Noise calculation device and sound collection system

Abstract

Th present invention addresses the problem of providing a noise calculation device and a sound collection system that enable a target sound included in ambient sound to be distinguished from noise. The noise calculation device 20 calculates the acoustic pressure level of noise included in the ambient sound around an analysis target 40 that includes a target sound emitted by the analysis target 40. The noise calculation device 20 uses a physical model, which defines the relationship between the distance to a microphone 209 from the analysis target 40 and the reference acoustic pressure level of the target sound, to specify the reference acoustic pressure level of the target sound corresponding to the distance. Furthermore, the noise calculation device 20 calculates an estimated acoustic pressure level of the noise included in the ambient sound, on the basis of a detected acoustic pressure level of the ambient sound detected by the microphone 209, and the reference acoustic pressure level of the target sound specified using the physical model.

Description

[Technical Field]

[0001] The present invention relates to a noise calculation device that calculates the sound pressure level of noise contained in ambient sound around an analysis target, including sound emitted by the analysis target (target sound), and a sound collection system that collects or records the target sound. [Background technology]

[0002] Conventionally, sound collection systems have been known that collect and analyze sounds (target sounds) emitted by an analysis target. In the sound collection system, the target sound of the analysis target is compared with normal sounds, and based on the comparison results, it is possible to analyze abnormalities in the analysis target.

[0003] Patent Document 1 aims to provide a diagnostic device, diagnostic system, and program that can acquire sound at an appropriate position according to the characteristics of each model of device that acquires the sound when acquiring the generated sound and analyzing the cause of the sound (

[0007] , Abstract).

[0004] To achieve this object, a movement distance measurement unit (38) in Patent Document 1 (Abstract) measures the movement distance of the device from a reference position based on acceleration information obtained by an acceleration sensor. An attitude change amount measurement unit (39) measures the amount of attitude change of the device from a reference state based on angular velocity information obtained by a gyro sensor. A control unit (33) calculates the movement distance and attitude change of the device until the device reaches an appropriate position where the sound signal acquired by the sound acquisition unit (31) has a magnitude appropriate for acquisition, using the movement distance measurement unit (38) and the attitude change amount measurement unit (39), and generates this as appropriate position information. Then, when attempting to acquire a sound signal using the sound acquisition unit (31), if appropriate position information has already been generated, the control unit (33) guides the device to the appropriate position using the generated appropriate position information and the measurement results by the movement distance measurement unit (38) and the attitude change amount measurement unit (39). [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2017-111293 Summary of the Invention [Problem to be solved by the invention]

[0006] As described above, in Patent Document 1, the device is guided to an appropriate position so that the sound signal acquired by the sound acquisition unit (31) has a size appropriate for acquisition (Summary). In other words, in Patent Document 1, the position of the device is adjusted so that the sound pressure level of the sound falls within a certain range.

[0007] However, for example, at a construction site, in addition to the sound emitted by the object to be analyzed (target sound), noise such as sounds emitted by other machines on the site and surrounding work noise is generated. In other words, the sound detected around the object to be analyzed (ambient sound) includes noise in addition to the target sound. If an attempt is made to use the technology of Patent Document 1 in such an environment where the ambient sound includes noise, the position of the device itself will have to be adjusted based on the sound pressure levels of the target sound and the ambient sound including the noise, making it difficult to accurately analyze the target sound, which will ultimately affect the accuracy of determinations such as anomaly analysis.

[0008] The present invention has been made in consideration of the above-mentioned problems, and aims to provide a noise calculation device and a sound collection system that can distinguish between target sounds contained in ambient sounds and noise. [Means for solving the problem]

[0009] In order to achieve the above object, a noise calculation device according to the present invention calculates a sound pressure level of noise contained in ambient sound of an analysis target, including a target sound that is a sound emitted by the analysis target, The noise calculation device a storage device for storing a program; a computing device configured to execute the program; The computing device specifying a reference sound pressure level of the target sound corresponding to the distance using a physical model that defines the relationship between the distance from the analysis target to an ambient sound detection device that collects the ambient sound of the analysis target and the reference sound pressure level of the target sound; a noise calculation unit that calculates an estimated sound pressure level of the noise contained in the ambient sound based on the detected sound pressure level of the ambient sound detected by the ambient sound detection device and the reference sound pressure level of the target sound identified using the physical model; It is characterized by:

[0010] The sound collection system according to the present invention collects target sounds emitted by an analysis target, and the sound collection system includes a server device and a mobile terminal connected to the server device; The mobile terminal a storage device for storing a program; a computing device configured to execute the program; The computing device of the mobile terminal an ambient sound detection unit that measures a distance to the analysis target and detects a sound pressure level of ambient sound generated around the analysis target; The server device a storage device for storing a program; a central processing unit configured to execute the program; a central processing unit of the server device; a physical model storage unit that stores a physical model that defines the relationship between the distance and the reference sound pressure level of the target sound; a noise calculation unit that calculates an estimated sound pressure level of noise included in the ambient sound based on the detected sound pressure level of the ambient sound and a reference sound pressure level of the target sound that corresponds to the distance and is identified using the physical model; The present invention is characterized by having the following. [Effects of the Invention]

[0011] According to the present invention, it is possible to distinguish between target sounds contained in ambient sounds and noise, and to collect sounds that can be used for abnormality analysis, etc. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a diagram showing the overall configuration of a sound collection system according to a first embodiment of the present invention in a simplified manner; [Figure 2A] FIG. 2 is a diagram showing an example of the hardware configuration of a mobile terminal according to the first embodiment. [Figure 2B] FIG. 2 is a diagram showing an example of the functional configuration of the mobile terminal according to the first embodiment. [Figure 3A] FIG. 2 is a diagram showing an example of the hardware configuration of a server device according to the first embodiment. [Figure 3B] FIG. 2 is a diagram showing an example of the functional configuration of the server device according to the first embodiment. [Figure 4] 5 is a flowchart showing a process executed by the mobile terminal of the first embodiment. [Figure 5] 10 is a flowchart for guiding the mobile terminal to an appropriate recording position and posture in the first embodiment (details of S105 in FIG. 4). [Figure 6] 6(A) to 6(C) are diagrams showing examples of first to third screens used in the flowchart of FIG. [Figure 7] Figure 7(A) is a diagram showing the relationship between the machine to be diagnosed and measurement points P1 to P3 in the first embodiment, and Figure 7(B) is a diagram showing the relationship between the distance between the machine to be diagnosed and each of the measurement points P1 to P3 in Figure 7(A) and the sound pressure level of the target sound. [Figure 8] 5 is a flowchart showing a process for calculating noise in the first embodiment (details of S106 in FIG. 4). [Figure 9] FIG. 10 is a diagram showing the overall configuration of a sound collection system according to a second embodiment of the present invention. [Figure 10] FIG. 10 is a diagram showing an example of the functional configuration of a mobile terminal according to a second embodiment. [Figure 11] 10 is a flowchart showing a process executed by the mobile terminal according to the second embodiment. [Embodiment for Carrying Out the Invention]

[0013] <A. First Embodiment> [A-1. Configuration] (A-1-1. Overall Configuration) FIG. 1 is a configuration diagram schematically showing the overall configuration of a sound collection system 10 according to the first embodiment of the present invention. The sound collection system 10 includes a mobile terminal 20 and a server device 30. The mobile terminal 20 collects or records sounds (ambient sounds) generated around a diagnostic target machine 40. The diagnostic target machine 40 can be, for example, a construction machine (such as an excavator, a wheel loader, a bulldozer, etc.). The ambient sounds include noises such as the operating sounds of other construction machines generated from outside the diagnostic target machine 40, the surrounding work sounds, and the environmental sounds, in addition to the sounds (target sounds) generated by the diagnostic target machine 40 itself.

