JP7491788B2 - Point machine status monitoring device and point machine status monitoring method - Google Patents

Description

The present invention relates to a point machine status monitoring device, etc.

In order to ensure safe railway transport, various methods have been developed to monitor the status of point machines, which are a type of railway equipment, and to quickly detect abnormalities and their signs. For example, Patent Document 1 discloses a technology for monitoring the status of point machines, which determines whether or not an abnormality has occurred in the point machine based on the motor voltage and motor current.

JP 2011-131882 A

However, it was thought that it might not be possible to determine whether or not a point machine abnormality has occurred using only sensors that measure electrical quantities, such as a voltage sensor for measuring motor voltage and a current sensor for measuring motor current. Therefore, there was a need for a method for determining whether or not a point machine abnormality has occurred using a method other than measuring electrical quantities.

The present invention was made in consideration of the above circumstances, and its purpose is to provide a new technology that makes it possible to determine whether or not an abnormality has occurred in a point machine using a method different from the conventional method.

The first invention for solving the above problem is:
An acoustic analysis means (e.g., the acoustic analysis unit 202 in FIG. 6 ) that analyzes acoustic data, which is data on the operation sound of the point machine acquired by the acoustic sensor;
A determination means (for example, the determination unit 206 in FIG. 6 ) that determines whether or not an abnormality has occurred in the point machine using an analysis result of the acoustic analysis means;
The point machine condition monitoring device is provided with the above.

Other inventions include:
an acoustic analysis step of analyzing acoustic data, which is data on the operation sound of the point machine acquired by the acoustic sensor;
a determination step of determining whether or not an abnormality has occurred in the point machine by using an analysis result in the acoustic analysis step;
A point machine condition monitoring method including the above may be configured.

According to the first invention, etc., it is possible to provide a new technology for determining whether or not a point machine abnormality has occurred, such as determining whether or not a point machine abnormality has occurred based on acoustic data, which is data on the operation sound of the point machine acquired by an acoustic sensor.

The second invention is the first invention,
An identification means (e.g., the identification unit 204 in FIG. 6 ) that identifies a switching operation period of the point machine in the acoustic data using an analysis result of the acoustic analysis means;
Further comprising:
the determination means determines whether or not an abnormality has occurred in the point machine by using an analysis result of the acoustic analysis means relating to the switching operation period.
This is a point machine condition monitoring device.

According to the second invention, the period of the point machine switching operation in the acoustic data can be identified, and it can be determined whether or not an abnormality has occurred in the point machine related to the switching operation.

A third aspect of the present invention relates to the second aspect of the present invention,
the identifying means identifies the switching operation period by detecting a generation timing of a sound that matches a relay operation acoustic condition based on an operation sound of the control relay of the point machine from the analysis result of the acoustic analysis means.
This is a point machine condition monitoring device.

According to the third invention, the switching operation period of the point machine can be identified by detecting the operation sound of the control relay from the analysis result of the acoustic data. The contact of the control relay is turned on by a switching command signal from the interlocking device, and the contact is turned off by the completion of locking. Therefore, the occurrence timing of the sound that matches the relay operation acoustic conditions based on each of the on operation sound and off operation sound of the control relay is detected, and the switching operation period can be identified by taking the occurrence timing of the on operation sound as the timing of the start of switching and the occurrence timing of the off operation sound as the timing of the end of switching. In other words, it is possible to identify the switching operation period related to one switching operation of the point machine based on the acoustic data acquired by one acoustic sensor installed on the point machine, and to determine whether or not an abnormality has occurred in the point machine.

A fourth aspect of the present invention is the second or third aspect of the present invention,
the specifying means detects occurrence timings of sounds that match respective acoustic conditions based on operation sounds generated at the boundary between the unlocking process and the switching process of the point machine and acoustic conditions based on operation sounds generated at the boundary between the switching process and the locking process, thereby dividing the switching operation period into an unlocking process, a switching process, and a locking process;
The determination means determines whether or not an abnormality has occurred in the point machine for each of the unlocking process, the switching process, and the locking process.
This is a point machine condition monitoring device.

According to the fourth invention, since the operation of the switch machine differs for each of the unlocking process, switching process, and locking process, the characteristics of the sounds generated are different. Therefore, by determining whether an abnormality has occurred for each process that divides the switching operation period, it is possible to improve the accuracy of the determination.

A fifth aspect of the present invention is the method according to any one of the second to fourth aspects of the present invention,
The determination means determines whether or not an abnormality has occurred in the point machine by using a time length of the switching operation period.
This is a point machine condition monitoring device.

According to the fifth invention, since the duration of the switching operation period for each switching operation is approximately constant for the same point machine installed in the same location, the presence or absence of a point machine abnormality can be determined using the duration of the switching operation period in the acoustic data. For example, if the duration of the switching operation period in the acoustic data is outside a predetermined time range, it can be determined that a point machine abnormality has occurred.

A sixth aspect of the present invention is the method for producing a liquid crystal display device according to any one of the first to fifth aspects of the present invention,
the determination means determines whether or not an abnormality has occurred in the point machine based on the analysis result in which it was previously determined that there was no abnormality and the current analysis result.
This is a point machine condition monitoring device.

According to the sixth invention, if the same point machine is installed in the same place, the analysis results of the acoustic data for each switching operation under normal conditions will be almost the same, so it is possible to determine whether or not a point machine abnormality has occurred based on the analysis results of the acoustic data that were previously determined to be normal and the current analysis results. For example, if the current analysis results do not match the analysis results of the acoustic data that were previously determined to be normal, it can be determined that a point machine abnormality has occurred.

A seventh aspect of the present invention is the method according to any one of the first to sixth aspects of the present invention,
the determination means determines whether or not an abnormality has occurred in the point machine by using any one of the number of occurrences, the occurrence frequency, and the occurrence timing of sounds corresponding to impulse sounds.
This is a point machine condition monitoring device.

According to the seventh invention, it is possible to determine whether or not a point machine abnormality has occurred based on the number, frequency, and timing of occurrence of impulse sounds in the acoustic data. During the switching operation of the point machine, impulse sounds, which are impact sounds caused by contact between various mechanical parts and various components, are generated. These impulse sounds occur even when the point machine is normal, but they also occur due to rattles caused by wear and damage to the various mechanical parts and various components. Therefore, it is possible to determine whether or not a point machine abnormality has occurred based on whether or not the number, frequency, and timing of occurrence of sounds corresponding to impulse sounds in the acoustic data match the number, frequency, and timing of occurrence of sounds corresponding to impulse sounds in acoustic data that was previously determined to be normal. For example, if they do not match, it can be determined that a point machine abnormality has occurred.

An eighth aspect of the present invention is the method according to any one of the first to seventh aspects of the present invention,
The acoustic analysis means analyzes a sound pressure waveform and a spectrogram of the acoustic data,
the determination means determines whether or not an abnormality has occurred in the point machine based on the sound pressure of the entire signal and the signal strength for each frequency.
This is a point machine condition monitoring device.

According to the eighth invention, the sound pressure waveform and spectrogram of the acoustic data are analyzed, and the presence or absence of an abnormality in the point machine can be determined based on the overall sound pressure of the signal in the sound pressure waveform and the signal strength for each frequency in the spectrogram.

An example of the application of a point machine condition monitoring device. An example of a switch machine configuration. 4 is a flowchart of a process performed by a point machine status monitoring device. An example of the results of acoustic data analysis. 13 is an example of the results of identifying the conversion operation period based on the analysis results. FIG. 2 is a functional configuration diagram of a point machine condition monitoring device.

Below, a preferred embodiment of the present invention will be described with reference to the drawings. Note that the present invention is not limited to the embodiment described below, and the forms to which the present invention can be applied are not limited to the following embodiment. In addition, in the description of the drawings, the same elements are given the same reference numerals.

[Application example]
FIG. 1 shows an application example of a point machine state monitoring device 1 in this embodiment. The point machine state monitoring device 1 is a device that determines whether or not an abnormality has occurred in the point machine 3 based on acoustic data, which is data on the operation sound of the point machine 3 acquired by an acoustic sensor 5 installed in the housing of the point machine 3. In this embodiment, the point machine state monitoring device 1 is configured to be able to acquire acoustic data acquired by one or more acoustic sensors 5 installed in the housing of the point machine 3 via a communication line N. The communication line N may be a wired communication path or a wireless communication path. The point machine state monitoring device 1 may be installed near the point machine 3, or may be installed in a location away from the point machine 3, such as an equipment room or a central control room in a station premises. The acoustic data acquired by the acoustic sensor 5 is output to a wireless or wired communication terminal (not shown) provided in association with the point machine 3, and the acoustic data is transmitted to the point machine state monitoring device 1 by this communication terminal connected to the communication line N.

[Example of switch machine configuration]
Fig. 2 shows an example of a point machine 3. Fig. 2 is a top view showing an example of the configuration of the internal mechanism of the point machine 3, with the top cover of the case 10 removed to show the inside. According to Fig. 2, the point machine 3 includes a circuit controller 32, a control relay 34, an external terminal board 36, a speed reduction mechanism 20, and a switch lock mechanism 22 in an openable and closable case 10 that serves as a housing. Cables required for signals to and from a power source and an external device (e.g., an interlocking device, etc.) are collected on the external terminal board 36 and then pulled out of the case 10 as a cable bundle 38.

