JP2001025819A - Method and device for diagnosing abnormality in rolling mill - Google Patents

Abstract

PROBLEM TO BE SOLVED: To speed up a factor estimating process of abnormality (inferior quality) occurring in a coil and to improve the accuracy. SOLUTION: This device is provided with a data collecting part 102 for collecting values observed at the time of operation, quality deciding part 103 for judging the quality of a produced sheet by using the collected data, part 104 for deciding the abnormality factor of detected abnormal parts by dividing the sheet into some regions in accordance with the state of the velocity of the sheet over the entire length of the coil, primary diagnosing part 105 for specifying an abnormal signal by collating it with diagnostic threshold values set by a threshold value constructing part 107 for every value featured by observed signal values and statistic values, and secondary diagnosing part 106 for estimating each abnormal factor by selecting a diagnostic logic of the abnormality factor which is supposed for every regions. In the case that the result determined by the quality deciding part 103 is contrary to the diagnostic result determined by the diagnosing part 105 or 106, the latter diagnostic threshold value is corrected by a threshold value adjusting part 108 so that the diagnostic accuracy improves.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for diagnosing equipment failure and operation failure in a steel mill for rolling mills composed of multi-stage stands.

[0002]

2. Description of the Related Art In a rolling line such as a cold rolling mill in a steel plant, various defects such as equipment defect, material defect, control defect, etc., which are considered as causes of abnormalities such as plate thickness variation and shape variation. Quickly extract the factors,
It is necessary to improve the yield. For this reason, conventionally, as described in Japanese Patent Application Laid-Open No. 4-9214, after detecting a failure of a plate, the cause of occurrence is specified by diagnostic logic such as an expert system using various operation data and equipment data. Was.

[0003]

In the above prior art, a portion where an abnormality has occurred and the type of the abnormality (local / entire area) are specified in the entire length of the coil, and an abnormality factor estimating process is performed for the entire area. Therefore, the processing speed tends to be slow. Also, there is no consideration that the abnormal factors differ depending on the part, and since all the abnormal factors assumed for all regions are searched, it is easy to include noise compared to the diagnosis limited to the part, There was a problem that the diagnostic accuracy was also reduced. Furthermore, when the characteristics of the target change due to aging or the like, the diagnostic threshold used to derive the diagnostic result does not become an appropriate value reflecting the current state, but there is no means for appropriately correcting this. There was also a problem of a decrease in diagnostic accuracy due to a threshold failure.

[0004] It is an object of the present invention to provide a method and apparatus for overcoming the problems of the prior art and for diagnosing abnormalities in a rolling mill with high speed and high accuracy.

[0005]

SUMMARY OF THE INVENTION In order to achieve the above object, according to the present invention, the quality of a plate is determined by using operation data collected during rolling, and when the plate is determined to be defective, the quality of the plate is determined. In a method for diagnosing abnormalities of a rolling mill for estimating an abnormal factor, an area is divided according to a characteristic of a sheet speed with respect to a total length of a coil, and a change in sheet thickness and / or shape is detected for each area, and an area having a change is detected as abnormal. It is characterized by being specified as a part. The area is an FGC (near welding) part, a low-speed part,
It is divided into an acceleration section, a high-speed section and a deceleration section.

[0006] Further, for the abnormal part, a state quantity observed at the time of operation or a statistical state quantity obtained by performing statistical processing on the state quantity, and a diagnostic threshold value set for each state quantity or statistical state quantity are provided. Is characterized in that a diagnosis for specifying an abnormal state quantity is performed by comparing. The state quantity or the statistical state quantity used in the “threshold value calculation” for deriving the conclusion of the presence or absence of an abnormality by comparing with the threshold value is hereinafter referred to as a feature value.

[0007] In addition, for the abnormal part, whether or not an abnormal factor assumed for each area is correct is diagnosed. These abnormal factors include individual factors such as variations in base material plate thickness, unevenness in base material hardness, slip between the plate and the roll, eccentricity of the rolls, and sensor failures. The diagnostic processing is performed by selecting logic.

[0008] A diagnostic apparatus to which the method of the present invention can be applied is a data collection unit for collecting values observed during operation and a quality judgment for judging the quality of a board using the data collected by the data collection unit. And a diagnostic unit for estimating the cause of quality abnormality when the quality determination unit determines that the plate is defective, and divides an area in accordance with the state of the sheet speed over the entire length of the coil. An abnormal content determination unit that detects the presence / absence and specifies an abnormal site in the area unit, wherein the diagnosis unit performs a diagnosis using the data related to the abnormal site specified by the abnormal content determination unit as a target. It is characterized by executing.

