JP2021096209A - System and program for detecting change of state - Google Patents

Abstract

To make a determination with a certainty without using a mechanical learning.SOLUTION: The present invention includes: a sound collecting device 10 for collecting sound data on sound generated from an object; a sending device 20 for sending the sound data collected by the sound collecting device 10 to a calculation device; a calculation device 40 for calculating a basic frequency of the sound data collected by the sound collecting device 10; and a determination device 50 for determining the presence or absence of a change in the state of the object on the basis of the basic frequency calculated by the calculation device 40 and the threshold value according to the object.SELECTED DRAWING: Figure 1

Description

The present invention relates to a state change detection system and a state change detection program, and more particularly to a state change detection system and a state change detection program that detects that a state change of an object has occurred based on a fundamental frequency of sound data.

Patent Document 1 discloses an abnormal noise determination method that accurately determines whether or not an abnormal noise is generated in an inspection target, including an unknown abnormal noise, without selecting an appropriate frequency component in advance. Has been done. This abnormal sound determination method records sound data when the inspection target travels, short-time Fourier transforms the recorded sound data of the inspection target, and decomposes it into multiple frequency bands such as every 100 Hz for each unit time. Then, the sound data decomposed into a plurality of frequency bands for each unit time is compared for each of the plurality of frequency bands, and the correlation coefficient (correlation coefficient matrix) indicating the degree of correlation between the two is calculated, and the calculated phase. The presence or absence of abnormal noise is determined by using machine learning or the like for the number of relations.
FIG. 2 of JP-A-2019-164107, etc.

However, since the determination method disclosed in Patent Document 1 uses machine learning or the like in determining the presence or absence of abnormal noise, a huge amount of data is collected for machine learning before the determination is made. It is necessary to keep it.

In addition, although it is possible to determine that there is a high probability that abnormal noise is occurring in the judgment using machine learning, this is despite the fact that abnormal noise is actually occurring. Since it is determined that no abnormal noise is generated and vice versa, the determination accuracy when machine learning is used is limited.

Therefore, an object of the present invention is to perform a reliable determination by an approach different from the determination method disclosed in Patent Document 1.

In order to solve the above problems, the state change detection system of the present invention is used.
A sound collector that collects sound data generated from an object,
A calculation device that calculates the fundamental frequency of the sound data collected by the sound collector, and
A determination device that determines whether or not there is a change in the state of the object based on the fundamental frequency calculated by the calculation device and the threshold value corresponding to the object.
To be equipped.

Since the present invention employs a judgment method such as mathematically analyzing the actually collected sound data without using a probabilistic judgment method such as machine learning, a state including the occurrence of abnormal noise in the object. It is possible to reliably determine whether or not there is a change.

Further, a transmission device for transmitting the sound data collected by the sound collecting device to the calculation device can also be provided. In this way, the sound collecting device and the transmitting device can be mounted on a portable information terminal, an unmanned aerial vehicle, or the like, or can be realized by the hardware possessed by them. In addition, as a side effect, according to the state change detection system for the object, it becomes possible to collect big data of sound data related to the state change.

Further, the calculation device performs autocorrelation processing on the acquisition means for acquiring the power spectrum of the waveform data of the sound data collected by the sound collector and the power spectrum acquired by the acquisition means, and the basics. It may have an indexing means for determining the frequency. There are several methods for determining the fundamental frequency, and by adopting such a method, a highly reliable fundamental frequency can be obtained.

The determination device is a detection means for detecting that the state is changed when the fundamental frequency appears at least three times or more with respect to a constant amplitude appearing in a unit time in the waveform data of the fundamental frequency a predetermined number of times or more. It is good to have. By doing so, it is possible to prevent erroneous detection that there is a state change due to the noise component included in the waveform data, while it is possible to reliably detect that there is a state change.

