JPS61189315A - Damage detector for bearing of rolling mill - Google Patents

Abstract

PURPOSE:To early detect damage of every bearing by detecting a supersonic signal generated by damage of a bearing and discriminating same from noise according to a characteristic of the signal. CONSTITUTION:Back-up rolls 3a, 3b disposed outside work rolls 2a, 2b are supported by bearings 4a, 4b, 5a, 5b, and an acoustic sensor 10 is provided on the bearing 4b. An acoustic signal received by the acoustic sensor 10 is input to a filter 12 through an amplifier 11. Subsequently, the signal processed by an envelope detection circuit 13 is input to a continuous signal discrimination circuit 14 and a sudden signal discrimination circuit 15. The signal processed by the respective discrimination circuits is input to a bearing abnormality discrimination circuit 16. The bearing abnormality is discriminated by the bearing abnormality discrimination circuit 16 and the result is displayed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Application of the Invention] The present invention relates to a damage detection device for sliding bearings and rolling bearings used in rolling mills.

[Background of the invention]

It is known that sliding bearings and rolling bearings are used in combination in rolling mills, and damage to either of them can cause the plant to shut down.

In sliding bearings, the causes include accidents in the lubricating oil system, excessive bearing load, rising lubricating oil temperature, and poor alignment between the bearing bushing and sleeve.On the other hand, in rolling bearings, grease deterioration and Damage phenomena occur due to fatigue damage, etc.

Conventionally, as a means of monitoring sliding bearing damage, the temperature of the lubricating oil drained was measured, and when the temperature of the drained oil exceeded a certain value, it was assumed that the bearing was damaged and the bearing was replaced.

In addition, vibration detection methods are used to monitor damage to roller υ bearings.
If the amount of vibration exceeded a certain value, it was assumed that the bearing was damaged and the bearing was replaced. Even if any of the above damage occurs, the energy consumed and the loss of production opportunities associated with bearing replacement will be enormous, such as having to uselessly keep the sujip warm in a heating furnace for the time required to replace the bearing. It was a cine economy.

By the way, the above-mentioned drain oil temperature measurement method is a method of determining the temperature rise of the bearing that occurs when the bearing is damaged through the drain oil temperature, so it has a slow response and cannot be detected unless the damage is very advanced. In many cases, an increase in oil temperature cannot necessarily be said to be a sign of bearing damage, and there were problems in terms of reliability. On the other hand, the vibration measurement method similarly has the problem that it cannot be detected unless rolling bearing damage has progressed to a certain extent. Furthermore, since vibration is generated when a speci bearing is damaged, it also has the drawback of receiving this vibration and erroneously diagnosing damage to the rolling bearing. That was difficult.

As an improvement to the above technology, there is a method for monitoring acoustic signals (mainly in the ultrasonic range) that can detect early damage states in the case of sliding bearings. For example, damage diagnosis method for sliding bearings used in steam turbines and generators (Japanese Patent Laid-Open No. 56-53422
．． This is a method using Japanese Patent Application Laid-Open No. 57-54835). Since the above patent is a means of detecting signals that have periodicity and occur intermittently, there is an increase in continuous signals that occur due to damage to rolling mill bearings, and sudden signals that occur randomly in addition to the continuous signals. It is difficult to detect features such as type signals. Furthermore, Japanese Patent Laid-Open No. 50-67182 discloses a technique for detecting damage to a roller bearing, but in this patent, the technique is a means for detecting a periodic vibration signal generated due to damage to a roller bearing. Acoustic signals generated due to damage to rolling bearings used in rolling mills are difficult to detect using the above method because sudden signals are generated randomly. Furthermore, in both of the above patents, it is difficult to separately determine damage to sliding bearings, but damage to sliding bearings.

Therefore, the plain bearings and rollers used in rolling mills are
There is a need for technology that can detect damage to bearings at an early stage and also determine which bearing is damaged.

[Purpose of the invention]

The purpose of the present invention is to detect damage to sliding bearings and rolling bearings used in rolling mills at an early stage, and to determine whether damage to sliding bearings is caused by damage to roller 9 bearings. An object of the present invention is to provide a bearing damage detection device.

