JP7330283B2 - Condition monitoring device, condition monitoring system and condition monitoring method - Google Patents

Description

The present invention relates to a state monitoring device, a state monitoring system, and a state monitoring method for monitoring the state of a target device based on acoustic emission (hereinafter referred to as AE) waves detected from the target device.

2. Description of the Related Art Conventionally, there is a technique of detecting an abnormality in a target device by monitoring the state of the target device based on AE waves detected from the target device. For example, the abnormality detection device described in Patent Document 1 discriminates an AE signal having a frequency component within a certain range from an AE signal detected from a rotary bearing by an AE sensor, and the level of the discriminated AE signal is greater than a threshold value. The duration from when the level becomes smaller than the threshold is calculated, and the rise time from when the level of the AE signal becomes larger than the threshold to when it reaches the maximum value is calculated. Then, the abnormality detection device determines that the calculated duration is longer than the first time and the calculated rise time is longer than a second time that is shorter than the first time, based on the number of times that the rotary bearing is determined. to detect anomalies in

JP 2006-226731 A

For example, in order to effectively perform maintenance on a target device such as a machine tool, it is necessary to continuously monitor changes in various states of the target device to determine if the target device is in a state (deterioration or failure) that requires maintenance. You need to know exactly what happened. However, the abnormality detection device described in Patent Document 1 detects a certain abnormal state that satisfies the conditions using the duration and rise time of the AE wave detected from the target device. It was not possible to monitor fluctuations in various conditions.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a state monitoring device, a state monitoring system, and a state monitoring method capable of monitoring various state fluctuations of a target device.

A state monitoring device according to the present invention receives a sine wave signal from an AE sensor that detects an acoustic emission wave that varies according to the state of a target device and outputs a sine wave signal of the acoustic emission wave. to digital data, a data extraction unit for extracting extreme values from the digital data, and the digital data to calculate the value of the parameter indicating the state of temporal fluctuation of the acoustic emission wave. A parameter calculation unit, a matrix generation unit that generates matrix data in which data is associated with a combination of a plurality of parameter ranges of parameter values and a plurality of extreme value ranges of extreme values, and matrix data generated by the matrix generation unit. and an output processing unit that outputs a parameter based on a portion of digital data in which extreme values belonging to one of a plurality of extreme value ranges are continuously extracted. , and the matrix generation unit adds a combination of the extreme value range corresponding to the part and the parameter range to which the parameter value calculated based on the part among the plurality of parameter ranges belongs to Associate the number of occurrences of extreme values as data .

According to the present invention, it has a plurality of regions in which a plurality of ranges of parameter values and a plurality of ranges of extreme values are associated, and the range of parameter values calculated by the parameter calculation unit and the data extraction unit Matrix data in which the number of occurrences of the extreme value is counted in the region corresponding to the range of the extracted extreme value is generated and output. The number of occurrences of extreme values in the matrix data corresponds to the number of occurrences of the state of the AE wave indicated by the parameter value and the state of the target device corresponding to the extreme value of the AE wave. As a result, by referring to the matrix data, it is possible to grasp the transition of the number of occurrences of the state of the target device, and to monitor various changes in the state of the target device.

1 is a block diagram showing the configuration of a condition monitoring system according to Embodiment 1; FIG. FIG. 2A is a diagram showing Example 1 of the output waveform of the AE sensor, and FIG. 2B is a diagram showing digital data of the output signal of the AE sensor. 4 is a flowchart showing a state monitoring method according to Embodiment 1; FIG. 10 is a diagram showing Example 2 of an output waveform of an AE sensor; It is a figure which shows the example of matrix data. FIG. 10 is a diagram showing example 3 of an output waveform of an AE sensor; FIG. 4 is a diagram showing an example of a matrix image; FIG. 7 is a diagram showing an example of extraction processing of local maximum values according to Embodiment 1; 9A is a block diagram showing a hardware configuration that implements the functions of the condition monitoring device according to Embodiment 1, and FIG. 9B shows a hardware configuration that implements the functions of the condition monitoring device according to Embodiment 1. FIG. 3 is a block diagram showing a hardware configuration; FIG.

Embodiment 1.
FIG. 1 is a block diagram showing the configuration of a condition monitoring system 1 according to Embodiment 1. As shown in FIG. A state monitoring system 1 is a system for monitoring the state of a target device 2 based on vibrations generated in the target device 2. As shown in FIG. Prepare. The target device 2 is, for example, a rotating machine such as a motor, a speed reducer, a cutting machine, a pump and a turbine. The state monitoring device 4 includes an A/D converter 41 , a data extractor 42 , a parameter calculator 43 , a matrix generator 44 and an output processor 45 .

The AE sensor 3 is attached to the target device 2 and detects AE waves corresponding to vibrations generated in the target device 2 . AE is a phenomenon in which elastic energy stored inside a solid is released as an elastic wave (AE wave) when the solid deforms or breaks. Since the vibration generated in the target device 2 fluctuates according to the state of the target device 2 , it can be said that the AE wave fluctuates according to the state of the target device 2 .

Also, the AE sensor 3 has a cantilever structure that outputs a sine wave signal of the AE wave detected from the target device 2 . A cantilever structure is an oscillating structure made of a piezoelectric material with a high Q value, and has, for example, a plurality of cantilevers. A resonance frequency included in the frequency band of the AE wave is set for each of the plurality of cantilevers. When the target device 2 is a rotating machine, the cantilever structure in the AE sensor 3 detects AE waves corresponding to the resonance frequency among wideband AE waves (frequency components from several kHz to several MHz) generated by the rotation of the rotating machine. is detected, and a sine wave signal of the detected AE wave is output.

2A is a diagram showing an output waveform example 1 of the AE sensor 3, and FIG. 2B is a diagram showing digital data of the output signal of the AE sensor 3. FIG. During the period A shown in FIG. 2A, the vibration generated in the target device 2 is small, and the AE wave detected from the target device 2 by the AE sensor 3 at this time is also small. For example, if the target device 2 is a cutting machine, the vibration generated in the cutting machine is small during the period in which the spindle to which the cutting blade is attached is idle.

During the period B, the vibration generated in the target device 2 is larger than during the period A, and the AE wave detected from the target device 2 by the AE sensor 3 at this time also has a large amplitude. For example, in a cutting machine, when a cutting blade starts cutting a workpiece, the vibration generated between the cutting blade and the workpiece suddenly increases, and accordingly the amplitude of the AE wave also increases. Change.

