DE102021214275A1 - Plant control apparatus, plant control method and program - Google Patents

Abstract

[Objection] The control effects and operation efficiency are improved by predicting occurrence of operation abnormality in a factory to be controlled and appropriate operation of the control operating member with which no operation abnormality occurs. [Solution of the problem] For the factory to be controlled, a first operation processing at a predetermined response speed to the operation, and a second operation processing at a slower response speed to the operation than the first operation processing is performed. The first operation processing is performed in a first operation element, and the second operation processing is performed in a second operation element. Here, there is provided a safety manipulation range determination section that determines, based on the result of the first manipulation, a safety manipulation range of the first manipulation processing by the first manipulation. The command to the second operation processing is corrected or changed, and a result position of the first operation processing by the first operation member is moved to a result position in which the occurrence of an operation abnormality is not estimated when the safety operation area determination section determines that the first operation processing by the first Operating element is not in the safety operating area.

Description

[Technical Field of Application]

The present invention relates to a plant control apparatus, a plant control method and a program.

[State of the art]

When rolling mill control, which is one of the controls of a factory, is performed, fuzzy control and neurofuzzy control have conventionally been applied to the shape control for controlling the wavy state of a plate. The fuzzy control is applied to the shape control using coolant, and the neurofuzzy control is applied to the shape control of a Sendzmir mill.

In the shape control to which the neurofuzzy control is applied, as shown in Patent Literature 1, the processing is performed in which the difference between the result shape pattern detected by a shape detector and the target shape pattern and the similarity ratio of the result shape pattern to a predefined one Reference shape patterns are determined. In the shape control to which the neurofuzzy control is applied, the amount of the control output for the manipulation member is determined from the obtained similarity ratio by the control rule expressed by the control manipulation member manipulation amount for the predefined reference shape pattern.

The shape control has plural control handles, and the control is performed by the difference in the characteristics of plural control handles. The shape is a wavy state of the disk in the direction of the disk width, and the control actuator can change the shape of a certain area in the direction of the disk width. For example, the AS-U roll may change shape near the saddle position to be serviced, and the intermediate roll shift may change the shape of the plate end portion. When the shape control is performed, each control handle is operated in combination with each other depending on the actual shape so that a shape difference is limited.

When the rolling mill carries out rolling, a cooling material for lubricating between the material to be rolled and the rolls of the rolling mill and for cooling the heat generated by rolling (hereinafter referred to as coolant) is required. This cooling material serves as an actuator for shape control, and the shape can be changed in the entire range in the plate width direction by adjusting the injection amount of the coolant in the plate width direction. As shown in Patent Literature 2, the 6-stage rolling mill has an adjustment mechanism for the coolant injection amount in the plate width direction, and the shape control is performed by changing the injection amount using the result shape. However, in the Sendzimir mill, the mill rolls are immersed in the coolant during rolling, so the coolant takes more time than AS-U or the inter-roll shift until the effects on the shape appear. Also, since the coolant injection amount cannot be automatically adjusted, it may be necessary for the operator to operate the flow rate adjustment valve, etc. In such a case, adjustment is possible only before rolling starts.

While rolling is being performed in a rolling mill, an operational abnormality may occur in which the material to be rolled is broken. The breakage of the material to be rolled is often caused by the material to be rolled, but the breakage may also occur due to the meandering of the material to be rolled. The meandering of the material to be rolled means that the material to be rolled is shifted to one side of the rolling mill, and rolling is usually performed at the middle part of the rolling mill.
Such meandering is expected to occur depending on the position of the AS-U roll or the inter-roll shift. In the shape control, the AS-U roll and the intermediate roll shift are operated to keep the shape of the material to be rolled in the target shape, and as a result, the mechanical state may come to the state where meandering of the material to be rolled Material easily occurs.

Conventionally, in ordinary rolling mills such as 4-stage mill, 6-stage mill, etc., in addition to a mechanical operation section, namely, bender, leveller, an operation section for changing the coolant injection amount in the plate width direction is provided and used for shape control. Unlike the Sendzimir mill, in a conventional rolling mill, the roll of the mill is not immersed in the coolant, and effects on the shape of the coolant can be obtained in the same time as a mechanical operating means. Therefore, in the shape control in an ordinary rolling mill, the mechanical operating means and the coolant are treated the same, and the coolant is used as the control operating member de. Also in this case, since the coolant acts in the entire range in the plate width direction, it competes with the mechanical operating section, and the resultant positions of the mechanical operating section are often different even if the resultant shape of the material to be rolled is the same.

the 15 Fig. 12 shows a schematic construction of a control device for a conventional Sendzimir rolling mill.

First, the arithmetic unit 901 determines a difference between the target shape d1 and the resultant shape d2 of the material to be rolled obtained in the rolling mill 990, and gives this difference to the first shape control section 902 and the second shape control section 903. The first shape control section 902 controls the mechanical operating element 904, which is a mechanical operation section. The second shape control section 903 controls the coolant actuator 905 that changes the coolant injection amount.

The rolling mill 990 performs mechanical operation processing by the mechanical operation member 904 and operation processing of the coolant injection amount by the coolant operation member 905, and rolling processing of the material to be rolled, so that the shape result d2 and the rolling result d3 of the material to be rolled are obtained. In this case, in the mechanical operation processing by the mechanical operation member 904, the shape of the material to be rolled is changed with a relatively quick response by controlling the operation amount. On the other hand, the manipulation processing of the coolant injection quantity by the coolant manipulation member 905 has a slower response than the mechanical manipulation processing even when the manipulation quantity is controlled.

At the in the 15 As already described, the mechanical manipulation processing by the mechanical manipulation member 904 and the manipulation processing of the coolant injection amount by the coolant manipulation member 905 compete with each other as already described. Here, even if the shape result d2 is the same, the mechanical operation amount by the mechanical operation member 904 may not always be the same. In addition, the response speed of the mechanical operation processing by the mechanical operation member 904 and the response speed of the coolant injection amount by the coolant operation member 905 differ from each other, and the effective range in the plate width direction is different in each operation processing. Therefore, there is a problem that it is difficult to appropriately control both the mechanical operation processing by the mechanical operation member 904 and the operation processing of the coolant injection amount by the coolant operation member 905 . For example, when the mechanical actuator 904 with a high-speed response is in a state near the upper limit of the operating range, the shape result d2 and the rolling result d3 cannot be stably controlled and shrinkage occurs, and the rolling mill 990 which is the factory to be controlled , may not operate stably.

As a prior art for appropriately controlling the shape of such a Sendzimir rolling mill, as described in Patent Literature 3, for example, there is a technique for control in which the relationship between the resultant shape deviation of the material to be rolled and the control operation amount is learned from result data by machine learning. In the technique described in Patent Literature 3, the control output is output depending on the shape deviation, and the shape control for operating the control actuator is performed.