[Summary of the invention]

A feature of the present invention is that an acoustic sensor detects ultrasonic signals generated by damage to sliding bearings and rolling bearings, and distinguishes them from noise based on characteristics such as the shape and size of these signals. The key point is that damage can be detected early.

[Embodiments of the invention]

The present invention detects damage to the above-mentioned sliding bearing and rolling bearing.One embodiment of the present invention will be described below with reference to the drawings.

FIG. 2(A) shows a simplified configuration of a rolling mill and a partial detailed structure of a bearing for the rolling mill. Rolled material 1 is work roll 'la
, 2b, the work rolls 2a,
There are backup rolls 3a and 3b on the outside of roll 2b.

Bearings 4a are provided at both ends of the backup rolls 3a and 3b.
, 4b and 5a.

5b is installed. The bearings 4a, 4b and 5a.

The structure of 5b is different, as shown in the same figure (B).
A sliding bearing 6 for supporting the load in the radial direction and a roller bearing 7 for supporting the load in the thrust direction are installed on the bearings 8 and 4b. On the other hand, 5a,
No rolling bearing is installed in the bearing 5b, and it consists only of a sliding bearing.

Among the bearings of a rolling mill, an example will be described in which an acoustic sensor 10 is installed on the bearing 4b shown in FIG. 2 to diagnose damage to the bearing. The acoustic signal received by the acoustic sensor 10 is
The signal is input to a filter 12 via an amplifier circuit 11. The filter 12 removes unnecessary frequency components, and is composed of, for example, a band-pass filter that passes signals in the frequency range from 100 kHz to 2 MH1. Next, the signal passed through the envelope detection circuit 13 and subjected to envelope detection processing is passed to the continuous type signal discrimination circuit 14 and the sudden type signal discrimination circuit 15. The signals processed by the respective discrimination circuits are input to the bearing abnormality discrimination circuit 16.
The determination of bearing abnormality and its results are displayed.

Next, the detection operation in the above configuration will be explained in detail with reference to the output waveform diagrams of each circuit shown in FIGS. 3 and 4. FIG. 3 shows an example of the operation of each circuit when a sliding bearing is damaged; FIG. 4 shows an example of the operation of each circuit when a rolling bearing is damaged. The envelope detection circuit 1 shown in FIGS. 3 and 4
As shown in the output waveform No. 3, even when the bearing is normal, a sudden signal is periodically generated once per revolution of the bank unroll in the rolling process. This is a characteristic unique to rolling mills, as it receives acoustic signals generated from sliding bearings. That is, in the sliding bearing of a rolling mill, as shown in FIG. Sliding. In the sliding bearing of a rolling mill, since the key 101 is present, the m-P of the sleeve 102 protrudes slightly compared to the others, as shown in FIG. 1(B). During the rolling process, when the protrusion P reaches the pressure-receiving position, that is, when the maximum rolling load is applied to the noon part, a slight contact phenomenon occurs with the bushing 104. This causes a periodic sudden sound signal to be generated once per rotation of the backup roll.

By the way, if the sliding bearing mentioned above is damaged, @3
As seen in the output waveform of the envelope detection circuit 13 shown in Figure (A), the amplitude of the DC component gradually increases in the initial stage. This is because the metal of the sliding bearing gradually undergoes plastic deformation and its area gradually expands, resulting in an increase in the continuous acoustic signal. As the damage progresses further and the damage state enters the middle stage, as shown in the figure, a sudden type signal with a large amplitude is added to the continuous type signal. This is due to deformation or peeling of the bearing metal. A characteristic of the damage process of sliding bearings for rolling mills is that the above-mentioned signals are generated over time.