Further, the state change detection program of the present invention is
Steps to collect sound data generated from an object,
The step of transmitting the sound data to the fundamental frequency calculation device, and
A step of notifying the presence or absence of a change in the state of the object based on the change of the fundamental frequency and the threshold value corresponding to the object, and
Is executed by the information processing device.

In this way, as in the case described above, it is possible to cause the information processing device provided in the portable information terminal, the unmanned aerial vehicle, or the like to perform the required processing for the sound collection step and the sound transmission step.

Embodiment of the invention

Hereinafter, the state change detection system 1000 according to the embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram showing a schematic configuration of the state change detection system 1000 of the present embodiment. FIG. 1 shows a state change detection system 1000 in which a smartphone 100 and a personal computer 200 are connected via a network 300.

The smartphone 100 includes a sound collecting device 10, a transmitting device 20, and a notification device 30, which will be described below. Further, the personal computer 200 includes a calculation device 40, a determination device 50, and a reply device 60, which will be described below.

The sound collecting device 10 collects sound data generated from an object. The sound collector 10 typically includes a microphone. Although FIG. 1 shows an example in which the sound collecting device 10 is realized by a microphone mounted on the smartphone 100, the sound collecting device 10 may be prepared separately from the smartphone 100.

The transmission device 20 transmits the sound data collected by the sound collection device 10 to the personal computer 200 through the network 300. The transmission device 20 may not be realized by the communication function mounted on the smartphone 100, but may be prepared separately from the smartphone 100.

The notification device 30 notifies the presence or absence of a state change of the object returned from the personal computer 200 as a result of the sound data being transmitted by the transmission device 20. The notification device 30 typically includes a display and a speaker. The notification device 30 may also be prepared separately from the smartphone 100. Further, the presence or absence of a change in the state of the object may be notified by printing with a printer built in or connected to the notification device 30.

The smartphone 100 is merely a typical example of a smartphone 100 equipped with an information processing device, and if the smartphone 100 is equipped with an information processing device, the smartphone 100 can be replaced with a mobile phone, a PDA, a tablet terminal, or the like. It can also be an unmanned aerial vehicle such as a notebook personal computer or a drone.

Here, the object means an object that generates some kind of sound data, for example, an electric product having a built-in AC motor (blower, boiler, automatic door, escalator), an engine such as an automobile (bearing, etc.), and a building structure. It refers to things that are subject to crime prevention (window glass, floors, etc.) and parts of real estate (tunnels, bridges, walls, etc.).

The state change detection system 1000 of the present embodiment can mainly prompt the user to replace parts of the abnormal sound source when the sound data generated by these objects is abnormal sound data.

The calculation device 40 calculates the fundamental frequency of the sound data collected by the sound collector 10. The specific configuration of the calculation device 40 will be described later with reference to FIG. 2. For example, the calculation device 40 calculates the fundamental frequency of the sound data by performing a filtering process, a discrete Fourier transform process, or the like on the sound data.

The determination device 50 determines whether or not there is a change in the state of the object based on the fundamental frequency calculated by the calculation device 40 and the threshold value corresponding to the object. The specific determination method performed by the determination device 50 will be described later with reference to FIGS. 3 to 5. For example, in the waveform data of the fundamental frequency, when a peculiar fundamental frequency appears more than a predetermined number of times, the object is an object. It is determined that there is a state change in.

The reply device 60 returns the determination result by the determination device 50 to the smartphone 100 through the network 300. The reply destination of the determination result from the reply device 60 may be addressed to a user of a smartphone (not shown) or an administrator of an object, etc. together with the smartphone 100 or instead of the smartphone 100.

In order to realize this, it is necessary to register a pair of e-mail addresses indicating the reply destinations of the determination results in the reply device 60 together with the name information indicating the users of the smartphone 100 and the like in a database (not shown). Good.

The network 300 connects the smartphone 100 and the personal computer 200. The network 300 is a general term for various networks such as a mobile phone network, a public telephone network, the Internet, an intranet, and a dedicated line.