The diagnosis section describes a diagnosis process related to each part specified by the abnormality content determination section, and selectively executes the diagnosis processing corresponding to the part specified by the abnormality content determination.

Further, the abnormality diagnosis device has a threshold adjustment unit, and when the quality determination unit outputs a good result and the diagnosis unit outputs an abnormality, the diagnosis unit which has derived the abnormality has a threshold value adjustment unit. If the diagnostic threshold value is corrected in a direction in which an abnormality determination is unlikely to occur, and the quality determination unit outputs a poor determination result and the diagnostic unit outputs a normal diagnostic result, the diagnostic threshold value and the magnitude relationship therewith Then, the diagnostic threshold value whose absolute difference from the threshold value for determining normality / abnormality is equal to or smaller than a certain value is corrected in a direction in which anomaly determination is likely to occur.

[0011]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG. 1 shows a configuration of an abnormality diagnosis apparatus for a rolling mill according to an embodiment of the present invention. Fault diagnosis device 101
Is a data collection unit 102, a quality judgment unit 103, and an abnormal content judgment unit 10.
4.Primary diagnostic unit 105, secondary diagnostic unit 106, threshold construction unit 10
7. It comprises a threshold adjustment unit 108 and a guidance unit 109.

The data collection unit 102 is, for example, a plant 100 which is a tandem cold rolling mill composed of five stands.
And collects the operation data and sends it to the quality judgment unit 103.
The quality determination unit 103 determines the quality of the coil every time one coil is rolled, and determines the quality of the coil.

The abnormal content judging section 104 divides the coil, which has been judged as defective, into regions according to the state of the sheet speed,
A region where an abnormality has occurred is extracted as an abnormal site. The primary diagnosis unit 105 performs a statistical operation such as an average value or a variance value on the directly detected values such as the tension and the load on the abnormal part extracted by the abnormal content determination unit 104. Further, those that are greatly separated from those of the normal coil are detected as abnormal, and the results are transmitted to the guidance unit 109. The secondary diagnostic unit 106 performs a process of specifically estimating an assumed abnormal cause such as a base material plate thickness variation or a roll eccentricity, with respect to the abnormal portion extracted by the abnormal content determining unit 104, Is transmitted to the guidance unit 109.

The threshold value constructing unit 107 compares a value with a diagnostic threshold value to derive a conclusion on the presence or absence of an abnormality in order to construct a diagnostic threshold value used in the diagnostic processing of the primary diagnosis 105 and the secondary diagnosis 106. (This value is called a feature amount), and a threshold value is calculated using the average value and the variance value.

The threshold adjuster 108 maintains the consistency between the result of the determination by the quality determiner 103 and the result of the diagnosis by the secondary diagnostic unit 106, that is, when the result of the determination by the quality determiner 103 is good, The diagnostic result of the secondary diagnostic unit 106 is “no abnormality”, and if the determination result is bad, the diagnostic threshold is adjusted so that any variation factor is diagnosed as the cause of the abnormality.

The guidance unit 109 files the output results of the primary diagnostic unit 105 and the secondary diagnostic unit 106 as a diagnostic log,
This is stored and output to the display device 110 to notify the operator of the diagnosis result.

Next, a quality judgment unit 103, an abnormality content judgment unit 104, a primary diagnosis unit 105, which are main parts of the diagnosis apparatus of this embodiment,
Secondary diagnosis unit 106, threshold construction unit 107, and threshold adjustment unit
The details of the process of 108 will be described.

The quality judging section 103 executes processing according to the processing flow shown in FIG. 3 by using the operation data of one coil thickened by the data collecting section 102. In S3-1,
For the evaluation criteria of the board prepared in advance, values (quality 1 to 3) for determining the evaluation criteria are calculated. In S3-2,
A score is obtained for each item (evaluation criterion) using the evaluation table 200 shown in FIG. In S3-3, the scores of each item are summed up, and the final evaluation
~ D. A is the best, B and C are lower grade, D is
Becomes a defective coil.

The abnormality content judging section 104 basically pays attention to the plate speed, and divides the entire length of the coil into partial areas according to the state. That is, the data is divided and managed into a low-speed part, an acceleration part, a high-speed part, a deceleration part, and an FGC (Flying Gauge Changing; near the welded part). When there are a plurality of each area, they are managed as separate divided areas by the number of occurrences. For each of the divided regions, a change in plate and shape is detected, and a region having a change (abnormal) is specified as an abnormal part. In addition, it is possible to determine an abnormal state whether the abnormal site has occurred at the front end, the rear end, the midside, the local area, or the entire area.