Δv1 is a value obtained by subtracting the maximum signal level of the sine wave signal of the AE wave in the period A from the maximum value of the signal level of the sine wave signal of the AE wave in the period B. This means that the change of the target device 2 from the state in the period A to the state in the period B causes the signal level of the sine wave signal of the AE wave to change by Δv1 at maximum.

Digital data obtained by A/D converting the output signal of the AE sensor 3 in period B is time-series data of a plurality of data points, as shown in FIG. 2B. In A/D conversion, variation in the timing of sampling data points from an analog signal generally causes measurement errors in the cycle-to-cycle peak value of a sine wave. For example, among the digital data for each cycle of the sinusoidal signal shown in FIG. 2B, the maximum data points M1 to M4 have a measurement error of Δv2 (=signal level of M4−signal level of M3). In the sine wave signal of the AE wave output from the AE sensor 3, the amount of change Δv1 in the signal level of the AE wave is sufficiently larger than the measurement error Δv2 of the peak value.

The state monitoring device 4 receives the sine wave signal of the AE wave detected from the target device 2 by the AE sensor 3, converts the input sine wave signal of the AE wave into digital data as shown in FIG. Extrema (maxima or minima) data points are extracted from the digital data of the sinusoidal signal of the AE wave. As described above, in the sine wave signal of the AE wave output from the AE sensor 3, the amount of change Δv1 in the signal level of the AE wave is sufficiently larger than the measurement error Δv2 of the peak value. Without applying envelope processing to the output waveform of the AE sensor 3, data points of extreme values can be accurately extracted from the digital data of the sine wave signal of the AE wave. In the following, unless otherwise specified, "extracting extreme data points" is simply referred to as "extracting extreme values."

The state monitoring device 4 uses the digital data of the sine wave signal of the AE wave to calculate the value of the parameter indicating the state of the AE wave. The state of the AE wave is the state of temporal change in the signal level of the AE wave. Then, the state monitoring device 4 generates matrix data using the parameter values calculated using the digital data of the sine wave signal of the AE wave and the extreme values extracted from the digital data of the sine wave signal of the AE wave. do. The matrix data is data having a plurality of areas in which a plurality of ranges of parameter values and a plurality of ranges of extreme values are associated with each other. The number of occurrences of the extreme value is counted in the region where the range of the parameter value calculated using the AE wave corresponds to the range of the extreme value extracted from the digital data of the sinusoidal signal of the AE wave. The number of occurrences of extreme values corresponds to the number of occurrences of states of the target device corresponding to the state of the AE wave indicated by the parameter value and the extreme values generated in the AE waves. Therefore, by referring to the matrix data, it is possible to grasp the transition of the number of occurrences of the state of the target device, and to monitor the fluctuation of various states of the target device.
In the following description, a case of extracting a maximum value as an extreme value from digital data of a sine wave signal of an AE wave will be described. When extracting a minimum value, the sign of the extracted value can be reversed so that it can be handled in the same way as a maximum value.

The output unit 5 visually or audibly outputs the information input from the condition monitoring device 4 . The output unit 5 is, for example, a display unit such as a display or an audio output unit such as a speaker. The state monitoring device 4 and the output unit 5 are connected by wire or wirelessly. Information input from the state monitoring device 4 to the output unit 5 includes, for example, alert information indicating that the target device 2 behaves differently from the normal state, and AE wave detection from the target device 2 by the AE sensor 3. It is matrix image information in which digital data or matrix data of a sine wave signal is imaged.

FIG. 3 is a flowchart showing a state monitoring method according to Embodiment 1, showing a series of processes by the state monitoring device 4 shown in FIG. The AE sensor 3 detects AE waves corresponding to vibrations generated in the target device 2 . The A/D converter 41 receives the sine wave signal of the AE wave detected by the AE sensor 3, and converts the input sine wave signal into digital data (step ST1). Digital data converted by the A/D converter 41 is output to the data extractor 42 and the parameter calculator 43 .

The data extractor 42 extracts the maximum value from the digital data converted by the A/D converter 41 (step ST2). For example, the data extracting unit 42 determines a maximum value larger than a threshold among the maximum values of each cycle in the digital data of the sine wave signal of the AE wave, and among the determined maximum values, each of the maximum values in the matrix data Extract what is contained in a range. The maximum value extracted by the data extractor 42 is output to the parameter calculator 43 . When extracting the minimum value from the digital data of the sine wave signal of the AE wave, the data extraction unit 42 inverts the sign of the extracted minimum value before outputting it to the parameter calculation unit 43 . The sine wave signal of the AE wave is mainly superimposed with noise that changes on the positive side of the sine wave. can be reduced.

The threshold value for extracting the maximum value is determined using the maximum value for each cycle in the digital data of the sine wave signal of the AE wave acquired when the target device 2 is in a normal state. For example, the state in period A shown in FIG. 2A before the target device 2 deteriorates over time is regarded as a normal state. The A/D converter 41 converts the sine wave signal of the AE wave detected from the target device 2 by the AE sensor 3 into digital data when the target device 2 is in a normal state. The data extraction unit 42 extracts the maximum value for each cycle from the digital data converted by the A/D conversion unit 41, and sets the average value of the extracted maximum values as the threshold value.

The parameter calculator 43 calculates the value of the parameter indicating the state of the AE wave using the digital data of the sine wave signal of the AE wave (step ST3). The parameter indicating the state of the AE wave is a parameter indicating the state of temporal fluctuation of the signal level of the AE wave. Parameters that indicate the state of the AE wave include, for example, rise time, duration, interval between occurrences of local maxima, and period energy. Note that any parameters other than these may be used as long as they indicate the state of temporal fluctuations in the signal level of the AE wave.

FIG. 4 is a diagram showing Example 2 of the output waveform of the AE sensor 3. In FIG. The output waveform shown in FIG. 4 is the output waveform (voltage waveform) of the AE wave detected from the target device 2 by the AE sensor 3 when the state of the target device 2 fluctuates. For example, in a rotating machine, when some kind of impact is applied to a rotating shaft, the vibration generated in the rotating machine due to the rotation of the shaft abruptly changes, and gradually returns to the original state over time. The waveform of the AE wave also changes as shown in FIG. 4 in accordance with this change in vibration. Note that FIG. 4 shows only the plus side of the output waveform of the AE sensor 3 .

The rise time Tr is the time until the data extraction unit 42 extracts the maximum maximum value within each range of maximum values in the matrix data. In FIG. 4, the range of maximum values in the matrix data is the range of voltage V0 or more and less than voltage V1. The parameter calculator 43 calculates the rise time Tr until the data extractor 42 extracts the maximum and maximum value A within the range of voltage V0 or more and less than voltage V1.

For example, the parameter calculation unit 43 calculates the time t0 at which the data extraction unit 42 extracts the maximum value of the voltage V0 from the digital data of the sine wave signal of the AE wave and the time t1 at which the data extraction unit 42 extracts the value A. Tr, which is a time interval obtained by subtracting time t0 from time t1, is calculated.

The duration Tk is the time during which the data extraction unit 42 continues to extract values within each range of maximum values in the matrix data. For example, the parameter calculation unit 43 determines the time t0 at which the data extraction unit 42 extracts the maximum value of the voltage V0 from the digital data of the sine wave signal of the AE wave, and the data extraction unit 42 extracts The time t2 at which the maximum value below the deviating voltage V0 was extracted is obtained, and Tk, which is the time interval obtained by subtracting the time t0 from the time t2, is calculated.

The generation interval of the maximum value is the generation interval at which the maximum value extracted by the data extraction unit 42 occurs in each range of the maximum value in the matrix data.
For example, the parameter calculation unit 43 determines the time when the data extraction unit 42 extracts the maximum value of voltage V0 or more from the digital data of the sine wave signal of the AE wave, and and the time when the next maximum value of is extracted are calculated, and the occurrence interval, which is the time interval between these times , is calculated, and the number of occurrences of the maximum value during the occurrence interval is counted .

The period energy is the waveform area of the sinusoidal signal of the AE wave in a certain period. For example, the parameter calculator 43 calculates the waveform area of the sine wave signal of the AE wave during the duration Tk as the period energy during the duration Tk.

Next, the matrix generation unit 44 generates matrix data in which the number of occurrences of local maximum values is counted (step ST4). FIG. 5 is a diagram showing an example of matrix data. In FIG. 5, matrix data (1) is data having a plurality of regions in which a plurality of ranges of rise times Tr and a plurality of ranges of maximum values are associated with each other. Matrix data (2) is data having a plurality of regions in which a plurality of ranges of duration Tk and a plurality of ranges of maximum values are associated with each other. Matrix data (3) is data having a plurality of regions in which a plurality of ranges of occurrence intervals Ti of maximum values and a plurality of ranges of maximum values are associated with each other. Matrix data (4) is data having a plurality of regions in which a plurality of ranges of period energy Ve and a plurality of ranges of maximum values are associated with each other.

The plurality of ranges of maximum values in the matrix data (1) to (4) are the maximum range (V4-) defined by the maximum value of the maximum values and the maximum value obtained when the target device 2 is in a normal state. The minimum range (V0-V1) defined by the average value of the values, and the intermediate range (V1-V2, V2-V3, V3-V4) defined by equally dividing the minimum range and the maximum range . For example, the maximum range (from V4) is defined by the maximum value (voltage V4) that can be taken by the maximum value extracted by the data extraction unit 42 when the target device 2 is in a normal state. It is a voltage range of V4 or higher. The minimum range (V0 to V1) is equal to or greater than the average value (voltage V0) of the maximum values of the sinusoidal signals of the AE waves detected by the AE sensor 3 when the target device 2 is in a normal state. is a voltage range less than V1, which is higher than V1.

The intermediate ranges (V1-V2, V2-V3, V3-V4) are, for example, equally divided ranges between the minimum range (V0-V1) and the maximum range (V4-). The intermediate range (V1 to V2) is a voltage range equal to or higher than voltage V1 and lower than voltage V2, which is higher than voltage V1. The intermediate range (V2 to V3) is a voltage range equal to or higher than voltage V2 and lower than voltage V3, which is higher than voltage V3. The intermediate range (V3 to V4) is a voltage range equal to or higher than voltage V3 and lower than voltage V4, which is higher than voltage V3.

In FIG. 5, the range of the rise time Tr in the matrix data (1) includes a time range (0 to Tr1) in which the rise time Tr is 0 or more and less than Tr1, and a time range (0 to Tr1) in which the rise time Tr is Tr1 or more and less than Tr2, which is longer than Tr1. , a time range (Tr2 to Tr3) in which the rise time Tr is equal to or greater than Tr2 and less than Tr3, which is longer than Tr2, and a time range (Tr3 to Tr3) in which the rise time Tr is equal to or greater than Tr3.

For example, when the rise time Tr calculated by the parameter calculation unit 43 is included in the time range (Tr1 to Tr2) and the maximum value extracted by the data extraction unit 42 is included in the intermediate range (V2 to V3), the matrix The generator 44 increments the number of occurrences of the local maximum value by +1 in the area corresponding to the range (Tr1 to Tr2) and the intermediate range (V2 to V3) in the matrix data (1). In the matrix data (1) shown in FIG. 5, the maximum number of occurrences is 10 in the region where the range (0 to Tr1) and the minimum range (V0 to V1) correspond. In the region where the range (Tr1-Tr2) and the intermediate range (V1-V2) correspond, the number of occurrences of the maximum value is 10, and in the region where the range (Tr1-Tr2) and the intermediate range (V2-V3) correspond The number of occurrences of the maximum value is 1000, and the number of occurrences of the maximum value is 10 in the region where the range (Tr1 to Tr2) and the intermediate range (V3 to V4) correspond. In the region where the range (Tr2-Tr3) and the intermediate range (V2-V3) correspond, the number of occurrences of the maximum value is 10, and in the region where the range (Tr2-Tr3) and the intermediate range (V3-V4) correspond The number of occurrences of the maximum value is ten. In addition, the number of occurrences of local maximum values is 0 in regions other than the above.

In addition, the range of values of the duration Tk in the matrix data (2) includes a time range (0 to Tk1) in which the duration Tk is 0 or more and less than Tk1, a time range (0 to Tk1) in which the duration Tk is Tk1 or more and less than Tk2, which is longer than Tk1. , a range (Tk2 to Tk3) in which the duration Tk is equal to or greater than Tk2 and less than Tk3, which is longer than Tk2, and a range (Tk3 to) in which the duration Tk is equal to or greater than Tk3.

For example, when the duration Tk calculated by the parameter calculation unit 43 is included in the time range (Tk1 to Tk2) and the maximum value extracted by the data extraction unit 42 is included in the intermediate range (V2 to V3), the matrix The generation unit 44 counts up the number of occurrences of the maximum value by +1 in the region where the range (Tk1 to Tk2) and the range (V2 to V3) in the matrix data (2) correspond. In the matrix data (2) shown in FIG. 5, the maximum number of occurrences is 10 in the region where the range (0 to Tk1) and the minimum range (V0 to V1) correspond. In the area where the range (Tk1 to Tk2) and the intermediate range (V1 to V2) correspond, the number of occurrences of the maximum value is 10, and in the area where the range (Tk1 to Tk2) and the intermediate range (V2 to V3) correspond The number of occurrences of the maximum value is 1000, and the number of occurrences of the maximum value is 10 in the region where the range (Tk1 to Tk2) and the intermediate range (V3 to V4) correspond. In the area where the range (Tk2 to Tk3) and the intermediate range (V2 to V3) correspond, the number of occurrences of the maximum value is 10, and in the area where the range (Tk2 to Tk3) and the intermediate range (V3 to V4) correspond The number of occurrences of the maximum value is ten. In addition, the number of occurrences of local maximum values is 0 in regions other than the above.

The range of the occurrence interval Ti of the maximum value in the matrix data (3) includes a time range (0 to Ti1) in which the occurrence interval Ti is 0 or more and less than Ti1, and a time range (0 to Ti1) in which the occurrence interval Ti is Ti1 or more and less than Ti2, which is longer than Ti1. There is a certain time range (Ti1 to Ti2), a time range (Ti2 to Ti3) in which the occurrence interval Ti is Ti2 or more and less than Ti3, which is longer than Ti2, and a time range (Ti3 to Ti3) in which the occurrence interval Ti is Ti3 or more.

For example, when the occurrence interval Ti calculated by the parameter calculation unit 43 is included in the time range (Ti1 to Ti2) and the maximum value extracted by the data extraction unit 42 is included in the intermediate range (V2 to V3), matrix generation The unit 44 increments the number of occurrences of the maximum value by +1 in the region corresponding to the range (Ti1 to Ti2) and the intermediate range (V2 to V3) in the matrix data (3). In the matrix data (3) shown in FIG. 5, the maximum number of occurrences is 10 in the region where the range (0 to Ti1) and the minimum range (V0 to V1) correspond. In the region where the range (Ti1 to Ti2) and the intermediate range (V1 to V2) correspond, the number of occurrences of the maximum value is 10, and in the region where the range (Ti1 to Ti2) and the intermediate range (V2 to V3) correspond The number of occurrences of the maximum value is 1000, and the number of occurrences of the maximum value is 10 in the region where the range (Ti1 to Ti2) and the intermediate range (V3 to V4) correspond. In the region where the range (Ti2 to Ti3) and the intermediate range (V2 to V3) correspond, the number of occurrences of the maximum value is 10, and in the region where the range (Ti2 to Ti3) and the intermediate range (V3 to V4) correspond The number of occurrences of the maximum value is ten. The number of occurrences of the maximum value in regions other than the above is zero.

The range of values of the period energy Ve in the matrix data (4) includes a range (Ve0 to Ve1) where the period energy Ve is Ve0 or more and less than Ve1 which is higher than Ve0, and a range (Ve0 to Ve1) where the period energy Ve is Ve1 or more and is higher than Ve1. There are a range (Ve1 to Ve2) in which the period energy Ve is less than Ve2, a range (Ve2 to Ve3) in which the period energy Ve is higher than Ve2 and less than Ve3, and a range (Ve3 to) in which the period energy Ve is higher than or equal to Ve3.

For example, when the period energy Ve calculated by the parameter calculation unit 43 is included in the range (Ve1 to Ve2) and the maximum value extracted by the data extraction unit 42 is included in the intermediate range (V2 to V3), the matrix generation unit 44 counts up the number of occurrences of the local maximum value by +1 in the area corresponding to the range (Ve1 to Ve2) and the intermediate range (V2 to V3) in the matrix data (4). In the matrix data (4) shown in FIG. 5, the maximum number of occurrences is 10 in the region where the range (Ve0 to Ve1) and the minimum range (V0 to V1) correspond. In the area where the range (Ve1-Ve2) and the intermediate range (V1-V2) correspond, the number of occurrences of the maximum value is 10, and in the area where the range (Ve1-Ve2) and the intermediate range (V2-V3) correspond The number of occurrences of the maximum value is 1000, and the number of occurrences of the maximum value is 10 in the region where the range (Ve1 to Ve2) and the intermediate range (V3 to V4) correspond. In the area where the range (Ve2-Ve3) and the intermediate range (V2-V3) correspond, the number of occurrences of the maximum value is 10, and in the area where the range (Ve2-Ve3) and the intermediate range (V3-V4) correspond The number of occurrences of the maximum value is ten. In addition, the number of occurrences of local maximum values is 0 in regions other than the above.

The matrix generator 44 repeats the matrix data generation described above each time the AE sensor 3 detects an AE wave generated in the target device 2 . As a result, in the matrix data, the number of occurrences of the maximum value corresponding to the normal state of the target device 2 is counted up. For example, in the area corresponding to the range (Tr1-Tr2) and the range (V2-V3) in the matrix data (1), the number of occurrences of the maximum value is counted 1000 times, and the range (Tr1-Tr2) and the range (V2) are counted 1000 times. . . . V3) corresponds to the state of the target device 2, which is presumed to be the normal state (normal operating state) of the target device 2. FIG.

Next, the output processing section 45 outputs the matrix data generated by the matrix generating section 44 to the output section 5 (step ST5). For example, the output processing unit 45 outputs alert information to the output unit 5 when the number of occurrences of the maximum value is counted in an area in which the number of occurrences of the maximum value has not been counted in the matrix data.

In the matrix data (1), regions in which the number of occurrences of local maximum values has not been counted are, for example, regions corresponding to the range (Tr1-Tr2) and the range (V4-). As described above, in the matrix data (1), the regions corresponding to the normal state of the target device 2 are the range (Tr1 to Tr2) and the range (V2 to V3) in which the number of occurrences of the maximum value is counted 1000 times. is the corresponding region. A region other than the region corresponding to the normal state of the target device 2 and in which the number of occurrences of the maximum value was not counted is a state that occurred suddenly in the target device 2 (for example, a failure that occurred suddenly in the target device 2). ).

The output processing unit 45 outputs alert information to the output unit 5 when the number of occurrences of the maximum value is counted in the area corresponding to the state that occurred suddenly in the target device 2 in the matrix data. The output unit 5 visually or audibly outputs the alert information input from the output processing unit 45 . A maintenance worker can perform maintenance related to a state that has suddenly occurred in the target device 2 based on the alert information output from the output unit 5 .

In addition, the output processing unit 45 counts the number of occurrences of the local maximum value in a region where the number of occurrences of the local maximum value is less than a predetermined number of times in the matrix data generated by the matrix generation unit 44 when the target device 2 is in the normal state. If so, the alert information is output to the output unit 5 . As the deterioration of the target device 2 progresses over time, the number of occurrences of the maximum value is counted up in the region where the number of occurrences of the maximum value is small among the regions corresponding to the normal state of the target device 2 in the matrix data. become.

For example, in the matrix data (1), the range (Tr1 to Tr2) and the range (V3 to V4) correspond to the corresponding region (the number of occurrences of the maximum value=10 times). ) corresponds to the region (the number of occurrences of the maximum value=1000), the counted-up number of occurrences of the local maximum value is smaller, and it is estimated that this corresponds to the state when the target apparatus 2 is shaken. As the deterioration of the target device 2 progresses over time, such deviations from normal operation frequently occur. Therefore, when the number of occurrences of the maximum value is counted up in the area corresponding to the range (Tr1 to Tr2) and the range (V3 to V4) in the matrix data (1), the output processing unit 45 outputs the alert information. Output to the output unit 5 . The output unit 5 visually or audibly outputs the alert information input from the output processing unit 45 . A maintenance worker can perform maintenance for deterioration over time of the target device 2 based on the alert information output from the output unit 5 .

Further, the output processing unit 45 can output matrix image information obtained by imaging the matrix data to the output unit 5 . FIG. 6 is a diagram showing Example 3 of the output waveform of the AE sensor 3, and shows the case where the target device 2 is a cutting machine. A period S1 shown in FIG. 6 is a period before the cutting machine starts cutting the workpiece, and the spindle to which the cutting blade is attached is idling without the cutting blade coming into contact with the workpiece. . At this time, vibration is generated in the cutting machine due to the rotation of the main shaft.

A period S2 is a machining start period in which the cutting blade is brought into contact with the workpiece and cutting of the workpiece is started. During the period S2, when the cutting blade contacts the workpiece, a large vibration is suddenly generated between the cutting blade and the workpiece, and the AE wave also varies greatly according to this vibration. A period S3 is a machining period during which the workpiece is cut. During period S3, the cutting machine generates vibration due to the cutting of the workpiece by the cutting blade.

A period S4 is a period in which the cutting of the workpiece is completed. In the period S4, the cutting blade is separated from the workpiece, and both of them oscillate due to the recoil of the cutting blade being pressed against the workpiece, and a large vibration is generated between the cutting blade and the workpiece. . The AE wave also fluctuates greatly according to this vibration.

For example, in period S1, the data extraction unit 42 extracts a maximum value within a range of voltage V0 or more and voltage V1 or less. The parameter calculator 43 calculates the duration during which the data extractor 42 continues to extract the maximum value within the range between the voltage V0 and the voltage V1.

In period S2, the data extractor 42 extracts the maximum value within the range of voltage V3 or more and voltage V4 or less. The parameter calculator 43 calculates the duration during which the data extractor 42 continues to extract the maximum value within the range of voltage V3 or more and voltage V4 or less. In the period S2, the output voltage of the AE sensor 3 greatly changes up to around the voltage V4 due to the large vibration generated between the cutting blade and the workpiece. The duration, which is a value within the range above voltage V3 and below voltage V4, is long.

In period S3, the data extractor 42 extracts the maximum value within the range of voltage V2 or more and voltage V3 or less. The parameter calculator 43 calculates the duration during which the data extractor 42 continues to extract the maximum value within the range between the voltage V2 and the voltage V3. Since the output voltage of the AE sensor 3 changes around the voltage V2 in the period S3, the maximum value sequentially extracted by the data extracting unit 42 lasts for a short period of time equal to or higher than the voltage V2.

In the period S4, the data extraction unit 42 extracts the maximum value within the range of the voltage V2 or more and the voltage V3 or less, as in the period S3. The parameter calculator 43 calculates the duration during which the data extractor 42 continues to extract the maximum value within the range between the voltage V2 and the voltage V3. In the period S4, the output voltage of the AE sensor 3 greatly changes up to around the voltage V3 due to the large vibration generated between the cutting blade and the workpiece. The duration, which is a value within the range of voltage V2 or more and voltage V3 or less, is long.

The matrix generation unit 44 generates matrix data by counting up the number of occurrences of the maximum value in the region where the range of the duration calculated by the parameter calculation unit 43 and the range of the maximum value extracted by the data extraction unit 42 correspond. Generate. As the cutting machine repeatedly executes the operation from the period S1 to the period S4, the number of occurrences of the maximum value corresponding to the state of each period is counted up in the matrix data. The matrix image shown in FIG. 7 is obtained by imaging the matrix data generated by the matrix generation unit 44 in this manner.

In the matrix image shown in FIG. 7, the distribution of the number of occurrences of the maximum value corresponding to each state of the cutting machine from the period S1 to the period S4 on a two-dimensional plane in which the horizontal axis is the duration and the vertical axis is the maximum value. A1 to A4 are displayed. Distribution A1 is the distribution of the number of occurrences of maximum values corresponding to the state of the cutting machine during period S1 shown in FIG. 6. Distribution A2 is the number of occurrences of the maximum value corresponding to the state of the cutting machine during period S2. is the distribution of Distribution A3 is the distribution of the number of occurrences of the maximum value corresponding to the state of the cutting machine during period S3, and distribution A4 is the distribution of the number of occurrences of the maximum value corresponding to the state of the cutting machine during period S4.