The basic configuration of the state change determination system of the present embodiment is as described above, but the state change determination system is not essential to include the transmission device 20 and the network 300.

For example, the sound collector 10 and the calculation device 40, etc., excluding these, are realized by one dedicated machine, and the sound data is collected in the vicinity of the object using this dedicated machine, and the fundamental frequency of the sound data is collected. It is also possible to calculate, determine whether or not there is a change in the state of the object, and notify the determination result.

FIG. 2 is a functional block diagram of the calculation device 40 shown in FIG. In FIG. 2, as described below, the acquisition means 40A for acquiring the power spectrum of the waveform data of the sound data collected by the sound collector 10 and the power spectrum acquired by the acquisition means 40A are subjected to autocorrelation processing. The autocorrelation means 40B for detecting the fundamental frequency of the sound data is shown.

The acquisition means 40A includes a digital filter 41 that extracts only components in a desired frequency band from the sound data, and a window processing means 42 that performs window function processing on the components of the sound data extracted by the digital filter 41. And a first conversion processing means 43 that performs a discrete Fourier transform processing on the components of the sound data window-processed by the window processing means 42.

The frequency band of the component to be extracted by the digital filter 41 may be determined by the frequency of the sound data generated from the object. For example, the object is
-In the case of a combustion device such as a boiler, the AC motor and the target (for example, a relay) operated by the AC motor usually generate sound data in the frequency band of 10 Hz to 700 Hz.
・ For bridges, sound data in the frequency band of 20 Hz or less is generated in order to remove sound data generated from automobiles, etc.
-In the case of the footsteps of a trespasser, sound data in the frequency band of 100 Hz to 500 Hz is generated from the feet of the trespasser.
-If the window glass is used for crime prevention and security, sound data in the frequency band of 500 Hz to 1000 Hz generated when the window glass is broken is generated.
Therefore, the digital filter 41 under the condition that each of these components can be extracted may be used.

The window processing means 42 can perform window processing on the components of the sound data extracted by the digital filter 41 by using various window functions. The type of the window function is not particularly limited, and an appropriately selected one from a rectangular window function, a Gaussian window function, a Hanning window function, a Blackman window function, and the like can be used.

The first conversion processing means 43 performs the discrete Fourier transform processing on the component of the sound data window-processed by the window processing means 42 in a sampling frequency band that is, for example, twice the frequency of the component. Therefore, if the sound data in the frequency band of 700 Hz or less is the target of the discrete Fourier transform, the sampling frequency band of 1.4 KHz may be used. When the sound collecting device 10 is realized by the microphone of the smartphone 100, since the sampling frequency of this microphone is usually 44.1 KHz, the discrete Fourier transform process can be performed as it is as in the example described later.

The number of sampling points N may be appropriately determined according to the frequency of the object, for example, N = 2048. However, the present invention is not limited to this, and for example, N = 1024 or N = 4096 may be used.

The autocorrelation means 40B is converted by the second conversion processing means 44 that performs the inverse Fourier transform processing on the power spectrum acquired as a result of the discrete Fourier transform by the first conversion processing means 43 and the second conversion processing means 44. It has a peak detecting means 45 for detecting a peak of a frequency with respect to the collected data.

Next, the operation of the state change detection system 1000 shown in FIG. 1 will be described. Here, an agricultural hot air blower will be described as an example of a state change detection target. First, when the power of the agricultural hot air blower is turned on, the fan is then started to rotate by driving the motor connected to the fan. As a result, sound data will be generated from the agricultural hot air blower.

When the user of the smartphone 100 switches the sound collecting device 10 to the on state, sound data generated from the agricultural hot air blower is collected. The sound data collected by the sound collecting device 10 is transmitted to the personal computer 200 by the transmitting device 20 through the network 300.

When the personal computer 200 receives the sound data transmitted from the transmission device 20 of the smartphone 100, the calculation device 40 calculates the fundamental frequency of the sound data. Specifically, first, the power spectrum of the waveform data of the sound data is acquired by the acquisition means 40A in the following manner.

