JPS63309824A - Abnormality diagnosing method for machine - Google Patents

Abstract

PURPOSE:To enable an abnormal position to be decided, to increase deciding ability, and to improve the reliability of diagnosis by finding the effective values of acoustic spectra at respective positions from the detection results of acoustic sensors installed at plural positions inside or outside a machine to be diagnosed, and finding the ratios too a preset reference value. CONSTITUTION:An acoustic sensor 1 set at a reference position detects and amplifies a noise generated by the machine to be diagnosed and a band-pass filter 3 passes a signal in a certain frequency band. Then an effective value circuit 4 finds the effective value of an acoustic spectrum and a microcomputer 6 finds the ratio of the effective value and the reference value of a preset and stored acoustic spectrum. Then it is decided whether the machine is abnormal or not from whether or not the ratio exceeds a certain permissible value. Here, the reference value is set by finding the mean value of acoustic spectrum effective values of many machines.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a method for diagnosing abnormalities in machinery, and in particular, a method suitable for diagnosing failures of rotating machinery, information terminal equipment, etc. using acoustic sensors. This invention relates to a method for diagnosing machine abnormalities.

[Prior Art] Development of failure margin technology for rotating machines, etc. has been progressing, and for example, the technology described in Japanese Patent Application Laid-open No. 1983-63528 is known as a method for diagnosing rotating machines in large plants and equipment. ing.

This technology detects the acceleration of a rotating machine, obtains a vibration spectrum in a predetermined frequency range based on the detection results, and compares the spectrum with a reference spectrum, or obtains the effective value of the spectrum and compares it with the reference value. , an abnormality was diagnosed based on whether the ratio exceeded a predetermined value.

[Problems to be Solved by the Invention] Since the above-mentioned conventional technology uses a vibration sensor that detects the acceleration of a rotating machine, it is possible to identify an abnormal region based on the installation position of the vibration sensor.

However, with respect to machines that are not suitable for measurement such as detecting the acceleration of rotating machines, such as information terminal equipment, it has not been possible to diagnose abnormalities using the above-described diagnosis method for rotating machines.

Therefore, if a similar method is performed using the acoustic sensor output, the following problem arises.

In general, machine noise is caused by the vibration of its structural members, but vibrations do not come from only specific structural members, and noise can also come from, for example, cooling fans installed for cooling the machine. Occur. For this reason, when an acoustic sensor is used, it is necessary to separate unnecessary noise from the detected noise. However, in the conventional technology, detection by an acoustic sensor is also a fixed point observation, and does not include information necessary to separate unnecessary noise.

The present invention has been made in order to solve the above-mentioned problems of the prior art, and aims to provide a method for diagnosing mechanical abnormalities that enables the determination of abnormal parts, has high determination ability, and improves the reliability of diagnosis. This is the purpose.

[Means for Solving the Problems] In order to achieve the above object, the machine abnormality diagnosis method according to the present invention detects the sound pressure of the machine, and based on the detection result, detects the sound pressure within a predetermined frequency range. In a machine abnormality diagnosis method that calculates the effective value of the acoustic spectrum and compares this effective value of the acoustic spectrum with a preset reference value of the acoustic spectrum to determine the abnormality of the machine, the internal or external space of the machine to be diagnosed is Acoustic sensors are installed at multiple positions, and the effective value of the acoustic spectrum at each position is determined based on the detection results of these acoustic sensors. This is a method in which the abnormal part of the machine to be diagnosed is determined based on the difference.

In addition, the above purpose is to detect the sound pressure (acoustic) of the machine to be diagnosed, that is, the equipment to be monitored, at multiple points, calculate the effective value of the acoustic spectrum of the output, and calculate the effective value of the acoustic spectrum of the output at one point (reference position). ) by comparing the effective value of the acoustic spectrum with the reference value of the acoustic spectrum, and further comparing the respective ratios of the effective value of the acoustic spectrum observed at multiple points and the reference value to determine the abnormal part, achieved.

[Operation] First, in the first step, the noise generated from the machine to be diagnosed is detected by an acoustic sensor set at a specific position (reference position), and after amplifying it, the effective value of the acoustic spectrum is determined. The ratio between the effective value of the spectrum and the preset reference value of the acoustic spectrum is determined, and an abnormality in the machine is determined based on whether the ratio exceeds a certain tolerance value.

Here, the reference value is set in advance by calculating the average value of the effective values of the acoustic spectrum of a large number of machines.

Generally, when a machine deteriorates, parts wear out, breakage, etc. occur, and when looking at machine elements, bearings, gears, etc. are the targets of diagnosis. Abnormality determination is possible.

By the way, in mechanical equipment, for example, multiple units that perform the same function may be installed in a housing, but in such machines, even if an abnormality occurs in one of the units, the acoustic spectrum will remain the same. , it is not possible to determine in which unit the abnormality has occurred using only the method of the first stage.

Therefore, in the method of the present invention, acoustic sensors are placed at multiple points close to each unit, and the effective value of each acoustic spectrum is determined based on the detection results of the acoustic sensors at multiple points set according to the machine to be diagnosed. By determining the ratio of each acoustic spectrum to a reference value set in advance at the reference position and comparing these ratios, it is possible to separate and identify, for example, that a unit with a large ratio is abnormal. .

