Device for detecting abnormal sounds in equipment, method for detecting abnormal sounds, and program for detecting abnormal sounds

Abstract

To reduce the influence of noise from equipment to achieve highly accurate abnormal sound detection.SOLUTION: In the abnormal sound detection of an instrument, it determines whether to start abnormal sound detection and the start timing, and starts measurement by acquiring equipment information of equipment that emits sound outside the device.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to an abnormal sound detection device, an abnormal sound detection method, and an abnormal sound detection program for equipment. [Background technology]

[0002] A technology is known that acquires sound waveforms from a device to be evaluated or the results of analyzing the sound waveforms as sound data, acquires sensing data on any of the temperature, air pressure, humidity, speed, acceleration, pressure, tension, or load related to the device to be evaluated, generates reference sound data based on the sensing data, and compares the reference sound data with the sound data acquired during diagnosis to determine whether there is an abnormality in the sound of the device to be evaluated (see Patent Document 1). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2013-200143 Summary of the Invention [Problem to be solved by the invention]

[0004] The device and method described in the above-mentioned JP 2013-200143 A do not take into account noises emitted from facilities such as air conditioners, robots, free rollers, AGVs (automated guided vehicles), and public address equipment when detecting abnormal sounds from equipment, which poses a problem of affecting abnormal sound detection.

[0005] The problem to be solved by the present invention is to reduce the influence of noise from equipment and achieve highly accurate detection of abnormal sounds. [Means for solving the problem]

[0006] The present invention provides Detection target In detecting abnormal sounds from equipment, Abnormal sound detection Device and different from the detection target deviceBased on the operation plan information of the equipment, the input conditions for sound data for judging abnormal sounds are set.

[0007] The present invention also provides Detection target In detecting abnormal sounds from equipment, based on the planning data for determining abnormal sounds, Abnormal sound detection Device and different from the detection target device Generates and outputs control data for equipment. [Effects of the Invention]

[0008] According to the present invention, sound data outside the abnormal sound detection device is acquired, Abnormal sound detection By obtaining operation plan information for equipment that is external to the device and generates sound, or by transmitting control data based on an abnormal sound detection plan to the equipment and controlling the operation of the equipment, it is possible to reduce the impact of sound fluctuations on abnormal sound detection, enabling highly accurate abnormal sound detection. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a diagram showing a configuration according to a first embodiment of the present invention. [Figure 2] 4 is a flowchart showing a control procedure according to the first embodiment of the present invention. [Figure 3] 1A and 1B are diagrams showing display examples according to the first and second embodiments of the present invention. [Figure 4] FIG. 10 is a diagram showing a configuration according to a second embodiment of the present invention. [Figure 5] 10 is a flowchart showing a control procedure according to a second embodiment of the present invention. [Figure 6] FIG. 10 is a diagram showing a configuration according to a third embodiment of the present invention. [Figure 7] 10 is a flowchart showing a control procedure according to a third embodiment of the present invention. [Figure 8] FIG. 10 is a diagram showing a display example according to a third embodiment of the present invention. [Figure 9] FIG. 10 is a diagram showing a configuration according to a fourth embodiment of the present invention. [Figure 10]10 is a flowchart showing a control procedure according to a fourth embodiment of the present invention. [Figure 11] FIG. 10 is a diagram showing a display example according to a fourth embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an embodiment for carrying out the present invention will be described with reference to the drawings.

[0011] First Embodiment The first embodiment will be described with reference to the configuration diagram of FIG. 1, the flowchart of FIG. 2, and the display example of FIG.

[0012] The abnormal sound detection device according to this embodiment will be described with reference to Figure 1. (1) The abnormal sound detection device has a (10) sound data input unit, a (20) vehicle operation acquisition unit, a (30) equipment information acquisition unit, a (40) sound fluctuation prediction unit, a (50) control unit, and an (90) output unit.

[0013] The (10) sound data input unit is a sound data input unit that acquires sound or voice information and converts it into sound data such as an electrical signal. For example, it is a microphone. The microphone's directivity and frequency characteristics are selected according to the measurement target, etc. The (10) sound data input unit acquires both (1) sounds external to the abnormal sound detection device and sounds from the equipment being detected as noise when abnormal sounds are detected. Here, external sounds mean sounds emitted from sources other than the components that make up the (1) abnormal sound detection device. For example, if the (1) abnormal sound detection device is integrated with equipment such as a conveyor, sounds emitted from the conveyor are acquired as external sounds.

[0014] The (50) control unit is a control block that has the functions of the (51) sound measurement control unit and (52) abnormal sound detection unit described below, and this control unit may be configured as an integrated piece of hardware, or the sound measurement control unit and the abnormal sound detection unit may be configured as separate units.

[0015] The (51) sound measurement control unit is a control unit for controlling the characteristics and operations related to the measurement of the (10) sound data input unit. For example, it can perform various controls such as the timing of sound measurement, the time length of the sound data to be measured, the sound pressure (volume) level, directivity, measurement direction, and frequency characteristics. These are just examples, and other characteristics and parameters of sound measurement not listed here may also be controlled.

[0016] (20) The vehicle operation acquisition unit is a unit for acquiring the operating state of a vehicle, which is one of the targets of sound measurement. The operating state of a vehicle indicates, for example, the vehicle speed, engine speed, steering angle, brake pedal force, and shift lever operation during the finished vehicle inspection process in a vehicle production factory.

[0017] The (52) abnormal sound detection unit is an abnormal sound detection unit that uses sound data acquired using the (10) sound data input unit and the (51) sound measurement control unit, and vehicle operation information acquired by the (20) vehicle operation acquisition unit, such as vehicle speed, engine RPM, steering angle, brake pedal force, shift lever operation, etc. as tags to determine whether the sound acquired for each tag is normal or abnormal. For example, the unit may perform FFT (Fast Fourier Transform) processing on the sound data to derive a frequency waveform and determine whether the sound is normal or abnormal based on changes in the waveform, or may use machine learning algorithms or deep learning or other artificial intelligence (AI) generated using multiple types of such features to determine whether the sound is normal or abnormal.

