined angle, and a sound intensity is obtained from the microphone output signal. The sound source search device according to claim 1, wherein the direction of the sound intensity vector that indicates the maximum of the sound intensity vectors obtained for each stop angle of the sound source is the sound source direction. 4. A driving device for changing the direction of the sound receiver, wherein the driving device is stopped every time the direction of the sound receiver changes by a predetermined angle, and a sound intensity is obtained from an output signal of the microphone. The sound source search device according to claim 1, wherein the direction of the sound intensity vector obtained as a result of adding the sound intensity vectors obtained for each stop angle of the device in a vector manner is a sound source direction. 5. The optical axis of the camera is directed to the direction of the sound source with the result of three-dimensional vector synthesis of the respective sound intensities obtained by arranging four or more microphones three-dimensionally. A sound source detection device according to claim 1.