The envelope detection signal obtained by the phenomenon described above is then input to the continuous type signal discrimination circuit 14 and the sudden type signal discrimination circuit 15. As shown in FIG. 2, the continuous signal discrimination circuit 14 includes two comparison circuits 17a and 17b with different comparison voltages.
Nuclear comparison circuit 1 shown in FIG. 3(A) is provided.
7H comparison voltage Els comparison voltage E2 of the comparison circuit 17b
It is compared with the envelope detection waveform. (B) and (
C) shows the output waveforms of the comparison circuits 17a and 17b. Next, the outputs of the comparison circuits 17a and 17b are output from the rising edge detection circuit 1.
8a and 18b, respectively, and as shown in FIGS. 3(E) and 3(G), only the rising edge of the outputs of the comparison circuits 17a and 17b is detected. Furthermore, the comparison circuit 1
The outputs of 7a and 17b are input to integration circuits 19a and 19b. Furthermore, the outputs of the rise detection circuits 18a and 18b are input as reset signals for the integration circuits 19a and 19b. By inputting the output of the comparison circuit and the reset signal, the integration circuit 19a.

The waveforms shown in (F) and (H) in the figure are output from 19b. Next, the output of the integration circuit is a comparison circuit 20a,
20b, and is compared with the comparison voltage E4 of the comparison circuit 20H and the comparison voltage E5 of the comparison circuit 20b shown in FIG. ) and (J) are obtained.

Further, the output of the comparison circuit is inputted to an AND circuit 21, whereby a logical output shown in FIG. 2(K) is obtained from the AND circuit 21. That is, by performing the above-described processing, it is possible to detect a continuous signal generated in the initial process of sliding bearing damage.

Although two types of comparison circuits 17a and 17b with different comparison voltages are used in the continuous signal discrimination circuit 14 described above, it is possible to detect a continuous signal even if only the comparison circuit 17a is used. Furthermore, it goes without saying that if two or more types of comparison circuits with different comparison voltages are used, it becomes possible to determine continuous signal generation with even higher accuracy.

On the other hand, the output of the envelope detection circuit 13 is input to the sudden signal discrimination circuit 15, and then to the comparison circuit 22 shown in FIG. The comparison voltage of the comparison circuit 22 is shown in FIG. 3(A).
As shown in FIG. 2, since the comparison circuits 17a and 17b are set at a higher level (comparison voltage El), the continuous signal that occurs at the initial stage of bearing damage is not detected, and it is not possible to detect sliding bearing damage. Only sudden signals that occur after the initial stage, ie, in the middle stage of damage, are detected. The output waveform of the comparator circuit 22 is shown in (D) of the same figure. Next, the output of the comparison circuit 22 is input to the counter 23. The counter 23 counts the pulses output from the comparator circuit 22 at regular intervals, and outputs a value corresponding to the counted value to the abnormality detection circuit 24 as shown in FIG. 3(L).

As shown in FIG. 3, the number of sudden sound signals generated in the middle stage of sliding bearing damage is several times greater than the sudden sound signals generated during normal conditions. The abnormality detection circuit 24 detects when the value output from the counter 23 is several times higher than the normal value and outputs the logic signal shown in the figure. That is, at the time when the logic signal is output, it is notified that the sliding bearing has reached the intermediate stage of damage.

Next, we will explain the operation of each circuit when the roller υ bearing is damaged. When the rolling bearing is damaged, sudden signals will randomly occur as shown in the output waveform of the envelope detection circuit 13 shown in FIG. 4(A). This is a characteristic of the acoustic signal generated when a roller used in a rolling mill is damaged. The envelope detection signal obtained by the above phenomenon is input to the continuous type signal discrimination circuit 14 and the sudden type signal discrimination circuit 15. The output waveforms of each circuit constituting the continuous signal discrimination circuit 14 are shown in FIGS. 4(B) to 4(K). In the acoustic signal generated by rolling bearing damage, a sudden type signal is generated, but a continuous amount is not generated, so the integration circuit 1
The amplitudes of the output waveforms (F) and (H) of 9a and 19b do not become large.

Therefore, since no signals are output from the comparison circuits 20a and 20b, no logic signal is output from the AND circuit 21 as shown in FIG. 4(K).