[0014] The mobile terminal 20 determines the S / N ratio (signal-noise ratio) of the target sound and the noise included in the ambient sound, and determines whether the ambient sound is suitable for the analysis of the target sound. When determining the S / N ratio, various information (described later) is acquired from the diagnostic target machine 40 via a serial Wi-Fi converter 401 of the diagnostic target machine 40. Therefore, it is also possible to recognize the diagnostic target machine 40 as part of the sound collection system 10.

[0015] When the ambient sound is suitable for the analysis of the target sound, the mobile terminal 20 collects or records the ambient sound. Thereafter, the mobile terminal 20 transmits data of the ambient sound (ambient sound data) etc. to the server device 30 via a communication network 50 such as the Internet or a LAN (Local Area Network). The server device 30 analyzes the received ambient sound data and performs a diagnosis (analysis) of the diagnostic target machine 40. The diagnosis (analysis) here includes, for example, an abnormality diagnosis of the diagnostic target machine 40. In the mobile terminal 20, since the ambient sound is collected or recorded when the ambient sound is suitable for the analysis of the target sound, the server device 30 can extract a high-quality target sound, improve the analysis accuracy, and perform a highly accurate abnormality determination.

[0016] In the first embodiment, the target sound is assumed to be operating sound generated from the entire machine 40 to be diagnosed, but it may also be operating sound generated from a specific part (e.g., a pump or an injector) of the machine 40 to be diagnosed (this point will be described later). Furthermore, noise refers to sounds other than the target sound, including environmental sounds such as wind and human voices.

[0017] (A-1-2. Mobile terminal 20) (A-1-2-1. Overview) In the first embodiment, the mobile terminal 20 functions as a noise calculation device that calculates the sound pressure level of noise contained in the ambient sound of the diagnosis target machine 40, and also functions as an ambient sound detection device that collects or records the ambient sound. The mobile terminal 20 is a portable terminal such as a smartphone, a tablet personal computer (tablet PC), or a notebook PC.

[0018] (A-1-2-2. Hardware configuration of the mobile terminal 20) 2A is a diagram showing an example of the hardware configuration of the mobile terminal 20 in the first embodiment. As shown in FIG. 2A, the mobile terminal 20 has a processing unit (CPU) 201, a memory 202, a storage device 203, a communication interface (I / F) 204, an acceleration sensor 205, a gyro sensor 206, a geomagnetic sensor 207, a camera 208, a microphone 209, an input device 210, and a display device 211.

[0019] The CPU 201 is a computer that executes various programs installed in the storage device 203. The memory 202 functions as a main storage unit that stores various programs, data, etc. required for the CPU 201 to execute the various programs stored in the storage device 203. Specifically, the memory 202 stores boot programs such as a BIOS (Basic Input / Output System) and an EFI (Extensible Firmware Interface).

[0020] The storage device 203 stores various programs and data. Each function of the mobile terminal 20, which will be described later, is realized by the CPU 201 loading the program stored in the storage device 203 into the memory 202 and executing it. Furthermore, various data required to realize the functions, such as data held in a physical model storage unit 241 (FIG. 2B), which will be described later, and data generated during processing, are stored in the storage device 203, etc.

[0021] The acceleration sensor 205 detects acceleration in three axial directions of the mobile terminal 20. The gyro sensor 206 detects the rotation or change in orientation of the mobile terminal 20 as an angular velocity using Coriolis force, thereby detecting the attitude (orientation) of the mobile terminal 20. The geomagnetic sensor 207 detects the attitude (orientation) of the mobile terminal 20 using geomagnetism. The camera 208 is a color camera, and is used to capture images of the machine 40 to be diagnosed. The acceleration sensor 205, the gyro sensor 206, the geomagnetic sensor 207, and the camera 208 function as means for acquiring position information and attitude information of the mobile terminal 20.

[0022] The microphone 209 functions as a means for detecting ambient sounds of the diagnosis target machine 40. The input device 210 accepts input from the user. The display device 211 displays guidance instructions for the mobile terminal 20 to the user of the mobile terminal 20.

[0023] (A-1-2-3. Functional configuration of the mobile terminal 20) Fig. 2B is a diagram showing an example of the functional configuration of the mobile terminal 20 in the first embodiment. Each function of the mobile terminal 20 shown in Fig. 2B is realized by, for example, executing a program stored in the storage device 203 by the CPU 201. In other words, each function shown in Fig. 2B is configured as a function of the arithmetic unit of the mobile terminal 20. The CPU 201 (arithmetic unit) of the mobile terminal 20 includes a communication unit 231, an ambient sound detection unit 232, a sound recording control unit 233, an object information acquisition unit 234, a measurement condition acquisition unit 235, a display unit 236, a distance measurement unit 237, an attitude measurement unit 238, a noise calculation unit 239, a position / attitude guidance unit 240, and a physical model storage unit 241.

[0024] The communication unit 231 transmits and receives signals to and from external devices via the communication network 50 (FIG. 1). In the first embodiment, the communication unit 231 transmits and receives signals to and from the server device 30 and the diagnosis target machine 40. The communication unit 231 is realized via the communication I / F 204. The ambient sound detection unit 232 detects ambient sound generated around the diagnosis target machine 40 and generates an ambient sound signal. The ambient sound detection unit 232 is realized via the microphone 209.

[0025] The recording control unit 233 controls the overall collection or recording of ambient sounds. The target information acquisition unit 234 acquires information (target information) of the diagnosis target machine 40 required for collecting or recording ambient sounds from the diagnosis target machine 40. The measurement condition acquisition unit 235 acquires measurement conditions from the server device 30 via the communication unit 231. Details of the measurement conditions will be described later with reference to FIG. 4.

[0026] The display unit 236 displays various data on the display device 211 based on control by the sound recording control unit 233. The distance measurement unit 237 measures the distance between the diagnosis target machine 40 and the mobile terminal 20 (microphone 209) from information obtained from the acceleration sensor 205, gyro sensor 206, geomagnetic sensor 207, and camera 208 (details will be described later with reference to FIG. 5). The attitude measurement unit 238 measures the attitude of the mobile terminal 20 and the diagnosis target machine 40 from information obtained from the acceleration sensor 205, gyro sensor 206, geomagnetic sensor 207, and camera 208 (details will be described later with reference to FIG. 5).

[0027] The noise calculation unit 239 evaluates noise contained in the ambient sound detected by the microphone 209. Specifically, the noise calculation unit 239 calculates an estimate of the sound pressure level of the noise (estimated sound pressure level) based on the sound pressure level of the ambient sound detected by the microphone 209 (detected sound pressure level), etc. The position / posture guidance unit 240 generates information for guiding the position and posture of the mobile terminal 20 (microphone 209) for the user. The physical model storage unit 241 stores a physical model, which will be described later with reference to FIGS. 7(A) and 7(B).

[0028] (A-1-3. Server device 30) The server device 30 (FIG. 1) of the first embodiment provides measurement conditions to the mobile terminal 20, and diagnoses (analyzes) the diagnosis target machine 40 based on the ambient sound data received from the mobile terminal 20, etc.

[0029] 3A is a diagram showing an example of the hardware configuration of the server device 30 in the first embodiment. As shown in FIG. 3A, the server device 30 has a central processing unit (CPU) 311, a memory 312, and a storage device 313.

[0030] The central processing unit 311 is a computer that executes various programs installed in the storage device 313. The memory 312 functions as a main storage unit that stores various programs, data, etc. that are necessary for the central processing unit 311 to execute the various programs stored in the storage device 313. The storage device 313 stores various programs and data.

[0031] Fig. 3B is a diagram showing an example of the functional configuration of the server device 30 in the first embodiment. Each function of the server device 30 shown in Fig. 3B is realized by, for example, executing a program stored in the storage device 313 in the central processing unit 311. In other words, each function shown in Fig. 3B is configured as a function of the central processing unit 311 of the server device 30.

[0032] 3B, server device 30 has a data collection unit 301, a data analysis unit 302, a measurement condition storage unit 303, and a data storage unit 304. A data collection program and a data analysis program are installed in the storage device of server device 30. Server device 30 functions as data collection unit 301 and data analysis unit 302 by executing these programs.

[0033] The data collection unit 301 transmits the measurement conditions extracted from the measurement condition storage unit 303 to the mobile terminal 20. The data collection unit 301 also receives ambient sound data, target information, distance / posture data, and noise data transmitted from the mobile terminal 20, and stores them in the data storage unit 304.

[0034] The data analysis unit 302 performs analysis based on the sound data, machine information data, distance / posture data, and noise data stored in the data storage unit 304. For example, the data analysis unit 302 compares the data with a previously prepared ambient sound model for normal conditions and calculates the degree of similarity to determine whether the data is normal or abnormal. The data analysis unit 302 may also display the analysis results via the display unit 236 of the mobile terminal 20. This allows the analysis results to be checked on the spot.

[0035] The server device 30 can collect ambient sound data measured under conditions that satisfy the measurement conditions, extract high-quality target sounds, improve the accuracy of analysis in the server device 30, and perform highly accurate abnormality determinations. Furthermore, the server device 30 uses the target information, distance / posture data, and noise data collected in association with the ambient sound data to understand the circumstances under which the ambient sound data was acquired, for example, when the analysis results differ from reality, and the analyst can obtain useful insights into the cause.

[0036] The measurement condition storage unit 303 stores the measurement conditions. The data storage unit 304 stores various data such as ambient sound data.

[0037] (A-1-4. Machine to be diagnosed 40) As described above, when collecting or recording the ambient sounds of the diagnosis target machine 40, the mobile terminal 20 acquires specific information from the diagnosis target machine 40. Therefore, the diagnosis target machine 40 can be considered as part of the sound collection system 10. Therefore, the configuration of the diagnosis target machine 40 will be explained below.

[0038] As described above, the machine 40 to be diagnosed is, for example, a construction machine such as a shovel, a wheel loader, or a bulldozer. The machine 40 to be diagnosed has a serial-to-Wi-Fi converter 401 that can communicate with the mobile terminal 20. Bluetooth (registered trademark), wireless LAN, or the like may be used instead of the serial-to-Wi-Fi converter 401. The machine 40 to be diagnosed is also provided with a control device (not shown). Information about the machine 40 to be diagnosed (target information) is sent from this control device to the mobile terminal 20.

[0039] [A-2. Control of the First Embodiment] (A-2-1. Overview) Next, various controls in the sound collection system 10 of the first embodiment will be described, focusing on the mobile terminal 20. As described above, in the sound collection system 10 of the first embodiment, a user operates the mobile terminal 20 to collect or record ambient sounds (target sounds) of the diagnosis target machine 40. The collected or recorded ambient sounds are then transmitted from the mobile terminal 20 to the server device 30 based on the user's operation. The server device 30 analyzes (diagnoses abnormalities) the diagnosis target machine 40 using the ambient sounds (target sounds). The user referred to here is assumed to be, for example, an employee of the company that owns the diagnosis target machine 40 or an employee of the manufacturer of the diagnosis target machine 40.

[0040] Furthermore, the user can carry the mobile terminal 20 and move it to any location. Therefore, after recording ambient sound in an offline environment, the user can move to an online environment and transmit the ambient sound data to the server device 30.

[0041] (A-2-2. Overall flow when recording ambient sounds) 4 is a flowchart showing the processing executed by the mobile terminal 20 of the first embodiment. When the mobile terminal 20 receives a command from the user on a menu screen (not shown) to start an abnormality diagnosis (analysis) of the diagnosis target machine 40, the mobile terminal 20 acquires measurement conditions from the server device 30 (step S101).

[0042] The measurement conditions here include three types: measurement conditions related to the diagnosis target machine 40, measurement conditions related to ambient sound, and measurement conditions related to noise. The measurement conditions related to the diagnosis target machine 40 include, for example, the engine speed [rpm] of the diagnosis target machine 40. Furthermore, the measurement conditions related to ambient sound include, for example, the distance and attitude between the diagnosis target machine 40 and the mobile terminal 20 (microphone 209). The measurement conditions related to noise include, for example, the S / N ratio of the target sound and noise.

[0043] When acquiring the measurement conditions from the server device 30, the mobile terminal 20 notifies the server device 30 of the diagnostic content selected by the user. Then, the server device 30 transmits the measurement conditions corresponding to the notified diagnostic content to the mobile terminal 20. In addition, the mobile terminal 20 may acquire the model as target information from the diagnosis target machine 40 and notify the server device 30. Then, the server device 30 may transmit the measurement conditions to the mobile terminal 20 in association with the combination of the diagnostic content and the model.

[0044] Next, the mobile terminal 20 determines whether or not the measurement conditions received from the server device 30 include a measurement condition related to the diagnosis target machine 40 (step S102 in FIG. 4). As described above, the measurement condition related to the diagnosis target machine 40 includes a condition related to the engine speed (for example, the engine speed being within a predetermined range (for example, 300 to 400 rpm)). If there is a measurement condition related to the diagnosis target machine 40 (S102: YES), the mobile terminal 20 performs processing to guide the user to satisfy the measurement condition (step S103).

