JP2007112023A - Sheet state data estimation apparatus of corrugator and sheet state data estimation method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce the cost or maintenance load of a sheet state data estimation apparatus of a corrugator to precisely grasp the same, in the sheet state data estimation apparatus of the corrugator and the sheet state data estimation method of the corrugator. <P>SOLUTION: In the corrugator 1 for manufacturing corrugated cardboard sheets 7 and 8 from paper sheets 3, 4 and 6, the sheet state data estimation apparatus is constituted so as to use the specific state data (temperature) among a plurality of state data (the temperature, moisture, warpage and adhesion failure of sheets) related to the running paper sheets 3, 4 and 6 or the corrugated cardboard sheets 7 and 8 to set other data (moisture, warpage and adhesion failure) as estimate targets to estimate them. Sensors 31a-31k and 31m for detecting the specific state data (temperature) and estimate models 25a-25c being the correlation of the specific state data with the state data of the estimate targets are used to estimate the state data of the estimate targets from the data detected by the sensors. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an apparatus and method for estimating state information such as sheet temperature, sheet moisture, sheet warpage, and sheet adhesion failure in a corrugator relating to a pre-manufactured paper sheet and a corrugated cardboard sheet during or after production.

In a corrugator (corrugated cardboard sheet manufacturing apparatus) that manufactures various types of cardboard such as single-sided cardboard and double-sided cardboard, it is required that the manufactured cardboard does not warp or have poor adhesion. The warpage of the cardboard depends on the temperature and moisture state in each manufacturing process of the sheet constituting the cardboard, and the poor adhesion of the cardboard mainly depends on the temperature in the sheet bonding process.
For this reason, technology has been developed to directly detect sheet warpage and adhesion failure in each manufacturing process, or directly detect the temperature state and moisture state in each manufacturing process that affects these product defects, Used in the manufacture of cardboard.

  As a technique for detecting the seat temperature on the corrugator, a technique for actually measuring the surface temperature of the traveling seat in a non-contact manner using a radiation thermometer or the like has been proposed. For example, in Patent Document 1, non-contact type thermometers that detect the temperature of the back liner paper, the front liner paper, and the back liner paper of the single-sided cardboard sheet are installed at the heater roll outlet and the double facer inlet, respectively. (Refer to paragraph 0031 of Patent Document 1 and FIGS. 1-3).

Further, as a technique for detecting sheet moisture in a corrugator, a technique for actually measuring the surface moisture of a traveling sheet in a non-contact manner using an infrared moisture meter or the like has been proposed. For example, Patent Document 1 describes that a non-contact type moisture meter that detects the moisture content of the preheated back liner paper is installed at the entrance of the single facer (paragraph 0031, Patent Document 1). See FIG. 1-3).
Furthermore, with regard to the warpage of the sheet, there is no technique for directly measuring on the corrugator, and in general, the degree of visual check by the operator immediately after the manufacture (adhesion) of the sheet is present, but in Patent Document 2, there is a corrugator. The amount of vertical displacement of the corrugated sheets stacked on the stacker after manufacturing is detected along the flow direction of the corrugated sheets by a displacement sensor (or CCD camera), and the flow direction of the corrugated sheets is warped based on this detection information. A technology for calculating the amount and a technology for detecting the vertical displacement amounts of the four corners of the corrugated cardboard manufactured by the corrugator and calculating the twist warpage amount of the corrugated cardboard based on this detection information have been proposed (Patent Document 2). (See Claims 1 and 2, paragraphs 0040-0050, FIGS. 1 and 5-8).

  Patent Document 1 describes that the warp height of a corrugated cardboard sheet depends on the moisture content difference between the front and back liner paper before bonding the front liner paper with a double facer (see paragraph 0020). In other words, if the difference in moisture content between the front and back liner papers at the time of bonding is not within the proper range, the amount of expansion and contraction from the steady moisture state at atmospheric humidity at which the moisture content of both liner papers is equal will be different between the front liner paper and the back liner paper. Both are pasted together in a state that is greatly different from each other, and the adhesion point is fixed. Therefore, if the moisture content of the back liner paper on the upper surface side exceeds the moisture content of the front liner paper beyond the appropriate range, the moisture content of the back liner paper is greatly reduced until the steady moisture content is reached. If the amount of shrinkage due to the decrease in moisture content increases, warping occurs, and conversely, if the moisture content of the front liner paper on the lower surface side exceeds the moisture content of the back liner paper, the downward warping occurs. It happens.

Furthermore, a technique for measuring the sheet warpage before the stacker with a laser has been developed (see Patent Document 2 and Non-Patent Document 1).
In addition, regarding the sheet adhesion failure detection, a technique for directly measuring on the corrugator is not seen, and generally, an operator visually checks the sheet immediately after sheet manufacture (adhesion). Before the sheet bonding process, the speed sensor detects the running speed or running amount of the front liner paper and single-sided corrugated cardboard sheet, finds the difference between the two, compares it with the reference value, and the difference exceeds the reference value. There has been proposed a technique for detecting a bonding failure of a corrugated cardboard sheet by detecting the fact (see claim 1 of Patent Document 3).

In addition, one using matrix control for corrugator control has been proposed (see Patent Document 4).
That is, as shown in FIG. 11, the corrugator 1 according to this technique includes an upstream (back) liner preheater 11, a core preheater 12, a single facer 13, a single stage preheater 14, a glue machine 15, and a lower side. (Table) A liner preheater 16, a double facer 17, a slitter scorer 18, a cut-off 19 and a stacker 20 are provided.

  The upper (back) liner 3 is introduced into the upper liner preheater 11 and heated, and the core 4 is introduced into the core preheater 12 and heated, and the heated upper liner 3 and the heated core 4 are single facers. The single-stage sheet (single-sided corrugated cardboard sheet) 5 is formed by being introduced into 13 and glued. Further, the single-stage sheet 5 is introduced into the single-stage preheater 14 and heated, and then introduced into the glue machine 15 to be glued. In addition, the lower (front) liner 6 is introduced into the lower liner preheater 16 and heated, and the heated single-stage sheet 5 and the lower liner 6 are introduced into the double facer 17 and pressed and glued to form a double-sided sheet (both sides) Corrugated cardboard sheet) 7 is formed. Thereafter, the double-sided sheet 7 is cut (cut) and ruled along the sheet conveying direction by a slitter scorer 18, and further cut (cut) to a predetermined length in the sheet conveying direction by a cut-off 19. A sheet (one double-sided sheet) 8 is stacked on the stacker 20. Thereafter, the single-sided double-sided sheet 8 is unloaded from the stacker 20 as a final product.

The production management device 202 that controls the corrugator 1 has a control for calculating control amounts such as sheet heating, glue gap amount, and pressing force from the corrugator production state information, the corrugator machine state information, and the sheet state information. An amount calculation unit 221 and a process controller 222 for controlling the sheet heating, the glue gap amount, and the pressing force are respectively provided. The control amount calculation unit 221 and the process controller 222 provide a sheet based on the production state information and the operation state information. Matrix control is performed in accordance with a matrix condition table that is set so as not to cause warpage or poor sheet adhesion.
Japanese Patent No. 3629591 JP 2003-266666 A JP-A-10-29253 Japanese Patent Laid-Open No. 2003-231193 International Paper Board Industry March 2005, pp. 84-88

However, the above-described conventional technology has the following problems.
(1) About sheet temperature detection in a corrugator (a) A sensor (non-contact type thermometer) is expensive.
(B) Calibration by paper type is required (particularly, in the case of color print paper, the detected value may not be matched with the actual sheet temperature even if it is calibrated).
(C) Since it is weak at high temperature (50 ° C. or higher), cooling such as air purge is required.
(D) Since paper dust adheres to the lens surface, it is necessary to remove the paper dust by air purge or the like.
(E) Regular maintenance is required, and a maintenance burden is applied.
(F) Since it is a hard product, it has a lifetime and may be deteriorated or damaged.

(2) Sheet moisture detection in a corrugator (a) The sensor (non-contact type moisture meter) is expensive (more expensive than a thermometer).
(B) Calibration by paper type is required (particularly, in the case of color print paper, the detected value may not be matched with the actual sheet temperature even if it is calibrated).
(C) Since it is weak at high temperature (50 ° C. or higher), cooling such as air purge is required.
(D) Since paper dust or the like adheres to the lens surface, it is necessary to remove the paper dust or the like by air purge or the like. (E) Periodic maintenance is required and a maintenance burden is applied.
(F) Since it is a hard product, it has a lifetime and may be deteriorated or damaged.

(3) Sheet warpage detection in the corrugator (a) Since the amount of warpage of a single sheet in the corrugator cannot be quantitatively grasped, for example, the correspondence with the temperature and moisture content of the sheet on the corrugator is clear In addition, it is difficult to appropriately adjust the temperature and moisture content of the sheet for correcting warpage. For this reason, warpage of a single sheet (a sheet obtained by cutting a sheet on the web in the flow direction) cannot be sufficiently suppressed.
(B) Since this is not a technique for grasping the amount of warpage over time after sheet stacking, adjusting the sheet temperature and water content on the corrugator to suppress warpage that occurs over time after sheet stacking is not possible. Have difficulty. For this reason, it is not possible to sufficiently suppress warping over time.
(C) When a measurement system using a laser or a CCD camera is used, periodic maintenance is required, and a maintenance burden is applied.

(4) Detection of sheet adhesion failure in corrugator (a) In Patent Document 3, the speed difference between the upper and lower liners is detected by a speed sensor to detect a cardboard sheet adhesion failure. It is difficult to detect defects with a sufficient system.
(B) The maintenance of the speed sensor is necessary, and a maintenance burden is applied.

  The present invention has been devised in view of such problems, and provides a corrugator sheet state information estimation apparatus and method that can reduce the cost and maintenance burden of the apparatus and can be accurately grasped. For the purpose.

  In order to achieve the above object, a corrugator sheet state information estimation apparatus (Claim 1) according to the present invention is a corrugator that manufactures a corrugated cardboard sheet from a paper sheet. 1 is a sheet state information estimation device for estimating other information from the specific state information as an estimation target, the detection means (sensor) for detecting the specific state information, the specific state information, and the estimation target And an estimation unit that estimates the state information of the estimation target from the specific state information detected by the detection unit using a prediction model that is a correlation with the state information.

The specific state information is a sheet temperature measurement value of the paper sheet or the cardboard sheet, the detection means is a temperature sensor, and the state information of the estimation target is the sheet moisture of the paper sheet or the cardboard sheet, It is preferable that the information relates to either sheet warpage or sheet adhesion failure.
Alternatively, the specific state information is a sheet moisture of the paper sheet or the cardboard sheet, the detection means is a moisture sensor, and the state information of the estimation target is a sheet warp of the paper sheet or the cardboard sheet, It is preferable that the information is related to any sheet adhesion failure.

Alternatively, the specific state information is a sheet temperature measurement value and a sheet moisture measurement value of the paper sheet or the corrugated cardboard sheet, the detection means is a temperature sensor and a moisture sensor, and the estimation target state information is the It is preferable that the information relates to either a sheet warpage or a sheet adhesion failure of the paper sheet or the corrugated cardboard sheet.
In addition, it is preferable that the information on the sheet warp includes information on the warpage of the corrugated sheets stacked as a single sheet on the stacker at the downstream end of the corrugator (Claim 5).

Further, it is preferable that storage means (memory) for storing the prediction model is further provided, and the estimation means calls the prediction model from the storage means to estimate the state information of the estimation target.
In addition, it is preferable that a knowledge database storing the prediction model under various conditions is further provided, and the estimation unit estimates the state information of the estimation target by calling the predetermined prediction model from the knowledge database. Item 7).

Alternatively, it is preferable to provide a neural network that constitutes the prediction model, and the estimation means estimates the state information of the estimation target using the neural network.
Or, further comprising a knowledge database storing the prediction model under various conditions, and further comprising a neural network constituting the prediction model, wherein the estimating means includes the predetermined prediction model called from the knowledge database and the neural network. It is preferable that the state information of the estimation target is estimated using (Claim 9).

The neural network is preferably a hierarchical neural network.
The corrugator sheet state information estimation method according to the present invention (Claim 11) is a corrugator that manufactures a corrugated cardboard sheet from a paper sheet. From the specific state information among a plurality of state information on the paper sheet or the corrugated cardboard sheet that travels. A sheet state information estimation method for estimating other information as an estimation target, wherein a prediction model that is a correlation between the specific state information and the state information of the estimation target is prepared in advance, and the specific state information is It is characterized by comprising a detection step of detecting and an estimation step of estimating the state information of the estimation object from the specific state information detected in the detection step using the prediction model.

  According to the corrugator sheet state information estimation apparatus (Claim 1) and method (Claim 11) of the present invention, a prediction model that is a correlation between specific state information relating to a paper sheet or cardboard sheet and state information to be estimated. Is used to estimate the state information of the estimation target from the detected specific state information, so that it is possible to estimate without providing a detection unit for the state information of the estimation target only by providing a detection unit for the specific state information. The status information of the target can be grasped.

  For example, by preparing a prediction model for the correlation between the sheet temperature on the corrugator and any of sheet moisture, sheet warpage, and sheet adhesion failure, the sheet temperature on the corrugator as specific state information is detected, By using a prediction model to estimate any of sheet moisture, sheet warpage, and sheet adhesion failure from the sheet temperature, the status of sheet moisture, sheet warpage, and sheet adhesion failure can be ascertained without providing each detection means. This can reduce the cost of the apparatus. By enhancing the sheet temperature detection means and creating a prediction model with high accuracy, various status information related to paper sheets or cardboard sheets can be accurately grasped while reducing the equipment cost and maintenance burden. (Claim 2).

  Also, prepare a prediction model for the correlation between the sheet moisture on the corrugator and any of sheet warp or sheet adhesion failure, detect the sheet moisture on the corrugator as specific state information, By using the sheet moisture to estimate either sheet warpage or sheet adhesion failure, it is possible to grasp the state of the sheet warpage or sheet adhesion failure without providing each detection means, thereby reducing the cost of the apparatus. This makes it possible to suppress various types of state information related to the paper sheet or the corrugated cardboard sheet while accurately reducing the apparatus cost and the maintenance burden (claim 3).

  Also, prepare a prediction model for the correlation between the sheet temperature and sheet moisture on the corrugator and any of sheet warp and sheet adhesion failure, and the sheet temperature and sheet moisture on the corrugator as specific status information By detecting and estimating either sheet warpage or sheet adhesion failure from sheet moisture using a prediction model, it is possible to grasp the state of sheet warpage or sheet adhesion failure without providing each detection means. The cost of the apparatus can be suppressed, and various state information relating to the paper sheet or cardboard sheet can be accurately grasped while reducing the apparatus cost and the maintenance burden as a whole. ).

As information regarding sheet warpage, if it is possible to estimate information on the stacking warpage of the corrugated sheets stacked as a single sheet on the stacker at the downstream end of the corrugator, it is less likely to cause warpage of stacking and higher quality. Corrugated cardboard sheets can be produced (claim 5).
In addition, if a knowledge database storing prediction models under various conditions is prepared and state information to be estimated is estimated by calling a predetermined prediction model from the knowledge database, a detection unit for specific state information is provided. Thus, it is possible to easily grasp the state information of the estimation target even under various conditions (claims 7 and 9).

  In addition, by configuring the prediction model with a neural network (especially a hierarchical neural network), the prediction model can be updated by learning and a more accurate prediction model can be constructed, which contributes to improved estimation accuracy. (Claims 8, 9, 10).

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[A. First Embodiment]
1 to 5 show a corrugator and a seat state information estimation device according to a first embodiment of the present invention. FIG. 1 is a schematic configuration diagram thereof, FIGS. 2 to 4 are detailed configuration diagrams thereof, and FIG. It is a perspective view which shows the curvature simple model.

  As shown in FIG. 1, the corrugator 1 according to the present embodiment includes an upstream (back) liner preheater 11, a core preheater 12, a single facer 13, a single stage preheater (single sheet preheater) 14, A glue machine 15, a lower (table) liner preheater 16, a double facer 17, a slitter scorer 18, a cut-off 19, and a stacker 20 are provided.

  The upper (back) liner 3 is introduced into the upper liner preheater 11 and heated, and the core 4 is introduced into the core preheater 12 and heated, and the heated upper liner 3 and the heated core 4 are single facers. The single-stage sheet (single-sided corrugated cardboard sheet) 5 is formed by being introduced into 13 and glued. Further, the single-stage sheet 5 is introduced into the single-stage preheater 14 and heated, and then introduced into the glue machine 15 to be glued. Further, the lower (front) liner 6 is introduced into the lower liner pre-heater 16 and heated, and the heated single-stage sheet 5 and the lower liner 6 are introduced into the double facer 17 and pressurized and heated to be bonded to each other. (Double-sided cardboard sheet) 7 is formed. Thereafter, the double-sided sheet 7 is cut (cut) and ruled along the sheet conveying direction by a slitter scorer 18, and further cut (cut) to a predetermined length in the sheet conveying direction by a cut-off 19. A sheet (one double-sided sheet) 8 is stacked on the stacker 20. Thereafter, the single-sided double-sided sheet 8 is unloaded from the stacker 20 as a final product.

  Here, the detailed configurations of the upper liner preheater 11, the core preheater 12, the single-stage preheater 14, the lower liner preheater 16, the single facer 13, the glue machine 15, and the double facer 17 will be described with reference to FIGS. I will explain. 2 is a schematic diagram showing the configuration of the upper liner preheater 11, the core preheater 12, and the single facer 13. FIG. 3 shows the one-stage preheater 14, the lower liner preheater 16, the glue machine 15, and the double facer 17. FIG. 4 is a schematic diagram showing a configuration of a part of the double facer 17.

  As shown in FIG. 2, the lower liner preheater 11 includes back liner heating rolls 111 </ b> A and 111 </ b> B arranged vertically in two stages. The back liner heating rolls 111A and 111B are heated to a predetermined temperature by supplying steam therein. The back liner 3 guided in order by the guide rolls 115, 114A, 116, and 114B is wound around the peripheral surfaces of the back liner heating rolls 111A and 111B, and preheated by the back liner heating rolls 111A and 111B.

  Of the guide rolls 115, 114A, 116, and 114B, the guide roll 114A provided in the vicinity of one back liner heating roll 111A is the tip of an arm 113A that is swingably attached to the shaft of the back liner heating roll 111A. The guide roll 114B provided in the vicinity of the other back liner heating roll 111B is supported at the tip of an arm 113B that is swingably attached to the shaft of the back liner heating roll 111B. Each arm 113A, 113B can be moved to an arbitrary position within an angle range indicated by an arrow in the drawing by a motor (not shown). Here, the guide roll 114A, the arm 113A, and a motor (not shown), and the guide roll 114B, the arm 113B, and a motor (not shown) constitute the winding amount adjusting devices 112A and 112B, respectively.

  With such a configuration, in the back liner preheater 11, the vapor pressure supplied to the back liner heating rolls 111A and 111B and the winding amount of the back liner 3 on the back liner heating rolls 111A and 111B by the winding amount adjusting devices 112A and 112B. The water content of the back liner 3 can be adjusted by (wrapping angle). Specifically, the greater the vapor pressure and the greater the amount of wrapping, the greater the amount of heat applied from the back liner heating rolls 111A and 111B to the back liner 3, and the drying of the back liner 3 proceeds and the water content is increased. Will drop.

  The single facer 13 includes a pressure belt 133 wound around a belt roll 131 and a tension roll 132, and an upper roll 134 that has a wave-like surface and is in contact with the pressure belt 133 in a pressurized state. A lower roll 135 having a wave-like surface and meshing with the upper roll 134 is provided. The back liner 3 heated by the back liner preheater 11 is wound around the liner preheating roll 137 and preheated on the way, and then guided by the belt roll 131 and together with the pressure belt 133 and the upper belt 133. It is transferred to the nip portion with the roll 134.

On the other hand, after the core 4 heated by the core preheater 12 is wound around the core preheating roll 138 and preheated, and is wound around the meshing portion of the upper roll 134 and the lower roll 135. Then, it is guided by the upper roll 134 and transferred to the nip portion between the pressure belt 133 and the upper roll 134.
A gluing device 136 is disposed in the vicinity of the upper roll 134. This gluing device 136 adjusts the amount of glue 30 attached to the peripheral surface of the gluing roll 136b, the gluing roll 136b that glues the gluing dam 136a storing the gluing 30, the middle core 4 conveyed by the upper roll 134, and the gluing roll 136b. A meter roll 136c and a glue scraping blade 136d for scraping glue from the meter roll 136c are configured. The core 4 wound around the meshing part of the upper roll 134 and the lower roll 135 is glued to each top of the stage by the gluing roll 136b, and is attached to the back liner 3 at the nip portion between the pressure belt 133 and the upper roll 134. It is pasted together. Thereby, the single-stage sheet | seat 5 is formed.

  With such a configuration, in the single facer 13, the moisture content of the back liner 3 is controlled by the gap amount between the gluing roll 136b and the upper roll 134 and the gap amount between the gluing roll 136b and the meter roll 136c. It can be adjusted. Specifically, as the gap amount is larger, the amount of glue on the bonding surface between the core 4 and the back liner 3 is increased, and the moisture content of the back liner 3 is increased by the moisture contained in the glue. Each of the gap amounts can be adjusted by moving the gluing roll 136b or moving the meter roll 136c.

  The core preheater 12 has the same configuration as the back liner preheater 11, and the core heating roll 121 heated to a predetermined temperature by supplying steam therein, and the core heating roll 121 A winding amount adjusting device 122 that adjusts the winding amount (winding angle) of the core 4 is provided. The winding amount adjusting device 122 includes a guide roll 124 around which the core 4 is wound, an arm 123 that is swingably attached to the shaft of the core heating roll 121 and supports the guide roll 124, and rotates the arm 123. It is composed of a motor that does not.

  The single-stage sheet preheater 14 and the front liner preheater 16 are vertically arranged in two stages as shown in FIG. These preheaters 14 and 16 have the same configuration as the above-described back liner preheater 11. The single-stage sheet preheater 14 includes a single-stage sheet heating roll 141 and a winding amount adjusting device 142. The single-stage sheet heating roll 141 is heated to a predetermined temperature by supplying steam therein. Around the circumferential surface of the single-stage sheet heating roll 141, the back liner 4 side of the single-stage sheet 5 guided in order by the guide rolls 145 and 144 is wound and preheated by the single-stage sheet heating roll 141.

  The winding amount adjusting device 142 includes one guide roll 144, an arm 143 that is swingably attached to the shaft of the one-stage sheet heating roll 141, and supports the guide roll 144, and a motor (not shown) that rotates the arm 143. Has been. Then, the guide roll 144 is moved to an arbitrary position within the angle range indicated by the arrow in the drawing by the control of the motor so that the winding amount (winding angle) of the single-stage sheet 5 around the single-stage sheet heating roll 141 can be adjusted. It has become.

  With such a configuration, in the preheater 14 for a single-stage sheet, the moisture content of the back liner 4 depends on the vapor pressure supplied to the single-stage sheet heating roll 141 and the winding amount (winding angle) of the single-stage sheet 5 around the single-stage sheet heating roll 141. The amount can be adjusted. Specifically, the greater the vapor pressure and the greater the amount of wrapping, the greater the amount of heat applied from the single-stage sheet heating roll 141 to the back liner 4, and the drying of the back liner 4 proceeds and the water content decreases. Will do.

  The front liner preheater 16 includes a front liner heating roll 161 and a winding amount adjusting device 162. The front liner heating roll 161 is heated to a predetermined temperature by supplying steam therein. A front liner 6 guided in order by guide rolls 165 and 164 is wound around the peripheral surface of the front liner heating roll 161 and preheated by the front liner heating roll 161.

  The winding amount adjusting device 162 includes one guide roll 164, an arm 163 that is swingably attached to the shaft of the front liner heating roll 161 and supports the guide roll 164, and a motor (not shown) that rotates the arm 163. Has been. Then, the guide roll 164 is moved to an arbitrary position within the angle range indicated by the arrow in the drawing by the control of the motor so that the winding amount (winding angle) of the front liner 6 around the front liner heating roll 161 can be adjusted. It has become.

  With such a configuration, in the front liner preheater 16, the moisture content of the front liner 6 depends on the vapor pressure supplied to the front liner heating roll 161 and the winding amount (winding angle) of the front liner 6 around the front liner heating roll 161. The amount can be adjusted. Specifically, the greater the vapor pressure and the greater the amount of wrapping, the greater the amount of heat applied from the surface liner heating roll 161 to the surface liner 6, and the drying of the surface liner 6 proceeds and the water content decreases. Will do.

  The glue machine 15 includes a gluing device 151 and a pressure bar device 152. The single-stage sheet 5 heated by the single-stage sheet preheater 14 is preheated by the single-stage preheating roll 155 and then guided in order in the glue machine 15 by the guide rolls 153 and 154. The gluing device 151 is disposed on the lower side (the center core 4 side) of the single-stage sheet 5 between the guide rolls 153 and 154, and the pressure bar device 152 is disposed on the upper side of the traveling line (the back liner 3 side). Has been.

  The gluing device 151 includes a gluing dam 151a in which the gluing 31 is stored, a gluing roll 151b disposed in the vicinity of the traveling line of the single-stage sheet 5, and a doctor roll 151c that contacts the gluing roll 151b and rotates reversely with the gluing roll 151b. Has been. On the other hand, the pressure bar device 152 includes a pressure bar 152a disposed so as to sandwich the single-stage sheet 5 between the pressure bar 151a and an actuator 152b that presses the pressure bar 152a toward the glue roll 151b. Yes. The single-stage sheet 5 is pressed against the gluing roll 151b by the pressure bar 152a, and when passing between the pressure bar 152a and the gluing roll 151b, the gluing roll 151b glues the top of each step of the core 4. It has come to be. The single-stage sheet 5 glued to the middle core 4 is bonded to the front liner 6 by a double facer 17 in the next process.

  With such a configuration, in the glue machine 15, the moisture content of the front liner 6 is determined by the gap amount between the gluing roll 151b and the pressure bar 152a (that is, the gap amount of the gluing roll 151b with respect to the travel line of the single-stage sheet 4). It can be adjusted. Specifically, as the gap amount increases, the amount of glue on the bonding surface between the core 4 and the front liner 6 increases, whereby the amount of water applied to the front liner 6 increases and the water content of the front liner 6 increases. Will increase. The gap amount can be adjusted by adjusting the position of the pressure bar 152a by the actuator 152b.

  The single-stage sheet 5 glued by the glue machine 15 is transferred to the double facer 17 of the next process. The front liner 6 heated by the front liner preheater 16 is also transferred to the double facer 17 through the glue machine 15. At that time, the front liner 6 is preheated from the liner preheating roll 156 while being guided by the liner preheating roll 156 disposed in the glue machine 15.

  At the entrance of the double facer 17, a first shower device (back liner wetting device) 171 </ b> A as a wetting device or a moisture applying device is disposed on the back liner 3 side along the traveling line of the single-stage sheet 5, and on the front liner 6 side. A second shower device (front liner wetting device) 171B as a wetting device or a moisture applying device is disposed. These shower devices 171A and 171B are for adjusting the water content of the back liner 3 and the front liner 6, and are directed from the shower device 171A toward the back liner 3, and from the shower device 171B toward the front liner 6. Water is sprayed. And the moisture content of the back liner 3 increases according to the shower amount from the shower device 171A, and the moisture content of the front liner 6 increases according to the shower amount from the shower device 171B. The shower devices 171A and 171B are controlled independently of each other.

In this embodiment, a moisture applying device for the upper and lower liners is provided before the double facer as described above. However, this moisture providing device is usually provided as an option and is not essential. Further, the moisture applying device may be equipped and controlled for the upper and lower liners after the double facer.
As shown in FIG. 4, the double facer 17 is divided into an upstream heating section 17 </ b> A and a downstream cooling section 17 </ b> B along the travel lines of the single-stage seat 5 and the front liner 6. Among these, a plurality of heating plates 172 are arranged in the heating section 17 </ b> A, and the front liner 6 passes over these heating plates 172. The hot platen 172 is heated to a predetermined temperature by supplying steam therein.

  In addition, a loop-shaped pressure belt 173 travels on the heating plate 172 in synchronism with the single-stage sheet 5 and the front liner 6 across the travel line. The pressure unit 174 is disposed so as to face the hot platen 172. The pressure unit 174 includes a pressure bar 174 a that is in sliding contact with the back surface of the pressure belt 173, and an actuator 174 b that presses the pressure bar 174 a against the heating plate 172 side.

  The single-stage sheet 5 glued by the glue machine 15 is carried between the pressure belt 173 and the heating plate 172 from the pressure belt 173 side. On the other hand, the front liner 6 heated by the front liner preheater 16 is preheated by the liner inlet preheating roll 175 and then carried between the pressure belt 173 and the hot platen 172 from the hot platen 172 side. . Then, after the single-stage sheet 5 and the front liner 6 are carried between the pressure belt 173 and the heating platen 172, respectively, the single-stage sheet 5 and the front liner 6 are integrally transferred to the cooling section 17B in a state where they overlap each other. . During the transfer, the single-stage sheet 5 and the front liner 6 are heated from the front liner 6 side while being pressurized by the pressure unit 174 via the pressure belt 173, thereby being bonded to each other to form the cardboard sheet 7. The entire width or end of the corrugated cardboard sheet 7 is cut by a rotary shear 176 provided at the outlet of the cooling section 17B, and is transferred to a slitter scorer 18 in the next process.

  With such a configuration, in the double facer 17, the moisture content of the front liner 6 can be adjusted by the vapor pressure supplied to the hot platen 172 and the pressurizing force of the pressurizing unit 174. Specifically, the greater the vapor pressure and the greater the applied pressure, the greater the amount of heat applied from the hot platen 172 to the front liner 6, leading to the drying of the front liner 6 and the reduced water content. become. The moisture content of the front liner 6 can also be adjusted by the speed at which the single-stage sheet 5 and the front liner 6 pass through the double facer 17. In this case, the slower the passing speed, the longer the time during which the front liner 6 is in contact with the hot platen 172, and thus the drying of the front liner 6 proceeds and the water content decreases.

  In such a corrugator 1, the heating state of the sheet by each preheater of the upper (back) liner preheater 11, the core preheater 12, the single-stage preheater 14, and the lower (front) liner preheater 16, and the single facer 13 The glue gap amount in the glue machine 15 and the pressurizing force in the double facer 17 are controlled through the production management device (controller) 2.

  The heating state of the sheet by each preheater can be controlled by adjusting the amount of sheet wound around each heating roll of each preheater and the vapor pressure applied to each heating roll. The glue gap amount is a gap amount between the glue roll and the sheet running line. As described above, the single facer 13 can be controlled by adjusting the gap amount between the glue roll and the upper roll. The glue machine 15 can be controlled by adjusting the gap amount between the gluing roll and the pressure bar.

Other elements to be controlled are the shower amount provided by the shower device that humidifies the back liner 3 and the front liner, which is disposed at the entrance or exit of the double facer 17, and the travel line of the front liner 6. A heating state (amount of steam supplied to the inside) of a hot platen that is arranged along the surface and heats the front liner 6 may be added.
The production management device 2 includes a control amount calculation unit 21 that calculates control amounts such as the sheet heating, the glue gap amount, and the pressing force from the production state information of the corrugator, the machine state information of the corrugator, and the sheet state information. The process controller 22 for controlling the sheet heating, the glue gap amount, and the pressing force, the optimum state storage unit 23 for storing the optimum state of the sheet heating, the glue gap amount, and the pressing force, and the sheet state information. A sheet state information estimating unit (sheet state information estimating apparatus of the present invention) 24A for acquiring and estimating is provided.

Note that the sheet state information estimation device 24A is also referred to as a soft sensor 24A because the sheet state information is acquired not by hardware (physical sensor) but by software (computer logic).
The production status information of the corrugator includes various data such as the paper width, flute, configuration (base paper configuration), basis weight (base paper basis weight) of the base paper (the material of the upper liner 3, the core 4, and the lower liner 6). Input from a higher-level production management system (not shown).

  Corrugator machine status (operating status) information includes operating speed (conveying speed of each sheet) and pre-heaters 11, 12 for the upper (back) liner, for the core, for one stage, and for the lower (front) liner. 14 and 16 includes various data such as the sheet winding amount, the gap amount of the single facer 13 and the glue machine 15, and the pressurizing force in the double facer 17, and the production management device 2 sets the control amount for these machine states. Set and control.

  The control amount calculation unit 21 acquires the above-described production state information from a production management system (not shown), acquires sheet state information such as sheet temperature, sheet moisture, sheet warpage, and sheet adhesion failure from the soft sensor 24A, and a process controller. 22 is used to acquire machine state information such as sheet heating, glue gap amount, and pressing force of the corrugator 1, and sheet heating and glue gap amount so as not to cause sheet warpage or sheet adhesion failure according to the obtained information. , Each control amount such as a pressing force is calculated, and the calculation result is output to the process controller 22 as a control command.

  Further, the process controller 22 is based on a control command from the control amount calculation unit 21, such as the preheaters 11, 12, 14, 16, the single facer 13, the gap adjusting element of the glue machine 15, and the pressure adjusting element of the double facer 17. Each control element is controlled. Then, the control amount calculation unit 21 and the process controller 22 perform matrix control according to the matrix condition table set so as not to cause sheet warpage, sheet adhesion failure, or the like based on such production state information and operation state information. It is like that.

Matrix control is control for selecting an operating condition described in a matrix condition table created in advance and adjusting the selected operating condition to be the selected operating condition. For example, Patent Document 4 describes the corrugator control. It is a known technique.
Here, a matrix condition table for the sheet state information and the operation state information under each of the above production states has been created. Under each production state (base paper composition, base paper basis weight, paper width, flute, etc.), the sheet Corresponding to the state, the operating state (operating speed, amount of sheet winding of each preheater 11, 12, 14, 16 and the like, each gap amount, applied pressure, etc.) is controlled.

  In the optimum state storage unit 23, various production state information and various operation state information when the optimum state of the sheet heating, the glue gap amount, and the pressing force are output from the soft sensor 24 </ b> A are used as the control amount calculation unit 21. When the optimum state is output, the control amount calculation unit 21 compares the estimated values of the sheet heating, the glue gap amount, and the pressing force with the reference value from the soft sensor 24A. Then, if all the estimated values are within the reference value, it is determined that the optimum state is obtained, and various production state information and various operation state information at that time are output to the optimum state storage unit 23. It has become. In place of this, or in addition to this, an OK button (not shown) operated by the operator is provided, and the operator has the optimum sheet heating, glue gap amount, and pressure, and there is no warpage or poor adhesion. By pressing this OK button when confirming the above, various production state information and various operation state information at this time may be input from the control amount calculation unit 21 to the optimum state storage unit 23.

  Further, as described above, the process controller 22 always keeps track of each production status information. For example, when the production status is changed by switching the order of the cardboard sheets, the current production status and the production status match. [Here, the paper width, the flute, the base paper composition, and the base paper basis weight match each other (including not only completely matching but also substantially matching)] Search the optimum operating state storage unit 23 for the data set, If a desired data set is found, the operation state information of this data set is read as the optimum operation state information, and the corresponding control elements are controlled so as to be in the optimum operation state. Since it can be seen that the optimum operation state information is taught from the optimum operation state storage unit 23, this control is also referred to as teaching control. On the other hand, if the optimum operation state information corresponding to the current production state is not found in the optimum operation state storage unit 23, the process controller 22 performs normal matrix control.

  By the way, in the soft sensor (sheet state information estimation device) 24A according to the present embodiment, as shown in FIG. 1, the detection information of the sheet temperature of each part of the corrugator 1 (including the outlet part of the stacker 18) is installed in each part. Obtained from the temperature sensors (non-contact type sensors) 31a to 31m (reference numeral 31 when individual sensors are not distinguished), the sheet moisture of each part is predicted from these sheet temperatures, and the predicted sheet moisture of each part Therefore, the sheet warpage is predicted from the sheet temperature, and the sheet adhesion failure of each part is predicted from the sheet temperature of each part. In particular, these predictions are performed by taking a prediction model constructed in advance into the memory 25 and using it.

  The temperature sensor 31a includes an upstream portion and a downstream portion of the upper liner preheater 11, an upstream portion and a downstream portion of the core preheater 12, a downstream portion of the single facer 13, an upstream portion and a downstream portion of the single stage preheater 14, A downstream portion of the glue machine 15, an upstream portion and a downstream portion of the lower liner preheater 16, a downstream portion of the double facer 17, and a downstream portion of the stacker 20 are provided.

Here, the prediction model will be described.
First, the principle of predicting sheet moisture from sheet temperature will be described. When the sheet temperature of each part where the temperature sensors 31a to 31m are installed is in the respective reference temperature state, the sheet moisture content of each part becomes the reference moisture content, but if the sheet moisture content of each part is different from the reference moisture content, The sheet temperature of each corresponding part is also different from the reference temperature. Further, the change in the sheet moisture content in each part appears not only as the sheet temperature itself in each part but also as a change in sheet temperature between the parts. Therefore, the sheet temperature of each part and / or the sheet moisture content of each part against the sheet temperature change between the parts can be obtained in advance using experiments or simulations, and the temperature range of each part assumed is sufficiently In order to cover, the correspondence of the sheet moisture content of each part to the sheet temperature of each part and / or the sheet temperature change between the parts is obtained, and this is set in the prediction model 25a.

Next, the principle of predicting the sheet warp from the sheet moisture will be described. The sheet expands or contracts according to the sheet moisture amount and the applied tension, and according to the difference between the extension or contraction between the front liner 6 and the back liner 1. Warping occurs. As shown in FIG. 5, this warp differs between the MD direction and the CD direction when the traveling direction of the sheet is the MD direction and the width direction of the sheet is the CD direction.
For example, when a standard tension is applied to the sheet and the sheet moisture is changed by 1%, the MD direction elongation or shrinkage and the CD direction elongation or shrinkage due to moisture are as follows.
-MD direction moisture elongation or shrinkage: 0.03% / [1% moisture change]
・ Moisture elongation or shrinkage in CD direction: 0.1% / [1% moisture change]
When the sheet moisture is increased by 1%, the above numerical value is increased, and when the sheet moisture is decreased by 1%, the sheet is decreased by the above numerical value.

Moreover, the extension or shrinkage in the MD direction due to moisture is as follows.
・ MD direction tensile elongation or shrinkage: 0.07% / [tensile 1 kgf / cm change]
When the tension is increased by 1 kg, the above-mentioned numerical value is increased, and when the tension is decreased by 1 kg, the above-mentioned numerical value is decreased.
Thereby, it is possible to predict the expansion or contraction of the sheet in each direction when the sheet moisture and the sheet tension change from the steady state (sheet moisture: 8 to 10%, sheet tension: 0 to 0.2 kgf / cm). it can.

The warp factor of the sheet is the WF (warp factor) indicating the warp state in the MD direction and the CD direction according to the following formula, by obtaining the difference between the sheet expansion and contraction in the MD direction and the CD direction (expansion difference, distortion difference) of the upper and lower liners ).
WF = δ (610 / W) ² · (1/254)
^{= (610 2/254) ·} (ε / 8t)
However, δ = W ² ε / 8t, ε: upper and lower liner strain difference, t: corrugated board thickness, W: corrugated board width Therefore, if the sheet tension is a reference state, the sheet moisture state of each part, MD direction, and CD direction A correspondence relationship with the warped state can be obtained and set in the prediction model 25b.

In addition, the sheet adhesion failure is good if the temperature state at the bonding point is within an appropriate range, otherwise the sheet adhesion can be poor, and the correspondence between the temperature state of each part and the sheet adhesion failure A relationship can be obtained and set in the prediction model 25c.
In the present embodiment, such a prediction model 25a that associates sheet moisture with sheet temperature, a prediction model 25c that associates sheet adhesion failure with sheet temperature, and a prediction model 25b that associates sheet warp with sheet moisture. Are constructed using experiments or simulations according to various production states (for example, paper width, flute, base paper composition, and base paper basis weight), and are used as the knowledge database 26 as a database.

  The knowledge database 26 can be updated or augmented as necessary. That is, when it is determined that there is an error in the prediction result of the sheet moisture, sheet adhesion failure, and sheet warp by the soft sensor 24A (for example, when the control result based on the knowledge database 26 is not appropriate), the production state The necessary part of the knowledge database 26 can be updated by re-implementing the experiment or simulation and reconstructing the corresponding prediction model. In addition, the prediction model under the new production state can be strengthened in the knowledge database 26 by constructing a prediction model corresponding to the production state that is not stored in the knowledge database 26 by performing experiments or simulations. .

The corrugator seat state information estimation device (soft sensor) according to the first embodiment of the present invention is configured as described above, so that the soft sensor operates and the operation of the corrugator is controlled as follows. The
In other words, when the corrugator is started, the process controller 22 always grasps the machine state of the corrugator 1, and sends the current machine state to the control amount calculation unit 21 periodically or in response to a request from the control amount calculation unit 21. Output.

  Further, in the present soft sensor 24A, the sheet temperature detection information of each part of the corrugator 1 is obtained from the degree sensors 31a to 31m, and the sheet moisture of each part is determined from each sheet temperature using the corresponding prediction model called from the knowledge database 26. Further, the sheet warpage is predicted from the predicted sheet moisture of each part, the sheet adhesion failure of each part is predicted from the sheet temperature of each part, and these prediction results are output to the control amount calculation unit 21.

  In the control amount calculation unit 21, production state information from a production management system (not shown), sheet state information such as sheet temperature, sheet moisture, sheet warpage, and sheet adhesion failure from the soft sensor 24 A, and the corrugator 1 from the process controller 22. Depending on the machine status information such as sheet heating, glue gap amount, and pressing force, control amounts such as sheet heating, glue gap amount, and pressing force are calculated from the matrix condition table prepared in advance, and the calculation results are used as control commands. To the process controller 22. The process controller 22 controls the sheet heating, the glue gap amount, the applied pressure, and the like by matrix control based on the control amounts output from the control amount calculation unit 21.

  For example, when the order of the corrugated sheet is changed and the production state is changed, the process controller 22 matches the current production state with the production state [here, paper width, flute, base paper composition and base paper basis weight. Are matched (including not only a perfect match but also a substantial match)] The optimum operating state storage unit 23 is searched for a data set, and if a desired data set is found, the operating state of this data set is found. The information is read out as the optimum operation state information, and the corresponding control elements are controlled so as to be in the optimum operation state. On the other hand, if the optimum operation state information corresponding to the current production state is not found in the optimum operation state storage unit 23, the process controller 22 performs normal matrix control.

In this way, each information of sheet temperature, sheet moisture, sheet warpage, and sheet adhesion failure can be grasped, so that the corrugator can be operated so that sheet warpage and sheet adhesion failure do not occur.
In particular, according to the present apparatus, only the sheet temperature sensor is necessary, but the sheet moisture sensor, the sheet warpage sensor, and the sheet adhesion failure detection sensor are unnecessary, so that the apparatus cost and the maintenance burden of the apparatus are greatly reduced. is there.

  In addition, since the knowledge database 26 storing various prediction models is used to estimate sheet moisture, sheet warpage, and sheet adhesion failure, the degree of freedom of the form of the prediction model stored in the knowledge database 26 is high. Since it can be configured as a partial linear model for each condition area, the entire production state area can be created as a set of partial linear models, even if it is a nonlinear prediction model. A highly accurate prediction is possible with a simple linear model.

Further, since all know-how and knowledge are stored in the knowledge database in the form of a prediction model, there is no missing or lost data, and the safety is excellent.
In particular, even if the estimated value is deviated, such as when the soft sensor does not work well, if the knowledge database 26 is updated by re-learning and the knowledge database 26 is updated, the soft sensor can be corrected and the practicality is improved. high.

[B. Second Embodiment]
FIG. 6 is a schematic configuration diagram showing a corrugator and a seat state information estimation device according to the second embodiment of the present invention.
As shown in FIG. 6, the corrugator 1 according to the present embodiment is obtained by adding sheet moisture sensors (non-contact type sensors) 32 a to 32 m (reference numeral 32 when not individually distinguished) to those of the first embodiment. is there. In addition, each moisture sensor 32a-32m is installed in the same location as the temperature sensors 31a-31m.
Further, the soft sensor (sheet state information estimation device) 24B of the production management device 2 is different from that of the first embodiment.

  That is, the soft sensor 24B according to the present embodiment acquires the sheet temperature information of each part of the corrugator 1 from each temperature sensor 31 and also acquires the sheet moisture information of each part of the corrugator 1 from each moisture sensor 32. Estimate only poor adhesion. Note that, as described in the first embodiment, sheet warpage is predicted from sheet moisture, and sheet adhesion failure is predicted from sheet temperature.

  Therefore, the prediction model according to the present embodiment does not include the prediction model 25a (see FIG. 1) related to the correspondence between the sheet temperature state and the moisture state, but the prediction model related to the correspondence between the sheet moisture state and the warp state. 25b and a prediction model 25c relating to the correspondence between the sheet temperature state and the sheet adhesion failure. Therefore, the knowledge database 26 stores prediction model data related to the prediction models 25b and 25c.

Other configurations are the same as those of the first embodiment.
Since the corrugator seat state information estimation device (soft sensor) according to the second embodiment of the present invention is configured as described above, it can operate substantially the same as the first embodiment to obtain substantially the same effect. it can.
In this embodiment, since the sheet temperature sensor and the sheet moisture sensor are installed, the effect of reducing the apparatus cost and the maintenance burden of the apparatus is reduced as compared with the first embodiment. Since the defect detection sensor is unnecessary, there is an effect that the apparatus cost and the maintenance burden of the apparatus are reduced. However, in the first embodiment, since the sheet warpage is predicted based on the predicted sheet moisture (prediction based on prediction), it is difficult to ensure the accuracy of prediction of the sheet warpage. Since sheet warpage is predicted based on this, there is an advantage that it is easy to ensure prediction accuracy. In addition, since modeling for predicting sheet moisture is not necessary as compared with the first embodiment, the preparation burden is reduced.

[C. Third Embodiment]
FIG. 7 is a schematic configuration diagram showing a corrugator and a seat state information estimation device according to a third embodiment of the present invention.
As shown in FIG. 7, a soft sensor (sheet state information estimation device) 24C according to the present embodiment includes a neural network (particularly a hierarchical neural network) 27 instead of the knowledge database 26 of the first embodiment. Yes.

  As shown in FIGS. 7 and 8, the neural network 27 includes an input layer 27a, an intermediate layer 27b, and an output layer 27c. Sheet temperature detection data of each part of the corrugator 1 from the temperature sensors 31a to 31m is input to the input layer 27a. The information is input, and information on sheet moisture, sheet warpage, and sheet adhesion failure is output from the output layer 27c via the intermediate layer 27b. This neural network 27 is preliminarily learned by simulation or testing so as to obtain an appropriate output for the input, and the intermediate layer 27b is constructed. The prediction model 25a for predicting the sheet moisture from the sheet temperature, the sheet moisture The sheet has a function corresponding to a prediction model 25b that predicts sheet warpage from the sheet and a prediction model 25c that predicts sheet adhesion failure from the sheet temperature.

Other configurations are the same as those of the first embodiment.
Since the corrugator seat state information estimation device (soft sensor) according to the third embodiment of the present invention is configured as described above, it can operate substantially the same as the first embodiment to obtain substantially the same effect. it can.
In this embodiment, unlike a knowledge database, it is not necessary to create and maintain a large amount of database, so that it is easy in terms of data management. Further, since the neural network 27 is suitable for learning update, the neural network 27 can be learned and updated while using a machine, so that a more accurate neural network can be constructed.

[D. Fourth Embodiment]
FIG. 9: is a schematic block diagram which shows the corrugator and its sheet state information estimation apparatus concerning 4th Embodiment of this invention.
As shown in FIG. 9, a soft sensor (sheet state information estimation device) 24D according to this embodiment according to this embodiment is replaced with a neural network (in particular, a hierarchical neural network) instead of the knowledge database 26 of the second embodiment. ) 27 is provided.

  The neural network 27 includes an input layer 27a, an intermediate layer 27b, and an output layer 27c. The input layer 27a includes sheet temperature detection data of each part of the corrugator 1 from the temperature sensors 31a to 31m and each part of the corrugator 1 from the moisture sensors 32a to 32m. The sea moisture detection data is input, and information on sheet warpage and defective sheet adhesion is output from the output layer 27c via the intermediate layer 27b. This neural network 27 is constructed by learning in advance so as to obtain an appropriate output with respect to the input by simulation or test, and constructs an intermediate layer 27b. A prediction model 25b for predicting sheet warpage from sheet moisture, sheet temperature From this, it has a function corresponding to the prediction model 25c for predicting sheet adhesion failure.

Other configurations are the same as those of the second embodiment.
Since the corrugator seat state information estimation device (soft sensor) according to the fourth embodiment of the present invention is configured as described above, it can operate in substantially the same manner as in the second embodiment to obtain substantially the same effect. it can.
In this embodiment, unlike a knowledge database, it is not necessary to create and maintain a large amount of database, so that it is easy in terms of data management. Further, since the neural network 27 is suitable for learning update, the neural network 27 can be learned and updated while using a machine, so that a more accurate neural network can be constructed.

[E. Fifth Embodiment]
FIG. 10: is a schematic block diagram which shows the corrugator concerning the 5th Embodiment of this invention, and its sheet | seat state information estimation apparatus.
As shown in FIG. 10, a soft sensor (sheet state information estimation device) 24E according to this embodiment according to this embodiment includes a neural network (hierarchical neural network) 27 in addition to the knowledge database 26 according to the first embodiment. Is provided.
This neural network 27 is the same as that of the third embodiment.

  In this embodiment, the knowledge database 26 and the neural network 27 can be selected and used for estimating the sheet state information. In particular, the knowledge database 26 and the neural network 27 can be selected according to the production status information of the corrugator or according to the sheet status information.

Other configurations are the same as those in the first and third embodiments.
Since the corrugator seat state information estimation device (soft sensor) according to the fifth embodiment of the present invention is configured as described above, it operates in substantially the same manner as in the first and third embodiments to obtain substantially the same effect. be able to.
In particular, depending on the production state of the corrugator, there are a database suitable for the knowledge database 26 and a network suitable for the neural network 27. For example, if the neural network 27 is applied to the case where the correlation between the input and output of the prediction model is relatively linear, the software sensor 24E can be efficiently efficiently as a whole while saving the amount of information in the knowledge database 26. Can be configured.

[F. Others]
Although the embodiments of the present invention have been described above, the present invention is not limited to the above-described embodiments, and various modifications can be made without departing from the spirit of the present invention.
For example, in each embodiment, the temperature sensor 31 or the moisture sensor 32 is installed at each location of the corrugator 1, but may be provided only at some of the locations. Particularly important for estimating the sheet warpage are the upstream and downstream portions of the double facer 17, for example, the temperature sensors 31i, 31j, 31k or the moisture sensors 32i, 32j, 32k, and the detection locations are selected so as to include them. It is preferable to do.