On the other hand, the sudden type signal discriminating circuit 15 detects sudden type signals generated due to damage to the roller bearing.
A logic signal is output from the abnormality detection circuit 24 as shown in FIG. That is, at the time when the logic signal is output, it is notified that the roller υ bearing is damaged.

As described above, logical signals are output from the continuous type signal discrimination circuit 14 and the sudden type signal discrimination circuit 15, respectively, due to damage to the sliding bearing and the rolling bearing. Next, the logic signal is input to the bearing abnormality determination circuit 16 described above. The bearing abnormality determination circuit 16 is an AND circuit 30. NAN
D circuit 31a.

31b and an alarm 32 using, for example, an LED.
a, 32b, and 32c, and determines and displays the initial stage of the sliding bearing damage state and the middle stage of the damage, as well as whether the roller 9 bearing is in the damaged state.

Here, the operation of the bearing abnormality determination circuit 16 will be explained. For convenience of explanation, the logic output of the continuous signal discrimination circuit 14 is
It is assumed that the logic output of the sudden type signal discrimination circuit 15 is Y. The table shows the X and Y outputs and alarm operating conditions that occur due to bearing damage.

During normal conditions, no signals are output for both X and Y, so the alarm does not light up. On the other hand, in the initial state of sliding bearing damage, as described above, the X output is 1 and the Y output is 0, so the alarm 32a lights up. Further, in the middle stage of damage to the sliding bearing, 1 is output for both X and Y, so the alarm 32b lights up. Furthermore, if the rolling bearing is damaged, the X output becomes 0 and the Y output becomes 1, so the alarm 32C lights up. That is, since an alarm is displayed for each phenomenon, the operator can know in real time what condition the rolling mill bearing is in during operation. Furthermore, if an abnormality occurs, appropriate measures can be taken to avoid accidents.

〔Effect of the invention〕

As explained above, according to the present invention, damage to the sliding bearing used in a rolling mill can be detected as a phenomenon and at an early stage, thereby preventing accidents such as rolling plant stoppage due to bearing damage. It has extremely significant effects, such as reducing emissions and contributing to highly efficient plant operation.

[Brief explanation of drawings]

FIG. 1 is an embodiment of a bearing damage detection device for a rolling mill according to the present invention, FIG. 2(A) is a simplified configuration diagram of a rolling mill and a rolling mill bearing, FIG. 3. Figure 4 is the first
FIG. 5 is an output waveform diagram for explaining the operation of an embodiment of the present invention shown in the figure, and FIG. 5 is a structural diagram of a Pesci bearing for a rolling mill. 6...Sliding bearing, 7...Roller bearing, 10...
- Acoustic sensor, 11... Amplification circuit, 12... Filter, 13... Envelope detection circuit, 14... Continuous type signal discrimination circuit, 15... Sudden type signal discrimination circuit, 16...
Abnormality determination circuit. 2nd solid (c) (δ)

Claims (1)

[Scope of Claims] 1. In a bearing damage detection device for detecting damage to sliding bearings and rolling bearings installed in a bearing section of a rolling mill, an acoustic signal for detecting acoustic signals generated by each damage. It comprises a sensor and a means for detecting the output of the acoustic sensor, and a means for determining whether a continuous type signal is generated and a means for determining whether a sudden type signal is generated among the acoustic signals generated due to damage to the sliding bearing or the rolling bearing. and means for determining a damaged state of the bearing in response to output signals from the continuous type signal generation determination means and the sudden type signal generation determination means. Bearing damage detection device. 2. In claim 1, the determination means determines whether a continuous signal is generated by the continuous signal generation determination means and the sudden signal generation determination means, and determines whether a sudden signal generation occurs. If this is not done, it is determined that the sliding bearing is in an initial state of damage, and if no continuous type signal is generated and a sudden type signal is determined to be generated, it is determined that the rolling bearing is damaged and the continuous type signal is not generated. , a device that determines that damage to the sliding bearing has entered a medium-term state when it is determined that the occurrence of any of the sudden type signals, and a bearing abnormality determination device that issues an alarm in response to the respective determination results. A bearing damage detection device for a rolling mill, characterized in that: