JPH045518A - Abnormality diagnostic system for rolling facility - Google Patents

Abstract

PURPOSE:To improve the diagnostic accuracy by monitoring a facility and a processing state by a sensor group, analyzing the phenomenon of abnormality when the abnormality is detected, and estimating the cause of the abnormality according to the analytic result, and diagnosing a rolling line synthetically. CONSTITUTION:Some of signals detected with the process monitor sensor signal group 2 and facility monitor sensor signal group 3 of the facilities 1 are sent to an arithmetic storage part 8 after the primary signal processing of a signal processing part 4, and others are sent through a signal repeating part 5 and a signal transmitter 7 without being processed. The arithmetic storage part 8 performs diagnostic operation, data control, etc. The diagnostic result is stored in the arithmetic storage part 8 or stored in a storage part 10 which can shift in connection position on the transmitter 7. Further, a creation part 9 can structure a new diagnostic means based upon a knowledge base generated according to past diagnostic results, etc.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention particularly relates to a method for diagnosing abnormalities in equipment and processes in rolling equipment, particularly rolling equipment in which a plurality of rolling mills are arranged in tandem.

[Conventional technology]

In order to maintain stable production quality in hot or cold rolling equipment, a rolling line diagnostic system based on equipment diagnostic technology and process diagnostic technology derived therefrom is necessary.

Conventionally, as a diagnostic method that can be applied to such rolling mills, for example, Japanese Patent Publication No. 61-42300, "Materials and Processes J
VOl, 2. NO, 2Lfl, 475 (1989
) There is a method disclosed in .

[Problem to be solved by the invention]

However, in the conventional diagnosis method described above, diagnosis is only performed for specific specifications that are set in advance for the equipment to function; It was only a data analysis, and was not a diagnostic method that functionally linked rolling equipment and processes.

Therefore, it has not been possible to obtain a comprehensive diagnosis of rolling equipment and operating conditions, and it has been impossible to accurately predict how an emerging problem will affect subsequent rolling operations.

The problem to be solved by the present invention is to find an effective means for diagnosing, for example, abnormal thickness of rolled material during rolling operation from both the equipment itself and the operating process.

Unique means for solving the problems] The present invention consists of 1. The status of equipment and processes is monitored using a group of sensors, and when an abnormality is detected, the phenomenon is analyzed. Based on the analysis results, a knowledge base that functionally combines equipment diagnosis and process diagnosis that has been accumulated in advance is used. This is an abnormality diagnosis method for rolling equipment that estimates the cause of the abnormality and uses this as action guidance for handling the abnormality.

Specifically, it consists of a group of sensors that detect information about the equipment and the rolling operation process, a signal processing unit installed near the equipment that performs primary processing of the signals detected by the sensors, and a secondary processing unit that processes the signals detected by the sensors. It consists of a remotely installed computing and storage unit that performs processing, diagnosis, data management, etc., and a creation unit that constructs new diagnostic methods using a knowledge base. Each unit uses a diagnostic device that is connected by a signal transmission device.

As the sensor, we use a sensor dedicated to diagnosis and a sensor for rolling operation, and as the signal processing section, for example, we use a system based on a personal computer and a general-purpose signal analysis device.
For example, a process computer for operation is used in the calculation and storage department, and a workstation is used in the creation department.

Furthermore, a personal computer or the like whose connection position to a signal transmission device other than that of the signal processing section is variable is also used as the storage section.

[Effect]

Referring to FIG. 1 showing a functional image of the present invention, a group of signals from a process monitoring sensor group and an equipment monitoring sensor group in equipment are collected via a data way into a host computer which is a calculation/storage unit. These signals are analyzed by the process diagnosis and analysis section and the equipment diagnosis and analysis section, and the results are corrected by knowledge data processing using a knowledge base based on knowledge sources such as operators, and an abnormal operation avoidance action guide. Instructions will be given as action guides for avoiding equipment troubles and action guides for improving operational levels.

Therefore, the time from the occurrence of an abnormality to the return to normality can be shortened and accurate guides can be provided.Moreover, these guides can be stored in the host computer and used for the next inference.

〔Example〕

FIG. 2 shows an example of detailed diagnostic items in the present invention. As shown in the figure, equipment diagnosis can be broadly divided into drive system diagnosis, hydraulic/lubrication equipment diagnosis, electrical/instrumentation equipment diagnosis, inspection data management, etc., and each of them is further divided into:
It can be divided into vibration monitoring, hydraulic circuit troubleshooting, etc., and specific diagnosis for each can be performed using further subdivided items.