That is, first, the acquisition means 40A extracts only the components in the frequency band of 10 Hz to 700 Hz from the sound data by the digital filter 41 that extracts only the components in the frequency band of 10 Hz to 700 Hz, for example.

Then, the window processing means 42 performs window function processing on the sound data component digitally filtered by the digital filter 41 by using, for example, a Hanning window function.

Next, for the sound data component extracted by the window processing means 42 by the first conversion processing means 43, for example, the sampling point N = 2048 is set, and here, the microphone of the smartphone 100 is used as the sound collecting device 10. Therefore, the discrete Fourier transform process is performed in the sampling frequency band of 44.1 KHz. As a result, the acquisition means 40A can acquire the power spectrum of the sound data.

Subsequently, the autocorrelation means 40B performs an inverse Fourier transform process on the power spectrum acquired by the acquisition means 40A by the second conversion processing means 44. Then, the peak detecting means 45 detects the peak of the frequency with respect to the data converted by the second conversion processing means 44.

In the personal computer 200, when the fundamental frequency is calculated by the calculation device 40 in this way, the object is calculated by the determination device 50 based on the fundamental frequency calculated by the calculation device 40 and the threshold value corresponding to the object by the method described below. Judges whether or not there is a change in the state of.

After that, in the personal computer 200, the reply device 60 returns the determination result by the determination device 50 to the smartphone 100 or the like through the network 300.

3 to 5 are explanatory views of a determination method in the determination device 50 shown in FIG. FIGS. 3 to 5 are created by using three agricultural hot air blowers having different years of use as objects A to C, measuring the sound data generated from each of them, and calculating their frequencies and fundamental frequencies. The waveform data is shown. It should be noted that all of the objects A to C are of general frequency of use, and there is no particular circumstance that only one of the objects is worn.

The objects A to C are similar (same scale) products manufactured by the same manufacturer, although their models are different because their manufacturing and sales times are different. Here, in order to specify whether or not the fan motors provided in the objects A to C are to be replaced, and if so, a guideline for the timing, the determination device 50 is used based on the waveform data shown in FIGS. 3 to 5. The judgment method to be performed will be described.

FIG. 3A shows waveform data of sound data generated from the object A collected by the sound collecting device 10. The horizontal axis of FIG. 3A shows the time [s], and the vertical axis of FIG. 3A shows the amplitude of the waveform of the sound data. FIG. 3B shows the waveform data of the fundamental frequency of the sound data of the object A detected by the autocorrelation means 40B. The horizontal axis of FIG. 3 (b) indicates the time [s], and the vertical axis of FIG. 3 (b) indicates the fundamental frequency [Hz].

Similarly, FIG. 4A shows waveform data of sound data generated from the object B, and FIG. 4B shows waveform data of the fundamental frequency of the sound data of the object B. Further, FIG. 5A shows waveform data of sound data generated from the object C, and FIG. 5B shows waveform data of the fundamental frequency of the sound data of the object C.

The waveform data of FIGS. 3 (a), 4 (a), and 5 (a) are all level-adjusted so that the stable amplitude falls within a certain range. In all of FIGS. 3 (a), 4 (a) and 5 (a), the level is adjusted so as to be within ± about 0.4.

Even if the sound data generated from the sound source is at the same level, the closer the sound collecting device 10 is to the sound source, the higher the level of the collected sound data. Therefore, if the above level adjustment is performed, it is not necessary to strictly determine the distance between the objects A to C and the sound collecting device 10 when collecting the sound data generated from the objects A to C which are the sound sources. There is an advantage of becoming.

Next, with respect to FIG. 3A, when the object A is switched to the ON state and the fan of the agricultural hot air blower, which is the object A, starts operating, sound data is generated from the object A. On the right side of FIG. 3A, the amplitude of the sound data narrows because the object A is switched to the off state.

Next, in the section where the amplitude of the sound data generated from the object A in FIG. 3 (a) is generally within ± 0.4, as shown in FIG. 3 (b), the fundamental frequency corresponding to this is set. It can be seen that it is generally within ± 10 [Hz].

As a conclusion, the determination device 50 determines that the object A does not change its state during operation. Therefore, in this case, since no particular abnormal noise is generated from the fan motor of the object A, it is not necessary to replace the fan motor. In fact, the object A has a general usage frequency of about March 2013, and the fan motor of the object A has not reached the required replacement period.

Next, with respect to FIG. 4A, when the object B is switched to the ON state and the fan of the agricultural hot air blower starts operating, sound data is generated from the object B. Only humans have visually observed FIGS. 3 (a) and 4 (a), that is, the sound data generated from the object A and the sound data generated from the object B are visualized by adjusting their amplitudes (levels). It can be said that it is difficult to specify whether or not the fan motors of the objects A to B should be replaced only by comparing them with each other.

Next, in the section where the amplitude of the sound data generated from the object B in FIG. 4 (a) is generally within ± 0.4, as shown in FIG. 4 (b), the corresponding fundamental frequency is Although most of them are generally within ± 10 [Hz], it can be seen that the peaks of the fundamental frequency at least three times or more appear in a whisker-like manner in some places.

As a conclusion, the determination device 50 determines that the object B has a small change of state during operation. Then, in this case, a slight abnormal noise is sometimes generated from the fan motor of the object B, so that it is necessary to replace the fan motor in the near future. In fact, the object B has a general usage frequency of about September 20 and the fan motor of the object B is about to reach the required number of years to be replaced.

Next, with respect to FIG. 5A, when the object C is switched to the ON state and the fan of the agricultural hot air blower starts operating, sound data is generated from the object C. As shown in FIGS. 3A and 5A, when the sound data generated from the object A and the sound data generated from the object C are visualized by adjusting their amplitudes and compared with each other, It is not impossible to specify that the fan motor of the object C should be replaced, but it can be said that it is difficult to clearly specify the replacement time.

By the way, it is difficult to specify that the fan motor of the object C should be replaced by simply listening to the sound data generated from the object A and the sound data generated from the object C. ..

Next, in the section where the amplitude of the sound data generated from the object C in FIG. 5 (a) is generally within about ± 0.4 to about ± 10, this corresponds to this as shown in FIG. 5 (b). The fundamental frequency to be used is clearly different from that in FIGS. 3 (b) and 4 (b), and although about half of the frequency is within ± 10 [Hz], the remaining fundamental frequency is at least three times higher. It can be seen that is frequently appearing like a whiskers.

As discussed below, in conclusion, the determination device 50 determines that the object C has a large change of state during operation. Then, in this case, since the fan motor of the object C frequently generates a slight abnormal noise, it is necessary to replace the fan motor immediately. In fact, the object C has been used for 25 years at a general frequency of use, and the fan motor of the object C has reached the required number of years to be replaced.

As described with reference to FIGS. 3 to 5, in the state change determination system of the present embodiment, abnormal noise is generated from the fan motor when the fan of the agricultural hot air blower is rotated, which is very difficult to determine by human hearing. Whether or not it occurs, and if so, how often it occurs.

Specifically, based on the fundamental frequency of the sound data generated from the agricultural hot air blower and the threshold value corresponding to the agricultural hot air blower, it is determined whether or not there is an abnormal noise of the fan motor when the fan of the agricultural hot air blower rotates. Can be done.

Next, these waveform data will be considered in comparison with FIGS. 3 to 5. As described above, the object A does not show a sudden change in the fundamental frequency during the operation of the fan of the agricultural hot air blower, and it is not necessary to replace the fan motor.

What is important here is whether or not it is possible to prompt the replacement of the fan motor at the stage when the sound data of the fundamental frequency shown in FIG. 4B is generated. When the sound data of the fundamental frequency shown in FIG. 5B is generated, it can be said that the fan of the object C can be stopped at any time.