[0018] The (30) equipment information acquisition unit is an equipment information acquisition unit for acquiring information on equipment related to the finished vehicle inspection process in which inspections are performed in a vehicle production plant. The (30) equipment information acquisition unit acquires operation plan information including operation programs and plans for equipment related to the inspection process, such as free rollers (roller devices for vehicle running inspections), air conditioners, robots, conveyors for transporting parts, AGVs (automated guided vehicles) that run in the vicinity, and public address equipment (equipment that emits sounds such as chimes and announcements). In other words, the (30) equipment information acquisition unit acquires operation plan information from a predetermined drive mechanism that emits sounds to the surrounding area or equipment with an acoustic function that is located outside the (1) abnormal sound detection device.

[0019] The (40) sound fluctuation prediction unit is a sound fluctuation prediction unit that performs calculations based on the operation plan information of each piece of equipment obtained from the (30) equipment information acquisition unit to predict whether or not the sound acquired by the (10) sound data input unit is likely to fluctuate over time due to the sound emitted from each piece of equipment, and to predict the level, etc. The (40) sound fluctuation prediction unit, for example, creates a time chart from the operation program of each piece of equipment and calculates a spectrogram that shows the sound pressure and frequency characteristics of the sound that may be generated by each piece of equipment.

[0020] The image generation unit (91) is an image generation unit that generates an image showing the normal / abnormal sound determination result and the reason for the determination, with the aim of conveying the result output from the abnormal sound detection unit (52) to people in an easy-to-understand manner. The image generation unit (91) generates, for example, a display image having a user interface such as that shown in FIG. 3.

[0021] The (60) display unit is a display unit for displaying the images generated by the (91) image generation unit to relevant parties such as workers around the equipment, managers, and data scientists who utilize the data. The (60) display unit is, for example, a liquid crystal display or an organic EL display.

[0022] The (90) output unit is an output unit for outputting the results output from the abnormal sound detection unit, and may be composed of the aforementioned (91) image generation unit and (92) display unit, or may be composed of a sound source, amplifier, speaker, etc. for generating and emitting sound.

[0023] Next, (1) the control process of the abnormal sound detection device in the first embodiment will be described with reference to the flowchart in FIG. 2 and the display example of the first embodiment in FIG.

[0024] In this embodiment, the purpose is to use equipment such as a free roller to collect sounds of a vehicle in dynamic and static states during the finished vehicle inspection process at a vehicle production plant, and to detect whether the sounds are normal or abnormal. This is just one example, and is not limited to a particular vehicle, production plant, finished vehicle inspection, etc., and can be applied to similar phenomena.

[0025] First, in step S1, the abnormal sound detection unit (52) acquires an abnormal sound detection program for detecting abnormal sounds. The abnormal sound detection program is acquired from the device's internal memory or via communication or other means from an external device. This abnormal sound detection program records plan data indicating the abnormal sound detection plan, such as when and at what timing to measure, what measuring equipment to use, and the characteristics and parameters of the measuring equipment. For example, when measuring engine noise, information such as whether the vehicle is on a free roller and whether the target vehicle is an internal combustion engine or an electric vehicle is collected, and microphones are selected based on the sensitivity characteristics of multiple measuring microphones, and parameters such as a frequency equalizer are selected. Furthermore, the program records operations such as starting measurement three seconds before the vehicle's start switch is pressed and ending measurement when the vehicle has completely stopped (vehicle speed reaches zero) for five seconds or more.

[0026] Then, in step S2, the (30) equipment information acquisition unit acquires operation plan information for the equipment around the vehicle to be measured. This is because each piece of equipment and the (30) equipment information acquisition unit are connected by wire or wirelessly so that data can be collected.

[0027] Specific equipment information included in the operation plan information, for example, from the free rollers on the finished vehicle inspection line, includes a list of equipment operations, information on the sound pressure and frequency of the sounds emitted by the equipment during each operation, and information on their temporal characteristics, all of which are listed. (30) From the AGVs that automatically transport parts, the equipment information acquisition unit acquires the current location, route information and predicted time, characteristics of the alarm sounds emitted by the AGV during operation (sound pressure, frequency, time series fluctuations, etc.), and information on the timing of alarms in the event of an abnormality (e.g., sounds issued near intersections or when an obstacle is detected). (30) In the case of sound equipment in the factory, the equipment information acquisition unit collects information such as the timetable and broadcast duration of alarms such as chimes, the timetable and broadcast duration of other announcements, and the start time lag and broadcast duration (typical duration) of emergency voice announcements. (30) The equipment operation plan collected by the equipment information acquisition unit may be programmed into an operation program.

[0028] In step S3, the (40) sound fluctuation prediction unit uses the collected equipment operation plan information to calculate and predict the temporal fluctuation of the sound acquired by the (10) sound data input unit.

[0029] Specifically, the sound fluctuation prediction unit (40) creates a time graph of sound fluctuation based on a time table that summarizes the data collected from the equipment information so that the sound measurement control unit (51) can measure sound fluctuation at the lowest possible level.

[0030] Furthermore, (40) the sound fluctuation prediction unit may perform machine learning using both the sound fluctuation information (planned values) and the actual sound fluctuation information (past data) to construct a prediction model for predicting sound fluctuations.

[0031] Here, the sound fluctuation information (planned value) is a sound fluctuation timetable (planned value) generated based on the information obtained by the equipment information acquisition unit (30). On the other hand, the actual sound fluctuation (past data) is the past measurement results of sound fluctuation, which measure whether the sound actually fluctuated according to the plan.

[0032] The sound fluctuation information indicates the time of occurrence of the sound, the duration of the sound, the volume (sound pressure, sound pressure level), the time fluctuation of the sound, the frequency characteristics of the sound, and the like.

[0033] Machine learning predictive models (=algorithms) include common linear regression, SVM, decision tree, random forest, etc., as well as deep learning, etc. Models are not limited to these, and newly developed models or multiple models can be combined in an ensemble.

[0034] The machine learning prediction model learns the planned sound fluctuations, for example, the two values ​​of the actual sound pressure R (dBA) against the sound pressure P (dBA), as well as the equipment, event, time, etc. as labels. As a result, if, for example, there is an in-house announcement at 9:00 and the planned level is 60 dBA but it is 70 dBA, this learning model will convert the in-house announcements at the same time to correct it to 70 dBA.