On the other hand, for process diagnosis, for example, gauge (plate thickness)
, surface flaws, and each covers gage abnormality diagnosis, lubrication monitoring, etc.

FIG. 3 is a block diagram showing an example in which the present invention is applied to a tandem type cold rolling equipment 1. The equipment 1 is provided with a process monitoring sensor/signal group 2 for operation and an equipment monitoring sensor/signal group 3 dedicated for diagnosis, and the status of the process and equipment is monitored online. Some of the signals detected by the process monitoring sensor/signal group 2 and the equipment monitoring sensor/signal group 3 are subjected to primary signal processing in the signal processing unit 4 installed near the equipment, and some are subjected to signal processing. The signals are transmitted through the signal relay section 5, the signal transmission device 7, and the arithmetic/storage section 8 installed remotely. Diagnosis, data management, etc. are performed in this calculation/storage B8 using the signals, and at this time, rolling/operation information obtained from the rolling equipment control device group 6 may also be used. The diagnosis results are stored in the arithmetic/storage section 8, but can also be stored in the storage section 10, which can be connected to a variable location on the signal transmission device 7.

Furthermore, a creation unit 9 connected to the signal transmission device 7 can construct a new diagnostic method based on a knowledge base created from past diagnostic results and the like.

Note that the creation section 9 may be located on the same computer as the calculation/storage section.

FIG. 4(a) shows an example of the content of the abnormality cause estimation knowledge in the present invention, and Q)) shows an example of the content of the akunyong determination knowledge based on the estimation.

Using the above diagnostic items, a cold rolling diagnosis-action guide was obtained.

Example 1 Plate thickness abnormality diagnosis was performed as an item of process diagnosis.

In online monitoring, the plate thickness is constantly monitored using a sensor (plate thickness gauge), and it is determined whether the plate thickness is smaller than a preset abnormality level.If it is larger than the abnormality level, the occurrence of plate thickness abnormality is detected, and the Investigate the cause.

Examples of causes of the abnormality include the following items.

・AGC abnormality: This is an abnormality that occurs due to incorrect gain settings of the AGC (automatic plate thickness control device), and a phenomenon occurs where the abnormality is amplified.

・Roll eccentricity abnormality: This is an abnormality that occurs because the rolling roll is eccentric.
The abnormality occurs in accordance with the roll rotation period.

- Mill speed uniformity abnormality: This is an abnormality that occurs because the rotational speed ratio of the mill motor (rolling mill drive motor) of each stand is not constant, and a noticeable abnormality occurs during accelerated rolling or decelerated rolling.

・Schedule abnormality: This is an abnormality that occurs due to incorrect setting of the rolling schedule, and the abnormality with large fluctuations occurs the moment the incorrect schedule is set.

- Plate thickness gauge abnormality: This is an abnormality that occurs due to a circuit defect in the plate thickness gauge, and appears shockingly. Therefore, false data is given to the AGC,
A shocking anomaly is created.

Since the frequency with the maximum value of the frequency analysis result of the final stand plate thickness signal matched the roll rotation frequency of stand #3, the roll eccentricity of stand #3 was confirmed.

On the other hand, as a knowledge base, there is a judgment rule that says, ``If there is an abnormality in roll eccentricity of stand #N, check the hydraulic pressure reduction control system of stand #N.''

Based on this knowledge base, we performed a diagnosis of the hydraulic pressure reduction control system of #3 stand in equipment diagnosis using frequency response.
An abnormality in the servo amplifier was confirmed.

At this stage, as a knowledge base, we use the judgment rule that ``If #N stand roll eccentricity affects the plate thickness and there is an abnormality in the hydraulic pressure reduction control system, #N stand reduction control is defective.'' As the cause of the abnormality, we were able to obtain "#3 stand pressure reduction control system failure (medium failure)" and as action guidance, "Please lower the #3 stand pressure reduction control gain by 30%."

Example 2 Drive system diagnosis was mentioned as a diagnosis item for equipment diagnosis.

A simple diagnosis revealed an abnormality in the #2 pinion stand bearing vibration value.

Immediately, we performed a dimensional analysis as a detailed diagnosis, and as the results appeared as shown in Figure 6, we found that the meshing component 2 of the gears had increased, and the cause of this was the wear of the #2 pinion stand gear. I found out something. Here, the order analysis is a frequency analysis in which a unit called "order" is placed on the horizontal axis, for example, how many times the number of rotations is. If the frequency is 120 Orpm, the "rotation frequency" fr is
It is defined as f, -110jLL""=20 [Hz] 60 [sec]. Let this 20Hz be "1st order",
The horizontal axis is scaled. In normal frequency analysis,
The horizontal axis is "frequency", and the unit is, for example, [7].

As a knowledge base, ``If the pinion vibration is abnormal, CAU
Process diagnosis was performed in the same manner as in Example 1 based on the judgment rule "If it is at the TION level, check the #5D plate thickness deviation." As a result of the process diagnosis, it was found that the plate thickness deviation did not exceed the abnormal level, and there was no abnormality in the plate thickness. As a knowledge base, we have a judgment rule that states, ``If there is an abnormality in the vibration of the binion, and the vibration value does not suddenly increase and the plate thickness is not affected, inspect the abnormality on the next inspection/repair date.'' Based on this, the cause of the abnormality is "#2 Binion stand gear is worn (minor failure)" and the action guidance is "Tension between 1 and 2 is reduced to 3".
Please lower it by 0%. Please check the amount of gear wear on the next inspection/repair molar.''

Example 3 As a process diagnosis, plate fracture prediction diagnosis was mentioned.

The tension was primarily monitored using a sensor (tension meter), and it was observed that the tension in the #2 stand instantaneously exceeded the abnormal level several times. As a result, it was determined that there was a sudden change in the #2 stand tension. As a knowledge base, we have a judgment rule that says, ``If there is a sudden change in tension in stand #N, check the speed control system of stand #N,'' and this allows us to control the electrical control of stand #2 during equipment diagnosis. When the system was diagnosed using frequency response, it was determined that shaft resonance had occurred. As a knowledge base, we have a judgment rule that says, ``If #N stand tension suddenly fluctuates and there is an abnormality in the speed control system, #N stand shaft resonance is defective''. ``Shaft resonance has occurred (minor failure)'' and action guidance ``Please adjust the constants of the shaft resonance suppression circuit'' was obtained.

FIGS. 5(a) and 5(a) illustrate the steps from the occurrence of an abnormality in the conventional method and the method of the present invention in a temporal comparison.

In the conventional method, from abnormality detection to action (treatment), human processes such as operator judgment, data collection, analysis and consideration, and the work of experts based on experience and intuition were required, but with the present method, various Processes and equipment are monitored in real time using a group of sensors, and the system performs everything from data collection to inference and optimal guidance output.
Processes and time are significantly shortened, and once the system is constructed, diagnostic records can be stored in a knowledge base to make subsequent inferences more probable.

〔Effect of the invention〕

According to the present invention, the following effects can be achieved.

(1) Automatic monitoring can shorten the time required to discover anomalies, collect data, and analyze it.

(2) Integration of process diagnosis and equipment diagnosis enables more accurate cause inference.

(3) Appropriate guidance can be extracted, and even unskilled personnel can perform stable and sophisticated operations.

These have made it possible to comprehensively diagnose rolling lines. Furthermore, since abnormality diagnosis and data storage can be performed at the same time, the accuracy of diagnosis can be improved through data analysis at a later date, and the growth of the system can be promoted.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram showing a functional image of the present invention, FIG. 2 is a system diagram showing an example of detailed items of diagnosis in the present invention, and FIG.
The figure is a block diagram showing an example in which the present invention is applied to a tandem type cold rolling equipment, Figure 4 is an explanatory diagram showing part of the abnormality cause estimation knowledge and action decision knowledge in the present invention, and Figure 5 is a conventional method. FIG. 6 is a time chart illustrating the process from the occurrence of an abnormality between the method and the method of the present invention in a temporal comparison, and FIG. 6 is a graph showing the distribution of vibration speed with respect to rotational order ratio. 1: Cold rolling equipment 2: Process monitoring sensor/signal group 3: Equipment monitoring sensor/signal group 4; Signal processing section 5: Signal relay section 6: Rolling equipment control device group 7: Signal transmission device 8: Calculation/storage section 9: Creation Department 1O: Storage unit with variable connection position Patent applicant: Nippon Steel Corporation Representative: Masu Kobori Figures (a) (b)

Claims (1)

[Claims] 1. Monitor the status of equipment and processes using a group of sensors, analyze the phenomenon when an abnormality is detected, and use the analysis results to create a knowledge base that functionally combines pre-accumulated equipment diagnosis and process diagnosis. An abnormality diagnosis method for rolling equipment that estimates the cause of an abnormality based on the information and uses this as action guidance for abnormality treatment.