For example, the output processing unit 45 generates matrix image information shown in FIG. 7 by imaging the matrix data output from the matrix generation unit 44 and outputs the matrix image information to the output unit 5 . The output unit 5 displays the matrix image information on the screen. In the matrix image information shown in FIG. 7, a white square indicates that the number of occurrences of the maximum value is 0, meaning that the corresponding state of the target device 2 has not occurred. The number of occurrences of local maxima increases as the color of the squares darkens. An increase in the number of occurrences of the maximum value means that the state of the target device 2 corresponding to the occurrence of the maximum value will occur more frequently. Therefore, various states of the target device 2 can be monitored by referring to the matrix image information.

When the cutting machine repeatedly performs the operations from period S1 to period S4 and the cutting machine deteriorates over time, the state of the cutting machine corresponding to the duration and the maximum value gradually changes. At this time, the distributions A1 to A4 in the matrix image information shown in FIG. 7 move in the axial direction of the maximum value or the axial direction of the duration on the two-dimensional plane defined by the duration and the maximum value.

For example, the anomaly detection device described in Patent Literature 1 monitors only the state of the target device that meets certain conditions defined by the duration and rise time of the AE signal. On the other hand, in the state monitoring device 4 according to the first embodiment, by referring to the matrix image information, it is possible to grasp the transition of the number of times the state of the target device 2 occurs, and to Changes in state can be monitored. For example, a maintenance worker can recognize the progress of deterioration over time in the cutting machine by visually recognizing the movement of the distributions A1 to A4 in the matrix image information shown in FIG. FIG. 7 shows the matrix image information in which the parameter is the duration, but the matrix image information can be generated regardless of whether the parameter is the rise time, the occurrence interval of the local maximum value, or the period energy. is. When generating matrix image information corresponding to each of a plurality of parameters, the horizontal axis direction parameter You may want to adjust the range of values for . This makes it easier for the maintenance worker to visually recognize the number of occurrences of the maximum value, that is, the distribution of the number of occurrences of the states of the target device 2 .

Next, a modification of the maximum value extraction processing by the data extraction unit 42 will be described.
FIG. 8 is a diagram illustrating an example of extraction processing of local maximum values according to the first embodiment. The data extraction unit 42 sequentially extracts the maximum values of a plurality of cycles in the digital data of the sinusoidal signal of the AE wave from the highest value until they become equal to the voltage V0. As a result, 26 maximum values from 1020 to 875 (=voltage V0) are extracted as shown in FIG. 8, for example.

Subsequently, the data extraction unit 42 sequentially classifies the maximum values sequentially extracted from the digital data of the sine wave signal of the AE wave into a first group in which the value obtained by rounding off the least significant digit of the maximum value is the representative value. . Group (1) to group (6) are the first group. For example, in group G(1), the representative value 1020 is obtained by rounding off the least significant digits of the maximum values 1020, 1016 and 1015. FIG. In group G (2), the values obtained by rounding off the least significant digits of the maximum values are all representative values of 1010, and in group G (3), the values obtained by rounding off the least significant digits of the maximum values are all representative values. 1000. In group G (4), the values obtained by rounding off the least significant digits of the maximum values are all representative values of 900, and in group G (5), the values obtained by rounding off the least significant digits of the maximum values are all representative values. 890, and group G(6) has a representative value of 880, which is obtained by rounding off the least significant digit of the maximum value.

Subsequently, the data extraction unit 42 divides the classified groups G(1) to G(6) into the first groups where the difference between the immediately higher representative value and the immediately lower representative value is larger than the judgment value. are further classified into a plurality of second groups as Group G(A) and group G(B) are the second groups. In FIG. 8, group G(1) and group G(2), group G(2) and group G(3), group G(4) and group G(5), and group G(5) and group G(6) ) is 10. Also, the difference between the representative values of group G(3) and group G(4) is 100. For example, if the determination value is 20, the difference (=100) between the representative values of group G(3) and group G(4) is greater than 20, which is the determination value. Therefore, the data extracting unit 42 divides the group G(3) and the group G(4) into groups G(A) including the groups G(1) to G(3) and the group G(A), which includes the groups G(1) to G(3) 4) to group G(B) including group G(6).

Next, the data extracting unit 42 determines that there is a tendency of increase or decrease in the number of occurrences of local maximum values between the first group with the highest representative value and the first group with the lowest representative value among the first groups included in the second group. Extract the local maxima from the inverted first group. For example, in group G(A) shown in FIG. 8, the number of occurrences of maximum values in group G(1) is 3, the number of occurrences of maximum values in group G(2) is 6, and group G(3) Since the number of occurrences of the local maximum value in is 3, the increase/decrease trend of the number of occurrences of the local maximum value is reversed in the group G(2).

The data extracting unit 42 extracts the maximum value from the group G(2) in which the increase/decrease trend of the number of occurrences of the maximum value is reversed. Similarly, in group G(B), the data extracting unit 42 extracts the maximum value from group G(5) in which the increasing/decreasing trend of the number of occurrences of the maximum value is reversed. The matrix generation unit 44 generates matrix data in which the number of occurrences of the maximum values extracted from the second group by the data extraction unit 42 is counted. Even in this way, it is possible to generate matrix data with which it is possible to grasp the number of times the state of the target device 2 has occurred.

Further, the data extraction unit 42 extracts the maximum value for each second group based on the maximum value that becomes the boundary value between the second groups when the least significant digit is rounded off among the maximum values included in the first group. can do. For example, in group G (A), group G (1) has a representative value of 1020, and the first group in which the number of occurrences of the maximum value of 1025 or more is set (when the least significant digit of the maximum value is rounded off, , a typical value of which is 1030) is 1025. Group G(3) has a representative value of 1000, so it is the first group in which the number of occurrences of the maximum value less than 995 is set (the representative value becomes 990 when the least significant digit of the maximum value is rounded off). is 994.

In group G (B), group G (4) has a representative value of 900, and the first group in which the number of occurrences of a maximum value of 905 or more is set (when the least significant digit of the maximum value is rounded off, the representative value is value becomes 910) and the boundary value is 905. Group G(6) has a representative value of 880, so it is the first group in which the number of occurrences of local maximum values less than 875 is set (if the least significant digit of the local maximum value is rounded off, the representative value becomes 870). is 874. The data extraction unit 42 sets the boundary value 994 to the voltage V0, and extracts the maximum value from the group G(A) based on the voltage V0. The data extraction unit 42 also sets the boundary value 874 to the voltage V0, and extracts the maximum value from the group G(B) based on the voltage V0.

