JPH11248528A - Abnormal vibration detecting apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a general purpose abnormal vibration detecting apparatus at a low cost which performs vibration analysis of sufficiently high quality. SOLUTION: This apparatus is provided with a vibration sensor 1 outputting a signal corresponding to vibration acceleration, a plurality of bandpass filters BPF1-BPFn different in pass band which receive an output of the sensor 1 as the respective inputs, and a processor 5 which takes in the outputs of BPF1- BPFn and performs operation process. In the state that the vibration of a normal mode is applied to the sensor 1 from an object, an initially set mode is executed. In this mode, the processor 5 analyzes outputs of BPF1-FPFn, and sets an abnormality judgment condition for each band. After the initially set mode is executed, an abnormality judgment mode is executed. In this mode, the processor S judges normality/abnormality for each band by making the outputs of BPF1-BPFn refer to the abnormality judgment condition for each band, and generates an abnormal vibration detecting signal on the basis of the judgment results.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus for detecting abnormal vibrations of various machines such as motors and engines, and more particularly to an apparatus for performing vibration analysis of as high a performance as possible with a device as inexpensive as possible. About technology.

[0002]

2. Description of the Related Art An abnormal vibration detecting device of this type is equipped with a vibration sensor on a target machine to be monitored, and electrically analyzes and processes the sensor output to determine whether the vibration is within a normal range.
It is determined whether the abnormal vibration exceeds the range. If the target is determined, the vibration form is also specified, so that the object can be achieved with a very simple vibration sensor and level discriminating circuit.

However, in order to construct a general-purpose abnormal vibration detecting device whose target is not specified, a function of analyzing the target vibration into a plurality of frequency bands in detail and a vibration mode which is considered normal, On the other hand, a considerably high information processing capability such as a function of appropriately setting an abnormality determination condition is required.

Japanese Unexamined Patent Publication No. 6-229818 discloses the following high-performance type abnormal sound inspection apparatus.
That is, a target vibration is picked up by a sensor, and the output of the sensor is band-decomposed by a plurality of band-pass filters having different pass bands. Then, the output of each band-pass filter is taken into a computer, subjected to analog / digital conversion, fast Fourier-transformed and analyzed for vibration, and abnormal vibration is detected by finely discriminating the level of each frequency component.

[0005]

A conventional apparatus (Japanese Patent Laid-Open No.
-229818), a high-performance computer (DSP: digital signal processor) is required to perform the fast Fourier transform, and the product price becomes extremely high.

SUMMARY OF THE INVENTION The present invention has been made in view of the above background, and has as its object to solve the above-mentioned problems and to provide a general-purpose abnormal vibration detecting device which performs sufficiently high-performance vibration analysis. It is to be realized at low cost.

[0007]

Means for Solving the Problems In order to achieve the above-mentioned object, an abnormal vibration detecting device which is the basis of the present invention is configured to have the following requirements (1), (2) and (3) ( Claim 1). (1) A vibration sensor that outputs a signal corresponding to the vibration acceleration, a plurality of band-pass filters having different pass bands to which the output of the vibration sensor is input, and a processor that takes in the output of each band-pass filter and performs arithmetic processing In the embodiment, it corresponds to “microcomputer 5”). (2) The initial setting mode is executed in a state where the vibration in the normal mode is applied from the target to the vibration sensor. In the initial setting mode, the processor sets an abnormality determination condition for each band based on the output level of each band pass filter. (3) Execute the abnormality determination mode after executing the initial setting mode. In this mode, the processor has at least one
Based on the outputs of the two band-pass filters, normal / abnormal is determined by illuminating the abnormality determination condition set for the selected band, and an abnormal vibration detection signal is generated according to the determination results.

[0008] Preferably, the abnormality determination condition is:
The output of the band-pass filter is set as an upper threshold value and a lower threshold value for level discrimination, and is set to be different for each band (claim 2). The present invention is realized by the first embodiment.

In the initial setting mode, the gain of each band is individually set based on the output level of each band-pass filter, thereby normalizing the output level of each band provided to the processor. May be provided (claim 3). Here, as a method of setting the gain for each band, a method of adjusting the gain for the input signal of the band-pass filter (the method described in the second embodiment) and a method of adjusting the gain of the output signal of the band-pass filter Gain adjustment.