[Prior Art Documents]

[patent literatures]

• Patent Literature 1: JP-B-2804161
• Patent Literature 2: JP-B-2515028
• Patent Literature 3: JP-A-2018-005544

[Overview of the invention]

[Problem to be solved by the invention]

The technique described in Patent Literature 3 performs shape control by learning a control manipulation method for shape deviation without considering the position of the control manipulation member. In actually performing the control, the operation range is limited by the mechanical condition of the control actuator, the control output for that control actuator is stopped and only the operation with another control actuator is performed if possible.

If the time until the coolant as the shape control actuator affects the shape is longer than a mechanical actuator as in the Sendzimir mill, the Shape control by the coolant cannot be effectively performed because the shape is controlled by the mechanical operator in which influences are transmitted to the shape in a short time even if operation commands of the coolant are issued the same as the mechanical operator as in the conventional case.

In the case of rolling with a rolling mill, an operation abnormality such as plate breakage, etc. occurs depending on the properties of the material to be rolled, the rolling condition and the result position of the mechanical operator of the shape control, so that the limitation of the result position of the mechanical operator is required. However, it is extremely difficult to limit the resultant position of the mechanical actuator when the mechanical actuator and the coolant are used as the actuator for the shape control.
As stated above, in the conventional shape control, the coolant is controlled due to the shape deviation as in the case of the mechanical operator, so there is a problem that the resultant position of the mechanical operator, which leads to an operation abnormality such as plate breakage, etc., cannot be limited .

In the previous explanation, the problem of the shape control of the Sendzimir rolling mill was described, in different devices for controlling a plant, when the control operation with faster response and the control operation with slower response are carried out at the same time, the same problems arise if the two control operations are carried out at the same time.

The object of the present invention is to provide a factory equipment control apparatus, a factory equipment control method and a program to improve the control effects and operational efficiency by predicting the occurrence of operational abnormalities in the factory equipment to be controlled and appropriately manipulating the control operator with which no operation abnormality occurs can be improved.

[means for solving the task]

To solve the above problem, for example, the construction described in the claims is used.
The present application includes several measures to solve the above problem. An example of this is a factory equipment control apparatus which, for the factory equipment to be controlled, performs first operation processing at a predetermined response speed to the operation and second operation processing at a slower response speed to the operation than the first operation processing, wherein a first control section, the obtains a target state quantity of the factory to be controlled and issues an instruction to the first operation processing, a second control section which obtains a target state quantity of the factory to be controlled and issues an instruction to the second operation processing, a first operator which outputs according to the instruction the first control section executes the first operation processing of the factory equipment to be controlled, a second operator which executes the second operation processing of the factory equipment to be controlled according to the command from the second control section Safety operation range determination section that determines a safety operation range of the first operation processing by the first operation element based on the result of the first operation element, and a third control section that corrects or changes the instruction to the second operation processing from the second control section and a result position of the first operation processing by the first operation element operation member is moved to a result position in which occurrence of an operation abnormality is not estimated when the safety operation area determination section determines that the first operation processing by the first operation member is not in the safety operation area.

[Effects and effects of the invention]

According to the present invention, it is possible to avoid the resultant position of the operating member, which leads to an operation abnormality, by appropriately controlling the operation state of the factory equipment to be controlled. As a result, improvement in control accuracy of a factory to be controlled, improvement in operation efficiency, and limitation of occurrence of an operation abnormality can be expected.
In addition, other objects, constructions, and effects and effects will be apparent from the following description of embodiments.

character list

• the 1 Fig. 12 is a block diagram showing a construction example of a factory plant control apparatus according to an embodiment of the present invention.
• the 2 12 is a block diagram showing a construction example in which the plant control apparatus according to an embodiment of the present invention is applied to a rolling mill.
• the 3 Fig. 12 shows a construction diagram showing an example of a Sendzimir rolling mill.
• the 4 Fig. 12 is a construction diagram showing an example of rolling equipment of a single-stand rolling mill.
• the 5 FIG. 12 shows a schematic representation of a mechanical operating element according to an embodiment of the present invention.
• the 6 Fig. 12 is a block diagram showing a construction example of a security operation area determination section according to an embodiment of the present invention.
• the 7 Fig. 11 illustrates a construction example of a neural network according to an embodiment of the present invention.
• the 8th Fig. 12 shows a construction example of a neural network management table according to an embodiment of the present invention.
• the 9 Fig. 12 illustrates a construction example of a learning database according to an embodiment of the present invention.
• the 10 12 is a block diagram showing a construction example of a position limit control section for the mechanical operator according to an embodiment of the present invention.
• the 11 Fig. 12 is a schematic diagram showing a position abnormal area judging section for the mechanical operating member according to an embodiment of the present invention.
• the 12 12 shows a construction and an operation of a control output calculation section for the coolant actuator according to an embodiment of the present invention. the 13 12 shows an operation of a control output selection section for the coolant actuator according to an embodiment of the present invention.
• the 14 12 is a block diagram showing a construction example of hardware in which the factory plant control apparatus according to an embodiment of the present invention is composed of a computer.
• the 15 Fig. 12 is a block diagram showing a construction example of a conventional rolling mill control apparatus.

[Embodiments of the invention]

Hereinafter, an example (hereinafter referred to as “the present example”) of an embodiment of the factory equipment control apparatus will be described with reference to FIG 1 - 13 explained.

[Overall Construction of Factory Equipment Control Device]

the 1 Fig. 12 shows an example of the overall construction of the factory plant control apparatus according to the present example.
The one in the 1 The apparatus for controlling a factory shown controls a factory 190 to be controlled, and as the control of the factory tray 190 to be controlled, operation processing by a high-speed operation member (first operation member) 103 and operation processing by a low-speed operation member (second operation member) 104 are performed. The device for controlling a factory according to the present example receives a first desired state variable d11 and in an arithmetic unit 101 the difference from a first state variable d12. A first state quantity d12 is given as the result of operation processing by the high-speed operator 103 and the low-speed operator 104 from the factory equipment 190 to be controlled. The second state quantity d13 is also given as the result of the operation processing in the high-speed operation element 103, the low-speed operation element 104 and other operation elements applied to the factory equipment 190 to be controlled.

The control unit 110 of the plant control apparatus has a first control section 111 for controlling the operation processing of the high-speed operator 103 and a second control section 112 for controlling the operation processing of the low-speed operator 104 .
The first control output of the first control section 111 is directly supplied to the high-speed operator 103 and controls the operation processing of the high-speed operator 103.
The second control output of the second control section 112 is fed to a third control section 102 and, after any necessary corrections or changes, to the low-speed operating element 104 .

The factory plant control apparatus according to the present example further includes a safety operation area determination section 105 . The safety manipulation area determination section 105 obtains an operation result of the high-speed manipulation member 103 and judges whether or not the obtained operation result has a margin in the safety manipulation area, and supplies data of the safety manipulation area d14 to the third control section 102 as a judgment result.
Further, the safety operation area determination section 105 obtains the first state quantity d12 and the second state quantity d13 of the factory equipment 190 to be controlled. With reference to the first state quantity d12 and the second state quantity d13, the safety operation area determination section 105 judges whether the current operation result of the high-speed operation element 103 is in the safety operation area has a leeway or not.

Here, the safety manipulation area determination section 105 detects the occurrence of an operation abnormality and learns the first state quantity d12 and the second state quantity d13 and the resultant position of the high-speed operation member 103 at that time, and determines the safety manipulation area d14 which is an operable area such that no operation abnormality in the high-speed operation member 103 occurs. Since the operation abnormality is not a frequently occurring phenomenon, it is preferable that the safety manipulation area determination section 105 continuously acquires result data and carries out learning using machine learning and obtains the safety manipulation area d14.
The safety manipulation area determination section 105 supplies the data of the judged safety manipulation area d<b>14 to the third control section 102 .

When the third control section 102 judges that there is a margin in the safety manipulation area based on the data of the safety manipulation area d14, it supplies the control output of the second control section 112 to the low-speed manipulation member 104 as it is. On the other hand, when the third control section 102 judges that there is no margin in the safety manipulation area based on the data of the safety manipulation area d<b>14 , it corrects or changes the control output of the second control section 112 and supplies it to the low-speed manipulation member 104 .
The third control section 102 is supplied with the output of the arithmetic unit 101, the operation result of the high-speed operation element 103 and the second state variable d13, and the third control section 102 corrects or changes the control output of the second control section 112 based on these data.

According to the device for controlling a factory with in the 1 As shown in the construction, it is possible to efficiently operate the high-speed actuator 103 in the range where no operation abnormality occurs, and in addition to the increase in operation efficiency, the increase in control accuracy can be expected.

[Overall construction in application to the apparatus for controlling a Sendzimir rolling mill]

Next, the overall construction of the plant control apparatus according to the present example in the case of application to a Sendzimir rolling mill will be explained.
the 2 Fig. 12 shows the construction of the plant control apparatus according to the present example in the case of application to a Sendzimir rolling mill.
The one in the 2 The device shown for controlling a factory receives a first target shape d21 of the material to be rolled and in an arithmetic unit 201 the difference from the shape result d23 after rolling.

The apparatus for controlling a plant according to the present example executes, as the controller of a rolling mill 301 , operation processing by a mechanical operator 203 and operation processing by a coolant operator 204 . The operation processing by the mechanical operation member 203 mechanically changes the roll gap to perform rolling processing, with the reaction appearing in the shape result d23 of the material to be rolled being made at high speed. On the other hand, the manipulation processing by the coolant manipulation member 204 changes the injection amount of the coolant, the reaction occurring in the rolling result d24 of the material to be rolled is made at a slower speed than the manipulation by the mechanical manipulation member 203 .

The control unit 210 of the plant control apparatus has a first shape control section 211 for controlling the manipulation processing of the mechanical operator 203 and a second shape control section 212 for controlling the manipulation processing of the refrigerant manipulation elements 204 on. The first shape control section 211 and the second shape control section 212 carry out control so that the material to be rolled corresponds to the target shape d21. The target shape d21 is predetermined depending on the properties of the material to be rolled and so on.
The control output of the first shape control section 211 is directly supplied to the mechanical operator 203 and controls the operation processing of the mechanical operator 203.
The control output of the second shape control section 212 is supplied to a position limit control section 202 for the mechanical actuator, and is supplied to the coolant actuator 204 after corrections or changes in the control output are necessary, if necessary.
The construction of the mechanical actuator position limiting control section 202 will be described later with reference to FIG 10 explained.

The factory plant control apparatus according to the present example further includes a safety operation range determination section 205 for the mechanical operation member. The mechanical operation safety manipulation area determination section 205 obtains a mechanical operation position result d22 which is an operation result of the mechanical operation member 203, and judges whether the obtained mechanical operation position result d22 has a margin in the safety manipulation area or not. The mechanical operating safety manipulation area determination section 205 supplies the data of the mechanical operating safety manipulation area d<b>25 as a judgment result to the mechanical operating element position limitation control section 202 .

The mechanical operating member position limiting control section 202 obtains the shape result d23 and the rolling result d24 of the rolling mill 301. The mechanical operating member safety operating range determining section 205 judges, with reference to the shape result d23 and the rolling result d24, whether the obtained mechanical operating member position result d22 is im security operation area has a margin or not.

The mechanical operating member safety manipulation area determining section 205 detects the occurrence of an operation abnormality and learns the shape result d23 and the rolling result d24 and the mechanical operating member position result d22 at that time. Through this learning, the safe mechanical operator manipulation range determination section 205 determines the safe mechanical operator manipulation range d25, which is an operable range, so that no operation abnormality in the mechanical operator 203 occurs. Here, the mechanical operator safe manipulation range determination section 205 continuously takes in result data and carries out learning by means of machine learning, etc., and obtains the mechanical operator safe manipulation range d25.
The safe mechanical operating range determination section 205 supplies the data of the judged safe mechanical operating range d<b>25 to the mechanical operating position limit control section 202 .
The detailed construction for learning an operation abnormality of the mechanical operating member position limit control section 205 will be described later with reference to FIG 6 explained.

When the mechanical operator position limitation control section 202 judges that there is a margin in the safety operator section based on the data of the mechanical operator safety operation area d25, it supplies the control output of the second shape control section 212 to the coolant operator 204 as it is. On the other hand, when the mechanical operator position limiting control section 202 judges that there is no play in the safety operator section based on the data of the mechanical operator safety operation area d25, it corrects or changes the control output of the second shape control section 212 and supplies it to the coolant operator 204. The mechanical operating member position limiting control section 202 is supplied with the output of the arithmetic unit 201, the mechanical operating member position result d22 and the rolling result d24, and the mechanical operating member position limiting control section 202 corrects or changes the control output of the second shape control section 212 based on these data.

[Construction of a Sendzimir rolling mill]

A construction example of a Sendzimir rolling mill will be explained here.
the 3 shows a schematic construction, wherein in a Sendzimir rolling mill the shape control tion is made.
The actual shape of the rolled material to be rolled is detected with a shape detector 14 in a Sendzimir rolling mill. The actual shape detected by the shape detector 14 is subjected to pattern recognition pretreatment in a shape detection pretreatment section 11 of a controller 50, and then calculation is performed to determine which of reference patterns predefined in a pattern recognition section 12 is the actual shape on is most similar.
Based on the calculated reference shape pattern, the operator to be operated and the operation amount are judged in a control calculation section 13, and with these operator to be operated and the operation amount, the processing for controlling a Sendzimir rolling mill is performed.

the 4 shows an example of a rolling installation of a single-stand rolling mill. The Sendzimir rolling mill is a type of single-stand rolling mill.
The one in the 4 The rolling mill shown consists of a rolling mill 301, an input-side idler (hereinafter referred to as "TR") 302 and an output-side TR 303, wherein the material to be rolled 300 drawn out from the input-side TR 302 passes through the rolling mill 301 and is connected to the output-side TR 303 is wound.
The rolling mill 301 rolls the material to be rolled 300. The rolling is the processing of thinning the plate thickness of the material to be rolled 300 to a specified plate thickness.

In the rolling mill 301, a rolling speed control section 304 for adjusting the rolling speed and a roll gap control section 307 for adjusting the roll gap of the rolling mill 301 are installed. Further, in the input TR 302 and the output TR 303, an input TR control section 305 and an output TR control section 306 are provided for adjusting the clamping force generated by the input TR and the output TR, respectively.

The rolling processing is performed by applying the pressure to crush the material to be rolled 300 by adjusting the distance between an upper roll and a lower roll of the rolling mill 301 by the roll gap control section 307 and applying the material to be rolled 300 to the exit side by the rolling speed control section 304 is promoted. Here, on the entrance side and the exit side of the rolling mill 301, the processing in which the clamping force is applied to the material to be rolled 300 by means of the entrance-side TR 302 and the exit-side TR 303 is also performed.

Important for the rolling operation is the plate thickness (plate thickness on the exit side of the rolling mill) of the material to be rolled 300 as a product, with the roll gap, the input-side clamping force and the output-side clamping force being predefined so that the plate thickness of the material to be rolled 300 to a predetermined plate thickness is set.
An input-side clamping force current converting section 315 calculates the electric current required for generating the set input-side clamping force using the input-side clamping force set in a input-side clamping force setting section 311, and obtains the input-side clamping force by converting the electric current through the input-side TR control section 305 is given to the input TR 302.
An output-side clamping force current converting section 316 calculates the electric current required for generating the set output-side clamping force using the output-side clamping force set in an output-side clamping force setting section 312, and obtains the output-side clamping force by converting the electric current through the output-side TR control section 306 is given to the downstream TR 303.

The nip set in a nip setting section 319 is given to the nip control section 307 and the nip is detected by the nip control section 307 .
A rolling speed setting section 310 determines the speed of the rolling mill 301 based on the instruction from the operator of the rolling mill and sets the speed of the rolling mill 301 by the rolling speed control section 304 .

On the entrance side and the exit side of the rolling mill 301, there are installed an entrance-side tensiometer 308 and an exit-side tensiometer 309, and an entrance-side tensioning force control section 313 and an exit-side tensioning force control section 314 perform control such that the resultant tensioning force measured by the tensiometers are measured and the setting clamping force agree with each other. On the exit side of the rolling mill 301, an exit-side plate thickness gauge 317 is installed, and an exit-side plate thickness control section 318 performs control so that the resultant plate thickness measured by the plate thickness gauge coincides with the setting plate thickness.

In addition to the above construction, a shape detector 14 for detecting the shape of the material to be rolled is installed on the exit side of the rolling mill, and shape control is performed so that the detected shape coincides with the predefined shape, as shown in FIG 3 is shown.

As stated above, the shape represents the degree of waviness of the metal plate, that is, the material to be rolled . In general, since the clamping force is applied to the material to be rolled, when there is a damage such as crack, etc. at the plate end, a gap occurs from there, so that the case where the material to be rolled is cut in the plate width direction often occurs (plate fracture). Consequently, provision is often made such that the plate end is made wavy in order to avoid the stress concentration.

The waviness of the material to be rolled does not appear practically due to the application of the tension force to the material to be rolled, so that there appears to be no waviness, but the tension force distribution changes in the plate width direction.
The Indian 3 The shape detector 14 shown here concludes from the measurement of the clamping force distribution in the direction of the plate width on the waviness of the plate and records this as a result of the shape.

[Construction and processing of the mechanical operating element for shape control]

the 5A Fig. 12 shows the construction in operation processing with the mechanical operation member 203 of a Sendzimir rolling mill. the 5 Fig. 13 shows a cross section of the material to be rolled 300 in the plate width direction, showing only the structure on the upper side of the material to be rolled 300 and omitting the lower structure.
the 5B , 5C show the operation waveform with changes in the shape of the material to be rolled 300.

As in 5A As shown, the Sendzimir rolling mill consists of a work roll 401, a first intermediate roll 402, a second intermediate roll 403 and an AS-U roll 404 on the two sides of the material to be rolled 300.
The first intermediate roll 402 is tapered at the opposite side of the upper and lower rolls, respectively, and can affect the shape of the plate end of the material 300 to be rolled by shifting in the plate width direction.

The AS-U roller 404 has a structure in which a saddle 406 is interposed between the plurality of split rollers 405, and can be adjusted by changing the position of the saddle 406 (the vertical position in the 5 ) change the deflection of the AS-U roll 404 in the plate width direction. As for example in the 5B As shown, the shape of the center portion of the material to be rolled 300 can be manipulated as the center saddle 406 is lowered.

The lowest part of the 5B , 5C The operation waveforms shown show changes in the plate thickness distribution of the material to be rolled 300 upon a shift operation of the saddle 406 or the first intermediate roll 402. The change in shape is opposite to the plate thickness distribution.
The shape is the distribution of the degree of waviness in the plate width direction, and a large waviness shows that the material to be rolled 300 is expanded. This is because the expressions “the plate thickness on the exit side becomes thin”, “the elongation of the material to be rolled in the thinned portion is large”, and “the shape of the material to be rolled is increased” are synonymous.

For the operation abnormality, plate breakage of the material to be rolled 300 is a big problem. When plate breakage occurs, the work roll 401 and the first intermediate roll 402 of the rolling mill are damaged by the broken material 300 to be rolled. Further, the second intermediate roll 403 and the AS-U roll 404 may be damaged due to the occurrence of the plate breakage. When such breakage occurs, it is necessary to exchange these rolls, and the time for the removal processing of the material to be rolled 300 remaining in the rolling mill is required, resulting in extremely low operation efficiency.

Since the AS-U roller 404 presses the divided roller 405 against the second intermediate roller 403 under depression with the saddle 406 , the divided roller 405 and the second intermediate roller 403 may not abut against each other depending on the position of the saddle 406 . In such a state, the force exerted on the material to be rolled 300 from the work roll 401 in this section is suddenly reduced, it does not cause expansion of the material to be rolled 300, and the tension force exerted on the material to be rolled 300 in this section is increased.

If such a condition occurs at the plate end of the material to be rolled 300, it will result in plate breakage from the plate end. Further, since the clamping force applied to both ends of the material to be rolled 300 in the plate width direction is changed, a phenomenon occurs in which the center of the material to be rolled 300 in the plate width direction deviates from the center of the rolling mill in the plate width direction, and this can lead to plate breakage due to collisions with machines located before and after the rolling mill. In this way, an operation abnormality may occur depending on the resultant position of the mechanical operating member 203.

The result position where the operation abnormality occurs is calculated not from the machine design but by the plate thickness distribution of the material to be rolled 300 in the plate width direction, the plate thickness on the entry/exit side, the rolling condition such as clamping force, rolling load, etc., and changed the positional relationship to the other mechanical actuator for shape control, so prediction is difficult.
Therefore, in the present example, the mechanical operating member safety manipulation area determining section 205 stores these conditions when an operating abnormality occurs as result data, and finds the resultant position of the mechanical operating member for shape control where the operating abnormality frequently occurs by comparing the stored data with the result data in normal operation be compared.

The safety manipulation area determination section 205 for the mechanical operator in the present example determines the safety manipulation area for the mechanical operator by machine learning. For the result data in machine learning, the rolling condition such as plate thickness on the input/output side, tension force, rolling load, etc. and the result position of the mechanical operating member 203 are used, and for the teacher data, the information on the occurrence of rolling abnormalities is used.

As information about the occurrence of rolling abnormality, information about the plate breakage and the emergency stop of the rolling mill are used. The plate breakage can be detected by the decrease in the clamping force on the input/output side, and the emergency stop uses the information about the operation switch operated by the operator when there is any abnormality in the rolling state and the operation is stopped. The information about the operation switch can be acquired by a computer constituting the control device for controlling the rolling mill and used as one of the result information. The mechanical operating member safety manipulation area determining section 205 generates a neural network (N.N.) for judging the presence or absence of an operation abnormality using these result data and teacher data.

[Construction of the safety operation range determination section for the mechanical operator and construction of the neural network]

the 6 Fig. 12 shows the construction when the mechanical operating member safety operation range determining section 205 is realized by the machine learning. the 7 12 also shows the construction of the neural network 502 that the mechanical operating member safety manipulation range determining section 205 has.
Like in the 7 1, the neural network 502 receives the rolling result d24 and the mechanical operating member position result d22 from an input data generating section 501 at the input terminal 502a, and outputs an operation abnormality judgment value d32 from the output terminal 502b. The operation abnormality judgment value d32 represents the plate breakage information and the emergency stop information, which are the information of occurrence of rolling abnormalities. The neural network 502 performs learning from the combination of these input data and output data.

The Indian 6 The mechanical operating member safety operation area determination section 205 shown in FIG. 1 will be explained. The input data generation section 501 inputs the position result d22 for the mechanical operating member and the shape result d23. Further, a teacher data generation section 505 takes in the operation abnormality judgment value d32 judged by an operation abnormality judgment section 506 . Data acquisition by the input data generation section 501 and the teacher data generation section 505 is performed in a fixed time cycle under the control of the neural network learning control section 503, and a set of learning data is obtained for each operation cycle. The obtained learning data is sequentially stored in a learning data database 511 .
The operation abnormality judging section 506 judges from the rolling result d24 whether or not there is a plate breakage and an emergency stop of the rolling mill, which are an operation abnormality. The operation abnormality judgment value d32, that is, the judgment result, represents the infor information about a plate break and an emergency stop.

The rolling mill rolls various materials to be rolled 300 according to specifications, and products are obtained. Therefore, it is common for the rolling mill to be modified according to the specifications of the material to be rolled 300 by changing the combination of the specification of the work roll 401 (diameter distribution in the plate width direction), the draft specification of the first intermediate roll 402 and the split roll 405 of the AS- U-roll 404, representing the machine construction, is used. The plate width and the material of the material to be rolled 300 are not uniform. Thus, more efficient learning is possible when the neural network 502 is partitioned depending on the machine design and the specification of the material 300 to be rolled.
Accordingly, the mechanical operating member safety manipulation range determination section 205 in the present example has plural types of neural networks 502 and includes a control rule database 512 and a neural network selection section 504 to enable use by switching.

the 8th shows a construction example of the control rule database 512.
Like in the 8A As shown, the control rule database 512 stores a plurality of neural networks that have been learned using learning data consisting of a combination of input data and teacher data.
Then, the neural network learning control section 503 determines the neural network No. requiring learning. After determining the neural network No. required for the learning of the neural network learning control section 503, the neural network selection section 504 extracts the neural network concerned from the control rule database 512 and sets it to the neural network 502.
According to the rolling conditions from the actual rolling result d24 and the machine design, the neural network selection section 504 extracts the neutral network of the relevant neural network No. from the control rule database 512 and sets it as the neural network d33 for the control of the mechanical operating member position limitation control section 202 Celebration.

the 8B FIG. 12 shows the construction of the neural network management table stored in the control rule database 512. FIG. The management table is sorted by (B1) plate width, (B2) steel type and (A) machine construction. As (B1) plate width, for example, 4 sorts of 3 foot width, meter width, 4 foot width and 5 foot width are used. As the (B2) kind of steel, about 10 sorts of kinds of steel (1) - (10) are used. For (A), the distinction between (A1) and (A2) is made according to the length of the neck portion, which is neck specification of the first intermediate roll 402, for example.
The above table distinction is an example and needs to be adjusted in time according to the rolling equipment and the type of rolling material.
The mechanical operator safety operation range determination section 205 uses these neural networks according to the rolling conditions and machine construction.

The neural network learning control section 503 combines the data shown in FIG 8A shown learning data, which is a combination of the input data and the teacher data, according to the in the 8B shown neural network management table with the relevant neural network No. and stores this learning data in the learning data database 511.

the 9 FIG. 12 shows an example of learning data stored in the learning data database 511. FIG. Like in the 9 As shown, the learning data database 511 stores the learning data associated with each neural network No.

The neural network learning control section 503 instructs the input data generating section 501 and the teacher data generating section 505 to extract the input data and the teacher data from the management table corresponding to the neural network concerned from the learning data database 511. The neural network 502 carries out learning by using these. Various learning methods for the neural network have been proposed conventionally, and any learning method can be used.

A large number of learning data sets are required for machine learning, and the neural network 502 performs the learning when a certain amount of them (eg, 10,000 sets) are stored in the learning data database 511 .
When the learning of the neural network 502 is completed, the neural network learning control section 503 rewrites the neural network 502, which is the learning result, in the position of the neural network No. of the control rule database 512, and the learning is thereby completed.

The neural network 502 that has completed the learning outputs the operation abnormality judgment value by inputting the rolling result d24 and the mechanical operating member position result d22. Therefore, the neural network 502 can predict the presence of an operation abnormality by giving the expected future shape result d23 and the position result d22 for the mechanical operator, and can search for the safety manipulation area d25 for the mechanical operator.

[Construction of Position Limiting Control Section for Mechanical Operator]

the 10 Fig. 12 shows the construction of the position limit control section 202 for the mechanical operator.
The mechanical operator position limit control section 202 includes a mechanical operator position abnormality range judgment section 610 and a mechanical operator position abnormality limit control section 620 .
The position abnormal area judging section 610 for the mechanical operating member concludes using the method shown in FIG 7 declared neural network 502 to the mechanical operating member 203 in which an operation abnormality is expected to occur. The neural network 502 used here is a neural network d33 for the control received from the mechanical operating member safety manipulation range determining section 205 .
Based on the judgment result in the mechanical operator position abnormality range judgment section 610, the mechanical operator position abnormality limitation control section 620 generates operation commands for the coolant operator 204.

During the rolling operation, the shape result d23 of the rolling mill 301 keeps changing, and the first shape control section 211 to maintain it in the target shape d21 operates the mechanical operating member 203, and the mechanical operating member position result d22 also keeps changing. A mechanical operating member position abnormal area searching section 611, which the mechanical operating member position abnormal area judging section 610 has, generates the input data of the neural network 502 in the input data generating section 612 using the mechanical operating member position result d22.

the 11 FIG. 12 shows processing performed by the position abnormal area judging section 610 for the mechanical operating member.

Die n Arten des mechanischen Bedienungselementes 203 entsprechen hierbei beispielsweise der Gesamtzahl der Anzahl der Sattel 406 der AS-U-Walze 404 und der Anzahl der ersten Zwischenwalzen 402, die in Richtung der Plattenbreite verschoben werden können. Im Fall des in der 5 gezeigten Beispiels ist n = 7, da die Anzahl der Sattel 5 beträgt und die Anzahl der ersten Zwischenwalzen zwei, d.h. die obere Walze und die untere Walze, beträgt.In the example of 11 , when the mechanical operating portion 203 has n types (n is an integer), the position result d22 for the mechanical operating member is as follows: $$\text{POS}\left(\text{k}\right),\text{k}=\mathrm{1,}\mathrm{2,}...,\text{n}$$

Here, the n types of the mechanical operating member 203 correspond to, for example, the total number of the saddles 406 of the AS-U roller 404 and the number of the first intermediate rollers 402 which can be shifted in the plate width direction. In the case of the 5 In the example shown, n=7 because the number of saddles is 5 and the number of first intermediate rolls is two, ie, the top roll and the bottom roll.

When the mechanical operating member 203 is operated by the first shape control section 211, this operation is performed up to a certain amount in each control cycle, so that the mechanical operating member position abnormal area searching section 611 finds the estimated position of the mechanical operating member position result d22 of each mechanical operating member 203 generated as described below. Here, the amount of change of the position result, which can be operated by multiple controls, for example, is set to ΔPOS, and three cases, ie, the case where the position result is not changed, the case where the result value is changed in the positive direction, become , and the case where the result value is changed in the negative direction is considered. $$\text{POS}\left(\text{k}\right),\text{POS}\left(\text{k}\right)\pm \mathrm{\Delta }\text{POS},\text{k}=\mathrm{1,}\mathrm{2,}...,\text{n}$$

Thereby, the estimated position results of each mechanical operating member 203 can be generated in 3 ⁿ ways (for example, 2187 ways in the case of n=7). For example, if n=7, 2187 estimated position results can be generated. The estimated position results are output to the input data generation section 612 sequentially. The input data generating section 612 generates input data for the neural network 502 from the rolling result d24 and the estimated position d31, and outputs this input data to the neural network 502. FIG.

The neural network 502 returns the in the 11D shown operation abnormality judgment value d32. The operation abnormality judgment value d32 is the degree of disc breakage and emergency stop, and an output data judging section 613 receives the operation abnormality judgment value d32 output from the neural network 502 and the two degrees are added after weighting and set as an operation abnormality judgment value d26. Generally, when an operation abnormality occurs, the operator performs an emergency stop, and this also applies to the case where a sign of disk breakage occurs. Meandering of the material 300 to be rolled, for example, can be a sign of plate breakage.
If there is no emergency stop and a disk break occurs, it means that the disk break occurred unsigned and the priority for suppressing the disk break is higher. Thus, the weighting of the plate breakage degree is increased.

The mechanical operating member position abnormal area searching section 611 stores the output estimated position d31 and the returned operation abnormality evaluation value d26, and searches for the estimated value at which the operation abnormality evaluation value d26 becomes maximum. When, as a result of the search, the maximum value of the operation abnormality evaluation value d26 exceeds a predetermined threshold value, the mechanical operating member position abnormal area search section 611 outputs the amount of change of the resultant position at the estimated position d31 in this case as the operating member resultant position abnormal area evaluation value d27. The operation abnormality evaluation value d26 in this case also belongs to the result position abnormality range evaluation value d27 for the operating member as the operation abnormality evaluation maximum value.

In the above-described example, the mechanical operating member position abnormal area searching section 611 generates the estimated position d31 as three kinds of the change amounts from the resultant position of the mechanical operating member 203, but these may be generated with other processings. For example, the position abnormal area search section 611 for the mechanical operating member precisely controls the amount of change of the estimated position d31. The mechanical operating member position abnormality range search section 611 can be timely changed depending on the situation, for example, when no search is performed in a direction in which an operation abnormality clearly does not occur. Here, the direction in which the operation abnormality clearly does not occur may be, for example, a movement toward the center of the resultant position for the mechanical operating member.

The position abnormality limitation control section 620 for the mechanical operator generates from the result position abnormality range judgment value d27 for the operator, which is the output from the position abnormality range judgment section 610 for the mechanical operator, and the control commands to the coolant operator 204 of the second shape control section 212, the control output to the coolant operator 204 .

When it is judged that no operation abnormality occurs even if the mechanical operating member position result d22 changes by the control, the operation using the coolant operating member 204 by the normal second shape control section 212 is possible. When the operation abnormality evaluation value d26 is not so large, the output is possible by combining the control output of the second shape control section 212 with the abnormality limitation output d28 for limiting the operation abnormality. Of course, when the operation abnormality evaluation value d26 is large, the abnormality limitation output d28 is preferentially output to the coolant actuator 204 .

In a coolant control rule database 623, actions for each mechanical actuator 203(k) and the influencing coolant actuator 204 are predetermined. These measures can be achieved by actually operating the mechanical operator 203 and the coolant operator 204 during the rolling operation, and can also be achieved from the result data through the machine learning. Here, the case where the actions are obtained from the result in the actual operation and registered in the coolant control rule database 623 is considered.

[Construction and operation of a control output calculation section for the coolant actuator]

the 12 FIG. 12 shows the construction and operation of a coolant actuator control output calculation section 621. FIG.
In the coolant control rule database 623 ( 10 ) the required amount of change in refrigerant flow rate that can achieve the same effect as the operation of each mechanical operator 203 (k) is registered. the inside 12A The database search section 631 shown in FIG is achieved.

Then, the output synthesizing section 632 adds the required change amount of the refrigerant flow rate for each extracted mechanical actuator 203(k), and obtains the abnormality limit output d28.
For example, if in the in the 11A , 11B shown position result and estimated position in the 11D result position abnormality range judgment value d27 shown for the operating member is obtained, the one shown in FIG 12B shown abnormality limit output d28 is reached.
Since the coolant actuator 204 performs the lubrication and cooling required for the rolling operation, mostly the maximum flow rate and the minimum flow rate at each point in the plate width direction are determined as the coolant flow rate, and the abnormality limit output d28 at which the operation abnormality occurs is obtained.

[Output selection processing in a coolant actuator control output selection section]

the 13 12 shows the output selection processing performed by a coolant actuator control output selection section 622 .
The coolant actuator control output selecting section 622 selects or synthesizes the data shown in FIG 10 shown abnormality limitation output d28 and the shape control output d29 of the second shape control section 212 according to the magnitude of the operation abnormality evaluation maximum value ( 11D ) in the result position abnormality range judgment value d27 for the operating member and outputs them. The output selected or synthesized in the coolant-manipulator control output selection section 622 is supplied to the coolant-manipulator 204 as the coolant-manipulator output d30.

If the in the 11D operation abnormality evaluation maximum value shown is small, an operation abnormality is unlikely to occur even if the shape control is performed as it is, so the coolant operation member control output selecting section 622 supplies the shape control output d29 as it is to the coolant operation member 204 as the coolant operation output d30. That is, the coolant manipulation output d30 = the shape control output d29.

On the other hand, when the operation abnormality evaluation maximum value is large, the coolant actuator control output selection section 622 determines that there is a high possibility that an operation abnormality occurs. In this case, as in the 13B 12, the abnormality limitation output d28 is generated, and the coolant actuator control output selection section 622 sets the one shown in FIG 13A shown coolant manipulation output d30 so that within the maximum and minimum values of the coolant flow rate, the effects of the abnormality limiting output d28 are maximized.

When the operation abnormality evaluation maximum value is not so large, the coolant actuator control output selecting section 622 adds the shape control output d29 to the abnormality limit output d28, as in FIG 13C shown and outputs them. At this time, the abnormality limit output d28 is set and added so that the refrigerant control element flow rate is and becomes within the maximum and minimum values as shown in FIG 11D shown, represents.
Here, the judgment as to whether the operation abnormality evaluation maximum value is large, not so large, or small is determined using a threshold value predefined in the coolant actuator control output selection section 622 . Further, the coolant actuator control output selection section 622 can always add the shape control output d29 to the abnormality limitation output d28 and output it. In this case, however, it is desirable to prevent the added abnormality limiting output d28 from becoming so large.
Furthermore, in the addition in the coolant operation member control output selection section 622, weighted addition may be used instead of simple addition, and the weight may be changed depending on the operation abnormality judgment maximum value.

As described above, according to the apparatus for controlling a plant of the present example, it is possible while preventing an operation drop of the mechanical operator 203 caused by the position result d22 for the mechanical operator normality to perform good gestalt control.

[modification example]

The present invention is not limited to the above embodiment but includes various modification examples. For example, the detailed explanation of the above embodiment has been made for easy understanding of the present invention, and the present invention is not limited to the construction including all explained elements.

For example, in the embodiment described above, the safety manipulation range determination section 205 for the mechanical operator is realized by the machine learning, but it is also possible to realize the safety manipulation range determination section 205 for the mechanical operator by formulating with the formulas based on the experience of the operator. Alternatively, it is also possible to store the rolling state when an operation abnormality occurs in a database, and by judging the existence of the corresponding case, to realize the mechanical operating member safety manipulation range determining section 205 .
Further, the mechanical operating member safe operation range determination section 205 can be used by creating a numerical model or a symbolic logic model based on the knowledge of an operator or an operating engineer in machine learning.

Further, in the embodiment described above, it is provided that the positional abnormality limit control section 620 for the mechanical operating member can store and use the results obtained through experiments, etc. in advance in the coolant control rule database 623 . On the other hand, the position abnormality limit control section 620 for the mechanical actuator can create a rule base from the result data by means of machine learning.
Further, in the embodiment described above, the shape control of the rolling mill is targeted, but the present invention can also be applied to a general control of a factory.

In the block diagrams, as in the 1 etc., only the control lines and information lines necessary for the explanation are shown, all the control lines and information lines necessary in a product are not necessarily shown. Actually, it can be assumed that almost all constructions are connected to each other.

The processing sections such as the control section etc. described in the above embodiment can each be configured by their own hardware, but it is also possible to implement the function of the processing sections described in the above embodiment in a computer by implementing a program (application). to realize.
the 14 Fig. 12 shows a hardware configuration example when the device is configured to control a factory by a computer.
In the 14 Apparatus (computer) 100 for controlling a factory shown in FIG. The factory plant control apparatus 100 further includes a non-volatile memory 100d, a network interface 100e, an input/output device 100f, and a display device 100g.

The CPU 100a is an arithmetic processing section that reads and executes a program code of software for realizing the function performed by the plant control apparatus 100 from the ROM 100b.
In the RAM 100c, the variables, parameters, etc. generated during the arithmetic processing are temporarily written.
For the non-volatile memory 100d, for example, a large-capacity information storage medium such as an HDD (Hard Disk Drive) or an SSD (Solid State Drive) is used. A program (plant control program) that executes a processing function to be performed by the plant control apparatus 100 is recorded in the non-volatile memory 100d. Further, data required for performing machine learning is recorded in the non-volatile memory 100d.

The network interface 100e transmits and receives various information to and from the outside via a LAN (Local Area Network), a leased line, etc.
The input/output device 100f inputs various information from the factory equipment 190 (rolling mill 301) to be controlled and outputs information about the command to each operator 103, 104 (203, 204).
The display device 100g displays the control state of the factory 190 (rolling mill 301) to be controlled.

Information about a program that realizes each processing function executed by the plant control apparatus 100 can be stored in a storage medium such as a semiconductor memory, an IC card, an SD card, in addition to a nonvolatile memory such as an HDD or an SSD or an optical disk, etc.

When a part or the whole of each processing section of the factory equipment control apparatus is configured with hardware, FPGA (Field Programmable Gate Array) or ASIC (Application Specific Integrated Circuit) can be used.