  The temperature sensor 31 or the moisture sensor 32 is most simply provided at the center in the sheet width direction at each point. However, when estimating the sheet warpage in the C direction (width direction) from the sheet moisture, It is necessary to provide a plurality of sensors 31 or moisture sensors 32 in the sheet width direction at the same point in the sheet traveling direction. In this case, it is provided at two places so as to make a pair on the left and right in the sheet width direction, or at a total of three places in the center and a pair of left and right in the sheet width direction.

  Although not mentioned in the above embodiment, it is also possible to use a genetic algorithm, linear regression, principal component regression, multiple regression, exact model by calculation, etc. as a prediction model used for the soft sensor. You may make it apply to a prediction model combining the database 26 and the neural network 27 suitably.

It is a lineblock diagram showing the corrugator concerning the 1st embodiment of the present invention, and its sheet state information estimating device. It is the schematic which showed the structure of the preheater for upper liners of a corrugator concerning each embodiment of this invention, a single facer, and a center core preheater. It is the schematic which showed the structure of a part of corrugator's single stage sheet | seat preheater, front liner preheater, glue machine, and double facer concerning each embodiment of this invention. It is the schematic which showed the structure of the double facer of the corrugator concerning each embodiment of this invention. It is a perspective view which shows the curvature simple model concerning each embodiment of this invention. It is a block diagram which shows the corrugator concerning 2nd Embodiment of this invention, and its sheet | seat state information estimation apparatus. It is a block diagram which shows the corrugator concerning the 3rd Embodiment of this invention, and its sheet | seat state information estimation apparatus. It is a schematic diagram which shows the neural network concerning 3rd, 4th embodiment of this invention. It is a block diagram which shows the corrugator concerning 4th Embodiment of this invention, and its sheet | seat state information estimation apparatus. It is a block diagram which shows the corrugator and its sheet | seat state information estimation apparatus concerning 5th Embodiment of this invention. It is a block diagram which shows the management structure of the conventional corrugator.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Corrugator 2 Production management apparatus 3 Upper (back) liner 4 Center core 5 Single-stage sheet (single-sided cardboard sheet)
6 Lower (table) liner 7 Double-sided sheet (double-sided cardboard sheet)
8 Single-sided sheet 11 Upper (back) liner preheater 12 Center core preheater 13 Single facer 14 Single stage preheater 15 Glue machine 16 Lower (table) Liner preheater 17 Double facer 17A Upstream heating section 17B Downstream Side cooling section 18 Slitter scorer 19 Cut-off 20 Stacker 21 Control amount calculation unit 22 Process controller 23 Optimal state storage unit 24A to 24E Soft sensor as sheet state information estimation unit (sheet state information estimation device) 25 Memory 25a, 25b, 25c Prediction model 26 Knowledge database 27 Neural network 31, 31a to 31m Temperature sensor 32, 32a to 32m Moisture sensor 30, 31 Glue 111A, 111B Back liner heating roll 11 A, 112B Wound amount adjusting device 113A, 113B Arm 114A, 114B, 115, 116 Guide roll 121 Core heating roll 122 Wound amount adjusting device 123 Arm 124 Guide roll 131 Belt roll 132 Tension roll 133 Pressure belt 134 Upper roll 135 Lower Roll 136 Gluing device 136a Gluing dam 136b Gluing roll 136c Meter roll 136d Glue scraping blade 137 Preheating roll for liner 138 Preheating roll for center core 141 Single-stage sheet heating roll 142 Winding amount adjusting device 143 Arm 145, 144 Guide roll 151 Gluing device 151a Dam 151b Gluing roll 151c Doctor roll 152 Pressure bar device 152a Pressure bar 152b Actuator 153, 154 Dror 155 pieces amount wrapped stage preheat roll 161 Table liner heating roll 162 adjuster 163 arms 164 and 165 guide roll 171A first shower device (rear liner wetting apparatus)
171B Second shower device (front liner wetting device)
172 Heating plate 173 Pressure belt 174 Pressure unit 174a Pressure bar 174b Actuator

Claims (11)

In a corrugator that manufactures a corrugated cardboard sheet from a paper sheet, a sheet state information estimation device that estimates other information as an estimation target from specific state information among a plurality of state information related to the traveling paper sheet or the corrugated cardboard sheet. ,
Detecting means for detecting the specific state information;
Estimating means for estimating the state information of the estimation target from the specific state information detected by the detection means, using a prediction model that is a correlation between the specific state information and the state information of the estimation target; A corrugator seat state information estimation device characterized by comprising: The specific state information is a sheet temperature measurement value of the paper sheet or the cardboard sheet, and the detection means is a temperature sensor,
The corrugator sheet state information estimation device according to claim 1, wherein the state information of the estimation target is information on any of sheet moisture, sheet warpage, and sheet adhesion failure of the paper sheet or the corrugated cardboard sheet. . The specific state information is sheet moisture of the paper sheet or the cardboard sheet, and the detection means is a moisture sensor,
2. The corrugator sheet state information estimation device according to claim 1, wherein the state information to be estimated is information relating to sheet warpage or sheet adhesion failure of the paper sheet or the cardboard sheet. The specific status information is a sheet temperature measurement value and a sheet moisture measurement value of the paper sheet or the cardboard sheet, and the detection means is a temperature sensor and a moisture sensor,
2. The corrugator sheet state information estimation device according to claim 1, wherein the state information to be estimated is information relating to sheet warpage or sheet adhesion failure of the paper sheet or the cardboard sheet. The information on the sheet warpage includes information on the warpage of the corrugated sheets stacked as a single sheet on a stacker at a downstream end of the corrugator, according to any one of claims 2 to 4. Corrugator seat state information estimation device according to claim. Further comprising storage means for storing the prediction model,
The corrugator seat state information estimation apparatus according to any one of claims 1 to 5, wherein the estimation unit calls the prediction model from the storage unit to estimate the state information of the estimation target. Further comprising a knowledge database storing the prediction model under various conditions;
The corrugator sheet state information according to any one of claims 1 to 6, wherein the estimation unit estimates the state information of the estimation target by calling a predetermined prediction model from the knowledge database. Estimating device. A neural network constituting the prediction model is provided,
The corrugator seat state information estimation apparatus according to any one of claims 1 to 6, wherein the estimation unit estimates state information of the estimation target using the neural network. A knowledge database storing the prediction model in various conditions is further provided,
A neural network constituting the prediction model is provided,
The said estimation means estimates the state information of the said estimation object using the said predetermined prediction model and the said neural network called from the knowledge database, The any one of Claims 1-6 characterized by the above-mentioned. Corrugator seat state information estimation device. 10. The corrugator sheet state information estimation apparatus according to claim 8, wherein the neural network is a hierarchical neural network. In a corrugator that manufactures a corrugated cardboard sheet from a paper sheet, a sheet state information estimation method for estimating other information as an estimation target from specific state information among a plurality of state information relating to the traveling paper sheet or the corrugated cardboard sheet, ,
Prepare in advance a prediction model that is a correlation between the specific state information and the state information of the estimation target,
A detecting step for detecting the specific state information;
A corrugator seat state information estimation method comprising: an estimation step of estimating state information of the estimation target from the specific state information detected in the detection step using the prediction model.

Images