The data extraction unit 42 can also extract the maximum value for each second group based on the standard deviation σ of the average value of the maximum values included in the second group. For example, the data extraction unit 42 calculates the average value of the maximum values included in the group G(A), and further calculates the standard deviation σ of the average value. Then, the data extraction unit 42 sets the maximum value corresponding to ±6σ or ±3σ in the distribution of maximum values in group G(A) as a threshold, and extracts the maximum value from group G(A) based on this threshold. Extract. For example, local maxima greater than the threshold are extracted from group G(B). Similarly for group G (B), the data extraction unit 42 sets the maximum value corresponding to ±6σ or ±3σ of the distribution of maximum values in group G (B) as a threshold, and based on this threshold, group G Extract the maxima from (B).

Next, a hardware configuration that implements the functions of the state monitoring device 4 according to the first embodiment will be described. Each function of the A/D converter 41, the data extractor 42, the parameter calculator 43, the matrix generator 44, and the output processor 45 in the state monitoring device 4 is implemented by a processing circuit. That is, the state monitoring device 4 has a processing circuit that executes the processing from step ST1 to step ST5 in FIG. The processing circuit may be dedicated hardware, or may be a CPU (Central Processing Unit) that executes a program stored in memory.

9A is a block diagram showing a hardware configuration for realizing the functions of the state monitoring device 4, and FIG. 9B is a block diagram showing a hardware configuration for executing software realizing the functions of the state monitoring device 4. FIG. 9A and 9B, input interface 100 relays the sinusoidal signal output from AE sensor 3 to condition monitor 4. In FIG. The output interface 101 relays data output from the state monitoring device 4 to the output unit 5 . Note that the output interface 101 may be a communication interface that transmits data to the output unit 5 by wireless communication.

If the processing circuit is the dedicated hardware processing circuit 102 shown in FIG. 9A, the processing circuit 102 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an Application Specific Integrated Integrated Circuit (ASIC), Circuit), FPGA (Field-Programmable Gate Array), or a combination thereof. The functions of the A/D conversion unit 41, the data extraction unit 42, the parameter calculation unit 43, the matrix generation unit 44, and the output processing unit 45 in the state monitoring device 4 may be realized by separate processing circuits. may be collectively realized by one processing circuit.

When the processing circuit is the processor 103 shown in FIG. 9B, the functions of the A/D converter 41, the data extractor 42, the parameter calculator 43, the matrix generator 44, and the output processor 45 in the state monitoring device 4 are software, It is implemented by firmware or a combination of software and firmware. Note that software or firmware is written as a program and stored in the memory 104 .

The processor 103 reads out and executes the program stored in the memory 104 so that the A/D conversion unit 41, the data extraction unit 42, the parameter calculation unit 43, the matrix generation unit 44, and the output processing unit 45 in the state monitoring device 4. to realize the function of For example, the state monitoring device 4 includes a memory 104 for storing a program that, when executed by the processor 103, results in the processing from step ST1 to step ST5 in the flowchart shown in FIG. These programs cause the computer to execute the procedures or methods of the A/D converter 41 , data extractor 42 , parameter calculator 43 , matrix generator 44 and output processor 45 . The memory 104 is a computer-readable storage medium storing a program for causing the computer to function as the A/D converter 41, the data extractor 42, the parameter calculator 43, the matrix generator 44, and the output processor 45. good too.

The memory 104 includes, for example, non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically-EPROM), magnetic Discs, flexible discs, optical discs, compact discs, mini discs, DVDs, and the like are applicable.

Some of the functions of the A/D converter 41, data extractor 42, parameter calculator 43, matrix generator 44, and output processor 45 in the state monitoring device 4 are realized by dedicated hardware, and partly by software or firmware. can be realized with For example, the A/D converter 41 is realized by the processing circuit 102, which is dedicated hardware. The functions are realized by reading and executing the programs stored in 104 . As such, the processing circuitry may implement the above functions through hardware, software, firmware, or a combination thereof.

As described above, the state monitoring device 4 according to the first embodiment generates a maximum data point extracted from the digital data of the sinusoidal signal of the AE wave corresponding to the value of the parameter indicating the state of the AE wave. Generates and outputs matrix data in which the number of times is set. The number of times the local maximum data point occurs is the number of times the target device state corresponding to the value of the parameter and the local maximum data point occurs. As a result, by referring to the matrix data, it is possible to grasp the transition of the number of occurrences of the state of the target device, and to monitor various changes in the state of the target device.

It should be noted that the present invention is not limited to the above-described embodiments, and any constituent elements of the embodiments may be modified or omitted within the scope of the present invention.

A condition monitoring device according to the present invention can be used, for example, for condition monitoring of a rotating machine.

1 state monitoring system 2 target device 3 AE sensor 4 state monitoring device 5 output unit 41 A/D conversion unit 42 data extraction unit 43 parameter calculation unit 44 matrix generation unit 45 output processing unit 100 input interface, 101 output interface, 102 processing circuit, 103 processor, 104 memory.

Claims (16)

A to input the sine wave signal from an AE sensor that detects an acoustic emission wave that fluctuates according to the state of the target device and outputs a sine wave signal of the acoustic emission wave, and converts the sine wave signal into digital data. a /D conversion unit;
a data extraction unit for extracting extreme values from the digital data;
a parameter calculation unit that uses the digital data to calculate a value of a parameter that indicates a state of temporal fluctuation of the acoustic emission wave;
a matrix generation unit that generates matrix data in which data is associated with combinations of a plurality of parameter ranges of the parameter values and a plurality of extreme value ranges of the extreme values;
an output processing unit that outputs the matrix data generated by the matrix generation unit;
The parameter calculation unit
calculating the value of the parameter based on the portion of the digital data in which the extreme values belonging to the extreme value range, which is one of the plurality of extreme value ranges, are continuously extracted;
The matrix generator is
The combination of the extreme value range corresponding to the portion and the parameter range to which the parameter value calculated based on the portion among the plurality of parameter ranges belongs, the extreme value in the portion A condition monitoring device, wherein the number of occurrences is associated as the data. The extreme value is a maximum value,
The plurality of extreme ranges is each range of the extreme maximum,
The rise time is the time until the data extraction unit extracts the maximum and maximum value within the range of the maximum value,
The duration is the time during which the maximum value belonging to the range of the maximum value has been continuously extracted by the data extraction unit,
The occurrence interval is an interval at which the maximum value occurs in the range of the maximum value,
period energy is the waveform area of the sinusoidal signal at the duration;
The condition monitoring device according to claim 1, wherein the parameter is at least one of the rise time, the duration , the occurrence interval , or the period energy. . The output processing unit controls the number of occurrences of the extreme value in the combination in which the number of occurrences of the extreme value is not counted in the matrix data generated by the matrix generation unit when the target device is in a normal state. 2. The state monitoring device according to claim 1, wherein alert information is output when is counted. The output processing unit generates the extreme value in the combination in which the number of occurrences of the extreme value is less than a predetermined number in the matrix data generated by the matrix generation unit when the target device is in a normal state. 2. The state monitoring device according to claim 1, wherein alert information is output when the number of times is counted. 2. The condition monitoring device according to claim 1, wherein the output processing unit outputs matrix image information obtained by imaging the matrix data. The data extraction unit extracts the maximum value in the portion based on a threshold determined using the maximum value detected by the AE sensor when the target device is in a normal state. 3. The condition monitoring device according to claim 2, characterized by: The extreme value is a maximum value,
The extreme value range is the range of the extreme values,
The plurality of ranges of local maxima are
a maximum range that is a range up to the maximum value of the local maximum extracted from the entire digital data ;
a minimum range that is a range of average values of the maximum values extracted from the entire digital data when the target device is in a normal state;
2. The condition monitoring device according to claim 1, wherein the range of said maximum values between said minimum range and said maximum range is an equally divided intermediate range. The extreme value is a maximum value,
The extreme value range is the range of the extreme values,
The number of occurrences of the extreme value is the number of occurrences of the maximum value,
The data extractor is
sequentially extracting from the highest local maximum value in the portion to the local maximum value equal to a threshold;
The plurality of sequentially extracted maximum values are sequentially classified into a first group whose representative value is a value obtained by rounding off the least significant digit of the numerical value representing the maximum value with multiple digits ,
The plurality of classified first groups are divided into a plurality of second groups by setting the first group where the difference between the immediately higher representative value and the immediately lower representative value is larger than a judgment value as a boundary. further classify
The matrix generator is
generating and outputting classification data in which the first group, the second group, and the number of occurrences of the local maximum values classified into these groups are associated with each other;
2. The condition monitoring device according to claim 1, wherein: The data extractor is
Among the first groups included in the second group, the tendency of increase/decrease in the number of occurrences of the local maximum is reversed between the first group having the highest representative value and the first group having the lowest representative value. extracting the maxima from the first group;
The matrix generator is
extracted from the first group into the combination of the maximum value range corresponding to the portion and the parameter range to which the parameter value calculated based on the portion of the plurality of parameter ranges belongs Associate the number of occurrences of the local maximum value as the data
9. The condition monitoring device according to claim 8, wherein: The data extracting unit obtains a boundary value between the second groups by rounding off the least significant digit of each of the plurality of maximum values included in the first group, each representing the maximum value with a plurality of digits. The condition monitoring device according to claim 8, wherein the maximum value is extracted from the second group based on the maximum value. 9. The state according to claim 8, wherein the data extraction unit extracts the maximum value from the second group based on a standard deviation of an average value of the plurality of maximum values included in the second group. surveillance equipment. an AE sensor that detects an acoustic emission wave that fluctuates according to the state of a target device and outputs a sine wave signal of the acoustic emission wave;
an A/D converter that receives the sine wave signal from the AE sensor and converts the sine wave signal into digital data;
a data extraction unit for extracting extreme values from the digital data;
a parameter calculation unit that uses the digital data to calculate a value of a parameter that indicates a state of temporal fluctuation of the acoustic emission wave;
a matrix generation unit that generates matrix data in which data is associated with combinations of a plurality of parameter ranges of the parameter values and a plurality of extreme value ranges of the extreme values;
an output processing unit that outputs the matrix data generated by the matrix generation unit;
The parameter calculation unit
calculating the value of the parameter based on the portion of the digital data in which the extreme values belonging to the extreme value range, which is one of the plurality of extreme value ranges, are continuously extracted;
The matrix generator is
The combination of the extreme value range corresponding to the portion and the parameter range to which the parameter value calculated based on the portion among the plurality of parameter ranges belongs, the extreme value in the portion A condition monitoring system characterized by associating the number of occurrences as the data. An output unit that visually or aurally outputs information input from the output processing unit,
The output processing unit controls the number of occurrences of the extreme value in the combination in which the number of occurrences of the extreme value is not counted in the matrix data generated by the matrix generation unit when the target device is in a normal state. 13. The state monitoring system according to claim 12, wherein alert information is output to said output unit when is counted. An output unit that visually or aurally outputs information input from the output processing unit,
The output processing unit generates the extreme value in the combination in which the number of occurrences of the extreme value is less than a predetermined number in the matrix data generated by the matrix generation unit when the target device is in a normal state. 13. The state monitoring system according to claim 12, wherein the alert information is output to the output unit when the number of times is set. An output unit that visually or aurally outputs information input from the output processing unit,
13. The state monitoring system according to claim 12, wherein the output processing section outputs a matrix image obtained by imaging the matrix data to the output section. An A/D converter receives the sine wave signal from an AE sensor that detects an acoustic emission wave that varies according to the state of the target device and outputs a sine wave signal of the acoustic emission wave, and converts the sine wave signal into into digital data; and
a data extractor extracting extreme values from the digital data;
a step in which a parameter calculation unit calculates, using the digital data, a value of a parameter indicating a state of temporal fluctuation of the acoustic emission wave;
a matrix generation unit generating matrix data in which data are associated with combinations of a plurality of parameter ranges of the parameter values and a plurality of extreme value ranges of the extreme values;
an output processing unit outputting the matrix data generated by the matrix generation unit;
The parameter calculation unit
calculating the value of the parameter based on the portion of the digital data in which the extreme values belonging to the extreme value range, which is one of the plurality of extreme value ranges, are continuously extracted;
The matrix generator is
The combination of the extreme value range corresponding to the portion and the parameter range to which the parameter value calculated based on the portion among the plurality of parameter ranges belongs, the extreme value in the portion A condition monitoring method, wherein the number of occurrences is associated as the data.

Images