Reference List

11

Shape capture pre-processing section

12

pattern recognition section

13

tax calculation section

14

shape detector

50

control device

100

Device for controlling a factory (computer)

101

calculator

102

third control section

103

high-speed operating element

104

Low Speed Control

105

Security Service Area Determination Section

110

control unit

111

first control section

112

second control section

190

factory to be controlled

201

calculator

202

Position limit control section for the mechanical operator

203

mechanical control element

204

Coolant Control

205

Safety operating range determination section for the mechanical operating element

210

control unit

211

first shape control section

212

second shape control section

300

material to be rolled

301

rolling mill

302

input-side tensioner (input-side TR)

303

output-side tensioner (output-side TR)

304

rolling speed control section

305

Control section for the input TR

306

Control section for the output TR

307

roll gap control section

308

upstream tensiometer

309

outlet-side tensiometer

310

rolling speed setting section

311

Adjustment section for the input side clamping force

312

Adjustment section for output side clamping force

313

Input clamping force control section

314

Output clamping force control section

315

Current conversion section for the input side clamping force

316

Current conversion section for the output side clamping force

317

output-side plate thickness gauge

318

Exit side plate thickness control section

319

roll gap adjustment section

401

stripper

402

first intermediate roll

403

second intermediate roll

404

AS-U roller

405

split roller

406

saddle

501

input data generation section

502

neural network

503

Learning control section for the neural network

504

Neural network selection section

505

teacher data generation section

506

operation abnormality judging section

511

Learning Data Database

512

control rules database

610

Position abnormality area judging section for the mechanical operating member

611

Position abnormality area searching section for the mechanical operating member

612

input data generation section

613

Output data judgment section

620

Position abnormality limiting control section for the mechanical operating member

621

Coolant actuator control output calculation section

622

Coolant actuator control output selection section

623

Coolant control rules database

631

database search section

632

output synthesis section

901

calculator

902

first shape control section

903

second shape control section

904

mechanical control element

905

coolant control

990

rolling mill

d11

first target state variable

d12

first state variable

d13

second state variable

d14

security service area

d21

target shape

d22

Position result for the mechanical control

d23

shape result

d24

rolling result

d25

Safety operating area for the mechanical operating element

d26

operation abnormality evaluation value

d27

Result position abnormality range judgment value for the operating member

d28

abnormality limitation output

d29

shape control issue

d30

coolant service output

d31

estimated position

d32

Operational Abnormality Judgment Value

QUOTES INCLUDED IN DESCRIPTION

This list of the documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• JP 2804161B [0012]
• JP 2515028B [0012]
• JP 2018005544 A [0012]

Claims (7)

A factory equipment control apparatus which performs, for the factory equipment to be controlled, first operation processing having a predetermined response speed to the operation and second operation processing having a slower response speed to the operation than the first operation processing, wherein a first control section which obtains a target state quantity of the plant to be controlled and gives an instruction to the first operation processing, a second control section that obtains a target state quantity of the factory equipment to be controlled and gives an instruction to the second operation processing, a first operator which, according to the command from the first control section, executes the first operation processing of the factory equipment to be controlled, a second operator which, according to the command from the second control section, executes the second operation processing of the factory equipment to be controlled, a safety operation area determination section that determines a safety operation area of the first operation processing by the first operation element based on the result of the first operation element, and a third control section that corrects or changes the command to the second manipulation processing from the second control section and moves a result position of the first manipulation processing by the first manipulation member to a result position in which the occurrence of an operation abnormality is not estimated when the safety manipulation area determination section determines, that the first operation processing by the first operation element is not in the safety operation area, are provided. Device for controlling a factory claim 1 , wherein in the first operation processing by the first operation member, a control time response has a higher speed than the second operation processing and influences on the state quantity to be controlled of the factory equipment to be controlled is limited, and in the second operation processing by the second operation member, a control time response has a lower speed than has the first operation processing and influences are given to the whole range of the state quantity to be controlled of the plant to be controlled. Device for controlling a factory claim 1 wherein the safety operation area determining section recognizes occurrence of an operation abnormality using result data of the factory to be controlled, and learns the relationship between the result data and the operation abnormality by using result data when an operation abnormality occurs as teacher data, and determines a safety operation area. Device for controlling a factory claim 3 , wherein the relationship between the result data and the operation abnormality is given by the machine learning based on the collected result data. Apparatus for controlling a factory according to any one of Claims 1 until 4 , wherein the plant to be controlled is a rolling mill, wherein the first operation processing is mechanical design operation processing in which the shape is changed by a mechanical design, wherein the second operation processing is coolant shape operation processing in which the shape is changed by changing the spray amount of the coolant is changed in the direction of the plate width, wherein the safety manipulation range determination section determines a safety manipulation range of the first manipulation processing based on a result value of a mechanical position of the first manipulation member in which no operation abnormality occurs, and the third control section by changing the control output of the coolant shape control of the second operating member in the output for performing the coolant shape control in which no operation abnormality occurs, the result position of the operation processing for the mechanical configuration is moved to the result position in which the occurrence of an operational abnormality is not estimated by the first operating member when the occurrence of an operational abnormality is estimated in the safety operational area determination section. Method for controlling a factory, which performs, for a factory to be controlled, first operation processing with a predetermined response speed to the operation and second operation processing with a slower response speed to the operation than the first operation processing in an arithmetic processing section by arithmetic processing, wherein the method comprises: a first control step in which the arithmetic processing section obtains a target state quantity of the plant to be controlled and gives an instruction to the first operation processing, a second control step in which the arithmetic processing section obtains a target state quantity of the factory to be controlled and gives an instruction to the second operation processing, a first operation execution step in which the arithmetic processing section, according to the instruction from the first control step, executes the first operation processing of the to factory to be controlled, a second operation execution step in which the arithmetic processing section executes the second operation processing of the factory to be controlled in accordance with the command from the second control step, a safety operation area determination step in which the arithmetic processing section determines, based on the result of the first operation processing, a safety operation area of the first operation processing determined by the first operation execution step, and a third control step in which the arithmetic processing section sends the instruction to the second Be corrects or changes service processing from the second control step and moves a result position of the first operation processing by the first operation execution step to a result position in which the occurrence of an operation abnormality is not estimated when it is determined in the safety operation area determination step that the first operation processing by the first operation execution step is not is in the safety operating area. A program that performs, for a factory to be controlled, first operation processing at a predetermined response speed to the operation and second operation processing at a slower response speed to the operation than the first operation processing by means of a computer, the program having a first control step in which a target state quantity of the factory to be controlled is obtained and an instruction is given to the first operation processing, a second control step in which a target state quantity of the plant to be controlled is obtained and an instruction is given to the second operation processing, a first operation execution step of executing the first operation processing of the plant to be controlled according to the command from the first control step, a second operation execution step in which, according to the command from the second control step, the second operation processing of the factory to be controlled is executed, a safety operation area determination step in which, based on the result of the first operation processing, a safety operation area of the first operation processing is determined by the first operation execution step, and a third control step in which the instruction to the second operation processing from the second control step is corrected or changed, and a result position of the first operation processing is moved to a result position in which the occurrence of an operation abnormality is not estimated by the first operation execution step when in the safety operation area- determination step it is determined that the first operation processing by the first operation execution step is not in the safety operation area, carried out by means of the computer.

Images