FIG. 4 shows a processing flow of the abnormality content judging section 104, and a specific processing flow will be described. In S4-1, an FGC portion is extracted from the welding point of the target coil before and after a fixed length before and after the start of feedback control (AGC; Automatic Gauge Control). In S4-2, to calculate the movement differential value v _md of the delivery side speed v of the final stand. The v _md plurality of state determination threshold value for the target (acceleration state determining threshold value: th _acc, deceleration state determining threshold value: t
A constant speed area, an acceleration area, and a deceleration area are extracted using h _dcc and a constant speed state determination threshold: th _cnst ). S4-4
In, the constant speed region is divided into a high speed portion and a low speed portion using a high / low determination threshold value (th _LH ). In S4-5, the variation of the final stand exit side plate thickness and the variation of the shape are detected for each of the extracted regions, and the region where the variation has occurred is specified as the variation region.

FIG. 5 shows the configuration of the primary diagnostic section. The primary diagnosis unit 105 includes a feature amount calculation unit 500, a feature amount table 501, a primary diagnosis execution unit 502, and a threshold table 503. The primary diagnostic unit 105 is a statistic (average value, variance value, moving variance maximum value, etc.) of values that the operator usually observes, such as plate thickness, tension, and load, as shown in the feature amount table 501 in FIG. Are compared with those of the normal coil,
Determine the abnormality. In the primary diagnosis unit 105, these statistical values correspond to feature amounts. The threshold value table 503 stores a threshold value V _th of each characteristic amount obtained by accumulating the characteristic amounts of a plurality of normal coils. Here, the average value μ of the feature amount, the standard deviation σ of the feature amount, and the coefficient α for obtaining the threshold value
Then, V _th is calculated from equation (1).

[0022]

(Equation 1)

FIG. 7 shows a processing flow of the primary diagnosis unit 105. In S7-1, the feature amount (name) corresponding to the variable region extracted by the abnormal content determination unit 104 is extracted from the feature amount table 501. In S7-2, the value of the extracted feature amount (name) is calculated. For example, when the FGC section is selected as a variable area, a feature quantity with FGC in the feature quantity table 501 is selected, and the average, variance,
Calculate the maximum value of the moving variance. In S7-3, a threshold value V _th corresponding to the calculated feature value is stored in a threshold value table.
Extract from 503. In S7-4, for each feature amount calculated in S7-2, normality / abnormality is determined using the threshold value _Vth obtained in S7-3. That is, when a certain feature value is equal to the threshold value V _th
Is exceeded, the feature value becomes abnormal, and if not, it becomes normal. Hereinafter, the process of comparing the feature amount with the diagnostic threshold value and concluding the normality / abnormality by exceeding / not exceeding the threshold value is referred to as “threshold value calculation”.

FIG. 8 shows the configuration of the secondary diagnostic section. The secondary diagnosis unit 106 includes an area-variation factor table 800, a factor diagnosis group 80
1. It comprises a factor diagnosis selection unit 802, a threshold table 803, and a detailed information calculation unit 804. The secondary diagnosis unit 106 prepares a factor diagnosis logic 805 for estimating the possibility of the factor for each assumed variation factor, and selectively executes the logic to investigate the variation factor.

The area-variation factor table 800 records the correspondence between the areas and the factor diagnosis that occurs in each area. The factor diagnosis group 801 includes factor diagnosis logics 805 prepared by the number of assumed variation factors. The threshold value table 803 stores a threshold value used by each factor diagnosis logic 805 to execute a diagnosis, and the factor diagnosis logic 805 refers to the threshold value table 803 to make a diagnosis.

FIG. 9 shows a processing flow of the secondary diagnostic section 106. In S9-1, the factor diagnosis selection unit 802 extracts the factor diagnosis logic corresponding to the variable region specified by the abnormal content determination unit 104 with reference to the region-variation factor table 800.

FIG. 10 shows an example of the area-variation factor table 800. The table 800 stores the fluctuation factors for each area. For example, there are two fluctuation factors in the low speed region: base material plate thickness and uneven hardness, and five fluctuation factors in the high speed region: base material plate thickness, uneven hardness, roll eccentricity, slip, and detector failure.

At S9-2, the selected factor diagnosis logic is selected from the factor diagnosis group 801 and activated. In S9-3, the activated factor diagnosis logic 805 refers to the threshold value table 803 and executes the factor diagnosis.

FIG. 11 shows an example of the threshold value table 803. In the threshold value table 803, threshold values obtained from the equation (1) are set for each area and each factor. For example, F
The preform thickness variation, which is one factor of GC unit, threshold VH_ the feature quantity (moving variance value of the base plate thickness) V _Hmax
th is the feature value (the maximum value of the frequency spectrum of the base material thickness)
The threshold value P_th is stored for Pmax.

In S9-4, if the result diagnosed by the factor diagnosis logic 805 is abnormal, the detailed information calculation unit 804 adds detailed information on the abnormal part, the abnormal actuator, and the like, and transmits the result to the guidance unit 109.

As described above, the primary diagnostic section 105 notifies the operator of an abnormality in operation data such as tension and load that the operator is usually accustomed to, and in addition to this, the secondary diagnostic section 106 specifies an assumed variable factor name. By performing the notification, it is possible to perform a multi-faceted diagnosis such as a verification of the secondary diagnosis result using the primary diagnosis result, and it is possible to provide the operator with valuable information that can improve the diagnosis accuracy.

Next, a specific example of the factor diagnosis logic 805 will be described. In the present embodiment, the base material thickness variation diagnosis logic when the variation of the base material thickness is a factor, the uneven hardness abnormality diagnosis logic when the hardness variation of the base material is a factor, the slip between the roll and the plate is a factor. Slip diagnosis logic when roll eccentricity, roll eccentricity diagnosis logic when roll eccentricity is a factor,
A description will be given of a total of five cases of the detector failure diagnosis logic when either the thickness gauge or the plate speedometer is the cause.

FIG. 12 is a schematic diagram of the processing of the base material thickness fluctuation diagnosis logic, and FIG. 13 is a processing flow. The variation of the base material thickness is a factor of variation over almost the entire region. The logic will be described according to the processing flow. In S13-1,
The moving variance value V _h of the final stand delivery side thickness h (2) formula,
It is calculated using equation (3).

[0034]

(Equation 2)

Further, a maximum value V _Hmax of V _h is calculated.

[0036] In S13-2, the search of the base material plate moving variance V _Hmax thick portions corresponding, the maximum value of which is calculated in the same manner as S13-1. In S13-3, frequency conversion of the final stand exit side plate thickness is performed, and n points at which the power spectrum exceeds the threshold value gth1 are extracted. In S13-4, the maximum value of the frequency spectrum of the base metal plate thickness existing near the width w of the frequency extracted in S13-3 is determined as _Pmax .

[0037] In S13-5, the magnitude of V _Hmax and P _max, and a threshold operation using a threshold stored in the threshold value table 803, normal that each result to an OR operation / Determine the abnormality. That is, when one of them is abnormal, it is determined to be abnormal. By the above processing, the behavior when the base material variation is the cause is observed from both the time domain and the frequency domain, so that a highly accurate estimation of the base material variation factor becomes possible.

FIG. 14 is a schematic diagram of the irregularity abnormality diagnosis logic, and FIG. 15 shows a processing flow. This logic calculates the ratio of the calculated load to the actually measured load (hereinafter Zp)
When there is no uneven hardness, since the Zp is constant, it is diagnosed that the part where the Zp fluctuates is a part where the uneven hardness occurs.

[0039] In S15-1, the moving variance value V _h of thickness at delivery side of the final stand (2) is calculated by equation, this maximum value V
_{Find Hmax} . In S15-2, Zp of the site corresponding to the V _Hmax
Is obtained, and the maximum value V _zpmax of the moving variance is obtained. S1
In 5-3, a threshold value calculation of V _zpmax is performed to derive a conclusion of normal / abnormal. According to the present logic, it is possible to highly accurately estimate the unevenness in hardness by using Zp in which the behavior of the hardness can be observed remarkably.

FIG. 16 shows a schematic diagram of the slip diagnosis logic, and FIG. 17 shows a processing flow. In this logic, the slip diagnosis is performed by using the actually measured advance rate fa and extracting a phenomenon in which the value fa, which is usually a value near 1, rapidly changes in the negative direction.

In S17-1, similarly to the base material thickness variation diagnosis logic, the moving variance value of the final stand exit side thickness is calculated, and the maximum value V _Hmax is calculated. In S17-2, it is confirmed whether or not there is fa satisfying fa <0 in the range of the width w of the portion corresponding to _VHmax . If so, the process proceeds to S17-3; otherwise, the process ends. In S17-3,
fa moving average fa _mv of obtained in (4), is calculated by the range of the width w of the portion corresponding to the V _Hmax absolute difference between the maximum value and the minimum value of fa _mv a (P / P).

[0042]

(Equation 3)

In S17-4, a threshold value calculation is performed using the P / P calculated in S17-3, and a conclusion is derived. According to this logic, by using the fa corresponding to the slip phenomenon one-to-one and smoothing the fluctuating fa by moving average, the qualitative behavior of this can be accurately detected, resulting in a highly accurate slip. Abnormality can be estimated.

FIG. 18 shows a schematic diagram of the roll eccentricity diagnosis logic, and FIG. 19 shows a processing flow. In S19-1, the frequency of the final stand exit side plate thickness is converted. In S19-2, a maximum power spectrum (Pmax (i) i; stand No.) is extracted for each roll near the rotation frequency w. In S19-3, a threshold value is calculated for Pmax, and a roll exceeding the threshold value is specified as an eccentric roll. The threshold value at this time is set by the threshold value construction unit 107, and is calculated using Pmax in a normal state as a feature amount as described later. Threshold construction section
By using the threshold value appropriately set by 107, it is possible to detect the eccentric roll with high accuracy.

FIG. 20 shows a processing flow of the detector failure diagnosis logic. This diagnostic logic focuses on the property that the mass flow value (= thickness x speed x width) is constant at each stand in a constant speed state in order to detect the failure of the thickness gauge or the speed gauge. Observe the mass flow value of each stand in the high speed section of the plate speed, and derive the conclusion that the detector (thickness gauge or plate speed meter) of the stand whose mass flow value is far apart from other stands is abnormal. I do.

First, in step S20-1, the moving variance value of the final stand exit side sheet thickness is calculated in the same manner as the base material sheet thickness fluctuation diagnosis logic, and the maximum value V _Hmax is calculated. In S20-2, V
The average value xi (i; stand N) of the mass flow value of each stand using the sheet speed and the exit side sheet thickness near time w corresponding to _Hmax
o. ) Is calculated. Here, a constant value is used for the plate width. S
In step 20-3, the average value E and the standard deviation σ of the mass flow values for the stands other than the stand i are calculated using the equations (5) and (6).

[0047]

(Equation 4)

In S20-4, it is determined whether the mass flow value of the stand i satisfies the expression (7).

[0049]

(Equation 5)

In S20-5, when the result of S20-4 is Yes, it is determined that the mass flow value of the i-stand is normal, that is, the detector is normal. In the case of No, it is determined that the detector of the i-stand is abnormal.

In S20-6, it is determined whether or not all the stands have been completed. If completed, the processing is terminated. If not, the processing is shifted to S20-3, and the same processing is executed in another stand.

According to this logic, the stand having the abnormality detector can be detected by paying attention to the disturbance of the mass flow value, and the distance of the i-stand mass flow value from other stands can be expressed by the following equations (5) and (5). By using the equation (6), highly accurate detector failure can be estimated.

FIG. 21 shows the configuration of the threshold value construction section. The threshold value construction unit 107 includes a coil A determination detection unit 2100, a feature value calculation unit 2101, a threshold value construction data update unit 2102, a threshold value construction data table 2103, and a threshold value table update unit 2104. You. Data table for threshold construction
2103, the average value μ, the standard deviation σ, and the number of coils N used for calculating these values of the feature amounts used for the threshold value calculation in the primary diagnosis unit 105 and the secondary diagnosis unit 106 are set for each threshold value. Is stored.

FIG. 22 shows a processing flow of the threshold value construction section 107. The startup process is performed for each coil.
The judgment detection unit 1900 checks the judgment result of the quality judgment unit 103. In S22-2, it is checked whether the determination result is A determination, and in the case of A determination, the process proceeds to S22-3. Otherwise, the process ends.

In S22-3, the characteristic amount calculation unit 2101 calculates the characteristic amount used for the threshold value calculation in the primary diagnosis unit 105 and the secondary diagnosis unit 106. In S22-4, the threshold value construction data update unit 2102 uses the feature value calculated in S22-2, and the average value Ave (N) of each feature value stored in the threshold value construction data table 2103, Standard deviation Stddev (N) (8)
The expression is updated by the expression (9). Here, Data (N) is the feature amount calculated in S22-2.

[0056]

(Equation 6)

In S22-5, whether there is an instruction from the operator,
Alternatively, if the number of coils N is equal to or greater than _Nmax , the threshold value table updating unit 2104 sends the threshold value construction data table 21
The threshold value _Vth is calculated by the equation (10) using the value of 03, and the threshold value table 503 of the primary diagnosis unit 105 and the threshold value table of the secondary diagnosis unit 106 are overwritten by overwriting this value. 80
Update 3 respectively.

[0058]

(Equation 7)

Here, β selects an appropriate value of about 3. According to the threshold value constructing unit 107, since the average value μ and the standard deviation σ are successively updated according to the equations (8) and (9), it is not necessary to accumulate the values of the feature amounts by the number of coils, and store them. It is possible to construct a threshold value using the area efficiently.