Therefore, it is an agricultural engagement to realize the replacement of the fan motor before the sound data of the fundamental frequency as shown in FIG. 5B is generated so that the agricultural hot air blower does not become inoperable. It is a great concern for those who are.

However, even if you adjust the amplitude of the sound data generated from the object B and visualize it, let alone hear this sound data, it is necessary for human beings to replace the fan motor of the object B soon. I can't figure it out.

On the other hand, the state change determination system of the present embodiment collects sound data generated from the object B, calculates the fundamental frequency of the sound data, and is based on the fundamental frequency and the threshold value corresponding to the object B. Since it is determined whether or not the state of the object B has changed, it is possible to understand that the fan motor of the object B needs to be replaced soon.

Next, the threshold value used in the above determination will be described. The threshold value is appropriately determined depending on the type of the object and the like. If the object is an agricultural hot air blower, it may fluctuate depending on the manufacturer, but according to the example of FIGS. 3 to 5, it appears in the waveform data in a unit time (for example, about 10 seconds), for example ± 10. The value may be set as to whether or not the occurrence of the fundamental frequency at least 3 times or more with respect to the amplitude of is a predetermined number of times (for example, 5 times) or more.

Specifically, in the case of FIG. 3, the "predetermined number of times" is zero. In the case of FIG. 4, since one scale on the horizontal axis is about 3 seconds, the "predetermined number of times" occurs 7 times in about 10 seconds. In the case of FIG. 5, since one scale on the horizontal axis is about 1.7 seconds, a considerable number of "predetermined times" are generated in about 10 seconds.

Therefore, under the condition that the predetermined number of times = 5, the state change determination system of the present embodiment determines that there is no state change in the case of FIG. 3, while the state change occurs in the cases of FIGS. 4 and 5. It will be determined that there is.

In the present embodiment, the method of determining the presence / absence of the state change of the object has been described by taking the presence / absence of the occurrence of abnormal noise as a typical example. Etc. are possible, so the effect is enormous.

It is a block diagram which shows the schematic structure of the state change detection system 1000 of this embodiment. It is a functional block diagram of the calculation device 40 shown in FIG. It is explanatory drawing of the determination method in the determination apparatus 50 shown in FIG. It is explanatory drawing of the determination method in the determination apparatus 50 shown in FIG. It is explanatory drawing of the determination method in the determination apparatus 50 shown in FIG.

10 Sound collector 20 Transmitter 30 Notification device 40 Calculation device 40A Acquisition means 40B Autocorrelation means 41 Digital filter 42 Window processing means 43 First conversion processing means 44 Second conversion processing means 45 Peak detection means 50 Judgment device 60 Replying device 100 Smartphone 200 Personal computer 300 Network

Claims (5)

A sound collector that collects sound data generated from an object,
A calculation device that calculates the fundamental frequency of the sound data collected by the sound collector, and
A determination device that determines whether or not there is a change in the state of the object based on the fundamental frequency calculated by the calculation device and the threshold value corresponding to the object.
A state change detection system equipped with. The state change detection system according to claim 1, further comprising a transmission device that transmits sound data collected by the sound collection device to the calculation device. The calculation device is
An acquisition means for acquiring the power spectrum of the waveform data of the sound data collected by the sound collector, and
An indexing means for determining the fundamental frequency by performing autocorrelation processing on the power spectrum acquired by the acquisition means.
The state change detection system according to claim 1. The determination device is
1. The first aspect of the present invention is a detection means for detecting a change in the state when the fundamental frequency appears at least three times or more a predetermined number of times with respect to a constant amplitude appearing in a unit time in the waveform data of the fundamental frequency. The described state change detection system. Steps to collect sound data generated from an object,
The step of transmitting the sound data to the fundamental frequency calculation device, and
A step of notifying the presence or absence of a change in the state of the object based on the change of the fundamental frequency and the threshold value corresponding to the object, and
A state change detection program that causes the information processing device to execute.

Images