That is, the frequency domain is divided into a plurality of bands, an initial setting mode is executed and learning is performed based on the output level of each band, and abnormality determination conditions are set. As a result, in each band, the determination is made based on the output level of the signal, so that the processing can be performed easily and at high speed. In addition, since the determination is performed for each band, the detection effect as a whole is similar to that using FFT. Can be expected.

Further, on the premise of each of the above-mentioned inventions,
The initial setting mode may be configured to have a function of determining that the output of a specific band-pass filter is not used for determination of normal / abnormal based on the obtained output level of each band-pass filter ( Claim 4).

That is, even in the normal state, the output level may be too low or too high depending on the band. In such a case, if the determination is made unconditionally based on all bands, erroneous determination may be likely to occur. Therefore, reliability is improved by not using an inappropriate band as a determination condition.

Further, on the premise of each of the above-mentioned inventions,
It is preferable that the processor further has a function of executing the capture of the output of each band-pass filter and the analysis processing for a certain period in accordance with the operation of the target, and disabling the other periods (claim 5). .

That is, some objects do not always operate and do not generate vibration in the normal mode. In such a case, by performing the learning process based on the period in which the normal vibration is likely to be suppressed or performing the process of determining the presence or absence of abnormalities, it is highly accurate. A determination process can be performed. Specifically, the operation start timing at the time of executing for a certain period of time is determined, for example, by receiving a trigger signal (external trigger) from the control unit when there is a control unit for controlling the drive of the target device. The processing may be performed for a certain period after receiving the trigger signal. Further, when the external trigger cannot be received, for example, the vibration level may be monitored, and when a vibration of a certain level or more is output, the process may be executed for a certain period based on the output.

[0015]

FIG. 1 schematically shows a circuit configuration of an abnormal vibration detecting apparatus according to a first embodiment of the present invention. As shown in FIG. 1, the vibration sensor 1 is attached to a target (a target for monitoring the presence or absence of abnormal vibration of a motor, various devices, and the like) and outputs a signal corresponding to a vibration acceleration. This sensor output is input in parallel to a plurality of bandpass filters BPF1 to BPFn having different passbands via the preamplifier 2. The output of each BPFi (i = 1 to n) is selectively input to the A / D converter 4 via the multiplexer 3, and the digitally converted signal is taken into the microcomputer 5. The microcomputer 5 is provided with an operation unit 6 for user interface and a memory 7.

When the vibration sensor 1 is attached to a specific object, and the object is vibrating in the normal mode, the operation unit 6
Causes the microcomputer 5 to execute the initial setting mode in response to a command from the microcomputer 5. In the initial setting mode, the microcomputer 5 sequentially switches the multiplexers 3 so that each band-pass filter BPFi
The digital value of the output of the above is sequentially taken in, and the vibration level of each band is analyzed. FIG. 2 shows an example of the frequency characteristic of the target vibration analyzed in the initial setting mode. Microcomputer 5
Sets an abnormality determination condition for each band based on the analyzed frequency characteristics.

Specifically, the abnormality determination condition for each band is set as follows. As shown in FIG. 2, when the peak value of the output level of the band-pass filter BPFi is Pi, the upper limit threshold value Ai for abnormality determination in that band is Ai = P
i × 2, and the lower threshold value Bi is set to Bi =
Set Pi ÷ 2.

When the setting of the upper threshold value Ai and the lower threshold value Bi for each band is completed, the mode automatically shifts to the abnormality determination mode. In the abnormality determination mode, the microcomputer 5
By sequentially switching the multiplexer 3, the digital value of the output of each band-pass filter BPFi is sequentially taken in, and it is determined whether or not the vibration level Xi for each band falls within the range between the upper threshold value Ai and the lower threshold value Bi. judge. Then, when the vibration level is out of the threshold range in any one band, an abnormal vibration detection signal is output. Note that other embodiments are also conceivable, such as abnormal vibration when the vibration level is out of the threshold range simultaneously in a plurality of predetermined bands.

FIG. 3 schematically shows a circuit configuration according to a second embodiment of the present invention. In this embodiment, in addition to the functions of the embodiment of FIG. 1, the microcomputer 5 switches and controls the gain of the preamplifier 2. In the initial setting mode, the microcomputer 5 sequentially switches the multiplexer 3 to sequentially capture the digital value of the output of each band-pass filter BPFi, and analyzes the vibration level for each band. First, the gain of the preamplifier 2 is kept constant. This process is performed while keeping it. Next, the gain of the preamplifier 2 for each band is set in accordance with the acquired vibration level for each band. Next, when taking in the output of each band-pass filter BPFi, the gain of the preamplifier 2 is variably adjusted based on the gain value set for each band. By doing so, the input level of each band-pass filter BPFi is normalized, and as a result, a frequency characteristic whose level has been corrected as shown in FIG. 4 can be obtained. In the characteristic diagram of FIG. 4, the input gains of the BPF1, BPF2, and BPF3 are made larger than the basic values, and the peak values of the vibration levels of the respective bands taken into the microcomputer 5 are made substantially the same. In this case, the upper threshold value Ai and the lower threshold value Bi for each band can all be set equal (FIG. 4).
Of course), and of course, they may be appropriately finely changed and set.

In the second embodiment shown in FIG. 3 and FIG.
Recognizes that the subject's vibration is within the normal range,
By analyzing the drift of the output of each band-pass filter BPFi, the abnormality determination condition for each band can be optimized.

That is, for example, when the amplification factor of the preamplifier 2 changes, the output may fluctuate. As an example,
When an input signal having a level of 1 is given to a preamplifier 2 having an amplification factor of 30, the output becomes 30. If you give it, the output should be 30, but
In fact it can be different. Therefore, by giving a correction coefficient in accordance with the amplification factor, accurate determination can be made.

The correction coefficient can be easily obtained from an output obtained when a known input signal for calibration is given as a true value, and the difference can be easily obtained. If the information is stored in the memory, the microcomputer 5 can automatically correct the data by referring to the table when performing the arithmetic processing.

The following table shows a specific example.

[0024]

[Table 1] Then, prior to the specification, an input of a calibration signal level is given in a state where each amplification factor is set, and a correction coefficient is obtained from the output. Further, when performing the actual processing, the microcomputer 5 recognizes an actual value based on the above-described correction coefficient, and makes various determinations based on the actual value. Then, the correction table actually stored is the one in which the amplification factor and the correction coefficient are associated with each other in the above table.

Further, in the above-described example, the amplification factor is divided into three stages for convenience of explanation, but the precision is further improved by making it finer. Further, if a calibration signal is input at the center frequency for each bandpass filter as well as the preamplifier and a correction table for the bandpass filter is created, the accuracy is further improved.

In each of the above-described embodiments, the abnormality determination conditions set for each band based on the analyzed frequency characteristics are all effective in the actual abnormality determination mode. The output of the bandpass filter is normal /
A function of determining not to use for determination of abnormality may be added. That is, for example, assuming that the frequency characteristics of the vibration waveform acquired in the initial setting mode are as shown in FIG. 5, the band-pass filter BPF1 having the lowest frequency
Since the vibration level of the band is too low, a normal range cannot be set (particularly on the lower threshold side). Similarly, since the band of the n-th bandpass filter BPFn is too high, it is impossible to set a normal range (especially on the upper threshold side).

Therefore, the microcomputer 5 determines whether or not the upper and lower threshold values in the normal range can be set within a certain range based on the detection result obtained in the initial setting mode. The function of making the pass band of the band-pass filter not used (disabled) for the determination of normal / abnormal is added. As a result, the vibration in the frequency band suitable for the target can be selectively used, and the reliability of the detection result is improved.

In each of the above-described embodiments and modifications, the microcomputer 5 fetches and analyzes the output of each band-pass filter BPFi in the initial setting mode and the abnormality determination mode in accordance with the periodicity of the target vibration. Can be synchronized automatically.

That is, when an object to be monitored for abnormality is rotating at a constant speed, such as a motor, vibration always occurs. On the other hand, when the target is various machine tools, the operation is intermittently driven, and the vibration is reduced as shown in FIG.
Some occur intermittently as shown in FIG. In such a case, if the normal / abnormal determination is always performed based on the output of the vibration sensor 1 during the operation of the abnormality determination mode, there is a period during which the vibration is reduced. There is a possibility that it becomes lower than the lower threshold value and is determined to be abnormal. In addition, if the learning is performed based on a small vibration level in the initial setting mode, the upper and lower thresholds become lower as a whole. There is a possibility that the level becomes high and it is erroneously determined to be abnormal.

Therefore, in the case of such an apparatus, since there is usually a trigger signal for driving the apparatus, the apparatus receives the trigger signal (external trigger) and performs processing for a certain period after receiving the trigger signal. I have to. Such processing corresponds to both the learning processing in the initial setting mode (setting of the upper and lower thresholds) and the determination processing in the abnormality determination mode. Except for processing for a certain period of time after receiving an external trigger, the processing is the same as in each of the above-described embodiments, and a detailed description thereof will be omitted.