[0035] The sound fluctuation prediction result is input to the sound measurement control unit (51). The sound fluctuation prediction result is expressed as a time chart, and the time chart of the prediction result corresponds to the time change of the sound calculated by the sound fluctuation prediction unit. In step S4, the sound measurement control unit (51) compares the sound fluctuation prediction result with a threshold value to determine whether the impact on sound measurement is small. If the sound fluctuation for the sound to be measured is equal to or less than a predetermined threshold value, the sound measurement control unit (51) determines that "the impact on sound measurement is small" and acquires the sound input to the sound data input unit (10). By acquiring the sound input to the sound data input unit (10), sound measurement, which is the control process of step S5, is performed. If the predetermined threshold value is exceeded and it is determined that "the impact on sound measurement is large," sound measurement is not performed and the control flow returns to step S2. In this manner, in this embodiment, the motion plan information is used to calculate and predict the temporal fluctuation of the sound acquired by the sound data input unit (10), and when the predicted result of the sound fluctuation is compared with a threshold value and the condition that the impact on sound measurement is small is met, the sound input to the sound data input unit (10) becomes the measurement target. In other words, this condition corresponds to the input condition of sound data based on the motion plan information, and the control process in which the sound measurement control unit compares the predicted result of the sound fluctuation with a threshold value to determine whether the impact on sound measurement is small corresponds to the setting control of the input condition by the control unit (50).

[0036] The acquired sound data is then sent to the (52) abnormal sound detection unit, which records the sound data as vehicle state data at that time based on the vehicle information acquired by the (20) vehicle operation acquisition unit, and determines whether the sound is normal or abnormal using a predetermined abnormal sound detection algorithm (step S6). This determines whether an abnormal sound is being emitted by a predetermined drive mechanism of the vehicle.

[0037] This (52) abnormal sound detection unit not only judges normality / abnormality, but also determines normality / abnormality by setting a threshold value for the value, for example, 0 to 1 (0 normal, 1 abnormal), and determines normality / abnormality. In addition to a simple binary judgment, it may also be configured to output a numerical value as an evaluation value of normality / abnormality between 0 and 1.

[0038] In step S7, the image generation unit (91) processes the results of the abnormal sound detection unit (52) as an image, for example, as shown in the display example of embodiment 1 in Figure 3. In addition to the characteristics of the measured sound, such as sound pressure and frequency, the sound characteristics and sound pressure plan of each piece of equipment, which are factors in sound fluctuations, are displayed. The conditions of the measuring instruments are also included in the image as reference information.

[0039] The image generated by the image generation unit (91) is then displayed as the measurement result to relevant parties such as workers and managers on a display device such as a liquid crystal display or organic electroluminescence display (92) (step S8).

[0040] After performing this processing, it is confirmed whether or not to continue detecting abnormal sounds (step S9). If it is to continue, the control flow performs the control processing from step S1 onwards, and once again acquires sound data through the process as described above to determine whether or not the sound is abnormal. If it is to end, a display (92) may be displayed indicating that measurement has ended, or an end buzzer or voice may be output.

[0041] As described above, the abnormal sound detection device or abnormal sound detection method according to embodiment 1 (1) acquires operation plan information for equipment other than the abnormal sound detection device, and (10) sets data input conditions for the sound data input unit based on the operation plan information. This makes it possible to reduce the impact of sound fluctuations on abnormal sound detection, enabling highly accurate abnormal sound detection.

[0042] Furthermore, the abnormal sound detection device or method according to embodiment 1 targets noise fluctuation prediction for equipment that is external to the device, such as free rollers, air conditioners, robots, conveyors, AGVs, public address equipment, other production lines, and equipment around other production lines that generate noise, and as such can utilize a large amount of equipment information and perform highly accurate abnormal sound detection.

[0043] Furthermore, the abnormal sound detection device or abnormal sound detection method according to embodiment 1 acquires sound data from a device that is the target of abnormal sound detection using the sound data input unit, which is the same input unit that acquires noise, so the device can be configured with fewer components.

[0044] Furthermore, in the abnormal sound detection device or abnormal sound detection method according to embodiment 1, the vehicle operation acquisition unit outputs operation status information such as the vehicle speed, engine rotation speed, steering angle, brake pedal force, and shift lever operation of the vehicle that is the target of abnormal sound detection to the abnormal sound detection unit, making it possible to detect abnormal sounds according to the vehicle status and the progress of the work process.

[0045] Furthermore, in the abnormal sound detection device or abnormal sound detection method according to embodiment 1, the sound measurement control unit receives data from the sound fluctuation prediction unit, so the calculation results of the change in sound fluctuation over time can be reflected in the decision as to whether or not to start abnormal sound detection, or in the start timing, making it possible to perform highly accurate abnormal sound detection.

[0046] Furthermore, in the abnormal sound detection device or abnormal sound detection method according to embodiment 1, the sound fluctuation prediction unit generates a time chart (future time changes) showing changes in sound fluctuations over time based on equipment information acquired from the equipment, so abnormal sound detection can be started at a timing that avoids noise emitted by the equipment. In other words, it is possible to determine when to start abnormal sound detection at a timing when sound fluctuations are small, and not to start abnormal sound detection at a timing when sound fluctuations are large, making it possible to perform highly accurate abnormal sound detection with reduced effects of sound fluctuations.

[0047] Furthermore, in the abnormal sound detection device or abnormal sound detection method according to embodiment 1, the equipment information acquisition unit outputs information including a plan for the times at which the equipment will emit sound, or an operation program in which the times at which sound will be emitted, to the sound fluctuation prediction unit as operation plan information, so the relative relationship between the sound of the device and the equipment becomes clear, enabling highly accurate abnormal sound detection.

[0048] Furthermore, when outputting the results of abnormal sound detection, the abnormal sound detection device or abnormal sound detection method according to embodiment 1 generates and displays images of the time variation of the sound emitted by the equipment, the sound characteristics, the equipment setting information, and the abnormal sound detection plan by the device, allowing workers, managers, and other parties involved in abnormal sound detection to visually understand the results and related information.