The point machine 3 also has an operating rod 16 and a locking rod 18 that are slidably installed through the case 10, and a lock piece 24 (24a, 24b) that is slidably installed in the case 10 and passes through the locking rod 18 in a direction intersecting the locking rod 18. The tongue rail of the switch is connected to the operating rod 16 via a point bar, and the tip of the tongue rail is connected to the locking rod 18 via a connection rod.

The point machine 3 is also equipped with a motor 12 on the outside of the case 10. When a switching command signal is received from the interlocking device, power is supplied to the motor 12 via the control relay 34, and the rotational power generated by the motor 12 is converted to an appropriate torque by the reduction mechanism 20 connected to the drive shaft 14 that penetrates the side of the case 10 and transmitted to the switching lock mechanism 22. The reduction mechanism 20 is a group of gears that receive the driving force of the motor 12, and includes a pinion gear 20a attached to the drive shaft 14 of the motor 12, a bevel gear 20b that meshes with the pinion gear 20a, a first reduction gear 20c attached to the rotation shaft, an intermediate gear 20d that meshes with the pinion gear 20a, a second reduction gear 20e provided on the rotation shaft, and a switching gear 20f that is the final gear that meshes with the pinion gear 20a.

The switching lock mechanism 22 is a mechanism that converts the rotational power decelerated by the reduction mechanism 20 into the linear motion of the operating rod 16 and locks and unlocks the locking rod 18. The switching by the switching lock mechanism 22 is realized by the engagement of the switching roller 22a protruding from the underside of the switching gear 20f with the switching cam groove 22b engraved on the operating rod 16 in a direction intersecting the movement direction of the operating rod 16. That is, the operating direction of the switching roller 22a moves clockwise/counterclockwise depending on the rotation direction of the motor 12, and the switching cam groove 22b engaging with the switching roller 22a can slide the operating rod 16 to the left/right in FIG. 2. The operating rod 16 is connected to the switch adjuster of the switch via the point rod, so the switch can be switched between the normal position and the reverse position by sliding the operating rod 16 to the left/right in FIG. 2. When the tongue rail is moved by the sliding movement of the operating rod 16, the locking rod 18, which is connected to the tongue rail via the front rod and the connecting rod, is moved to a position that corresponds to the opposite position of the tongue rail in the longitudinal direction.

Locking is achieved by the switching roller 22a and the first and second locking plates 22c and 22d, which have locking cam grooves engraved on their upper surfaces to engage with the switching roller 22a. The first locking plate 22c has a lock piece 24a extending toward the locking can 18, and the second locking plate 22d has a lock piece 24b extending toward the locking can 18. As the switching roller 22a moves clockwise/counterclockwise, one of the first and second locking plates 22c and 22d slides away from the locking can 18, and the other slides toward the locking can 18. As a result, one of the lock pieces 24a, 24b slides out of the locking can 18, and the other slides close to and fits into the locking can 18.

The locking rod 18 is connected to the tip of the tongue rail of the turnout via a connecting rod. Therefore, the locking rod 18 slides left or right in FIG. 2 in response to the switch between the normal and reverse positions of the turnout. Then, when either the lock piece 24a or 24b engages with the locking rod 18, the tongue rail after the switch is locked.

The acoustic sensor 5 is installed inside the case 10, which is the housing of the point machine 3. The installation location may be anywhere, but for example, the acoustic sensor 5 can be installed by attaching a circuit board on which the acoustic sensor 5 is mounted to the inside (recess on the back side) of the cover 30 that can be opened and closed to cover an inspection hole, which is an opening for visually inspecting the inside of the case 10 and is formed on the outer part of the case 10 on the side opposite the motor 12.

[process]
The point machine status monitoring device 1 judges whether or not an abnormality has occurred in the point machine 3 based on acoustic data acquired by an acoustic sensor 5 installed in the housing of the point machine 3, and the process related to this judgment will be described below. Fig. 3 is a flowchart explaining the flow of the process performed by the point machine status monitoring device 1. Fig. 3 shows an example of the process targeted at one point machine 3.

As shown in FIG. 3, the point machine condition monitoring device 1 first analyzes the sound pressure waveform and spectrogram of the acoustic data acquired from the acoustic sensor 5 (step S1).

Figure 4 is an example of the analysis results for acoustic data. In Figure 4, the horizontal axis represents a common time (time axis), with the sound pressure waveform shown on the top and the spectrogram shown on the bottom. It also shows the analysis results for acoustic data for a period including one switching operation of point machine 3 where there is no abnormality (normal). In the spectrogram shown in Figure 4, the signal strength (frequency components) for each frequency is shown in shades of gray, with the darker the shade, the stronger the signal strength.