[Example] Hereinafter, one embodiment of the present invention will be described with reference to the drawings.

The drawing is a block diagram of a preferred apparatus for carrying out a method for diagnosing a machine abnormality according to an embodiment of the present invention.

In the drawing, 1 is an acoustic sensor that detects the sound pressure of the machine to be diagnosed, 2 is an amplifier that amplifies the detection signal of the acoustic sensor 1, and 3 is a band-pass filter for band extraction that passes signals in a certain frequency band. , 4 is an effective value circuit that calculates an effective value, that is, the root mean square of the acoustic spectrum.
t MeanSquare calculation circuit (hereinafter referred to as RMS
circuit). 5 is an A/D converter, 6 is a microcomputer, 7 is a display section, 8 is a storage section, 9 is a D/A converter,
10 indicates a center frequency switching circuit.

The acoustic sensor 1 is connected to an amplifier 2 and further connected to a pandopass filter 3, where a signal in a certain frequency band is extracted, further averaged by an RMS circuit to calculate an effective value, and then sent to an A/D converter. 5, the data is taken into the microcomputer 6 as a digital quantity. The center frequency of the bandpass filter 3 is switched by a center frequency switching circuit 10, which is controlled by a microcomputer 6 via a D/A converter 9. After data collection within a predetermined frequency range, the effective value data of the acoustic spectrum is stored in the storage unit 8. The display section 7 is
Equipped with a teletype, it displays diagnostic results and measurement conditions, as well as inputting commands from the user.

In this embodiment, as a first step, a specific position (reference position) is
Although only one acoustic sensor 1 is installed in the sensor, there is no problem even if a plurality of sensors are used.

The storage unit 8 stores, as a reference value, an effective value of an acoustic spectrum collected when the machine is in a stable region.

Note that this reference value may be the average value of the effective values of the acoustic spectra of a large number of devices with the same specifications.

When the ratio between the detected effective value of the acoustic spectrum and the preset reference value of the acoustic spectrum exceeds a certain tolerance value, it is determined that the machine is abnormal, and the display unit 7 displays the abnormality. Then, the display section 7 displays that acoustic sensors are placed at other multiple points (multiple positions) to capture data.

Thereafter, as a second step, acoustic sensors are placed at a plurality of points, and the sound pressure of the machine is detected at each of the plurality of positions in the same manner as above, and the effective values of the acoustic spectrum are collected and stored in the storage unit 8. The microcomputer 6 calculates the ratio between the effective value of the acoustic spectrum at each of these positions and the reference value of the acoustic spectrum that has been collected and set in advance, finds an acoustic sensor position with a large ratio, and sets the sensor position. The abnormal part (abnormal part) corresponding to is displayed on the display section 7.

Here, a method of determining the ratio between the effective value of the acoustic spectrum and the reference value in the above will be explained.

Assuming that the sound pressure is P, the nth acoustic spectrum in one frequency domain is P (n), and the acoustic spectra to be compared are P□(n), P, and (n), respectively, in this example,
The ratio R is calculated by the following formula.

Here, ■ is the number of spectra.

In addition to the method described above, there is also a method of determining the ratio for each acoustic spectrum and averaging the ratios, and any method may be used.

According to this embodiment, the diagnostic device can be configured to the same extent as the conventional one, and can not only determine the abnormal state of the machine, but also determine (separate) abnormal parts (parts) inside the machine.

identification), which not only shortens the time required for machine inspection and inspection, but also improves the reliability of diagnosis, increases machine reliability, and significantly reduces economic losses.

Further, in this embodiment, not only is the diagnostic ability high, but also the abnormality diagnosis can be performed using the same procedure as the conventional method until an abnormality occurs in the machine, and the time required for the diagnosis does not change.

In addition, in the above-mentioned embodiment, an example is given in which the abnormal region is determined by calculating the ratio between the effective value of the acoustic spectrum at multiple positions and the effective value (reference value) of the acoustic spectrum when the machine is in the stable region in advance. Although described above, the present invention is not limited to this. For example, parts that are clearly in an abnormal state are installed in a machine in advance, and the effective values of the acoustic spectrum at multiple positions are determined,
It is also possible to determine abnormal areas by comparing with this.

[Effects of the Invention] As described above, according to the present invention, it is possible to provide a method for diagnosing an abnormality in a machine, which enables the determination of an abnormal part, has high determination ability, and improves the reliability of diagnosis.

[Brief explanation of drawings]

The drawing is a block diagram of a preferred apparatus for carrying out a method for diagnosing a machine abnormality according to an embodiment of the present invention. DESCRIPTION OF SYMBOLS 1...Acoustic sensor, 4...RMS circuit, 6...Microcomputer, 7...Display part, 8...Storage part.

Claims (1)

[Claims] 1. Detect the sound pressure of the machine, determine the effective value of the acoustic spectrum within a predetermined frequency range based on the detection result, and compare this effective value of the acoustic spectrum with a preset reference value of the acoustic spectrum. In a machine abnormality diagnosis method that determines the abnormality of a machine using A method for diagnosing an abnormality in a machine, characterized in that the ratio of each acoustic spectrum to a reference value set in advance is determined, and the abnormal part of the machine to be diagnosed is determined based on the difference between these ratios.