FIG. 23 shows the configuration of the threshold value correcting unit 108.
The threshold value correction unit 108 includes a coil determination detection unit 2300, a diagnosis execution unit 2301, a correction threshold value calculation unit 2302, a no-response detection unit 2303,
Correction target diagnosis factor extraction unit 2304, correction threshold value calculation unit 2305
And a threshold table updating unit 2306.

FIG. 24 shows a processing flow of the threshold value correcting unit 108, and details of the processing will be described with reference to the flow chart. S24-1
In the coil determination detection unit 2300, the quality determination unit 103
Check the judgment result of. In S24-2, it is determined whether the determination result is A or D. If the determination result is A, the process proceeds to S24-3. In the case of the D determination, the process proceeds to S24-5, and if neither is the case, the process ends.

In S24-3, the diagnosis execution section 2301 sets the secondary diagnosis section
The factor diagnosis logic 805 that activates 106, executes diagnosis for all areas, and determines an abnormality based on the result of the A determination.
Is extracted. In S24-4, the threshold value of the factor diagnosis logic 805 extracted in S24-3 is corrected.
Is corrected in a direction in which is difficult to respond (a direction in which an abnormality determination is not easily issued).

On the other hand, in the processing S24-5 for the D determination, it is verified whether the diagnosis result of the secondary diagnosis unit 106 is normal. If it is normal, the process proceeds to S24-6, and if not, the process ends. In S24-6, the correction target diagnosis factor extraction unit 2304 causes the factor diagnosis logic that satisfies Expression (11) to be satisfied.
Extract 805.

[0064]

(Equation 8)

Here, γ is the judgment coefficient, the feature amount is the value used for the threshold value calculation, and Stddev is the standard deviation used when constructing the threshold value.

In S24-7, the threshold value of the factor diagnosis lock 805 extracted in S24-6 is changed to a corrected threshold value calculation unit 2305.
Is corrected in a direction that makes it easier to react (a direction that makes it easier to make an abnormality determination). In S24-8, the threshold value table updating unit 2306
Is written into the threshold table 803, and is updated.

In the above flow, the correction procedure for the threshold value of the secondary diagnosis section 106 has been described.
By using the feature amounts described in the feature amount table 501 instead of 805, the threshold adjustment of the primary diagnosis unit 105 can be performed similarly.

[0068]

According to the rolling mill abnormality diagnosis apparatus of the present invention,
An abnormal site can be specified for each area classified according to the state of the plate speed, and a factor diagnosis suitable for the area can be selectively executed for the area, so that a high-speed and high-accuracy diagnosis can be performed.

Further, since the diagnostic threshold value is appropriately adjusted so as to maintain the consistency between the quality determination result of the board (coil) and the diagnostic result, the appropriate threshold value is maintained even when the characteristics of the object change, and the diagnostic threshold value is constantly increased. Accurate diagnosis can be performed.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram of a rolling mill abnormality diagnosis device according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram of a quality judgment evaluation table.

FIG. 3 is a processing flowchart of a quality judgment unit 103.

FIG. 4 is a processing flowchart of an abnormality content determination unit 104.

FIG. 5 is a configuration diagram of a primary diagnosis unit 105.

FIG. 6 is an explanatory diagram of a feature amount table 501.

FIG. 7 is a processing flowchart of the primary diagnosis unit 105.

FIG. 8 is a configuration diagram of a secondary diagnostic unit 106.

FIG. 9 is a processing flowchart of the secondary diagnosis unit 106.

FIG. 10 is an explanatory diagram of an area-variation factor table 800.

FIG. 11 is an explanatory diagram of a threshold table 803.

FIG. 12 is a schematic diagram of a base material thickness variation diagnosis logic.

FIG. 13 is a processing flowchart of a base material thickness fluctuation diagnosis lock.

FIG. 14 is a schematic diagram of an uneven hardness abnormality diagnosis logic.

FIG. 15 is a processing flowchart of uneven hardness abnormality diagnosis logic.

FIG. 16 is a schematic diagram of a slip diagnosis logic.

FIG. 17 is a processing flowchart of slip diagnosis logic.

FIG. 18 is a schematic diagram of a roll eccentricity diagnosis logic.

FIG. 19 is a processing flowchart of roll eccentricity diagnosis logic.

FIG. 20 is a processing flowchart of a detector failure diagnosis logic.

FIG. 21 is a configuration diagram of a threshold value construction unit 107.

FIG. 22 is a processing flowchart of the threshold value construction unit 107.

FIG. 23 is a configuration diagram of a threshold value correction unit 108.

FIG. 24 is a processing flowchart of the threshold value correcting unit 108.

[Explanation of symbols]

100: plant (tandem cold rolling mill), 101: abnormality diagnosis device, 102: data collection unit, 103: quality determination unit, 104: abnormality content determination unit, 105: primary diagnosis unit, 106: secondary diagnosis unit, 10
7: threshold construction unit, 108: threshold adjustment unit, 109: guidance unit, 110: display device, 200: evaluation table, 201 ...
Comprehensive evaluation table, 500: feature amount calculation unit, 501: feature amount table, 502: primary diagnosis execution unit, 503: threshold value table
(Primary), 800: area-variation factor table, 801: factor diagnosis group, 802: factor diagnosis selection unit, 803: threshold value table
(Secondary), 804: detailed information calculation unit, 805: factor diagnosis logic, 2100: coil A determination unit, 2101: feature amount calculation unit, 2102
… Threshold update data, 2103… threshold data table, 2104… threshold table update, 2300
... Coil determination detection unit, 2301 ... Diagnosis execution unit (secondary diagnosis unit activation unit), 2302 ... Correction threshold calculation unit, 2303 ... No response detection unit, 2304 ... Correction target diagnosis factor extraction unit, 2305 ... Correction threshold Value calculation unit, 2306... Threshold value table updating unit.

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Yoichi Sugita 7-1-1, Omika-cho, Hitachi City, Ibaraki Prefecture Inside Hitachi Research Laboratory, Hitachi, Ltd. (72) Inventor Masahiro Kayama 5-chome, Omika-cho, Hitachi City, Ibaraki Prefecture No. 1 Inside Hitachi, Ltd. Omika Plant (72) Inventor Kenichi Yoshioka 5-2-1, Omika-cho, Hitachi City, Ibaraki Prefecture Inside Hitachi Information & Control System Co., Ltd. (72) Inventor Naoki Hirota Omika, Hitachi City, Ibaraki Prefecture 5-2-1 Machi-cho Hitachi Information & Control Systems Co., Ltd. (72) Inventor Kim Yong Nam 700 South Korea, Jeollanam-do, Jeollanam-do South Korea Pohang Integrated Steel Works Co., Ltd. (72) Inventor Kak Ho Chul South Korea 700, Jinhu-dong, Gwangyang, Lunam-do Inside Pohang Integrated Steel Works Co., Ltd. 700 Gwanghu-dong, Gwangyang-si, Pohang, Sogo Steel Co., Ltd. (72) Inventor Shin-Song Kap 700, Jinhu-dong, Gwangyang-si, Jeollanam-do, Republic of Korea F-term in Pohang Sogo Steel, Ltd. FA01 FA06 FA11 4E024 AA01 AA07 BB02 BB06 CC03 EE01

Claims (15)

[Claims] 1. A method for diagnosing abnormalities in a rolling mill, which judges the quality of a sheet using operation data collected during rolling and estimates an abnormal factor when the sheet is judged to be defective. The area is divided according to the characteristics of the sheet speed for
Detecting variations in plate thickness and / or shape for each of these areas,
A method for diagnosing abnormalities in a rolling mill, wherein a region having a change is specified as an abnormal part. 2. The method according to claim 1, wherein the region is divided into an FGC (near welding) portion, a low speed portion, an acceleration portion, a high speed portion, and a deceleration portion. 3. The abnormal state according to claim 1, wherein the abnormal part is a state quantity observed during operation or a statistical state quantity obtained by performing statistical processing on the state quantity, and is set for each state quantity or statistical state quantity. An abnormality diagnosis method for a rolling mill, wherein a diagnosis for specifying an abnormal state quantity is performed by comparison with a diagnosis threshold value. 4. The method for diagnosing abnormalities in a rolling mill according to claim 1, wherein the abnormal site is diagnosed with respect to a possibility of an abnormal factor assumed for each area. 5. The abnormal condition according to claim 4, wherein the abnormal factors include individual factors such as variations in base material plate thickness, unevenness in base material hardness, slip between the plate and the roll, eccentricity of the roll, and sensor failure. A method for diagnosing abnormalities in a rolling mill, wherein a diagnosis is performed by selecting a diagnosis logic of an individual factor which is likely to occur. 6. An abnormality diagnosis for judging the quality of a plate using the collected operation data during rolling of a rolling mill including a plurality of stands, and estimating an abnormal factor when the plate is determined to be defective. In the method, a variation of the base material thickness corresponding to the variation portion of the final stand exit side thickness is observed, and if the base material thickness is different from the normal variation degree, and / or the final stand exit side thickness and the The frequency component of the base material thickness is compared, and when they are different from the normal degree of fluctuation, the fluctuation of the base material thickness is presumed to be an abnormal factor of the quality defect. Diagnostic method. 7. An abnormality diagnosis for judging the quality of a plate using the collected operation data during rolling of a rolling mill having a plurality of stands, and estimating an abnormal factor when the plate is determined to be defective. In the method, a zp value, which is a ratio of the calculated load to the actual load, is calculated. An abnormality diagnosis method for a rolling mill, wherein the abnormality is estimated to be an abnormal factor of quality defect. 8. An abnormality diagnosis for judging the quality of a plate using the collected operation data during rolling of a rolling mill including a plurality of stands, and estimating an abnormal factor when the plate is determined to be defective. In the method, a moving average fa of the actually measured advance rate is calculated, and when the behavior of the fa corresponding to the variation portion of the final stand exit side plate thickness changes in a negative direction, and the fa is negative, A method for diagnosing abnormalities in a rolling mill, wherein a slip between a plate and a roll is estimated to be an abnormal factor of poor quality. 9. An abnormality diagnosis for judging the quality of a plate using the collected operation data during rolling of a rolling mill including a plurality of stands, and estimating an abnormal factor when the plate is determined to be defective. In the method, the frequency component f of the final stand exit side plate thickness is determined, the maximum value fmax of the frequency component f near the rotation frequency of each roll is determined, and the fmax of the frequency component f of the plate having a normal size is determined. A method for diagnosing abnormalities in a rolling mill, wherein when the magnitude exceeds a threshold value, the eccentricity of the roll is estimated to be an abnormal factor of poor quality. 10. An abnormality diagnosis for judging the quality of a plate using collected operation data during rolling of a rolling mill having a plurality of stands, and estimating an abnormal factor when the plate is judged to be defective. In the method, the mass flow value given by the product of the sheet thickness detected by the thickness gauge and the sheet speed and the sheet width detected by the sheet speed meter is calculated for each stand, and the mass flow value is calculated relative to the mass flow values of the other stands. When the mass flow value indicating a different value is detected, the thickness gauge and / or the plate speedometer of the stand having the detected mass flow value is estimated to be defective. Abnormal diagnosis method. 11. A data collecting unit for collecting operation data during rolling, a quality judging unit for judging the quality of the plate using the data collected by the data collecting unit, and a defect of the plate by the quality judging unit. An abnormality diagnosis device for a rolling mill including a diagnosis unit for estimating a factor of quality abnormality when determined, in a rolling mill, a region is divided according to a state of a sheet speed in a total coil length,
An abnormal content determination unit that detects the presence or absence of an abnormality in each area and specifies an abnormal site in the area unit, wherein the diagnostic unit relates the abnormal site identified by the abnormal content determination unit to the site An abnormality diagnosis device for a rolling mill, wherein diagnosis is performed using data. 12. The diagnostic unit according to claim 11, wherein the diagnostic unit compares a feature quantity including a state quantity observed during operation or a statistical state quantity obtained by performing statistical processing on the state quantity with a feature quantity accumulated during normal rolling. , A primary diagnosis unit that specifies an abnormal state quantity from a feature amount that behaves abnormally based on the comparison result, and a secondary diagnosis unit that estimates the possibility of a possible abnormality factor. Diagnosis device for rolling mills. 13. The secondary diagnostic unit according to claim 12, wherein the secondary diagnostic unit has a plurality of assumed abnormal cause diagnostic logics, and selectively executes one or more abnormal cause diagnostic logics corresponding to each area. An abnormality diagnosis device for a rolling mill. 14. The method according to claim 12, wherein the characteristic value has an average value and a standard deviation from a certain point in time to the present time, the value being a value to be compared with a diagnostic threshold value to determine the presence or absence of an abnormality. At the timing when the data for the coil is collected, if the determination result of the plate by the quality determination unit is good, the average value and the standard deviation are sequentially updated using the newly acquired feature amount, and the updated value is updated. An apparatus for diagnosing abnormality of a rolling mill, comprising a threshold value construction unit for determining a diagnosis threshold value from an average value and a standard deviation. 15. The diagnosis unit according to claim 12, 13 or 14, wherein the diagnosis unit is executed when the judgment result of the quality judgment unit is good, and when the obtained diagnosis result is abnormal, the abnormality is derived. The diagnostic threshold of the diagnostic unit is corrected in a direction that is unlikely to cause an abnormality determination, and when the determination result is poor and the diagnostic result of the diagnostic unit is normal, the diagnostic threshold of the diagnostic unit that has derived normality is determined to be abnormal. An abnormality diagnosis apparatus for a rolling mill, comprising: a threshold value adjusting unit for correcting a direction in which the abnormality easily occurs.