Next, the point machine status monitoring device 1 detects the occurrence timing of the operating sound of the control relay 34 from the sound pressure waveform and spectrogram that are the analysis results, and identifies the switching operation period of the point machine 3 (step S3).

When the point 3 is switched, the contacts of the control relay 34 are turned on (closed) by a switching command (control current) from the interlocking device, and the drive current of the motor 12 is supplied and the switching operation is started. Then, when the locking is completed, the circuit controller 32 turns off (opens) the contacts of the control relay 34, and the supply of drive current to the motor 12 is stopped. In this way, in this embodiment, the operating sound when the contacts of the control relay 34 are turned on and off is detected, and the time of the on operation is set as the timing of the start of the switching, and the time of the off operation is set as the timing of the end of the switching, and the period from the start of the switching to the end of the switching is specified as the switching operation period of the point 3.

The operating sound of the contacts of the control relay 34 is detected by detecting the occurrence timing of sound that matches the relay operation acoustic conditions based on the operating sound of the control relay 34 from the sound pressure waveform and spectrogram that are the analysis results. The relay operation acoustic conditions are conditions that include a short-term sound pressure waveform and spectrogram that correspond to the analysis results of the operating sound of the control relay 34. For example, they can be created based on acoustic data acquired by performing a test operation of the point machine 3 when the point machine 3 is installed or during maintenance and inspection. In addition, since the operating sound generated when the control relay 34 is in on operation and when it is in off operation may differ, relay operation acoustic conditions are determined for each of the on operation and the off operation.

The contacts of the control relay 34 include a normal contact and a reverse contact. In this embodiment, the normal contact and the reverse contact are not distinguished as the contacts of the control relay 34, but the acoustic conditions for relay operation may be determined by distinguishing between them.

Figure 5 is an example of the result of identifying the switching operation period for the analysis result shown in Figure 4. Figure 5 also shows an example of the classification result of each process of the unlocking process, the switching process, and the locking process described later. As shown in Figure 5, various sounds are generated during the switching operation period of the point machine 3, which are mainly caused by the rotation sound of the motor 12 and the movement of each mechanism and various parts such as the reduction mechanism 20 and the switching locking mechanism 22. At the timing when the movement of each mechanism and various parts changes, such as the start and end of switching and the timing of the boundaries of each process, characteristic impact sounds are generated sporadically due to each mechanism and various parts coming into contact with another mechanism and various parts in accordance with the changing movement. Details will be described later, but the point machine status monitoring device 1 detects the timing when the movement of each mechanism and various parts changes based on these generated sounds.

Also, before the start of the conversion operation period, there is almost no sound. However, for a while after the conversion ends, a small sound is generated, but it is slightly louder than the sound before the conversion started. This sound is caused by the rotation of the motor 12 due to inertia and the rotation of the gear group of the reduction mechanism 20. Then, just before these rotations stop and the sound becomes almost silent, a relatively loud impact sound is generated (the impulse-like sound generated at the right end of Figure 5). This sound is generated when a stopper protruding from the operating rod 16 to restrict the movement of the operating rod 16 that penetrates the case 10 hits the inner wall of the case 10.

After identifying the switching operation period of the point machine 3, the point machine status monitoring device 1 then divides the switching operation period into an unlocking process, a switching process, and a locking process (step S5).

The switching operation of the point machine 3 consists of three steps: a locking step in which the motor 12 starts rotating and unlocks the switching locking mechanism 22 when the operating rod 16 is locked and stopped; a switching step in which the switching locking mechanism 22 drives the operating rod 16 to switch the tongue rail until it contacts the base rail, and then the tip of the tongue rail is brought into close contact with the base rail; and a locking step in which the switching locking mechanism 22 is locked, the operating rod 16 is stopped, and the operation of the motor 12 is stopped. The unlocking step and the switching step are distinguished by detecting the occurrence timing of a sound that matches the acoustic conditions based on the operating sound generated at the boundary between the unlocking step and the switching step from the sound pressure waveform and spectrogram, which are the analysis results of the period corresponding to the switching operation period. The switching step and the locking step are distinguished by detecting the occurrence timing of a sound that matches the acoustic conditions based on the operating sound generated at the boundary between the switching step and the locking step from the sound pressure waveform and spectrogram, which are the analysis results of the period corresponding to the switching operation period.

As shown in FIG. 5, when the unlocking process is switched to the conversion process, an impact sound is generated mainly due to the operation bar 16 starting to move. Also, when the conversion process is switched to the locking process, an impact sound is generated mainly due to the lock piece 24 starting to move. The acoustic conditions are conditions including short-term sound pressure waveforms and spectrograms that correspond to the analysis results of these characteristic impact sounds. For example, they can be created based on acoustic data acquired by test operating the point machine 3 when installing the point machine 3 or during maintenance and inspection, and include conditions related to thresholds and changes in sound intensity (signal strength) for each frequency that can determine each timing, and conditions related to thresholds and changes in the overall sound pressure level.