[0049] Second Embodiment Next, the operation of a second embodiment of the present invention will be described with reference to the configuration diagram of FIG. 4 and the flowchart of FIG. 5. This embodiment focuses on an air conditioner that adjusts the temperature, humidity, convection, etc. of the air, and differs from the first embodiment in that various characteristics of the (10) sound data input unit for abnormal sound detection are controlled in accordance with sound fluctuations estimated from the air conditioner's operation plan information. However, the other configuration is the same as that of the first embodiment. Below, the second embodiment will be mainly described in terms of differences from the first embodiment, and parts that are similar to those in the first embodiment will be assigned the same reference numerals and will not be described again. For parts for which description is omitted, the description of the first embodiment will be used as appropriate.

[0050] Regarding the influence of air on sound propagation, the following characteristics are known, for example: Koichi Yoshihisa, Yasuaki Okada, "Effect of air absorption on long-distance noise propagation", [online], Journal of the Acoustical Society of Japan, Vol. 69, No. 6 (2013), Acoustical Society of Japan <URL:https: / / www.jstage.jst.go.jp / article / jasj / 69 / 6 / 69_KJ00008722047 / _pdf / -char / ja>

[0051] The configuration and operation of a method for minimizing sound fluctuations and improving the accuracy of detecting abnormal sounds in the second embodiment will be described below.

[0052] First, a configuration diagram of the second embodiment is shown in Figure 4. Here, differences from the first embodiment will be explained.

[0053] (70) is an air conditioner used to control the temperature and humidity inside the factory, and also generates air convection.

[0054] (71) is an air conditioning control unit that controls the air conditioner (70), and has, for example, a timer, a temperature sensor, a humidity sensor, etc., and automatically controls the power ON / OFF, air volume, air direction, set temperature, humidity, etc. The operation plan information for the air conditioner can be acquired by outputting it from the air conditioning control unit (71) to the facility information acquisition unit (30).

[0055] Next, the operation of the second embodiment will be described with reference to the flowchart of FIG.

[0056] As in the first embodiment, in step S11, an abnormal sound detection program for detecting an abnormal sound is acquired from the abnormal sound detection unit (52).

[0057] Then, in step S12, the equipment information acquisition unit (30) acquires operation plan information of the air conditioner to be measured, such as timer information, set temperature, humidity, air volume, and wind direction.

[0058] Then, in step S13, the collected operation plan information is used by the sound fluctuation prediction unit (40) to calculate and predict the temporal fluctuation of the sound acquired by the sound data input unit (10).

[0059] Specifically, (51) creates a time graph of sound fluctuations based on the air conditioner control timetable so that the sound measurement control unit can measure sound fluctuations at the lowest level. Here, a prediction model as described in embodiment 1 may be constructed, which predicts sound fluctuations using a machine learning model generated from past performance data.

[0060] In step S14, the results of the sound fluctuation prediction are input to the sound measurement control unit (51), and the frequency characteristics of the sound data input unit (10), such as the acquisition start time, acquisition time length, acquired sound pressure level, and frequency filter control, are adjusted.

[0061] Specifically, it predicts and controls the convection and flow of air in the space between the sound sources, such as the vehicle engine, transmission, and tires, and the sound data input unit (10). For this reason, it switches according to the operation of the air conditioner, for example, by determining the measurement start time and acquisition time length based on the time characteristics of the air volume, wind direction, and blown air temperature of the air conditioner (70) to match the timing when the air flow is at its weakest and remains around 25°C, determining the sound pressure level according to the strength, direction, and temperature of the air flow, and determining the frequency characteristics of the sound data input unit (10) according to the temperature dependency of the air.

[0062] The above content differs depending on whether the object to be measured is an engine or a motor, and so the sound to be measured is determined and switched and controlled based on information from the (20) vehicle operation acquisition unit.

[0063] The above control is one example, and prediction, estimation, and control may also be performed by acquiring various other physical quantities, such as factors that cause fluctuations in air and sound, air pressure, and density. (10) The sound data input unit measures sound by acquiring sound data in an environment and / or at a timing that minimizes the impact on sound measurement (step S15). In this way, the input conditions for acquiring sound data in an environment and / or at a timing that minimizes the impact on sound measurement are indicated by the measurement start time, acquisition duration, strength and direction of air flow, temperature, sound pressure level, frequency characteristics, etc., and the sound data input unit can acquire data that minimizes the impact on sound measurement by acquiring sound data when the input conditions are met.

[0064] The acquired sound data is then sent to the abnormal sound detection unit (52), where it is recorded together with the vehicle information acquired by the vehicle operation acquisition unit (20) and the sound data of the vehicle state at that time, and a predetermined abnormal sound detection algorithm is used to determine whether the sound is normal or abnormal (step S16).

[0065] The results of the abnormal sound detection unit (52) are processed by the image generation unit (91) as an image such as the display example shown in FIG. 3 of the first embodiment (step S17). In addition to the characteristics of the measured sound such as sound pressure and frequency, the sound characteristics of each piece of equipment that are factors in sound fluctuations, such as sound pressure, frequency, and sound duration, are displayed. The conditions of the measuring instruments are also included in the image as reference information.

[0066] Then, as in the first embodiment, the image generated by the image generation unit (91) is displayed on a display unit (92) such as a liquid crystal display or an organic electroluminescence display, and the measurement results are presented in an easy-to-understand manner to relevant parties such as workers and managers (step S18).

[0067] After performing such processing, in step S19, it is confirmed whether or not to continue detecting abnormal sounds, and if so, the sound data is collected again through the above process and abnormal sound determination is performed. If so, a display (92) may be displayed indicating that measurement has ended, or an end buzzer or voice may be output.

[0068] As described above, the abnormal sound detection device or abnormal sound detection method according to embodiment 2 targets air conditioners, which are external to the device, for sound fluctuation prediction, and thereby enables highly accurate abnormal sound detection from limited equipment information.

[0069] Furthermore, the abnormal sound detection device or abnormal sound detection method according to embodiment 2 acquires sound data from a device that is the target of abnormal sound detection using the same sound data input unit that acquires noise, so the device can be configured with fewer components.

[0070] Furthermore, in the abnormal sound detection device or abnormal sound detection method according to the second embodiment, the vehicle operation acquisition unit outputs the operation state of the vehicle that is the target of abnormal sound detection to the abnormal sound detection unit, making it possible to detect abnormal sounds according to the vehicle state and the progress of the work process.

[0071] Furthermore, in the abnormal sound detection device or abnormal sound detection method according to embodiment 2, the sound measurement control unit receives data from the sound fluctuation prediction unit, so when detecting an abnormal sound, it is possible to control any one of the acquisition time, acquisition period, sound pressure level, and frequency characteristics of the sound data input unit, making it possible to detect abnormal sounds with high accuracy.

[0072] Furthermore, in the abnormal sound detection device or abnormal sound detection method according to embodiment 2, the sound fluctuation prediction unit generates a time chart based on equipment information acquired from the air conditioner, and therefore the sound measurement control unit can control the measurement time length, sound pressure level, and frequency characteristics in consideration of the effect of air conditioner operation on sound measurement, making it possible to perform highly accurate abnormal sound detection with reduced effects of sound fluctuations.

[0073] Furthermore, in the abnormal sound detection device or abnormal sound detection method according to embodiment 2, the equipment information acquisition unit outputs information including a plan for the time at which the air conditioner will emit sound, or an operation program in which the time at which the sound will be emitted, to the sound fluctuation prediction unit as operation plan information, so the relative relationship between the sound of this device and the air conditioner becomes clear, enabling highly accurate abnormal sound detection.

[0074] Furthermore, when outputting the results of abnormal sound detection, the abnormal sound detection device or abnormal sound detection method according to embodiment 2 generates and displays images of the time variation in the sound emitted by the air conditioner, the characteristics of the sound, the setting information for the air conditioner, and the abnormal sound detection plan by the device, allowing workers, managers, or anyone involved in abnormal sound detection to visually understand the results and related information.

[0075] Third Embodiment Next, the operation of a third embodiment of the present invention will be described with reference to the configuration diagram of FIG. 6, the flowchart of FIG. 7, and the display example of FIG. 8. This embodiment differs from the first embodiment in that the sound measurement control unit (51) controls the operation of the air conditioner (70) via the air conditioning control unit (71), but the other configurations are the same as those of the first embodiment. Below, the third embodiment will be mainly described with respect to the differences from the first embodiment, and parts that are the same as those of the first embodiment will be assigned the same reference numerals and will not be described again. For parts where description is omitted, the descriptions of the first and second embodiments will be used as appropriate.

[0076] In the third embodiment, sound fluctuations are minimized and the detection accuracy of abnormal sounds is improved by controlling various characteristics on the equipment side in accordance with the measurement conditions, state, etc. for abnormal sound detection. The configuration, operation, and display examples of this method are explained below.

[0077] First, a configuration diagram of the third embodiment is shown in Fig. 6. Here, differences from the first and second embodiments will be explained.

[0078] By sending a signal from the sound measurement control unit (51) to the air conditioning control unit (71), the sound measurement control unit (51) can operate the air conditioner (70) via the air conditioning control unit (71) in accordance with the abnormal sound detection program acquired from the abnormal sound detection unit (52).

[0079] Next, the operation will be described using the flowchart in FIG. 7 and the display example in FIG.

[0080] As in the first and second embodiments, in step S21, an abnormal sound detection program for detecting abnormal sounds is acquired from the abnormal sound detection unit (52). The content of this program may be changed depending on a signal from the vehicle operation acquisition unit (20).

[0081] Then, in step S22, the sound measurement control unit (51) sends control data to the air conditioning control unit (71) in accordance with the timing when the sound data input unit (10) starts collecting sound, and the operation plan for the air conditioner, such as the temperature, humidity, air volume, and air direction, is changed in accordance with the sound measurement timing. The control data may be, for example, instruction data such as turning off the air conditioner, setting the air volume to zero, directing the air conditioner's airflow upward so as to prevent airflow between the sound source and the sound data input unit (10), or sending air with a temperature of 20°C and humidity of 60% between the sound source and the sound data input unit (10), or start time data such as the time to start stopping the air conditioner's operation. This is just one example, and multiple combinations of these may be implemented. Any control method that minimizes sound fluctuations may be used. The sound measurement timing is also indicated by plan data for determining abnormal sounds.

[0082] In this way, when the sound fluctuations are minimized, the sound to be evaluated is acquired by the sound data input unit (10) according to instructions from the sound measurement control unit (51), and the sound is measured (step S23).

[0083] Thereafter, in the same manner as in the second embodiment, the acquired sound data is output to the abnormal sound detection unit (52), and the vehicle information acquired by the vehicle operation acquisition unit (20) and the sound data of the vehicle state at that time are recorded, and a predetermined abnormal sound detection algorithm is used to determine whether the sound is normal or abnormal (step S24).

[0084] The results of the abnormal sound detection unit (52) are processed by the image generation unit (91) as an image such as that shown in the display example of embodiment 3 in Fig. 8 (step S25). Fig. 8 shows an example in which there are three air conditioners (70) to be controlled. The setting conditions of each air conditioner, A to C, such as the power status, air volume, air direction, temperature, humidity, etc., are displayed as an illustration so that the placement and positional relationship of the measurement target and each air conditioner can be understood.

[0085] In addition to the sound pressure and frequency characteristics of the measured target sound, the image also displays the sound pressure, frequency, and time fluctuations of the sound characteristics of the air conditioner, which are factors that cause sound fluctuations.The measurement conditions of the measuring equipment are also included in the image as reference information. Furthermore, (10) the air flow between the sound data input unit and the sound source to be measured may be visualized as a three-dimensional image and displayed together as the air flow calculation result.

[0086] Then, as in the first and second embodiments, the image generated by the image generation unit (91) is displayed on a display device such as a liquid crystal display or an organic electroluminescence display (92) as a display unit, and the measurement results are presented in an easy-to-understand manner to relevant parties such as workers and managers (step S26).

[0087] When the above process is completed, the sound measurement control unit (51) notifies the air conditioning control unit (71) of the completion of abnormal sound detection, and the air conditioning control unit (71) controls the air conditioner (70) to return its state to the state before the control for abnormal sound detection (step S27).

