PLANT CONTROL DEVICE, PLANT CONTROL METHOD, AND PROGRAM

Abstract

To predict the occurrence of abnormal operation in a plant under control and properly operate a final control element so that the abnormal operation does not occur, thereby improving the control effect and operation efficiency.SOLUTION: A plant under control is subjected to a first operation process with a response rate to an operation being a predetermined response rate and a second operation process with a higher response rate to an operation than that of the first operation process. The first operation process is executed by a first final control element and the second operation process is executed by a second final control element. Here, a safe operation range determination unit is provided that determines the safety operation range of the first operation process by the first final control element on the basis of the records of the first final control element. If the safe operation range determination unit determines that the first operation process by the first final control element is not in the safe operation range, the instructions of the second operation process are corrected or changed, to move the record position of the first operation process by the first final control element to a record position where the occurrence of abnormal operation is not estimated.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a plant control device, a plant control method, and a program. [Background technology]

[0002] Fuzzy control and neuro-fuzzy control have been used for shape control, which is one type of plant control, to control the waviness of strips. Fuzzy control is used for shape control using coolant, and neuro-fuzzy control is used for shape control of Sendzimir rolling mills.

[0003] In shape control to which neuro-fuzzy control is applied, a process is performed to obtain a similarity ratio between the difference between an actual shape pattern detected by a shape detector and a target shape pattern, and a preset reference shape pattern, as shown in Patent Document 1. Then, in shape control to which neuro-fuzzy control is applied, a control output amount for the control element is obtained from the obtained similarity ratio according to a control rule expressed by the control element operation amount for the preset reference shape pattern.

[0004] Shape control has multiple control operation terminals, and control is performed based on the differences in the characteristics of these multiple control operation terminals. Shape refers to the waviness of the strip in the strip width direction, and the control operation terminals can change the shape of a specific area in the strip width direction. For example, the AS-U roll can change the shape near the operated saddle position, and the intermediate roll shift can change the shape of the strip end. When performing shape control, the respective control operation terminals are operated in combination according to the actual shape so that shape deviations are suppressed.

[0005] When a rolling mill performs rolling, a cooling material (hereinafter referred to as coolant) is required for lubrication between the material to be rolled and the rolls of the rolling mill and for cooling heat generated by rolling. This coolant is the operating end for shape control, and by adjusting the amount of coolant sprayed in the strip width direction, it is possible to change the strip shape over the entire strip width direction. In a six-high rolling mill, as shown in Patent Document 2, a coolant spray amount adjustment mechanism is provided in the strip width direction, and shape control is performed by changing the amount of sprayed using the actual strip shape. However, in a Sendzimir rolling mill, the rolls of the rolling mill are immersed in coolant during rolling, and it takes longer for the coolant to have an effect on the strip shape than with AS-U or intermediate roll shifting. In addition, since the amount of coolant sprayed cannot be adjusted automatically, there may be cases where an operator needs to operate the flow rate control valve or the like. In such cases, adjustment is possible only before rolling begins.

[0006] During rolling in a rolling mill, an operational anomaly may occur that causes the material being rolled to break. While breakage of the material being rolled often occurs due to the material being rolled, breakage may also occur due to the material meandering. The meandering of the material being rolled refers to the material being shifted to one side of the rolling mill, and rolling is usually performed in the center of the rolling mill. This kind of meandering is expected to occur due to the position of the AS-U rolls and intermediate roll shift. In shape control, the AS-U and intermediate roll shift are operated to maintain the shape of the rolled material at the target shape, but as a result, the mechanical state may become such that meandering of the rolled material is likely to occur.

[0007] Conventionally, in normal rolling mills such as 4-high rolling mills and 6-high rolling mills, in addition to mechanical operation parts such as bender and leveling, there is an operation part that changes the amount of coolant sprayed in the strip width direction, which is used for shape control. Unlike Sendzimir rolling mills, the rolls of the rolling mill are not immersed in the coolant, and the effect of the coolant on the strip shape is obtained in the same time as with mechanical operation means. Therefore, in shape control in normal rolling mills, the mechanical operation means and the coolant are treated equally, and the coolant is used as the control operation end. Even in this case, since the coolant has an effect over the entire strip width direction, it competes with the mechanical operation parts, and even if the actual shape of the rolled material is the same, the actual position of the mechanical operation parts is often different.

[0008] FIG. 15 shows a schematic configuration of a conventional control device for a Sendzimir rolling mill. First, a calculator 901 calculates the difference between a target shape d1 and an actual shape d2 of the material to be rolled obtained by a rolling mill 990, and provides this difference to a first shape control unit 902 and a second shape control unit 903. The first shape control unit 902 controls a mechanical operation terminal 904, which is a mechanical operation unit. The second shape control unit 903 controls a coolant operation terminal 905 that changes the amount of coolant injected.

[0009] The rolling mill 990 performs a rolling process on the rolled material by performing a mechanical operation process by the mechanical operation end 904 and an operation process of the coolant injection amount by the coolant operation end 905, and obtains a shape result d2 and a rolling result d3 of the rolled material. In this case, in the mechanical operation process by the mechanical operation end 904, the shape of the rolled material is changed with a relatively high-speed response by controlling the operation amount. On the other hand, the operation process of the coolant injection amount by the coolant operation end 905 has a slower response than the mechanical operation process, even if the operation amount is controlled.

[0010] In the case of the configuration shown in FIG. 15, as already explained, the mechanical operation process by the mechanical operation end 904 and the operation process of the coolant injection amount by the coolant operation end 905 compete with each other. In this case, even if the shape actual value d2 is the same, the mechanical operation amount by the mechanical operation end 904 is not necessarily the same. Moreover, the response speed of the mechanical operation process by the mechanical operation end 904 and the operation process of the coolant injection amount by the coolant operation end 905 are different, and further, the range of the effect in the strip width direction is different for each operation process. Therefore, there is a problem that it is difficult to appropriately control both the mechanical operation process by the mechanical operation end 904 and the operation process of the coolant injection amount by the coolant operation end 905. For example, when the high-speed response mechanical operation end 904 is in a state close to the upper limit of the operating range, the shape actual value d2 and the rolling actual value d3 cannot be stably controlled, and vibration occurs, which may prevent the rolling mill 990, which is the plant to be controlled, from operating stably.

[0011] Conventional techniques for appropriately controlling the shape of such a Sendzimir rolling mill include, for example, a technique for learning and controlling the relationship between the actual shape deviation of the rolled material and the manipulated variable of the control operating terminal from actual data using machine learning, as described in Patent Document 3. In the technique described in Patent Document 3, shape control is performed by issuing a control output in response to the shape deviation and manipulating the control operating terminal. [Prior art documents] [Patent documents]

[0012] [Patent Document 1] Patent No. 2804161 [Patent Document 2] Patent No. 2515028 [Patent Document 3] JP 2018-005544 A Summary of the Invention [Problem to be solved by the invention]

[0013] The technology described in Patent Document 3 executes shape control by learning a control operation method for shape deviation, and does not take into consideration the position of the control operation terminal. In actual control, the operation range is limited due to the mechanical conditions of the control operation terminal, but the control output for that control operation terminal is stopped and, if possible, operation is performed using another control operation terminal.

[0014] Furthermore, in cases such as a Sendzimir rolling mill, where the time it takes for the coolant as a shape control operating end to affect the shape is longer than that of a mechanical operating end, even if a coolant operation command is output in the same way as with a mechanical operating end as in the past, the shape will be controlled by the mechanical operating end, whose effect on the shape is transmitted in a short time, and shape control by the coolant cannot be effectively performed.

[0015] When rolling with a rolling mill, operational abnormalities such as plate breakage may occur depending on the characteristics of the material being rolled, the rolling state, and the actual position of the mechanical control terminal for shape control, so it is required to limit the actual position of the mechanical control terminal. However, if the mechanical control terminal and coolant are used as the control terminal for shape control, it is extremely difficult to limit the actual position of the mechanical control terminal. As described above, in conventional shape control, the coolant is controlled based on shape deviation in the same way as the mechanical operating element, and therefore there is a problem in that it is not possible to limit the actual position of the mechanical operating element that could lead to operational abnormalities such as plate breakage.

[0016] In the explanation so far, the problem of shape control of a Sendzimir rolling mill has been described. However, when various plant control systems simultaneously perform a control operation with a fast response and a control operation with a slow response, similar problems arise when both control operations are simultaneously and appropriately performed.

[0017] An object of the present invention is to provide a plant control device, a plant control method, and a program that can improve control effectiveness and operational efficiency by predicting the occurrence of operational abnormalities in a controlled plant and appropriately operating control terminals to prevent operational abnormalities from occurring. [Means for solving the problem]

[0018] In order to solve the above problems, for example, the configurations described in the claims are adopted. The present application includes a plurality of means for solving the above-mentioned problems. One example of such a means is a plant control device that performs a first operation process for performing a mechanical shape operation process for changing the thickness of the rolled material using a mechanical structure based on an operation of a controlled plant, which is a rolling mill that rolls the rolled material, and a second operation process for changing the thickness of the rolled material by changing the amount of coolant injected in the width direction based on the operation, and the second operation process has a slower response speed to the operation than the first operation process. A first control unit acquires a state quantity of a mechanical structure corresponding to a target plate thickness of the controlled plant and instructs a first operation process; A second control unit that acquires a state quantity of the coolant corresponding to a target plate thickness of the controlled plant and issues an instruction for a second operation process; a first operation terminal that executes a first operation process of the controlled plant in response to an instruction from a first control unit; a second operation terminal that executes a second operation process of the controlled plant in response to an instruction from the second control unit; a safe operation range determination unit that determines a safe operation range of a first operation process by the first operating end based on a performance of the first operating end; a third control unit that, when the first operation process by the first operating end is not within the safe operation range as determined by the safe operation range determination unit, corrects or changes an instruction of the second operation process by the second control unit based on the state quantity of the coolant, and moves an actual position of the first operation process by the first operating end to an actual position of a machine structure where the occurrence of an operational abnormality is not predicted, The safe operating range determination unit has a neural network that searches for and determines a safe operating range using learning data obtained based on past performance, The neural network used is selected from multiple types of available ones according to the specifications of the material to be rolled, including the plate width and steel type. The search for safe operating ranges using neural networks uses performance data of the plant to be controlled to recognize the occurrence of operational abnormalities, and uses the performance data at the time of the operational abnormality as training data to learn the relationship between the performance data and the operational abnormality, and determines the safe operating range using the training data obtained. The actual data when operational abnormalities occur shows the signs of plate breakage. Detection of meandering of rolled material This includes emergency stop data and emergency stop data. Effect of the Invention

[0019] According to the present invention, the operating state of the controlled plant can be appropriately controlled to suppress the actual position of the controlled element that may cause an operational abnormality. As a result, it is expected that the control accuracy of the controlled plant, the operational efficiency, and the occurrence of an operational abnormality can be improved. Problems, configurations and effects other than those described above will become apparent from the following description of the embodiments. [Brief description of the drawings]

[0020] [Figure 1] 1 is a block diagram showing an example of the configuration of a plant control device according to an embodiment of the present invention; [Diagram 2] 1 is a block diagram showing a configuration example in which a plant control device according to an embodiment of the present invention is applied to a rolling mill. [Diagram 3] FIG. 2 is a configuration diagram showing an example of a Sendzimir rolling mill. [Figure 4] FIG. 1 is a configuration diagram showing an example of rolling equipment for a single stand rolling mill. [Diagram 5] 1 is a diagram showing an overview of a mechanical operating end according to an embodiment of the present invention; [Figure 6] 4 is a block diagram showing an example of the configuration of a safe operation range determination unit according to an embodiment of the present invention. FIG. [Figure 7] FIG. 1 is a diagram showing an example of the configuration of a neural network according to an embodiment of the present invention. [Figure 8] FIG. 11 is a diagram showing an example of the configuration of a neural network management table according to an embodiment of the present invention. [Figure 9]FIG. 2 is a diagram showing an example of the configuration of a learning database according to an embodiment of the present invention. [Figure 10] 3 is a block diagram showing an example of the configuration of a mechanical operating element position suppression control unit according to an embodiment of the present invention; FIG. [Figure 11] 1 is a diagram showing an overview of a mechanical operating element position abnormality region determination unit according to an embodiment of the present invention; [Figure 12] FIG. 4 is a diagram illustrating a configuration and operation of a coolant operation terminal control output calculation unit according to an embodiment of the present invention. [Figure 13] FIG. 4 is a diagram illustrating the operation of a coolant operation terminal control output selection unit according to an embodiment of the present invention. [Figure 14] 1 is a block diagram showing an example of a hardware configuration when a plant control device according to an embodiment of the present invention is configured using a computer; [Figure 15] FIG. 1 is a block diagram showing a configuration example of a conventional control device for a rolling mill. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0021] A plant control device according to one embodiment of the present invention (hereinafter referred to as "this embodiment") will be described below with reference to Figs. 1 to 13.

[0022] [Overall configuration of plant control system] FIG. 1 shows an example of the overall configuration of a plant control device according to this embodiment. The plant control device shown in Figure 1 controls a controlled plant 190, and performs operation processing using a high-speed operating end (first operating end) 103 and operation processing using a low-speed operating end (second operating end) 104 to control the controlled plant 190. The plant control device of this example obtains a first state quantity target d11, and obtains the difference between the first state quantity d12 and the first state quantity d12 in the calculator 101. The first state quantity d12 is obtained from the controlled plant 190 as a result of operation processing by the high-speed operation terminal 103 and the low-speed operation terminal 104. In addition, a second state quantity d13 is obtained as a result of operation processing by the high-speed operation terminal 103, the low-speed operation terminal 104, and other operation terminals applied to the controlled plant 190.

[0023] The control unit 110 of the plant control device has a first control section 111 that controls the operation processing of the high speed control terminal 103 and a second control section 112 that controls the operation processing of the low speed control terminal 104. The control output of the first control unit 111 is directly supplied to the high speed operation terminal 103 and controls the operation processing of the high speed operation terminal 103 . The control output of the second control section 112 is supplied to the third control section 102, and the control output is corrected or changed as necessary, and then supplied to the low speed operating terminal 104.

[0024] The plant control device of this embodiment also includes a safe operating range determination unit 105. The safe operating range determination unit 105 acquires the operation record of the high speed control terminal 103, judges whether the acquired operation record has a margin within the safe operating range or not, and supplies data of the safe operating range d14 as the judgment result to the third control unit 102. Furthermore, the safe operating range determination unit 105 acquires the first state quantity d12 and the second state quantity d13 of the controlled plant 190. Then, the safe operating range determination unit 105 refers to the first state quantity d12 and the second state quantity d13 to determine whether or not the current operation record of the high speed operating end 103 has a margin within the safe operating range.

[0025] Here, the safe operating range determination unit 105 detects the occurrence of an operational abnormality, learns the first state quantity d12 and the second state quantity d13 at that time and the actual position of the high-speed control end 103, and determines a safe operating range d14, which is an operable range, so that an operational abnormality does not occur at the high-speed control end 103. Since operational abnormalities are not a phenomenon that occurs frequently, it is preferable that the safe operating range determination unit 105 continuously collects actual data and performs learning using machine learning or the like to obtain the safe operating range d14. Then, the safe operation range determination unit 105 supplies data of the determined safe operation range d14 to the third control unit 102.

[0026] When the third control unit 102 determines from the data of the safe operation range d14 that there is some margin in the safe operation range, it supplies the control output of the second control unit 112 as is to the low speed operation terminal 104. On the other hand, when the third control unit 102 determines from the data of the safe operation range d14 that there is no margin in the safe operation range, it corrects or changes the control output of the second control unit 112 and supplies it to the low speed operation terminal 104. The third control unit 102 is supplied with the output of the calculator 101, the operation results of the high-speed operating end 103, and the second state quantity d13, and based on this information, the third control unit 102 corrects or changes the control output of the second control unit 112.

[0027] According to the plant control device having the configuration shown in FIG. 1, it is possible to operate the high-speed control terminal 103 efficiently within a range in which no operational abnormalities occur, and this is expected to result in improved operational efficiency as well as improved control accuracy.

[0028] [Overall configuration when applied to a Sendzimir rolling mill control device] Next, the overall configuration of the plant control device of this embodiment when applied to a Sendzimir rolling mill will be described. FIG. 2 shows the configuration of the plant control device of this example when applied to a Sendzimir rolling mill. The plant control device shown in FIG. 2 obtains a target shape d21 of the material to be rolled, and a calculator 201 obtains the difference between the target shape d21 and the actual shape after rolling d23.

[0029] The plant control device of this example controls the rolling mill 301 by executing operation processing by the mechanical operation end 203 and operation processing by the coolant operation end 204. The operation processing by the mechanical operation end 203 mechanically changes the roll gap and the like that performs the rolling process, and the response that appears in the shape actual result d23 of the rolled material is high-speed. On the other hand, the operation processing by the coolant operation end 204 changes the amount of coolant injection, and the response that appears in the rolling actual result d24 of the rolled material is slower than the operation by the mechanical operation end 203.

[0030] The control unit 210 of the plant control device has a first shape control unit 211 that controls the operation processing of the mechanical operation end 203, and a second shape control unit 212 that controls the operation processing of the coolant operation end 204. The first shape control unit 211 and the second shape control unit 212 control the rolled material so that it has a target shape d21. The target shape d21 is set in advance according to the characteristics of the rolled material, etc. The control output of the first shape control unit 211 is directly supplied to the mechanical operation end 203 and controls the operation processing of the mechanical operation end 203 . The control output of the second shape control unit 212 is supplied to the mechanical operating end position suppression control unit 202, and the control output is corrected or changed as necessary, and then supplied to the coolant operating end 204. The configuration of the mechanical operation element position suppression control unit 202 will be described later with reference to FIG.

[0031] The plant control device of this example also includes a mechanical end safe operation range determination unit 205. The mechanical end safe operation range determination unit 205 acquires a mechanical end position record d22, which is the operation record of the mechanical end 203, and judges whether the acquired mechanical end position record d22 has a margin within the safe operation range. Then, the mechanical end safe operation range determination unit 205 supplies data on the mechanical end safe operation range d25 as the judgment result to the mechanical end position suppression control unit 202.

[0032] Furthermore, the mechanical operating end position suppression control unit 202 acquires the shape actual value d23 and the rolling actual value d24 of the rolling mill 301. Then, the mechanical operating end safe operation range determination unit 205 refers to the shape actual value d23 and the rolling actual value d24 and judges whether the acquired mechanical operating end position actual value d22 has a margin within the safe operation range.

[0033] The mechanical end safe operation range determination unit 205 detects the occurrence of an operational abnormality and learns the shape record d23 and rolling record d24 at that time, and the mechanical end position record d22. Through this learning, the mechanical end safe operation range determination unit 205 determines the mechanical end safe operation range d25, which is an operable range, so that an operational abnormality does not occur in the mechanical end 203. Here, the mechanical end safe operation range determination unit 205 continuously collects record data and performs learning using machine learning or the like to obtain the mechanical end safe operation range d25. Then, the mechanical element safe operation range determination unit 205 supplies data on the determined mechanical element safe operation range d25 to the mechanical element position suppression control unit 202. The detailed configuration of the mechanical operating element safe operating range determination unit 205 for learning operational abnormalities will be described later with reference to FIG.

[0034] When the mechanical operating end position suppression control unit 202 determines from the data of the mechanical operating end safe operating range d25 that there is a margin in the safe operating range, it supplies the control output of the second shape control unit 212 as is to the coolant operating end 204. On the other hand, when the mechanical operating end position suppression control unit 202 determines from the data of the mechanical operating end safe operating range d25 that there is no margin in the safe operating range, it corrects or changes the control output of the second shape control unit 212 and supplies it to the coolant operating end 204. In addition, the output of the calculator 201, the mechanical operating end position result d22, and the rolling result d24 are supplied to the mechanical operating end position suppression control unit 202, and the mechanical operating end position suppression control unit 202 corrects or changes the control output of the second shape control unit 212 based on this information.

[0035] [Configuration of Sendzimir rolling mill] Here, an example of the configuration of a Sendzimir rolling mill will be described. FIG. 3 shows a schematic configuration for performing shape control in a Sendzimir rolling mill. In the Sendzimir rolling mill, the actual shape of the rolled material is detected by shape detector 14. The actual shape detected by shape detector 14 is subjected to pre-processing for pattern recognition in shape detection pre-processing unit 11 of control device 50, and then a calculation is performed in pattern recognition unit 12 to determine which of preset reference shape patterns the actual shape is closest to. Then, based on the calculated reference shape pattern, control calculation unit 13 determines the operating element and operating amount to be operated, and a process is performed to control the Sendzimir rolling mill with the operating element and operating amount to be operated.

[0036] Figure 4 shows an example of a single-stand rolling mill. A Sendzimir rolling mill is a type of single-stand rolling mill. The rolling equipment shown in Figure 4 is composed of a rolling mill 301, an entrance tension reel (hereinafter referred to as "TR") 302, and an exit TR 303. The material to be rolled 300 pulled out from the entrance TR 302 passes through the rolling mill 301 and is wound up by the exit TR 303. The rolling mill 301 rolls the material to be rolled 300. The rolling here refers to a process of reducing the thickness of the material to be rolled 300 to a predetermined thickness.

[0037] The rolling mill 301 is provided with a mill speed control unit 304 for adjusting the rolling speed, and a roll gap control unit 307 for adjusting the roll gap of the rolling mill 301. Furthermore, the entry TR 302 and the exit TR 303 are provided with an entry TR control unit 305 and an exit TR control unit 306 for adjusting the tensions generated therein, respectively.

[0038] The rolling process is performed by applying pressure to crush the material 300 to be rolled by adjusting the distance between the upper and lower rolls of the rolling mill 301 with a roll gap control unit 307, and sending the material 300 to the exit side with a mill speed control unit 304. At this time, a process of applying tension to the material 300 to be rolled is also performed at the entry side and exit side of the rolling mill 301 using an entry TR 302 and an exit TR 303.

[0039] What is important for rolling operations is the thickness of the rolled material 300 that will become the product (the thickness at the exit of the rolling mill), and the roll gap, entry tension, and exit tension are preset so that the rolled material 300 will have a predetermined thickness. The entry-side tension current conversion unit 315 uses the entry-side tension set by the entry-side tension setting unit 311 to calculate the current required to obtain the set entry-side tension, and supplies this to the entry-side TR 302 via the entry-side TR control unit 305 to obtain the entry-side tension. Similarly, the exit side tension current conversion unit 316 uses the exit side tension set by the exit side tension setting unit 312 to calculate the current required to obtain the set exit side tension, and supplies this to the exit side TR 303 via the exit side TR control unit 306 to obtain the exit side tension.

[0040] The roll gap set by the roll gap setting unit 319 is given to the roll gap control unit 307, and the roll gap is set by the roll gap control unit 307. The rolling speed setting unit 310 determines the speed of the rolling mill 301 according to an instruction from an operator of the rolling mill, and the mill speed control unit 304 sets the speed of the rolling mill 301 .

[0041] An entry tension meter 308 and an exit tension meter 309 are installed on the entry and exit sides of the rolling mill 301, and an entry tension control unit 313 and an exit tension control unit 314 execute control so that the actual tensions measured by these units match the set tensions. Also, an exit thickness meter 317 is installed on the exit side of the rolling mill 301, and an exit thickness meter 318 executes control so that the actual thickness measured by these units matches the set thickness. Control Unit 318 performs the control.

[0042] In addition to the above configuration, as shown in FIG. 3 already explained, a shape detector 14 for detecting the shape of the rolled material is installed at the delivery side of the rolling mill, and shape control is performed so that the detected shape coincides with a preset target shape.

[0043] As already mentioned, the shape is the degree of corrugation of the metal plate to be rolled. Therefore, a target shape is set in advance based on the workability in downstream processes of the rolling mill and the efficiency of the rolling operation in the rolling mill. Generally, tension is applied to the material to be rolled, so if there is a flaw such as a crack at the edge of the plate, a tear will occur from that point, and the material to be rolled is likely to be torn in the plate width direction (plate breakage). For this reason, the edge of the plate is often made corrugated to prevent the tension from concentrating.

[0044] The waviness of the rolled material is not apparent because tension is actually applied to the rolled material, and there appears to be no waviness, but the tension distribution changes in the sheet width direction. Here, the shape detector 14 shown in FIG. 3 estimates the waviness of the plate by measuring the tension distribution in the plate width direction, and detects it as the actual shape.

[0045] [Configuration and processing of shape control mechanical operation terminals] Fig. 5(a) shows the configuration when operation processing is performed by the mechanical operation end 203 of the Sendzimir rolling mill. Fig. 5 shows a cross section in the width direction of the material to be rolled 300, and only shows the configuration on the upper side of the material to be rolled 300, omitting the configuration on the lower side. 5(b) and (c) respectively show operational waveforms when the shape of the material 300 to be rolled is changed.

[0046] As shown in FIG. 5( a ), the Sendzimir rolling mill is composed of a work roll 401 , a first intermediate roll 402 , a second intermediate roll 403 , and an AS-U roll 404 , which sandwich the material to be rolled 300 . The first intermediate roll 402 has a taper on the upper and lower rolls on opposite sides, and can affect the shape of the strip end of the material 300 to be rolled by shifting in the strip width direction.

[0047] The AS-U roll 404 is configured with a saddle 406 placed between multiple divided rolls 405, and by changing the position of the saddle 406 (the vertical position in Figure 5), the deflection of the AS-U roll 404 can be changed in the plate width direction. For example, as shown in FIG. 5(b), if the central saddle 406 is lowered, the shape of the central portion of the rolled material 300 can be affected.

[0048] 5(b) and (c) show the change in thickness distribution of the rolled material 300 when the saddle 406 or the first intermediate roll 402 is shifted. The change in shape is opposite to the thickness distribution. The shape is the distribution of the degree of waviness in the strip width direction, and large waviness means that the rolled material 300 is elongated. This is because "thinning of the strip thickness at the delivery side" is equivalent to "large elongation of the rolled material in the thinned portion" and "enlargement of the shape of the rolled material."

[0049] Regarding operational abnormalities, plate breakage of the rolled material 300 is a major problem. If plate breakage occurs, the rolled material 300 after the breakage will damage the work roll 401 and the first intermediate roll 402 of the rolling mill. In some cases, the plate breakage may also damage the second intermediate roll 403 and the AS-U roll 404. If such damage occurs, it becomes necessary to replace these rolls, and time is required to remove the rolled material 300 remaining in the rolling mill, resulting in an extreme decrease in operational efficiency.

[0050] Since the AS-U roll 404 presses the split roll 405 against the second intermediate roll 403 by holding it down with the saddle 406, the split roll 405 and the second intermediate roll 403 may not come into contact depending on the position of the saddle 406. When this happens, the force applied to that part of the work roll 401 on the rolled material 300 decreases suddenly, the rolled material 300 stops stretching, and the tension applied to that part of the rolled material 300 increases.

[0051] If such a state occurs at the end of the material to be rolled 300, the material will break from the end. Also, because the tension applied to both ends of the material to be rolled 300 in the width direction changes, a phenomenon occurs in which the center of the material to be rolled 300 in the width direction shifts from the center of the rolling mill in the width direction, and the material may collide with machinery before or after the rolling mill, causing the material to break. In this way, depending on the actual position of the mechanical operation terminal 203, operational abnormalities may occur.

[0052] The actual position where an operational abnormality will occur cannot be calculated from the machine configuration, and is difficult to predict in advance because it changes depending on the distribution of thickness in the width direction of the rolled material 300, the inlet and outlet thicknesses, rolling conditions such as tension and rolling load, and the positional relationship with other shape control mechanical operating terminals. Therefore, in this example, the mechanical control end safe operating range determination unit 205 stores these conditions when a rolling abnormality occurs as actual data, and by comparing it with the actual data under normal conditions, determines the actual positions of the shape control mechanical control ends at which rolling abnormalities are likely to occur.

[0053] The mechanical control element safe operation range determination unit 205 in this embodiment determines the safe operation range of the mechanical control element by using machine learning. When performing machine learning, the input and output plate thicknesses, rolling conditions such as tension and rolling load, and the actual position of the mechanical control element 203 are used as the record data, and rolling abnormality occurrence information is used as the teacher data. As rolling abnormality occurrence information, information on plate breakage and emergency stop of the rolling mill is used. Plate breakage can be determined by a decrease in the input / output tension, and emergency stop is determined by information on an operation switch operated by an operator when some abnormality occurs in the rolling state and operation is stopped. Information on the operation switch can be detected by a computer constituting the control device that controls the rolling mill, and can be used as one piece of performance information. The mechanical operating end safe operation range determination unit 205 uses this performance data and teacher data to create a neural network (NN) that determines whether or not an operational abnormality has occurred.

[0054] [Configuration of safe operating range determination unit for mechanical control elements and configuration of neural network] FIG. 6 shows a configuration in which the mechanical operation element safe operation range determination unit 205 is realized by machine learning. FIG. 7 shows the configuration of a neural network 502 included in the mechanical operation element safe operation range determination unit 205. As shown in Fig. 7, the neural network 502 receives the rolling results d24 and the mechanical operation end position results d22 from the input data creation unit 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 is information on plate breakage and emergency stop, which are rolling abnormality occurrence information. The neural network 502 executes learning from a combination of these input data and output data.

[0055] 6, the mechanical operating element safe operation range determination unit 205 will be described. An input data creation unit 501 collects the mechanical operating element position record d22 and shape record d23. A teacher data creation unit 505 collects the operation abnormality judgment value d32 judged by an operation abnormality judgment unit 506. The input data creation unit 501 and the teacher data creation unit 505 collect data at regular time intervals under the control of a neural network learning control unit 503, and one set of learning data is obtained for each operation cycle. The obtained learning data is stored in a learning data database 511 in sequence. The operational abnormality determination unit 506 determines the presence or absence of an operational abnormality such as a plate break or an emergency stop of the rolling mill based on the rolling performance d24. The operational abnormality determination value d32, which is the determination result, is information on a plate break or an emergency stop.

[0056] Incidentally, a rolling mill rolls various materials 300 to be rolled according to specifications to obtain products. For this reason, rolling mills generally respond to the specifications of the materials 300 to be rolled by changing the specifications of the work rolls 401 (diameter distribution in the sheet width direction), the taper specifications of the first intermediate rolls 402, and the combination of the split rolls 405 of the AS-U rolls 404, which are the mechanical configuration. Furthermore, the width and material of the materials 300 to be rolled are not uniform. For this reason, it is more efficient to learn by dividing the neural network 502 according to the mechanical configuration and the specifications of the materials 300 to be rolled. For this reason, the mechanical operating element safe operation range determination unit 205 of this embodiment has a control rule database 512 and a neural network selection unit 504 so as to have a plurality of types of neural networks 502 and be able to switch between them for use.

[0057] FIG. 8 shows an example of the configuration of the control rule database 512. As shown in FIG. As shown in FIG. 8(a), the control rule database 512 stores a plurality of neural networks trained using training data consisting of a combination of input data and teacher data. Then, the neural net learning control unit 503 designates the neural net number that needs to be learned. The neural net selection unit 504 receives the designation of the neural net number that needs to be learned from the neural net learning control unit 503, extracts the neural net from the control rule database 512, and sets it as the neural net 502. The neural network selection unit 504 extracts a neural network with a corresponding neural network number from the control rule database 512 according to the rolling conditions and machine configuration from the current rolling results d24, and sets it in the mechanical operating end position suppression control unit 202 as the control neural network d33.

[0058] 8(b) shows the configuration of the neural network management table stored in the control rule database 512. The management table is divided according to (B1) plate width and (B2) steel grade and machine configuration (A). For (B1) plate width, for example, four divisions are used: 3 feet width, meter width, 4 feet width, and 5 feet width. For (B2) steel grade, about 10 divisions are used: steel grade (1) to steel grade (10). For (A), for example, the division is made into (A1) and (A2) according to the length of the tapered portion, which is the tapered specification of the first intermediate roll 402. The above table classification is only an example, and it is necessary to set the table appropriately according to the rolling equipment and the type of rolled material to be produced. The mechanical operating element safe operation range determination unit 205 uses these neural networks selectively depending on the rolling conditions and the machine configuration.

[0059] The neural network learning control unit 503 associates the learning data, which is a combination of the input data and teacher data shown in Figure 8(a), with the corresponding neural network number in accordance with the neural network management table shown in Figure 8(b) and stores it in the learning data database 511.

[0060] FIG. 9 shows an example of the learning data stored in the learning data database 511. As shown in FIG. 9, the learning data database 511 stores learning data corresponding to each neural network number.

[0061] Neural network learning control unit 503 instructs input data creation unit 501 and teacher data creation unit 505 to retrieve input data and teacher data corresponding to the neural network from the management table in learning data database 511. Neural network 502 uses these to perform learning. Various neural network learning methods have been proposed in the past, and any learning method may be used.

[0062] Machine learning requires a large number of sets of training data, and once a certain number (for example, 10,000 sets) have been stored in the training data database 511, the neural network 502 begins learning. When the learning of the neural net 502 is completed, the neural net learning control section 503 writes the neural net 502, which is the learning result, back to the position of the neural net number in the control rule database 512, thereby completing the learning.

[0063] The neural network 502 which has completed learning outputs an operation abnormality judgment value by inputting the rolling result d24 and the mechanical control end position result d22. Therefore, the neural network 502 can predict the occurrence or absence of an operation abnormality by providing the predicted future shape result d23 and the mechanical control end position result d22, and can search for the safe operation range d25 of the mechanical control end.

[0064] [Configuration of the mechanical control element position suppression control unit] FIG. 10 shows the configuration of the mechanical operation element position suppression control unit 202. As shown in FIG. The mechanical element position suppression control section 202 includes a mechanical element position abnormality region determination section 610 and a mechanical element position abnormality suppression control section 620 . The mechanical control end position abnormality region determination unit 610 estimates the mechanical control end 203 where an operation abnormality is predicted to occur, using the neural network 502 explained in Fig. 7. The neural network 502 used here is the control neural network d33 received from the mechanical control end safe operation range determination unit 205. The mechanical operation element position abnormality suppression control unit 620 creates an operation command for the coolant operation element 204 based on the result of the determination by the mechanical operation element position abnormality area determination unit 610 .

[0065] During rolling operation, the shape actual value d23 of the rolling mill 301 changes from moment to moment, and the first shape control unit 211 operates the mechanical control element 203 to maintain it at the target shape d21, and the mechanical control element position actual value d22 also changes from moment to moment. The mechanical control element position abnormality area searching unit 611 provided in the mechanical control element position abnormality area determination unit 610 creates input data for the neural network 502 in the input data creation unit 612 using the mechanical control element position actual value d22.

[0066] FIG. 11 shows the process performed by the mechanical operation element position abnormality region determination unit 610. As shown in FIG. In the example of FIG. 11, when the mechanical operation element 203 has n types (n is an integer), the mechanical operation element position record d22 is as follows. POS(k), k=1,2,···,n The n types of mechanical operation ends 203 here refers to, for example, the total number of the saddles 406 of the AS-U roll 404 and the number of the first intermediate rolls 402 that can be shifted in the strip width direction. In the example shown in Fig. 5, there are five saddles and two first intermediate rolls, one on the top and one on the bottom, so n = 7.

[0067] When the mechanical operating end 203 is operated by the first shape control unit 211, the operation is limited to a certain amount per control cycle, so the mechanical operating end position abnormality area searching unit 611 creates an estimated position of the mechanical operating end position result d22 of each mechanical operating end 203 as follows. Here, for example, the position result change amount ΔPOS that may be operated in several control operations is taken as being possible, and three cases are considered: when there is no change in the position result and when the result value changes in the positive and negative directions. POS(k),POS(k)±ΔPOS,k=1,2,...,n

[0068] As a result, the estimated position results of each mechanical operation terminal 203 are n For example, when n=7, 2187 different estimated position results can be created. These estimated position results are output to the input data creation unit 612 in sequence. The input data creation unit 612 creates input data for the neural network 502 from the rolling results d24 and the estimated position d31, and outputs the data to the neural network 502.

[0069] The neural network 502 outputs an operation abnormality judgment value d32 shown in Fig. 11(d). The operation abnormality judgment value d32 is the degree of strip breakage and emergency stop, and here, the output data judgment unit 613 receives the operation abnormality judgment value d32 output from the neural network 502, weights the degree of both, and adds them together to obtain an operation abnormality evaluation value d26. In general, when an operation abnormality occurs, an operator performs an emergency stop, but also when a symptom of strip breakage occurs. An example of a symptom of strip breakage here is meandering of the rolled material 300. If a plate break occurs without an emergency stop, it means that the break occurred without any symptoms, and so the plate break has a higher priority for suppression. Therefore, the weighting of the degree of the plate break is increased.

[0070] The mechanical control element position abnormality region searching unit 611 stores the output estimated position d31 and the returned operation abnormality evaluation value d26, and searches for an estimated value that maximizes the operation abnormality evaluation value d26. If the search result shows that the maximum operation abnormality evaluation value d26 exceeds a predetermined threshold, the mechanical control element position abnormality region searching unit 611 outputs the actual position change amount in the estimated position d31 in that case as the actual position abnormality region control element judgment value d27. The operation abnormality evaluation value d26 in that case is also included in the actual position abnormality region control element judgment value d27 as the maximum operation abnormality evaluation value.

[0071] In the above-described example, the mechanical control element position abnormality area searching unit 611 creates the estimated position d31 with three variations from the actual position of the mechanical control element 203, but it may be created by a different process. For example, the mechanical control element position abnormality area searching unit 611 finely controls the variation of the estimated position d31. Note that the mechanical control element position abnormality area searching unit 611 may be changed as appropriate depending on the situation, such as not searching in a direction where no operational abnormality will obviously occur. Here, a direction where no operational abnormality will obviously occur may be, for example, a movement toward the center of the actual position of the mechanical control element.

[0072] The mechanical operating end position abnormality suppression control unit 620 creates a control output to the coolant operating end 204 from the actual position abnormality area operating end judgment value d27, which is the output of the mechanical operating end position abnormality area judgment unit 610, and a control command to the coolant operating end 204 from the second shape control unit 212.

[0073] When it is determined that no operational abnormality will occur even if the mechanical operating end position record d22 is changed by control, normal operation using the coolant operating end 204 by the second shape control unit 212 is possible. When the operation abnormality evaluation value d26 is not very large, it is also possible to output the control output of the second shape control unit 212 in a form in which the abnormality suppression output d28 for suppressing the operation abnormality is combined. Of course, when the operation abnormality evaluation value d26 is large, the abnormality suppression output d28 is preferentially output to the coolant operating end 204.

[0074] In the coolant control rule database 623, the correspondence between each mechanical operation end 203(k) and the coolant operation end 204 that it affects is determined in advance. This correspondence can be obtained by actually operating the mechanical operation end 203 and the coolant operation end 204 during rolling operation, or can be obtained from performance data by machine learning. Here, a case is considered in which the correspondence is obtained from the results of actual operation and registered in the coolant control rule database 623.

[0075] [Configuration and operation of the coolant operation terminal control output calculation unit] FIG. 12 shows the configuration and operation of the coolant operation terminal control output calculation unit 621. The coolant control rule database 623 (FIG. 10) stores required coolant flow rate changes that provide the same effect as operating each mechanical control end 203(k). A database search unit 631 shown in FIG. 12(a) extracts the required coolant flow rate change amount corresponding to the actual position change amount of the mechanical control end 203(k) at which an operational abnormality occurs, from the actual position abnormality area control end judgment value d27 obtained by the mechanical control end position abnormality area judgment unit 610.

[0076] Then, the output synthesis unit 632 adds up the required coolant flow rate change amounts for each of the extracted mechanical operation terminals 203(k) to obtain an abnormality suppression output d28. For example, when the actual position and estimated position shown in Figures 11(a) and (b) are used and the actual position abnormality region control end judgment value d27 shown in Figure 11(d) is obtained, the abnormality suppression output d28 shown in Figure 12(b) is obtained. The coolant operation terminal 204 performs the lubrication and cooling necessary for rolling operations, so in many cases the maximum and minimum coolant flow rates are determined for each point in the strip width direction, and an abnormality suppression output d28 is obtained to prevent operational abnormalities from occurring.

[0077] [Output selection process in the coolant operation terminal control output selection section] FIG. 13 shows the output selection process performed by the coolant operation end control output selection unit 622. The coolant operation end control output selection unit 622 selects or combines and outputs the abnormality suppression output d28 shown in Fig. 10 and the shape control output d29 of the second shape control unit 212 according to the magnitude of the operation abnormality evaluation maximum value (Fig. 11(d)) within the actual position abnormality region operation end judgment value d27. The output selected or combined by this coolant operation end control output selection unit 622 is supplied to the coolant operation end 204 as a coolant operation output d30.

[0078] 11(d) is small, there is a low possibility that an operation abnormality will occur even if shape control is continued, so the coolant operation terminal control output selection unit 622 supplies the shape control output d29 as is as the coolant operation output d30 to the coolant operation terminal 204. In other words, the coolant operation output d30=shape control output d29.

[0079] On the other hand, when the maximum operation abnormality evaluation value is large, the coolant operation end control output selection unit 622 judges that there is a high possibility of an operation abnormality occurring. In this case, the abnormality suppression output d28 is generated as shown in Fig. 13(b), and the coolant operation end control output selection unit 622 sets the coolant operation output d30 as shown in Fig. 13(a) so that the effect of the abnormality suppression output d28 is maximized within the maximum and minimum values ​​of the coolant flow rate.

[0080] In addition, when the maximum operation abnormality evaluation value is not so large, the coolant operating end control output selection unit 622 adds the shape control output d29 and the abnormality suppression output d28 as shown in Fig. 13(c) and outputs the result. At this time, the abnormality suppression output d28 is adjusted so that the coolant operating end flow rate falls within the maximum and minimum values, and is added, as shown in Fig. 11(d). Here, whether the maximum operation abnormality evaluation value is large, not so large, or small is determined using a threshold value preset in the coolant operation terminal control output selection unit 622. Also, the coolant operation terminal control output selection unit 622 may always add the shape control output d29 and the abnormality suppression output d28 and output the result. In this case, however, the sum of the abnormality suppression output d28 is made not to be too large. Furthermore, when the coolant operation element control output selection unit 622 performs addition, it is also possible to perform weighted addition instead of simple addition, and to vary the weighting according to the maximum operation abnormality evaluation value.

[0081] As described above, according to the plant control device of this embodiment, it is possible to perform good shape control while preventing operational abnormalities caused by the mechanical control terminal position record d22 of the mechanical control terminal 203.

[0082] [Variations] The present invention is not limited to the above-described embodiment, but includes various modifications. For example, the above-described embodiment has been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to those having all of the configurations described.

[0083] For example, in the above-mentioned embodiment, the mechanical end safe operation range determiner 205 is realized by machine learning, but the mechanical end safe operation range determiner 205 may be realized by expressing it by a mathematical formula based on the operator's experience. Alternatively, the rolling state when an operational abnormality occurs may be compiled into a database, and the presence or absence of a corresponding case may be determined to realize the mechanical end safe operation range determiner 205. In addition, the mechanical control element safe operation range determination unit 205 determines the safe operation range of the mechanical control element by using numerical models and symbols based on the knowledge of the operator and the operating engineer. logic You can create a model and use it during machine learning.

[0084] In the above-described embodiment, the mechanical operating element position abnormality suppression control unit 620 uses results previously obtained by experiments or the like that are stored in the coolant control rule database 623. Alternatively, the mechanical operating element position abnormality suppression control unit 620 may create a rule base from performance data using machine learning. Furthermore, in the above-described embodiment, the object is shape control of a rolling mill, but the present invention can also be applied to general plant control.

[0085] In addition, in block diagrams such as Figure 1, only control lines and information lines that are considered necessary for explanation are shown, and not all control lines and information lines in the product are necessarily shown. In reality, it can be considered that almost all components are interconnected.

[0086] In addition, the processing units such as the control unit described in the above-mentioned embodiment examples may each be configured with dedicated hardware, or the functions of each processing unit described in the above-mentioned embodiment examples may be realized by implementing a program (application) in a computer. FIG. 14 shows an example of a hardware configuration in the case where the plant control device is configured by a computer. 14 includes a central processing unit (CPU) 100a, a read only memory (ROM) 100b, and a random access memory (RAM) 100c, each of which is connected to a bus. The plant control device 100 further includes a non-volatile storage 100d, a network interface 100e, an input / output device 100f, and display The apparatus comprises 100g.

[0087] The CPU 100a is an arithmetic processing unit that reads out from the ROM 100b and executes program code of software that realizes the functions performed by the plant control device 100. The RAM 100c temporarily stores variables, parameters, and the like that are generated during the computation process. For the non-volatile storage 100d, for example, a large-capacity information storage medium such as a hard disk drive (HDD) or a solid state drive (SSD) is used. A program (plant control program) that executes the processing functions performed by the plant control device 100 is recorded in the non-volatile storage 100d. Furthermore, data required for machine learning is recorded in the non-volatile storage 100d.

[0088] The network interface 100e transmits and receives various information to and from the outside via a LAN (Local Area Network), a dedicated line, or the like. The input / output device 100f inputs various information from the controlled plant 190 (rolling mill 301) and outputs information for issuing instructions to each of the operation terminals 103, 104 (203, 204). The display device 100g displays the control state of the controlled plant 190 (rolling mill 301).

[0089] The information of the programs that realize the various processing functions performed by the plant control device 100 can be stored in non-volatile storage such as an HDD or SSD, as well as in recording media such as semiconductor memory, IC cards, SD cards, and optical disks.

[0090] When a part or all of each processing unit of the plant control device is configured by hardware, a field programmable gate array (FPGA) or an application specific integrated circuit (ASIC) may be used. [Explanation of symbols]

[0091] 11...shape detection pre-processing unit, 12...pattern recognition unit, 13...control calculation unit, 14...shape detector, 50...control device, 100...plant control device (computer), 101...calculator, 102...third control unit, 103...high-speed control end, 104...low-speed control end, 105...safe operation range determination unit, 110...control unit, 111...first control unit, 112...second control unit, 190...controlled plant, 201...calculator, 202...mechanical control end position suppression control unit, 203...mechanical control end, 204...coolant control end, 205...mechanical control end safe operation range determination unit, 210...control unit, 211...first shape suppression control unit, 212...second shape control unit, 300...rolled material, 301...rolling mill, 302...entry tension reel (entry TR), 303...exit tension reel (exit TR), 304...mill speed control unit, 305...entry TR control unit, 306...exit tension reel control unit, 307...roll gap control unit, 308...entry tension meter, 309...exit tension meter, 310...rolling speed setting unit, 311...entry tension setting unit, 312...exit tension setting unit, 313...entry tension control unit, 314...exit tension control unit, 315...entry tension current conversion unit, 316...exit tension current conversion unit, 317...exit plate thickness meter, 318...exit plate thickness Control Unit, 319...roll gap setting unit, 401...work roll, 402...first intermediate roll, 403...second intermediate roll, 404...AS-U roll, 405...divided roll, 406...saddle, 501...input data creation unit, 502...neural network, 503...neural network learning control unit, 504...neural network selection unit, 505...teaching data creation unit, 506...operation abnormality judgment unit, 511...learning data database, 512...control rule database, 610...mechanical operation end position abnormality area judgment unit, 611...mechanical operation end position abnormality area search unit, 612...input data creation unit, 613...output data judgment unit, 620...mechanical operation end position abnormality suppression control unit, 621...coolant operation end control output calculation unit, 622...cooler coolant control end control output selection unit, 623...coolant control rule database, 631...database search unit, 632...output synthesis unit, 901...calculator, 902...first shape control unit, 903...second shape control unit, 904...mechanical control end, 905...coolant control end, 990...rolling mill, d11...first state quantity target, d12...first state quantity, d13...second state quantity, d14...safe operation range, d21...target shape, d22...mechanical control end position result, d23...shape result, d24...rolling result, d25...mechanical control end safe operation range, d26...operation abnormality evaluation value, d27...actual position abnormality region control end judgment value, d28...abnormality suppression output, d29...shape control output, d30...coolant control output, d31...estimated position, d32...operation abnormality judgment value

Claims

1. A plant control device performs a first operation process for performing a mechanical shape operation process for changing the thickness of the rolled material using a mechanical structure based on an operation of a controlled plant which is a rolling mill that rolls the rolled material, and a second operation process for changing the thickness of the rolled material by changing an injection amount of coolant in the width direction based on the operation, the second operation process having a slower response speed to an operation than the first operation process, A first control unit that acquires a state quantity of a mechanical structure corresponding to a target plate thickness of the controlled plant and instructs the first operation process; A second control unit that acquires a state quantity of the coolant corresponding to a target plate thickness of the controlled plant and instructs the second operation process; a first operating end that executes the first operation process of the controlled plant in response to an instruction from the first control unit; a second operating end that executes the second operation process of the controlled plant in response to an instruction from the second control unit; a safe operation range determination unit that determines a safe operation range of the first operation process by the first operating end device based on a performance of the first operating end device; a third control unit that, when the first operation process by the first operating end is not within the safe operation range as determined by the safe operation range determination unit, corrects or changes an instruction of the second operation process by the second control unit based on the state amount of the coolant, and moves an actual position of the first operation process by the first operating end to an actual position of a machine structure where the occurrence of an operational abnormality is not predicted; Equipped with the safe operation range determination unit has a neural network that searches for and determines the safe operation range using learning data obtained based on past performance, The neural network is selected from a plurality of types prepared according to the specifications including the plate width and steel type of the rolled material, The search for the safe operation range by the neural network uses performance data of the controlled plant to recognize the occurrence of an operational abnormality, and uses the performance data at the time of the occurrence of the operational abnormality as training data to learn the relationship between the performance data and the operational abnormality, thereby determining the safe operation range using the training data, The performance data when the operational abnormality occurs includes detection data of meandering of the rolled material, which is a sign of plate breakage, and emergency stop data. Plant control device.

2. The actual data at the time of the occurrence of the operational abnormality was obtained by adding together the data on the detection of the meandering of the rolled material and the data on the emergency stop, with the degree of weighting of the data on the detection of the meandering of the rolled material being increased. The plant control device according to claim 1 .

3. A plant control method in which a calculation processing unit performs, by calculation processing, a first operation processing for performing a mechanical shape operation processing for changing the thickness of the rolled material by a mechanical structure based on an operation of a controlled plant which is a rolling mill that rolls a rolled material, and a second operation processing for changing the thickness of the rolled material by changing an injection amount of coolant in the plate width direction based on the operation, wherein the second operation processing has a slower response speed to the operation than the first operation processing, A first control procedure in which the calculation processing unit acquires a state quantity of the mechanical structure corresponding to a target plate thickness of the controlled plant and instructs the first operation processing; A second control procedure in which the calculation processing unit acquires a state quantity of the coolant corresponding to a target plate thickness of the controlled plant and issues an instruction for the second operation process; a first operation execution procedure in which the arithmetic processing unit executes the first operation process of the controlled plant in response to an instruction from the first control procedure; a second operation execution procedure in which the arithmetic processing unit executes the second operation process of the controlled plant in response to an instruction from the second control procedure; a safe operation range determination procedure in which the arithmetic processing unit determines a safe operation range of the first operation processing by the first operation execution procedure based on a performance record of the first operation processing; a third control procedure in which, when it is determined by the safe operation range determination procedure that the first operation processing by the first operation execution procedure is not within the safe operation range, the calculation processing unit corrects or changes an instruction of the second operation processing based on the state quantity of the coolant by the second control procedure, and moves an actual position of the first operation processing by the first operation execution procedure to an actual position of the machine structure where the occurrence of an operational abnormality is not predicted, the safe operating range determination step includes a neural network that searches for and determines the safe operating range using learning data obtained based on past performance, The neural network is selected from a plurality of types prepared according to the specifications including the plate width and steel type of the rolled material, The search for the safe operation range by the neural network uses performance data of the controlled plant to recognize the occurrence of an operational abnormality, and uses the performance data at the time of the occurrence of the operational abnormality as training data to learn the relationship between the performance data and the operational abnormality, thereby determining the safe operation range using the training data, The performance data when the operational abnormality occurs includes detection data of meandering of the rolled material, which is a sign of plate breakage, and emergency stop data. Plant control methods.

4. A program for causing a computer to execute a first operation process for performing a mechanical shape operation process for changing the thickness of the rolled material using a mechanical structure based on an operation of a controlled plant, the controlled plant being a rolling mill that rolls the rolled material, and a second operation process for changing the thickness of the rolled material by changing the amount of coolant sprayed in the width direction based on the operation, the second operation process being a process that has a slower response speed to the operation than the first operation process, A state quantity of the mechanical structure corresponding to a target plate thickness of the controlled plant is acquired, a first control procedure for instructing the first operation process; A second control procedure of acquiring a state quantity of the coolant corresponding to a target plate thickness of the controlled plant and instructing the second operation process; a first operation execution procedure for executing the first operation process of the controlled plant in response to an instruction from the first control procedure; a second operation execution procedure for executing the second operation process of the controlled plant in response to an instruction from the second control procedure; a safe operation range determination procedure for determining a safe operation range of the first operation processing by the first operation execution procedure based on a performance record of the first operation processing; a third control procedure for correcting or changing an instruction of the second operation process based on the state quantity of the coolant by the second control procedure when the first operation process by the first operation execution procedure is not within the safe operation range as determined by the safe operation range determination procedure, to move an actual position of the first operation process by the first operation execution procedure to an actual position of the machine structure where occurrence of an operational abnormality is not predicted; The method causes the computer to execute the following: the safe operating range determination step includes a neural network that searches for and determines the safe operating range using learning data obtained based on past performance, The neural network is selected from a plurality of types prepared according to the specifications including the plate width and steel type of the rolled material, The search for the safe operation range by the neural network uses performance data of the controlled plant to recognize the occurrence of an operational abnormality, and uses the performance data at the time of the occurrence of the operational abnormality as training data to learn the relationship between the performance data and the operational abnormality, thereby determining the safe operation range using the training data, The performance data when the operational abnormality occurs includes detection data of meandering of the rolled material, which is a sign of plate breakage, and emergency stop data. program.

Images