Then, the point machine status monitoring device 1 determines whether or not an abnormality has occurred in the point machine 3 based on the duration of the switching operation period (step S7).

If the same point machine 3 is installed in the same location, the length of time required for one switching operation is approximately constant, so a time range that includes the switching time defined in the specifications of that point machine 3 is defined as the time length condition, and the presence or absence of an abnormality in the point machine 3 is determined based on whether the length of the switching operation period satisfies this time length condition.

The point machine status monitoring device 1 also determines whether or not an abnormality has occurred in the point machine 3 based on the sound pressure waveform and spectrogram that are the results of the current analysis and the sound pressure waveform and spectrogram that are the results of past analysis in which no abnormality was found (step S9). This determination is made for each of the divided processes of the unlocking process, the switching process, and the locking process.

If the same switch machine 3 is installed in the same location, the acoustic data obtained for each switching operation will be approximately the same, and the sound pressure waveform and spectrogram that are the analysis results will also be approximately the same. For this reason, for example, the presence or absence of an abnormality in the switch machine 3 can be determined by comparing the sound pressure waveform and spectrogram that are the current analysis results with the sound pressure waveform and spectrogram that are the analysis results from a previous analysis in which the switch machine 3 installed in the same location was determined to be normal and determining whether they match.

The past analysis results that are used for the judgment may be the analysis results of the sound data related to any one conversion operation, or may be the results of statistical processing such as averaging the analysis results of the sound data related to each of the multiple conversion operations. Any method may be adopted for the comparison. For example, a graphical similarity judgment is adopted to calculate the similarity between the current analysis result and the past reference analysis result. If the similarity satisfies a threshold condition, it is determined that there is a match and that there is no abnormality, and if the threshold condition is not satisfied, it is determined that there is an abnormality. In addition, a feature analysis may be performed on each of the current analysis result and the past reference analysis result, and the presence or absence of an abnormality may be determined based on the difference in the features. For example, the time during which a sound pressure above a predetermined threshold is continuously generated in the conversion process or the frequency at which a signal above a predetermined threshold is generated is analyzed as a feature, and the presence or absence of an abnormality is determined based on the difference in the analyzed features. If a large signal occurs at a specific frequency for each cause of the abnormality, the cause of the abnormality can be estimated based on the signal conditions for each cause.

In addition, if the same point machine 3 is installed in the same location, the length of the switching operation period is approximately constant, and the length of each process is also approximately constant. Therefore, by making a judgment for each process of the unlocking process, switching process, and locking process, the influence of errors that occur when identifying the switching operation period and dividing each process is reduced, making it possible to make a more accurate judgment.

The point machine condition monitoring device 1 also detects the occurrence of impulse sounds from the analysis results, and determines whether or not an abnormality has occurred in the point machine 3 by using any of the number of occurrences, frequency of occurrence, and occurrence timing of sounds corresponding to impulse sounds (step S11).

During the switching operation of the point machine 3, impulse sounds are generated, which are impact sounds caused by contact between various mechanical parts and components. As shown in FIG. 5, these impulse sounds are generated even when the point machine 3 is normal, but they can also be generated due to rattles caused by wear and damage to the various mechanical parts and components. For this reason, the occurrence of impulse sounds is detected from the analysis results, and by determining whether the number, frequency, and timing of occurrence of these impulse sounds match those in past analysis results that determined there was no abnormality, it is possible to determine whether an abnormality has occurred in the point machine 3. If the number, frequency, and timing of occurrence differ for each cause of the abnormality, the cause of the abnormality can be estimated based on the signal conditions for each cause.

After performing such analysis and judgment, the switch status monitoring device 1 records and outputs the analysis and judgment results (step S13). After performing the above processing, this process ends.

[Functional configuration]
Fig. 6 shows an example of a functional configuration of the point status monitoring device 1. As shown in Fig. 6, the point status monitoring device 1 includes, as functional units, an operation unit 102, a display unit 104, a communication unit 106, a processing unit 200, and a storage unit 300, and can be configured as a kind of computer or CPU board.

The operation unit 102 is composed of input devices such as a keyboard, mouse, touch panel, and various switches, and outputs operation signals corresponding to operations performed to the processing unit 200. The display unit 104 is composed of a display device such as a liquid crystal display or touch panel, and performs various displays corresponding to display signals from the processing unit 200. The communication unit 106 is composed of a wired or wireless communication device, and is connected to a communication line N to communicate with an external device (e.g., the acoustic sensor 5).