[0088] After performing such processing, in step S28, it is confirmed whether or not to continue detecting abnormal sounds, and if so, the sound data is collected again through the above process and abnormal sound determination is performed. If so, a message indicating the end of measurement may be displayed on the display unit (92), or an end buzzer or voice may be output.

[0089] As described above, the abnormal sound detection device or abnormal sound detection method according to embodiment 3 generates and outputs control data for equipment external to the device based on plan data for determining abnormal sounds when detecting abnormal sounds in equipment. This makes it possible to reduce the impact of sound fluctuations on abnormal sound detection, enabling highly accurate abnormal sound detection.

[0090] Furthermore, the abnormal sound detection device or abnormal sound detection method according to the third embodiment controls an air conditioner that is located outside the device and emits sound, thereby enabling highly accurate abnormal sound detection with minimal equipment control.

[0091] Furthermore, the abnormal sound detection device or abnormal sound detection method according to embodiment 3 acquires sound data from a device that is the target of abnormal sound detection using the same sound data input unit that acquires noise, so the device can be configured with fewer components.

[0092] Furthermore, in the abnormal sound detection device or abnormal sound detection method according to the third embodiment, the vehicle operation acquisition unit outputs the operation state of the vehicle that is the target of abnormal sound detection to the abnormal sound detection unit, making it possible to detect abnormal sounds according to the vehicle state and the progress of the work process.

[0093] Furthermore, the abnormal sound detection device or abnormal sound detection method according to embodiment 3 is configured to transmit control data to an air conditioner external to the device based on an abnormal sound detection program, making it possible to control the air conditioner in accordance with the detection of an abnormal sound and enabling highly accurate abnormal sound detection.

[0094] Furthermore, the abnormal sound detection device or abnormal sound detection method according to the third embodiment includes, in the control data, instruction data that instructs the operating noise of the air conditioner to be reduced or stopped, thereby enabling highly accurate abnormal sound detection with reduced influence of the operating noise of the air conditioner.

[0095] Furthermore, the abnormal sound detection device or abnormal sound detection method according to the third embodiment includes in the control data start time data for starting to reduce or stop the operating noise of the air conditioner, so that the air conditioner can be controlled at the desired timing, enabling highly accurate abnormal sound detection.

[0096] Furthermore, when the abnormal sound detection device or abnormal sound detection method according to embodiment 3 outputs the results of abnormal sound detection, it generates and displays an image of the airflow between the sound data input unit and the equipment, the relative positions of the sound data input unit, the vehicle, and the air conditioner, and the setting information of the equipment, so that workers, managers, or anyone involved in abnormal sound detection can visually understand the results and related information.

[0097] Fourth Embodiment Next, the operation of a fourth embodiment of the present invention will be described with reference to the configuration diagram of FIG. 9, the flowchart of FIG. 10, and the display example of FIG. 11. This embodiment differs from the third embodiment in that it includes a robot (80) and a robot control unit (81) instead of the air conditioner (70) and air conditioning control unit (71). The remaining configuration is the same as that of the third embodiment. Below, the fourth embodiment will be mainly described in terms of differences from the third embodiment, and parts that are similar to those in the third embodiment will be assigned the same reference numerals and will not be described again. For parts for which description is omitted, the descriptions of the first to third embodiments will be used as appropriate.

[0098] In the fourth embodiment, we will explain the configuration, operation, and display examples when the form explained in the third embodiment, in which sound fluctuations are minimized by controlling various characteristics on the equipment side in accordance with the measurement conditions and state of abnormal sound detection, and the like, is applied to a robot.

[0099] First, a configuration diagram of the fourth embodiment is shown in Fig. 9. Here, differences from the first, second and third embodiments will be explained.

[0100] The configuration is similar to that of Fig. 6 of the third embodiment, but instead of the air conditioner (70), a robot (80) is used to weld the car body and install various parts on behalf of workers, and one or more than two robots are used in factories. The robot control unit (81) controls the operation of the robot (80).

[0101] By transmitting a signal from the (51) sound measurement control unit to the (81) robot control unit, the (51) sound measurement control unit can operate the (80) robot via the (81) robot control unit in accordance with the abnormal sound detection program acquired from the (52) abnormal sound detection unit.

[0102] Next, the operation will be described using the flowchart in FIG. 10 and the display example in FIG.

[0103] As in the first, second, and third embodiments, in step S31, an abnormal sound detection program for detecting abnormal sounds is acquired from the abnormal sound detection unit (52). The content of this program may be changed depending on a signal from the vehicle operation acquisition unit (20).

[0104] Then, in step S32, the sound measurement control unit (51) sends control data to the robot control unit (81) in accordance with the timing when the sound data input unit (10) starts collecting sound. The control data is, for example, instruction data such as turning off the power to the robot, stopping the robot's operation, changing the robot's working speed (speeding up or slowing down), changing the robot's working range or working process (narrowing the working range, not moving joint A and using joint B instead, stopping work step 2, or performing 5' instead of 5), or start time data such as the time to start stopping the robot's operation. This is just one example, and one or a combination of these may be implemented, and any control may be performed as long as it is a method that minimizes sound fluctuations.

[0105] In reality, it is difficult to interrupt the robot's operation and immediately change it during work, so the sound measurement control unit (51) determines whether the sound fluctuation is small enough to detect abnormal sounds (step S33). Specifically, the robot control unit (81) transmits response data in response to the control data transmitted by the sound measurement control unit (51).

[0106] This response data is used to transmit the propriety of the instructions, start time, etc. included in the control data, or the (80) robot operation plan information to the (51) sound measurement control unit, and the (51) sound measurement control unit determines whether or not to start abnormal sound detection based on the response data.

[0107] If it is determined that abnormal sound detection can be started, the robot (80) operates according to the instruction data. If it is determined that abnormal sound detection cannot be started, the sound measurement control unit (51) sends start time data changed according to the robot's operation plan information (80) to the robot control unit (81), and the robot (80) operates according to the start time data.

[0108] In this way, when the sound fluctuation is minimized, the sound to be evaluated is acquired by the sound data input unit (10) according to the instructions of the sound measurement control unit (51), and the sound is measured (step S34).

[0109] Thereafter, as in the first, second, and third embodiments, the acquired sound data is sent to the abnormal sound detection unit (52), and the vehicle information acquired by the vehicle operation acquisition unit (20) and the sound data of the vehicle state at that time are recorded, and a predetermined abnormal sound detection algorithm is used to determine whether the sound is normal or abnormal (step S35).

[0110] The results of the abnormal sound detection unit (52) are processed by the image generation unit (91) as an image such as that shown in the display example of embodiment 4 in Fig. 11 (step S36). Fig. 11 shows an example in which there are two robots (80) to be controlled. For each of robots A and B, the setting conditions of each robot, such as the power supply status, working speed, arm operating range, and work process list, are displayed as an illustration or the like so that the placement and positional relationship of the measurement target and robot can be understood.

[0111] Furthermore, in addition to the characteristics of the measured sound, such as sound pressure and frequency, the sound characteristics of each robot, which are factors that cause sound fluctuations, such as sound pressure, frequency, and sound time fluctuations, are also displayed.In addition, the measurement conditions of the measuring equipment are also included in the image as reference information.

[0112] Then, as in the first, second, and third embodiments, the image generated by the image generation unit (91) is displayed on a display unit (92) such as a liquid crystal display or an organic electroluminescence display, and the measurement results are presented in an easy-to-understand manner to relevant parties such as workers and managers (step S37).

[0113] When the above processing is completed, the sound measurement control unit (51) notifies the robot control unit (81) of the completion of abnormal sound detection, and the robot control unit (81) controls the robot (80) to return its state to the state before the control of abnormal sound detection (step S38).

[0114] After performing such processing, in step S39, it is confirmed whether or not to continue detecting abnormal sounds, and if so, the sound data is collected again through the above process and abnormal sound determination is performed. If so, a message indicating the end of measurement may be displayed on the display unit (92), or an end buzzer or voice may be output.

[0115] If it is determined in the control process of step S33 that abnormal sound measurement is impossible, the sound measurement control unit (51) sends a signal to the robot control unit (81) to change the robot's work content (80) (step S40). Then, after the robot's work is changed, the control process of step S32 is executed again.

[0116] As described above, the abnormal sound detection device or abnormal sound detection method according to embodiment 4 controls a robot that is located outside the device and emits sound, thereby enabling highly accurate abnormal sound detection with minimal equipment control.

[0117] Furthermore, the abnormal sound detection device or abnormal sound detection method according to embodiment 4 acquires sound data from a device that is the target of abnormal sound detection using the same sound data input unit that acquires noise, so the device can be configured with fewer components.

[0118] Furthermore, in the abnormal sound detection device or abnormal sound detection method according to the fourth embodiment, the vehicle operation acquisition unit outputs the operation state of the vehicle that is the target of abnormal sound detection to the abnormal sound detection unit, making it possible to detect abnormal sounds according to the vehicle state and the progress of the work process.

[0119] Furthermore, the abnormal sound detection device or abnormal sound detection method according to embodiment 4 is configured to transmit control data to a robot external to the device based on an abnormal sound detection program, making it possible to control the robot in accordance with the detected abnormal sound and enabling highly accurate abnormal sound detection.

[0120] Furthermore, the abnormal sound detection device or abnormal sound detection method according to the fourth embodiment includes, in the control data, instruction data that instructs the robot to reduce or stop its operating sound, thereby enabling highly accurate abnormal sound detection with reduced influence of the robot's operating sound.

[0121] Furthermore, in the abnormal sound detection device or abnormal sound detection method according to the fourth embodiment, the control data includes start time data for starting to reduce or stop the operating sound of the robot, so that the robot can be controlled at the desired timing and abnormal sounds can be detected with high accuracy.

[0122] Furthermore, the abnormal sound detection device or abnormal sound detection method according to embodiment 4 takes into consideration that there may be situations in which it is difficult to immediately stop the operation of a robot on a production line for safety reasons, and by having the robot return response data in response to the received control data, it is possible to adjust the start of safe abnormal sound detection. Furthermore, by having the robot return response data including whether or not to respond to abnormal sound detection control or operation plan information, the sound measurement control unit can schedule abnormal sound detection. Furthermore, by having the sound measurement control unit perform control based on the response data from the robot, it is possible to detect abnormal sounds while reducing the impact on the production line caused by stopping the robot or delaying its operation.

[0123] Furthermore, when the abnormal sound detection device or abnormal sound detection method according to embodiment 4 outputs the results of abnormal sound detection, the relative positions of the sound data input unit, vehicle, and equipment, as well as setting information for the equipment, are generated and displayed as images, allowing workers, managers, and other parties involved in abnormal sound detection to visually understand the results and related information.

[0124] These embodiments 1 to 4 are examples of the abnormal sound detection device and abnormal sound detection method according to the present invention, and the parameters, setting values, etc. described in embodiments 1 to 4 do not necessarily have to be the same, and the present invention is not limited to these.

[0125] Furthermore, in the above first to fourth embodiments, an abnormal sound detection device or an abnormal sound detection method has been described, but the present embodiments may also be applied to a program that causes a computer included in an abnormality detection device to execute the abnormal sound detection method. [Explanation of symbols]

[0126] 1...Sound data input section 10...Sound measurement control section 20...Vehicle operation acquisition unit 21...Abnormal sound detection unit 30…Equipment information acquisition department 40...Sound fluctuation prediction section 50...Image generation unit 60...Display section 70...Air conditioner 71...Air conditioning control unit 80...Robot 81...Robot control unit 90...Output section 100...Control unit

Claims

1. An abnormal sound detection device that detects abnormal sounds from a detection target device, a sound data input unit that acquires sound data from outside the abnormal sound detection device; an equipment information acquisition unit that acquires operation plan information for equipment other than the abnormal sound detection device and the detection target device; a control unit that determines an abnormal sound of the detection target device from the sound data acquired by the sound data input unit; an output unit that outputs the abnormal sound determination result determined by the control unit, The control unit sets input conditions for the sound data based on the operation plan information acquired from the facility information acquisition unit.

2. An abnormal sound detection device that detects abnormal sounds from a detection target device, a sound data input unit that acquires sound data from outside the abnormal sound detection device; a control unit that determines an abnormal sound of the detection target device from the sound data acquired by the sound data input unit; an output unit that outputs a determination result determined by the control unit; The control unit generates and outputs control data for the abnormal sound detection device and equipment other than the detection target device, based on plan data for determining abnormal sounds from the detection target device.

3. 3. The abnormal sound detection device according to claim 1, wherein the facility is located outside the abnormal sound detection device and has a predetermined drive mechanism that emits sound to the surroundings, or a facility that has an acoustic function.

4. 4. The abnormal sound detection device according to claim 1, wherein the detection target device is a vehicle, and the control unit determines an abnormal sound emitted to the surroundings by a predetermined drive mechanism of the vehicle.

5. a vehicle operation acquisition unit that acquires the operation state of the vehicle that is the target of abnormal sound detection and outputs the acquired state to the control unit; The abnormal sound detection device according to claim 4 , wherein the control unit determines whether an abnormal sound is generated in the vehicle based on the operational state of the vehicle acquired from the vehicle operation acquisition unit.

6. Further comprising a sound fluctuation prediction unit that calculates a time change in the sound emitted by the equipment, 6. The abnormal sound detection device according to claim 1, wherein the control unit controls whether or not to start abnormal sound detection or the start timing, or controls any one of the acquisition time, sound pressure level, or frequency characteristics of the sound data input unit, in accordance with the data received from the sound fluctuation prediction unit.

7. Further comprising a sound fluctuation prediction unit that calculates a time change in the sound emitted by the equipment, the control unit controls whether or not to start abnormal sound detection or the start timing according to the data received from the sound fluctuation prediction unit, or controls any one of the acquisition time, sound pressure level, and frequency characteristics of the sound data input unit, the sound fluctuation prediction unit calculates future time changes of the sound acquired from the sound data input unit based on the operation plan information acquired from the equipment information acquisition unit, 2. The abnormal sound detection device according to claim 1, wherein the control unit sets a start timing for abnormal sound detection so as to avoid sounds emitted by the equipment, according to a calculation result of the time change of the sound acquired from the sound fluctuation prediction unit.

8. 2. The abnormal sound detection device according to claim 1, wherein the input condition is determined by at least one of the acquisition time, acquisition period, sound pressure level, and frequency characteristics of the sound data.

9. 2. The abnormal sound detection device according to claim 1, wherein the operation plan information is information including a time at which the equipment is to emit a sound or an operation program in which a time at which the equipment is to emit a sound is specified.

10. Further comprising a sound fluctuation prediction unit that calculates a time change in the sound emitted by the equipment, the output unit includes an image generation unit that generates an image based on the results output from the sound fluctuation prediction unit and the control unit; a display unit that displays the image generated by the image generation unit, 2. The abnormal sound detection device according to claim 1, wherein the output unit outputs, as the image, a change over time in the sound emitted by the equipment, characteristics of the sound, setting information of the equipment, and a plan for abnormal sound detection by the abnormal sound detection device.

11. The abnormal sound detection device according to claim 2 , wherein the control unit transmits control data to the equipment based on the abnormal sound detection program acquired by the control unit.

12. 3. The abnormal sound detection device according to claim 2, wherein the control data is instruction data that instructs the operation of the equipment to be reduced or stopped during an abnormal sound detection period.

13. The abnormal sound detection device according to claim 2 , wherein the control data includes start time data for starting to stop the operation of the equipment.

14. The equipment transmits response data in response to the control data; The response data includes whether or not the control data is valid or invalid, or information on a previous operation plan for the equipment.

3. The abnormal sound detection device according to claim 2, wherein the control unit determines whether or not to start abnormal sound detection in accordance with the response data, and transmits the control data to change an operation plan of the equipment.

15. The output unit an image generating unit that generates an image based on the results output from the control unit; a display unit that displays the image generated by the image generation unit, 3. The abnormal sound detection device according to claim 2, wherein the output unit outputs, as the image, an image of the airflow between the sound data input unit and the equipment, a positional relationship between the sound data input unit, the detection target device, and the equipment, and setting information of the equipment.

16. An abnormal sound detection method for detecting an abnormal sound using an abnormal sound detection device having a sound data input unit, a facility information acquisition unit, a control unit, and an output unit, comprising: a step of inputting sound data from outside the abnormal sound detection device into the sound data input unit; a step in which the facility information acquisition unit acquires operation plan information for a facility other than the abnormal sound detection device and the detection target device; a step in which the control unit determines whether an abnormal sound is generated from the sound data acquired by the sound data input unit based on the information acquired from the equipment information acquisition unit; and a step in which the output unit outputs the result output from the control unit.

17. An abnormal sound detection method for detecting an abnormal sound using a device having a sound data input unit, a control unit, and an output unit, comprising: a step of inputting sound data from outside the abnormal sound detection device into the sound data input unit; a step in which the control unit generates and outputs control data for controlling the abnormal sound detection device and equipment other than the detection target device based on the abnormal sound detection plan, and determines whether an abnormal sound is heard from the sound data acquired by the sound data input unit; a step in which the output unit outputs the result output from the control unit; An abnormal sound detection method including:

18. An abnormal sound detection program for causing a computer included in an abnormal sound detection device having a sound data input unit, an equipment information acquisition unit, a control unit, and an output unit to execute steps for detecting an abnormal sound, a step of inputting sound data from outside the abnormal sound detection device into the sound data input unit; a step in which the facility information acquisition unit acquires operation plan information for a facility other than the abnormal sound detection device and the detection target device; a step in which the control unit determines whether an abnormal sound is generated from the sound data acquired by the sound data input unit based on the information acquired from the equipment information acquisition unit; a step in which the output unit outputs the result output from the control unit; A computer-readable abnormal sound detection program that is executed by the computer.

19. An abnormal sound detection program for causing a computer included in an apparatus having a sound data input unit, a control unit, and an output unit to execute steps for detecting abnormal sounds, a step of inputting sound data from outside the abnormal sound detection device into the sound data input unit; a step in which the control unit generates and outputs control data for controlling the abnormal sound detection device and equipment other than the detection target device based on the abnormal sound detection plan, and determines whether an abnormal sound is heard from the sound data acquired by the sound data input unit; a step in which the output unit outputs the result output from the control unit; A computer-readable abnormal sound detection program that is executed by the computer.

Images