On the other hand, there is a case where the above-mentioned external trigger cannot be obtained depending on the object. In such a case, FIG.
As shown in (1), an internal trigger is generated on the abnormal vibration detection device side, and learning and determination processing of the presence / absence of an abnormality are performed only for a predetermined time from the internal trigger. The internal trigger is generated when the vibration exceeds a certain level.

That is, for example, apart from the band filters that pass through individual narrow frequency bands used in each of the above-described embodiments and modifications thereof, band pass filters that pass through all of these frequency bands are prepared. The output of the broadband band-pass filter is monitored, and when the output (vibration level) of the output exceeds a predetermined value, an internal trigger is generated. Further, the output signal of the preamplifier 2 may be monitored as it is without using such a wide band pass filter, and an internal trigger may be output if a vibration (not abnormal) above a certain level is detected. Good. Furthermore, even if the frequency band is narrow, an internal trigger may be output based on the level of an output signal passing through a specific frequency band. Except for using the internal trigger, it is the same as the above-described case of the external trigger, and thus a detailed description thereof is omitted.

[0033]

As described above in detail, the abnormal vibration detecting apparatus according to the present invention does not require a high-grade and expensive processor capable of performing a fast Fourier transform, and is based on the outputs of a plurality of band-pass filters. By analyzing the frequency characteristics of the target vibration, normal vibration and abnormal vibration can be accurately distinguished. In addition, since the frequency of the normal vibration of the target is analyzed in detail and the abnormality judgment conditions for each band are automatically set, not only the specific target, but also a high performance and versatility that can be used for various target machines It becomes a high abnormal vibration detection device.

[Brief description of the drawings]

FIG. 1 is a circuit block diagram of an abnormal vibration detection device according to a first embodiment of the present invention.

FIG. 2 is a graph showing an outline of frequency analysis in the first embodiment.

FIG. 3 is a circuit block diagram of an abnormal vibration detection device according to a second embodiment of the present invention.

FIG. 4 is a graph showing an outline of frequency analysis in the second embodiment.

FIG. 5 is a graph showing an outline of frequency analysis for explaining an operation of a modification of the present invention.

FIG. 6 is a graph showing an outline of frequency analysis for explaining an operation of a modification of the present invention.

FIG. 7 is a graph showing an outline of frequency analysis for explaining an operation of a modification of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Vibration sensor 2 Preamplifier BPFi (BPF1-BPn) Band-pass filter 3 Multiplexer 4 A / D converter 5 Microcomputer (processor) 6 Operation part 7 Memory

Claims (5)

[Claims] 1. An abnormal vibration detecting device having the following requirements (1), (2) and (3). (1) A vibration sensor that outputs a signal corresponding to a vibration acceleration, a plurality of band-pass filters having different pass bands to which the output of the vibration sensor is input, and a processor that takes in the output of each band-pass filter and performs arithmetic processing. Having. (2) The initial setting mode is executed in a state where the vibration in the normal mode is applied from the target to the vibration sensor. In the initial setting mode, the processor sets an abnormality determination condition for each band based on the output level of each band pass filter. (3) Execute the abnormality determination mode after executing the initial setting mode. In this mode, the processor has at least one
Based on the outputs of the two band-pass filters, normal / abnormal is determined by illuminating the abnormality determination condition set for the selected band, and an abnormal vibration detection signal is generated according to the determination results. 2. The abnormality determination condition according to claim 1, wherein an upper threshold value and a lower threshold value for discriminating the level of the output of the band-pass filter are set differently for each band. An abnormal vibration detection device according to item 1. 3. A function of normalizing an output level of each band provided to the processor by individually setting a gain for each band based on an output level of each band-pass filter in the initial setting mode. The abnormal vibration detecting device according to claim 1, further comprising: 4. The initialization mode has a function of determining, based on the obtained output level of each band-pass filter, that the output of a specific band-pass filter is not used for determining whether the output is normal or abnormal. The abnormal vibration detection device according to claim 1. 5. The apparatus according to claim 1, further comprising a function of executing the processing of capturing and analyzing the output of each band-pass filter by the processor for a certain period in accordance with the operation of the object, and disabling the other period. Item 5. The abnormal vibration detection device according to any one of Items 1 to 4.