The processing unit 200 is realized by a calculation device such as a CPU (Central Processing Unit), and performs overall control of the point machine status monitoring device 1 by issuing instructions and transferring data to each component of the point machine status monitoring device 1 based on the programs and data stored in the memory unit 300. The processing unit 200 also executes the status monitoring program 302 stored in the memory unit 300, thereby functioning as each of the functional blocks of the acoustic analysis unit 202, the identification unit 204, and the determination unit 206. These functional blocks can also be configured as independent calculation circuits using an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array), etc.

The acoustic analysis unit 202 analyzes the sound pressure waveform and spectrogram of the acoustic data acquired by the acoustic sensor 5 installed in the housing of the point machine 3. The acoustic data acquired by the acoustic sensor 5 is stored as acquired acoustic data 317 for each point machine 3. The acoustic analysis unit 202 analyzes this acquired acoustic data 317. The sound pressure waveform and spectrogram that are the analysis results are stored as analysis result data 318 for the corresponding point machine 3.

The identification unit 204 detects the occurrence timing of sound that matches the relay operation acoustic conditions based on the operating sound of the control relay 34 of the point 3 from the analysis results of the acoustic analysis unit 202, and thereby identifies the switching operation period of the point 3 in the acoustic data.

Specifically, from the sound pressure waveform and spectrogram that are the analysis results, the occurrence timing of a sound that matches the relay ON operation acoustic condition 312 based on the operating sound when the control relay 34 is turned ON is detected and determined as the timing of the start of the changeover. Also, from the sound pressure waveform and spectrogram that are the analysis results, the occurrence timing of a sound that matches the relay OFF operation acoustic condition 313 based on the operating sound when the control relay 34 is turned OFF is detected and determined as the timing of the end of the changeover. Then, the period from the start of the changeover to the end of the changeover is identified as the changeover operation period of the point machine 3 (see FIG. 5).

The relay-on operation acoustic conditions 312 include conditions for sound pressure in a short-time sound pressure waveform corresponding to the analysis result of the operating sound when the control relay 34 is in an on-operation state, and conditions for the signal strength (frequency components) for each frequency in the spectrogram. Similarly, the relay-off operation acoustic conditions 313 include conditions for sound pressure in a short-time sound pressure waveform corresponding to the analysis result of the operating sound when the control relay 34 is in an off-operation state, and conditions for the signal strength (frequency components) for each frequency in the spectrogram.

The identification unit 204 also divides the switching operation period into an unlocking process, a switching process, and a locking process by detecting the timing of occurrence of sounds that match both acoustic conditions based on the operating sounds generated at the boundary between the unlocking process and the switching process of the point machine 3 and acoustic conditions based on the operating sounds generated at the boundary between the switching process and the locking process.

Specifically, among the sound pressure waveforms and spectrograms that are the analysis results for the period corresponding to the conversion operation period, the timing of occurrence of sound that matches the unlocking/conversion process boundary acoustic condition 314 is detected to distinguish between the unlocking process and the conversion process. In addition, among the sound pressure waveforms and spectrograms that are the analysis results for the period corresponding to the conversion operation period, the timing of occurrence of sound that matches the conversion/locking process boundary acoustic condition 315 is detected to distinguish between the conversion process and the locking process (see FIG. 5).

The unlocking/conversion process boundary acoustic conditions 314 include conditions for the sound pressure in the sound pressure waveform corresponding to the analysis result of the sound generated at the boundary between the unlocking process and the conversion process, and conditions for the signal strength (frequency components) for each frequency in the spectrogram. Similarly, the conversion/locking process boundary acoustic conditions 315 include conditions for the sound pressure in the sound pressure waveform corresponding to the analysis result of the sound generated at the boundary between the conversion process and the locking process, and conditions for the signal strength (frequency components) for each frequency in the spectrogram.

The determination unit 206 determines whether or not an abnormality has occurred in the point machine 3 for each of the unlocking process, the switching process, and the locking process based on the analysis results of the acoustic analysis unit 202 related to the switching operation period. The determination of whether or not an abnormality has occurred in the point machine 3 is made based on the analysis results from the past in which no abnormality was determined, and the current analysis results. The determination is also made based on the sound pressure of the entire signal and the signal strength for each frequency.

Specifically, the sound pressure waveform and spectrogram resulting from this analysis are compared with sound pressure waveforms and spectrograms resulting from previous analysis results of the point machine 3 that were determined to be abnormal, and the presence or absence of an abnormality in the point machine 3 is determined based on whether or not they match. The past analysis results are stored as analysis results in the conversion data 320 for the point machine 3.

The past analysis results that show no abnormalities used for the judgment may be the analysis results of the acoustic data related to any one switching operation, or may be the results of performing statistical processing such as averaging on the analysis results of the acoustic data related to each of multiple switching operations. This is because, if the same point machine 3 is installed in the same location, the acoustic data obtained for each switching operation will be approximately the same, and the sound pressure waveform and spectrogram that are the analysis results will also be approximately the same.