[0045] For example, if the measurement conditions for the diagnosis target machine 40 include that the engine speed of the diagnosis target machine 40 be within a predetermined range (e.g., 300 to 400 rpm), the mobile terminal 20 (display unit 236) displays a message on the display device 211 requesting that the engine speed be within the predetermined range. The user sees this message and requests the driver of the diagnosis target machine 40 to operate the diagnosis target machine 40 and achieve the above-mentioned engine speed. The engine speed may be set to, for example, a value during idling, so that the driver does not need to operate the accelerator pedal. Alternatively, the mobile terminal 20 may communicate with the diagnosis target machine 40 and automatically control the diagnosis target machine 40 so that the engine speed is within the above-mentioned predetermined range.

[0046] If there are no measurement conditions related to the diagnosis target machine 40 (S102: NO) or after step S103, the process proceeds to step S104. In step S104, the mobile terminal 20 determines whether or not there is a measurement condition related to the ambient sound among the measurement conditions received from the server device 30. As described above, measurement conditions related to the ambient sound include, for example, the distance and attitude between the diagnosis target machine 40 and the mobile terminal 20. For example, if there is a measurement condition related to the ambient sound, such as a distance of 2 m between the diagnosis target machine 40 and the mobile terminal 20 and an attitude in which the mobile terminal 20 is diagonally forward of the diagnosis target machine 40 (S104: YES), the mobile terminal 20 performs a process of guiding the user to satisfy the measurement condition (step S105). Details of the guidance will be described later with reference to FIG. 5 and FIGS. 6(A) to 6(C).

[0047] If there are no measurement conditions related to the ambient sound (S104: NO) or after step S105, the process proceeds to step S106. In step S106, the mobile terminal 20 evaluates the noise contained in the ambient sound. Specifically, the mobile terminal 20 calculates an estimated sound pressure level of the noise contained in the ambient sound (details will be described later with reference to FIGS. 7(A), 7(B), and 8).

[0048] Next, the mobile terminal 20 determines whether or not there is a measurement condition related to noise among the measurement conditions received from the server device 30 (step S107). As described above, an example of a measurement condition related to noise is the ratio between the sound pressure level of the target sound and the sound pressure level of the noise (S / N ratio). If there is a measurement condition related to noise (S107: YES), the mobile terminal 20 determines whether or not the noise satisfies the measurement condition (step S108). If there is no measurement condition related to noise (S107: NO) or if the noise satisfies the measurement condition (S108: YES), the mobile terminal 20 (ambient sound detection unit 232) starts recording the ambient sound (step S110). On the other hand, if the noise does not satisfy the measurement condition (S108: NO), the mobile terminal 20 does not record the ambient sound (step S109). Therefore, it is possible to actively collect higher quality ambient sound data.

[0049] (A-2-3. Guidance to proper position and posture (S105 in Figure 4)) FIG. 5 is a flowchart (details of S105 in FIG. 4) for guiding the mobile terminal 20 to an appropriate recording position and posture in the first embodiment. FIGS. 6(A) to 6(C) are diagrams showing examples of first to third screen displays used in the flowchart of FIG. 5. The mobile terminal 20 acquires appropriate position and posture information as measurement conditions related to ambient sound (step S201). Next, the mobile terminal 20 (sound recording control unit 233) measures the initial position and posture of the mobile terminal 20 using the distance measurement unit 237 and posture measurement unit 238 (step S202). Specifically, the mobile terminal 20 performs the following process.

[0050] First, in order to measure the initial position and attitude, the mobile terminal 20 (display unit 236) displays a screen 251 as shown in FIG. 6(A) and instructs the user to hold the camera 208 of the mobile terminal 20 over the diagnosis target machine 40. Then, after the user holds the camera 208 of the mobile terminal 20 over the diagnosis target machine 40, the distance measurement unit 237 and the attitude measurement unit 238 calculate the distance and attitude between the mobile terminal 20 and the diagnosis target machine 40. The calculation of the distance and attitude is performed by an algorithm for estimating the initial position and attitude based on information obtained from the acceleration sensor 205, the gyro sensor 206, the geomagnetic sensor 207, and the camera 208.

[0051] In the first embodiment, the attitude angle of the mobile terminal 20 is calculated based on the acceleration and geomagnetic values ​​obtained from the acceleration sensor 205 and the geomagnetic sensor 207, and the distance to the machine 40 to be diagnosed is calculated by triangulation using the elevation angle of the mobile terminal 20, thereby obtaining the distance and attitude.

[0052] Note that the method for measuring distance and orientation is not limited to this method, and it is also possible to use, for example, camera 208. That is, an image captured by camera 208 is used as an input, feature points are extracted, and matching is performed with feature points detected from previously captured images to obtain the amount of movement of the feature points. Then, the obtained amount of movement of the feature points may be used to estimate the camera position and camera orientation, thereby estimating the distance and orientation of mobile terminal 20. Furthermore, distance and orientation estimation using a plurality of methods may be combined.

[0053] Next, the mobile terminal 20 (distance measurement unit 237 and attitude measurement unit 238) tracks the amount of change in the position and attitude of the mobile terminal 20 from the initial position (step S203 in FIG. 5). The tracking is performed using an algorithm for self-position estimation based on information obtained from the acceleration sensor 205, gyro sensor 206, geomagnetic sensor 207, and camera 208.

[0054] In the first embodiment, an algorithm for self-location estimation is used that determines the direction and speed of movement by continuously capturing images with the camera 208 and comparing the positional difference of the same feature point in the two images. More specifically, for example, an algorithm based on a known method called SLAM (Simultaneous Localization and Mapping) is used as the algorithm.

[0055] In SLAM, the self-position of the mobile terminal 20 is sequentially estimated by the following process. That is, captured images are sequentially input from the camera 208, and a process of detecting feature points is sequentially executed for each input captured image. Then, feature points detected from a newly input captured image are compared with feature points detected from a previously input captured image, and the feature points are matched. In this matching, feature points whose difference in feature amount is equal to or less than a predetermined threshold are determined to be the same feature points. Then, based on the matching, the amount and direction of movement between the two times are estimated. Furthermore, the estimated values ​​of the amount and direction of movement and the estimated position and orientation of the mobile terminal 20 at the previous time are used to determine the current estimated position and orientation of the mobile terminal 20.

[0056] The algorithm for estimating the self-position is not limited to the above, and the self-position may be calculated by acquiring reference data in which a reference image prepared in advance is associated with the position and attitude of the mobile terminal 20 when the reference image was captured, and calculating the similarity between the captured image input from the camera 208 and the reference image. Alternatively, the moving distance from the initial position and the amount of attitude change from the initial attitude may be measured by accumulating the values ​​of the acceleration sensor 205 and the gyro sensor 206. Furthermore, estimation may be performed by combining a plurality of algorithms.

[0057] 5, the mobile terminal 20 (sound recording control unit 233) determines whether the current position and orientation of the mobile terminal 20 based on the result of step S203 has reached the appropriate position and orientation (step S204). If the appropriate position and orientation has not been reached (S204: NO), the mobile terminal 20 (sound recording control unit 233) displays guidance via the display unit 236 to move the mobile terminal 20 to the appropriate position and orientation (step S205).