The determination unit 206 also uses any one of the number of occurrences, frequency, and timing of occurrence of sounds corresponding to impulse sounds to determine whether or not an abnormality has occurred in the point machine 3. Specifically, the determination unit 206 detects the occurrence of impulse sounds from the sound pressure waveform and spectrogram, which are the analysis results, and determines whether or not the number of occurrences, frequency, and timing of occurrence match those of impulse sounds in past analysis results that determined no abnormality, thereby determining whether or not an abnormality has occurred in the point machine 3.

The determination unit 206 also uses the time length of the switching operation period to determine whether or not an abnormality has occurred in the point 3. Specifically, it determines whether or not the time length of the switching operation period satisfies the switching operation time length condition 316, and if not, it determines that an abnormality has occurred in the point 3. The switching operation time length condition 316 is a condition for a time range that includes the switching time defined as the specifications of the point 3. The switching operation time length condition 316 may further include a condition for the time length of each process of the unlocking process, the switching process, and the locking process. Then, it may be possible to determine whether or not the time length of each process in the switching operation period satisfies this condition to determine whether or not an abnormality has occurred in the point 3.

The memory unit 300 is realized by a storage device such as a hard disk, ROM (Read Only Memory), or RAM (Random Access Memory), and stores programs and data for the processing unit 200 to comprehensively control the point machine status monitoring device 1. It is also used as a working area for the processing unit 200, and temporarily stores the results of calculations performed by the processing unit 200 according to various programs, and input data via the operation unit 102 and communication unit 106. In this embodiment, the memory unit 300 stores a status monitoring program 302 and point machine data 310.

The status monitoring program 302 is a program executed by the processing unit 200 to realize a process (see FIG. 3) for determining whether or not an abnormality has occurred in the point machine 3 based on the acoustic data acquired by the acoustic sensor 5.

The point data 310 is generated for each point 3 monitored by the point status monitoring device 1, and stores relay-on operation acoustic conditions 312, relay-off operation acoustic conditions 313, unlocking/switching process boundary acoustic conditions 314, switching/locking process boundary acoustic conditions 315, switching operation duration conditions 316, acquired acoustic data 317, analysis result data 318, and switching data 320 in association with the point ID 311 of the point 3. The switching data 320 is data related to each switching operation, and includes the acquisition date and time of the acoustic data corresponding to the switching operation, the analysis result for the acoustic data, the switching operation period identified by the identification unit 204, and the judgment result of the presence or absence of an abnormality by the judgment unit 206. The analysis result includes a sound pressure waveform and a spectrogram.

[Action and Effect]
In this manner, the point machine status monitoring device 1 of the present embodiment judges whether or not an abnormality has occurred in the point machine 3 based on the acoustic data acquired by the acoustic sensor 5 installed in the housing of the point machine 3. The point machine status monitoring device 1 can provide a new technique for judging whether or not an abnormality has occurred in the point machine 3. That is, by detecting the operation sound of the control relay 34 from the analysis result of the acoustic data, which is data on the operation sound of the point machine 3, the switching operation period of the point machine 3 in the acoustic data can be specified. In addition, the switching operation period can be divided into an unlocking process, a switching process, and a locking process. Then, based on the analysis result of the acoustic data that was previously judged to be abnormal and the current analysis result, it is possible to judge whether or not an abnormality has occurred in the point machine 3. In addition, by judging whether or not an abnormality has occurred in the point machine 3 for each process into which the switching operation period is divided, it is possible to improve the accuracy of the judgment.

The applicable embodiments of the present invention are not limited to the above-mentioned embodiments, and can of course be modified as appropriate without departing from the spirit of the present invention.

(A) Division of Processes In the above-described embodiment, each process during the conversion operation period is divided based on the sound generated at the boundary between each process, but this may be achieved by other methods or by combining methods.

For example, the start of switching of the point machine 3 may be determined by a change in the switching control conditions from the interlocking device, and each process may be divided according to the time elapsed from the start of this switching. This is because, for the same point machine 3 installed in the same location, the switching time and the length of time for each process are approximately constant.

Alternatively, if a voltage sensor or current sensor is used to measure the motor current, motor voltage, etc., in the point machine 3, the start and end of switching in the point machine 3 and the boundaries of each process can be determined based on changes in the measured values.

Alternatively, if the motor 12 is a servo motor or is equipped with an optical or magnetic sensor that detects the amount of movement of the operating rod 16 or the amount of rotation of the gears of the reduction mechanism 20, and the stroke position, which is the displacement position of the operating rod 16, can be obtained, the start and end of conversion of the point machine 3 and the boundaries of each process can be determined based on the stroke position.