[0058] FIG. 6(B) is an example of a screen 252 used in step S205, displaying a guidance instruction to move the mobile terminal 20 away from the diagnosis target machine 40. Here, the sound recording control unit 233 changes the display content depending on the current position of the mobile terminal 20 and the remaining distance to the appropriate position. That is, the sound recording control unit 233 displays the current position of the mobile terminal 20 and the appropriate position condition, and if the distance is too close, displays the message "Please move away from the target machine," as shown in FIG. 6(B), to urge the user to move away. On the other hand, if the distance is too far, displays the message "Please move closer to the target machine," to urge the user to move closer. In this way, by changing the display content depending on the remaining distance to the appropriate position, the user who is moving the mobile terminal 20 while looking at the display can intuitively and objectively grasp the remaining distance.

[0059] Similarly, regarding the attitude of the mobile terminal 20, text is displayed to instruct the user to rotate the mobile terminal 20. In addition to or instead of such a display, a scale line display format may be used, in which case an interface may be used in which the closer to the correct attitude the scale increases and the further away from the correct attitude the scale decreases.

[0060] When the current position and attitude of the mobile terminal 20 has reached the appropriate position and attitude (S204: YES in FIG. 5), the mobile terminal 20 (sound recording control unit 233) displays that the appropriate position and attitude has been reached (step S206). FIG. 6(C) is an example of a screen 253 used in step S206, which displays that the mobile terminal 20 has reached the appropriate position (and appropriate attitude). The screen 253 in FIG. 6(C) includes an evaluation start button 261 that starts noise evaluation. When the user presses the evaluation start button 261, the process proceeds from step S105 in FIG. 4 to step S106.

[0061] (A-2-4. Noise Calculation (S106 in Figure 4)) (A-2-4-1. Noise calculation method) As described above, in the noise calculation, an estimated sound pressure level of noise contained in the ambient sound (hereinafter also referred to as "noise sound pressure level") is calculated. The concept of the method for calculating the noise sound pressure level will be described below.

[0062] The ambient sound generated by the diagnosis target machine 40 includes sound (target sound) generated by the diagnosis target machine 40 itself and other noises. Therefore, in the first embodiment, a reference value (reference sound pressure level) of the sound pressure level of the target sound corresponding to the specifications (model) of the diagnosis target machine 40, the operating status of the diagnosis target machine 40 (engine speed, etc.), and the positions of the diagnosis target machine 40 and the mobile terminal 20 is pre-established as a physical model and stored in the physical model storage unit 241. Then, when actually collecting or recording sound, the sound pressure level of the ambient sound (detected sound pressure level) is acquired under predetermined conditions of the specifications of the diagnosis target machine 40, the operating status of the diagnosis target machine 40, and the positions of the diagnosis target machine 40 and the mobile terminal 20. By setting the diagnosis target machine 40 and the mobile terminal 20 in this predetermined state, the conditions for acquiring the sound pressure level of the ambient sound (detected sound pressure level) can be always constant. The value obtained by subtracting the reference sound pressure level of the target sound from the detected sound pressure level of the ambient sound can be considered as an estimate of the sound pressure level of the noise (estimated sound pressure level).

[0063] Fig. 7(A) is a diagram showing the relationship between the diagnosis target machine 40 (sound source) and measurement points P1 to P3 in the first embodiment, and Fig. 7(B) is a diagram showing the relationship between the distance between the diagnosis target machine 40 and each of the measurement points P1 to P3 in Fig. 7(A) and the sound pressure level of the target sound. The horizontal axis of Fig. 7(B) is the distance from the diagnosis target machine 40 (sound source) to the mobile terminal 20 (microphone 209), and the vertical axis is the detected sound pressure level of the target sound displayed logarithmically.

[0064] In Figure 7(B), the sound pressure level at each of the measurement points P1 to P3 is a value detected in a noise-free environment (i.e., the detected sound pressure level of the ambient sound = the detected sound pressure level of the target sound). Line L is calculated by linear regression (here, the least squares method) using each of the measurement points P1 to P3, and can be expressed as a linear function. As can be seen from Figure 7(B), the sound pressure level of the ambient sound (target sound) decreases as the distance increases. In other words, the slope of line L represents the attenuation coefficient. By using the linear function shown as line L, it is possible to calculate the reference sound pressure level of the target sound even at points other than the measurement points P1 to P3, as long as the distance is specified.

[0065] In the first embodiment, the reference sound pressure level of the target sound is calculated using the straight line L shown in Fig. 7(B) as a linear function. When an actual recording is performed, if there is a difference between the detected sound pressure level of the ambient sound and the reference sound pressure level of the target sound, the difference is treated as the estimated sound pressure level of the noise.

[0066] Note that when generating a physical model, the number of measurement points is not limited to three, but may be more. Furthermore, instead of detecting the sound pressure level at one measurement point, multiple detections may be performed and the average value may be used. Furthermore, the method of estimating a physical model from measurement points is not limited to linear regression, but may also be curvilinear regression.

[0067] The physical model is created to correspond to the model of the machine 40 to be diagnosed and the engine speed at the time of creating the physical model. For example, a physical model of a 20-ton vehicle weighing at idling (the minimum rotation speed of the internal combustion engine) is created.

[0068] (A-2-4-2. Specific processing of noise evaluation) 8 is a flowchart (details of S106 in FIG. 4) showing how noise is calculated in the first embodiment. As described above, in calculating noise, the difference between the detected sound pressure level of the ambient sound and the reference sound pressure level of the target sound is calculated as the estimated sound pressure level of the noise.

[0069] When calculating noise, the mobile terminal 20 (object information acquisition unit 234) acquires information for selecting a physical model (model selection information) from the diagnosis target machine 40 (step S301 in FIG. 8). The model selection information is information (object information) including specification information (model information) of the diagnosis target machine 40 and operation information (engine speed) of the diagnosis target machine 40.

[0070] Next, the mobile terminal 20 selects a physical model from the physical model storage unit 241 using the model selection information (target information) (step S302). As described above, in the first embodiment, a linear function (straight line L in FIG. 7(B)) that defines the relationship between the distance between the diagnosis target machine 40 and the mobile terminal 20 (microphone 209) and the reference sound pressure level of the target sound is used as the physical model. Therefore, a physical model corresponding to the model selection information (target information) is selected.

[0071] Next, the mobile terminal 20 acquires the ambient sound using the microphone 209 (step S303). The acquisition of the ambient sound here is for the purpose of calculating the sound pressure level of noise, so the sound is not recorded. However, the sound may be recorded for future evaluation.

[0072] Next, the mobile terminal 20 acquires the current distance (current distance) between the diagnosis target machine 40 and the mobile terminal 20 (step S304). The current distance is calculated in the same manner as in steps S202 and S203 of FIG.

[0073] Next, the mobile terminal 20 calculates the noise, that is, calculates the estimated sound pressure level of the noise (step S305). As described above, the estimated sound pressure level of the noise is obtained by subtracting the reference sound pressure level of the target sound from the detected sound pressure level of the ambient sound. Therefore, the mobile terminal 20 first obtains the reference sound pressure level of the target sound corresponding to the current distance acquired in step S304 for the physical model selected in step S302. Then, the mobile terminal 20 calculates the estimated sound pressure level of the noise by subtracting the reference sound pressure level of the target sound from the detected sound pressure level of the ambient sound acquired in step S303.

[0074] (A-2-5. Post-recording processing) When recording of the ambient sound (target sound) (S110 in FIG. 4) is completed, the mobile terminal 20 (recording control unit 233) transmits data of the recorded ambient sound (ambient sound data) together with attached information to the server device 30 at a predetermined timing. The predetermined timing here may be, for example, immediately after recording is completed. Alternatively, if the diagnosis target machine 40 and the mobile terminal 20 are located deep in the mountains and the communication network 50 is unavailable, the predetermined timing may be the timing when the mobile terminal 20 is subsequently moved and the communication network 50 becomes available. Alternatively, the predetermined timing may be the timing when the user of the mobile terminal 20 operates the mobile terminal 20 when the communication network 50 is available.

[0075] In addition, the ancillary information transmitted together with the ambient sound data may include, for example, any or all of the target information, distance / attitude data, and noise data. As described above, the target information includes the specifications (model) and status (engine speed) of the diagnosis target machine 40. The distance / attitude data is data indicating the distance and attitude between the diagnosis target machine 40 and the mobile terminal 20 (microphone 209). The noise data includes the estimated sound pressure level of noise contained in the ambient sound.

[0076] [A-3. Effects of the First Embodiment] As is clear from the above description, the sound collection system 10 in the first embodiment guides the position of the mobile terminal 20 when the mobile terminal 20 collects ambient sound data based on the measurement conditions transmitted from the server device 30. Thereafter, the noise calculation unit 239 calculates noise, determines the measurement conditions when measuring the ambient sound data, and allows only ambient sound data that is determined to be in an appropriate measurement condition to be recorded.

[0077] Only ambient sound data that is determined to be in an appropriate measurement situation is transmitted to the server device 30, making it possible to evaluate noise and then collect high-quality ambient sound data (target sound data) that meets the measurement conditions. This improves the accuracy of analyzing the target sound data, enabling highly accurate abnormality determination.

[0078] Furthermore, because the attached information is also transmitted in addition to the ambient sound data, the server device 30 can analyze the ambient sound data measured under appropriate measurement conditions based on the attached information, making it possible to perform the analysis without reducing diagnostic accuracy.

[0079] According to the first embodiment, an estimated sound pressure level of noise contained in the ambient sound is calculated based on the detected sound pressure level of the ambient sound occurring around the diagnosis target machine 40 (analysis target) and the reference sound pressure level of the target sound that corresponds to the distance from the diagnosis target machine 40 to the microphone 209 (ambient sound detection means) and is identified using a physical model (S106 in FIG. 4, FIG. 7(B), FIG. 8). This makes it possible to calculate the estimated sound pressure level of noise even in environments where ambient sound is likely to contain noise, such as construction sites, and enables processing according to the estimated sound pressure level of noise.

[0080] In the first embodiment, the physical model storage unit 241 (FIG. 2B) stores a plurality of physical models for each operating state of the diagnosis target machine 40 (analysis target). Furthermore, the noise calculation unit 239 calculates the estimated sound pressure level of noise using a physical model corresponding to the operating state (engine speed) of the diagnosis target machine 40 (S106 in FIG. 4, FIG. 8). This makes it possible to calculate the estimated sound pressure level of noise based on the operating state of the diagnosis target machine 40, thereby improving the calculation accuracy of the estimated sound pressure level of noise.

[0081] In the first embodiment, the physical model storage unit 241 (FIG. 2B) stores a plurality of physical models for each specification of the diagnosis target machine 40 (analysis target). Furthermore, the noise calculation unit 239 calculates the estimated sound pressure level of noise using a physical model according to the specifications of the diagnosis target machine 40 (S106 in FIG. 4, FIG. 8). This makes it possible to calculate the estimated sound pressure level of noise based on the specifications (model, etc.) of the diagnosis target machine 40, thereby improving the calculation accuracy of the estimated sound pressure level of noise.

[0082] In the first embodiment, the physical model is expressed as a linear function in which the distance between the diagnosis target machine 40 (analysis target) and the microphone 209 (ambient sound detection means) is a function of the reference sound pressure level of the target sound (FIG. 7(B)). This makes it possible to calculate the reference sound pressure level of the target sound using the physical model with a relatively small load.

[0083] In the first embodiment, the sound collection system 10 has a recording control unit 233 (sound collection control unit) that determines whether to collect ambient sound based on the ratio (S / N ratio) between the reference sound pressure level of the target sound and the estimated sound pressure level of noise (FIG. 2B). As a result, in environments where ambient sound is likely to contain noise, such as construction sites, collection of the target sound is stopped if the noise is excessively loud. In other words, collection of the target sound is limited to cases where the noise is not excessively loud. This makes it possible to collect high-quality target sound, improving the accuracy of analysis of the target sound and enabling highly accurate abnormality determination. As a result, it becomes possible to appropriately use the collected target sound (for example, to determine the state of the analysis target).

[0084] In the first embodiment, the sound collection system 10 includes a distance measurement unit 237 that measures the distance between the diagnostic target machine 40 (analysis target) and the microphone 209 (ambient sound detection means), an attitude measurement unit 238 that measures the attitudes of the diagnostic target machine 40 and the microphone 209, and a position / attitude guidance unit 240 (guidance means) that determines the measurement status of the ambient sound based on the distance and attitude and guides the position or attitude of the microphone 209 according to the measurement status (FIG. 2B). Thereby, it becomes possible to more suitably adjust the position or attitude of the microphone 209.

[0085] <B. Second Embodiment> [B-1. Configuration (Differences from the First Embodiment)] FIG. 9 is a configuration diagram schematically showing the overall configuration of the sound collection system 10A according to the second embodiment of the present invention. The sound collection system 10 of the first embodiment had the mobile terminal 20 and the server device 30 (FIG. 1). In contrast, the sound collection system 10A of the second embodiment does not have the server device 30 and has only the mobile terminal 20a (FIG. 9). In other words, in the second embodiment, the mobile terminal 20a plays the role of the server device 30. Hereinafter, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed descriptions thereof are omitted. Note that the hardware configuration of the mobile terminal 20a is the same as that of the mobile terminal 20 shown in FIG. 2A.

[0086] FIG. 10 is a diagram showing an example of the functional configuration of the mobile terminal 20a in the second embodiment. Each function of the mobile terminal 20a shown in FIG. 10 is realized by executing a program stored in the storage device 203 shown in FIG. 2A. That is, each function shown in FIG. 10 is configured as a function of the CPU 201 (arithmetic unit) of the mobile terminal 20a. The arithmetic unit of the mobile terminal 20a includes a communication unit 231, an ambient sound detection unit 232, a recording control unit 233, a target information acquisition unit 234, a measurement condition acquisition unit 235, a display unit 236, a distance measurement unit 237, an attitude measurement unit 238, a noise calculation unit 239, a position / attitude guidance unit 240, a physical model storage unit 241, a data analysis unit 242, a measurement condition storage unit 243, and a data storage unit 244.

[0087] The data analysis unit 242, measurement condition storage unit 243, and data storage unit 244 of the mobile terminal 20a have the same configuration as the data analysis unit 302, measurement condition storage unit 303, and data storage unit 304 of the server device 30 in the first embodiment.

[0088] [B-2. Control of the Second Embodiment] The control of the mobile terminal 20a in the second embodiment is basically the same as the control of the mobile terminal 20 in the first embodiment. However, as described above, the mobile terminal 20a in the second embodiment is different from the mobile terminal 20 in the first embodiment in that it also performs the same functions as the server device 30 in the first embodiment.

[0089] FIG. 11 is a flowchart showing the processing performed by the mobile terminal 20a in the second embodiment. Steps S401, S402, S403, S404, S405, S406, S407, S408, S409, and S410 in FIG. 11 are basically the same as steps S101, S102, S103, S104, S105, S106, S107, S108, S109, and S110 in FIG. 4. However, in the second embodiment, there is no server device like the server device 30 in the first embodiment. Therefore, the acquisition of measurement conditions in step S401 is performed inside the mobile terminal 20a.

[0090] Also, in step S411, the mobile terminal 20a (data analysis unit 242) performs data analysis based on the recorded ambient sound and attached information.

[0091] [[ID=IS]] [B-3. Effects of the Second Embodiment] According to the second embodiment as described above, in addition to or instead of the effects of the first embodiment, the following effects can be achieved. That is, according to the second embodiment, since the data analysis unit 302 is provided in the mobile terminal 20a (FIG. 11), offline data analysis becomes possible.

[0092] [C. Modification Example] The present invention is not limited to the above-described embodiments, and various configurations can be adopted based on the contents of the present specification. For example, the following configurations can be adopted.

[0093] [C-1. Configuration] In the sound collecting system 10 of the first embodiment, the noise sound pressure level is calculated in the mobile terminal 20 to determine whether or not sound recording is possible (FIG. 4). However, the calculation of the noise sound pressure level and / or the determination of whether or not sound recording is possible can also be performed by the server device 30. In other words, instead of the mobile terminal 20, the server device 30 may be positioned as a noise calculation device or an ambient sound collecting device.

[0094] In the first embodiment, the physical model storage unit 241 is provided in the mobile terminal 20 (FIG. 2B). However, it is also possible to provide the physical model storage unit 241 in the server device 30. Providing the physical model storage unit 241 in the server device 30 does not put pressure on the capacity of the storage area of ​​the mobile terminal 20, and is therefore expected to contribute to improving the processing speed of the mobile terminal 20. Furthermore, by providing a server device 30 with higher accuracy than the mobile terminal 20, it is expected that the overall processing speed will also be improved.

[0095] [C-2. Control] (C-2-1. Physical Model) In the first embodiment, a physical model was selected using the specifications (model) and operating state (engine speed) of the machine 40 to be diagnosed (FIG. 8). However, for example, if a specific machine 40 to be diagnosed and its operating state are assumed, the selection of a physical model based on the specifications or operating state may be omitted. The same applies to the second embodiment.

[0096] The physical model of the first embodiment uses a linear function of the distance between the diagnosis target machine 40 and the mobile terminal 20 and the reference sound pressure level of the target sound (see FIG. 7(B)). In other words, the physical model of the first embodiment does not take into account the frequencies of the ambient sound, the target sound, and noise. The same is true for the second embodiment. However, for example, if it is known that the target sound occurs predominantly only at a specific frequency or in a specific frequency range, the frequency may be reflected in the physical model.

[0097] For example, if it is known that a target sound occurs predominantly only at a specific frequency and analysis of the target sound can be performed using only that specific frequency, the noise sound pressure level is calculated for only that specific frequency. That is, the physical model defines the relationship between the distance at that specific frequency and the reference sound pressure level of the target sound. Then, using this physical model, the reference sound pressure level (specific frequency component) of the target sound according to the distance is identified. Furthermore, only the component of that specific frequency is extracted from the detected sound pressure level of the ambient sound using Fourier analysis or the like. Then, the noise sound pressure level (specific frequency component) is calculated by subtracting the reference sound pressure level of the target sound (specific frequency component) from the extracted detected sound pressure level of the ambient sound (specific frequency component).

[0098] This allows the reference sound pressure level of the target sound to be estimated or calculated using not only the distance from the machine 40 to be diagnosed (object to be analyzed) to the microphone 209 (ambient sound detection means) but also the frequency of the ambient sound, making it possible to calculate the estimated sound pressure level of the noise with higher accuracy.

[0099] Alternatively, if it is known that the target sound occurs predominantly only in a specific frequency range (or the frequency range of the target sound is not specified), the physical model is a three-dimensional curved surface consisting of the distance between the diagnosis target machine 40 and the mobile terminal 20, the reference sound pressure level of the target sound, and the frequency. Then, by cutting this three-dimensional curved surface physical model at the distance, a two-dimensional physical model of the frequency corresponding to that distance and the reference sound pressure level of the target sound is obtained.

[0100] In this case, the noise calculation unit 239 calculates and sums up the reference sound pressure level of the target sound corresponding to the distance for each frequency included in the specific frequency range. Also, it acquires the sum of the detected sound pressure levels included in the specific frequency range of the ambient sounds detected by the microphone 209. Then, the noise sound pressure level may be calculated by subtracting the reference sound pressure level (total value) of the target sound from the detected sound pressure level (total value) of the ambient sound.

[0101] In the first embodiment, the physical model was selected based on the specifications (model) of the machine 40 to be diagnosed and the operating state (engine speed) of the machine 40 to be diagnosed (S302 in FIG. 8). In other words, the target sound in the first embodiment was assumed to be an operating sound generated from the entire machine 40 to be diagnosed. The same applies to the second embodiment. However, without being limited to this, a physical model may be constructed as an operating sound generated from a specific part of the machine 40 to be diagnosed (e.g., a pump or an injector).

[0102] For example, a physical model may be created based on the diagnosis target part, the specifications (model) of the diagnosis target machine 40, and the operating state (engine speed) of the diagnosis target machine 40 (or diagnosis target part). That is, a physical model may be created for each diagnosis target part, such as a physical model of a pump with a vehicle weight of 20 tons obtained in a benchmark test. When there are multiple parts that can be diagnosis target parts, the user may select the actual diagnosis target part using the screen of the mobile terminal 20. This makes it possible to calculate the estimated sound pressure level of noise according to the part of the diagnosis target machine 40 (analysis target), thereby improving the calculation accuracy of the estimated sound pressure level of noise.

[0103] (C-2-2. Position / attitude guidance) In the first embodiment, guidance to the appropriate position and posture is performed using a display on the display device 211 (S205, S206 in FIG. 5). However, guidance may be provided using audio in addition to or instead of the display. Also, for example, if attention is focused on the calculation of the estimated sound pressure level of noise, guidance to the user may be omitted. The same applies to the second embodiment.

[0104] In the first embodiment, when the mobile terminal 20 is guided to the correct position, the mobile terminal 20 is also guided to the correct posture (FIG. 5). However, for example, if it is obvious to the user how to adjust the posture of the mobile terminal 20, the guidance to the correct posture may be omitted. The same applies to the second embodiment.

[0105] In the first embodiment, the guidance instructions are displayed on the display device 211 (display unit 236) of the mobile terminal 20 (FIGS. 6(A) to 6(C)). However, for example, when the user performs an operation inside the diagnosis target machine 40, the guidance instructions may be configured to be displayed on a display device (not shown) of the diagnosis target machine 40. The same applies to the second embodiment.

[0106] (C-2-3. Noise calculation) In the sound collecting system 10 of the first embodiment, the mobile terminal 20 calculates the noise sound pressure level and determines whether or not to record (FIG. 4). However, the calculation of the noise sound pressure level and the determination of whether or not to record can also be performed by the server device 30. That is, the mobile terminal 20 transmits ambient sound data to the server device 30 regardless of the S / N ratio, and the server device 30 calculates the noise sound pressure level. Then, only when the S / N ratio is equal to or lower than a predetermined value, the server device 30 may record the ambient sound data in the data storage unit 304.

[0107] (C-2-4. Recording permission determination) In the first embodiment, the mobile terminal 20 determined whether or not to allow recording based on the ratio (S / N ratio) between the reference sound pressure level of the target sound and the estimated sound pressure level of the noise (S108 in FIG. 4). The same applies to the second embodiment (S408 in FIG. 11). However, this is not a limitation, for example, from the viewpoint of determining whether or not to allow recording depending on the level of noise. For example, it is also possible to determine whether or not to allow recording using the estimated sound pressure level of the noise itself.

[0108] Although the embodiments of the present invention have been described in detail above, the present invention is not limited to the above-described embodiments, and various design modifications can be made without departing from the spirit of the present invention as set forth in the claims. [Explanation of symbols]

[0109] 10 Sound Collection System 20 Mobile terminals (noise calculation devices, ambient sound collection devices) 40 Machine to be diagnosed (analysis target) 209 Microphone (Means for detecting ambient sound) 233 Recording control unit (sound collection control unit) 237 Distance measurement unit 238 Posture measurement section 239 Noise Calculation Unit 240 Position / attitude guidance unit (guidance means) 241 Physical Model Storage

Claims

1. A noise calculation device that calculates a sound pressure level of noise included in ambient sound of an analysis target, including a target sound that is a sound emitted by the analysis target, The noise calculation device a microphone for detecting the ambient sound of the analysis target; a storage device for storing a program; a computing device configured to execute the program; a display device, The computing device a measurement unit that measures a distance between the analysis target and the microphone and an attitude of the microphone relative to the analysis target; a measurement condition acquisition unit that acquires measurement conditions of the ambient sound of the analysis target, including the motion state of the analysis target, the distance and the posture measured by the measurement unit, and a ratio between a reference sound pressure level of the target sound and an estimated sound pressure level of the noise included in the ambient sound detected by the microphone; a display unit that displays, on the display device, a guidance display for guiding the motion state of the analysis object and the distance and the posture measured by the measurement unit to the measurement conditions; Using a physical model that defines the relationship between the distance from the analysis target to the microphone and the reference sound pressure level of the target sound, a reference sound pressure level of the target sound corresponding to the distance is identified; a noise calculation unit that calculates an estimated sound pressure level of the noise included in the ambient sound based on a detected sound pressure level of the ambient sound detected by the microphone and a reference sound pressure level of the target sound identified using the physical model; and a recording control unit that determines whether the distance and the attitude measured by the measurement unit, and a ratio between a reference sound pressure level of the target sound and an estimated sound pressure level of the noise included in the ambient sound detected by the microphone, satisfy the measurement conditions, and starts recording the ambient sound when it is determined that the measurement conditions are satisfied. A noise calculation device characterized by:

2. The noise calculation unit specifies a reference sound pressure level of the target sound using the physical model selected according to the operating state of the analysis target.

2. The noise calculation device according to claim 1.

3. The noise calculation unit specifies a reference sound pressure level of the target sound using the physical model selected according to the specifications of the analysis target.

2. The noise calculation device according to claim 1.

4. The physical model is represented by a linear function in which the distance is a function of the reference sound pressure level of the target sound.

2. The noise calculation device according to claim 1.

5. the physical model defines a relationship between the distance and a reference sound pressure level of the target sound according to a frequency of the ambient sound; The noise calculation device specifies a reference sound pressure level of the target sound using the physical model selected according to the distance and the frequency of the ambient sound.

2. The noise calculation device according to claim 1.

6. The noise calculation unit specifies a reference sound pressure level of the target sound using the physical model selected according to the part of the analysis target.

2. The noise calculation device according to claim 1.

7. A sound collection system that collects target sounds that are sounds emitted by an analysis target, the sound collection system includes a server device and a mobile terminal connected to the server device; The mobile terminal a storage device for storing a program; a computing device configured to execute the program; a display device, The computing device of the mobile terminal a measurement unit that measures the distance and attitude of the mobile device relative to the analysis target; an ambient sound detection unit that detects a sound pressure level of ambient sound generated around the analysis target; a measurement condition acquisition unit that acquires from the server device measurement conditions of the ambient sound, the measurement conditions including the motion state of the analysis target, the distance and the posture measured by the measurement unit, and a ratio between a reference sound pressure level of the target sound and an estimated sound pressure level of noise included in the ambient sound detected by the ambient sound detection unit; a display unit that displays, on the display device, a guidance display for guiding the motion state of the analysis object and the distance and the posture measured by the measurement unit to the measurement conditions; a recording control unit that determines whether the distance and the posture measured by the measurement unit, and a ratio between a reference sound pressure level of the target sound and an estimated sound pressure level of the noise included in the ambient sound detected by the ambient sound detection unit, satisfy the measurement conditions, and starts recording the ambient sound when it determines that the measurement conditions are satisfied; and The server device a storage device for storing the program and the measurement conditions for the ambient sound; a central processing unit configured to execute the program; a central processing unit of the server device; a physical model storage unit that stores a physical model that defines the relationship between the distance and the reference sound pressure level of the target sound; a noise calculation unit that calculates an estimated sound pressure level of noise included in the ambient sound based on the detected sound pressure level of the ambient sound and a reference sound pressure level of the target sound that corresponds to the distance and is identified using the physical model; A sound collection system comprising:

8. The computing device of the mobile terminal an attitude measurement unit that measures a distance to the analysis object and an attitude relative to the analysis object; a position / posture guidance unit that determines a measurement situation of the ambient sound based on the distance and the posture, and guides a position or posture of the analysis target in accordance with the measurement situation; a display unit that displays, on a display device of the mobile terminal, a guidance display for guiding the position or posture of the analysis target; 8. The sound collecting system of claim 7, further comprising:

Images