Alternatively, if a camera capable of photographing the inside of the housing of the point machine 3 is provided, the start and end of conversion of the point machine 3 and the boundaries of each process may be determined based on the operation of each part determined from an analysis of the images captured by the camera. Furthermore, if a camera capable of photographing the inside of the housing of the point machine 3 is provided, the analysis results of the images captured by the camera can be used as a clue to deduce the cause of the generated sound.

(B) Installation Position of Acoustic Sensor 5 In the above embodiment, the acoustic sensor 5 is installed inside the housing of the point machine 3, but it may be installed in a nearby position, such as around or near the point machine 3, instead of inside the housing. In this case, by giving directionality to the acoustic sensor 5, it is possible to acquire acoustic data, which is data on the operating sound of the point machine 3, in the same way as when the acoustic sensor 5 is installed inside the housing of the point machine 3. In addition, since the acoustic sensor 5 is not installed on the point machine 3, it is possible to eliminate the influence of vibrations transmitted to the point machine 3, such as train vibrations.

Reference Signs List 1: Point machine status monitoring device 200: Processing unit 202: Acoustic analysis unit 204: Identification unit 206: Determination unit 300: Storage unit 302: Status monitoring program 310: Point machine data 311: Point machine ID
312... Relay-on operation acoustic condition 313... Relay-off operation acoustic condition 314... Unlocking/switching process boundary acoustic condition 315... Switching/locking process boundary acoustic condition 316... Switching operation time length condition 317... Acquired acoustic data 318... Analysis result data 320... Switching data 3... Point machine 10... Case (housing)
34: Control relay 5: Acoustic sensor

Claims (9)

an acoustic analysis means for analyzing acoustic data, which is data on the operation sound of the point machine acquired by the acoustic sensor;
an identification means for identifying a switching operation period of the point machine in the acoustic data by using an analysis result of the acoustic analysis means;
a determination means for determining whether or not an abnormality has occurred in the point machine by using an analysis result of the acoustic analysis means relating to the switching operation period ;
A point machine condition monitoring device comprising: the identifying means identifies the switching operation period by detecting a generation timing of a sound that matches a relay operation acoustic condition based on an operation sound of the control relay of the point machine from the analysis result of the acoustic analysis means.
The point condition monitoring device according to claim 1 . the specifying means detects occurrence timings of sounds that match respective acoustic conditions based on operation sounds generated at the boundary between the unlocking process and the switching process of the point machine and acoustic conditions based on operation sounds generated at the boundary between the switching process and the locking process, thereby dividing the switching operation period into an unlocking process, a switching process, and a locking process;
The determination means determines whether or not an abnormality has occurred in the point machine for each of the unlocking process, the switching process, and the locking process.
The point condition monitoring device according to claim 1 or 2 . an acoustic analysis means for analyzing acoustic data, which is data on the operation sound of the point machine acquired by the acoustic sensor;
an identification means for identifying a switching operation period of the point machine in the acoustic data by using an analysis result of the acoustic analysis means;
A determination means for determining whether or not an abnormality has occurred in the point machine by using the time length of the switching operation period ;
A point machine condition monitoring device comprising: the determination means determines whether or not an abnormality has occurred in the point machine based on the analysis result in which it was previously determined that there was no abnormality and the current analysis result.
The point machine status monitoring device according to any one of claims 1 to 4 . the determination means determines whether or not an abnormality has occurred in the point machine by using any one of the number of occurrences, the occurrence frequency, and the occurrence timing of sounds corresponding to impulse sounds.
The point machine status monitoring device according to any one of claims 1 to 5 . The acoustic analysis means analyzes a sound pressure waveform and a spectrogram of the acoustic data,
the determination means determines whether or not an abnormality has occurred in the point machine based on the sound pressure of the entire signal and the signal strength for each frequency.
The point machine status monitoring device according to any one of claims 1 to 6 . an acoustic analysis step of analyzing acoustic data, which is data on the operation sound of the point machine acquired by the acoustic sensor;
an identifying step of identifying a switching operation period of the point machine in the acoustic data by using an analysis result in the acoustic analysis step;
a determination step of determining whether or not an abnormality has occurred in the point machine by using an analysis result in the acoustic analysis step related to the switching operation period ;
A point machine condition monitoring method comprising: an acoustic analysis step of analyzing acoustic data, which is data on the operation sound of the point machine acquired by the acoustic sensor;
an identifying step of identifying a switching operation period of the point machine in the acoustic data by using an analysis result in the acoustic analysis step;
a determination step of determining whether or not an abnormality has occurred in the point machine by using a time length of the switching operation period ;
A point machine condition monitoring method comprising:

Images