JP2751410B2 - Formation measurement method and formation control method using the formation measurement method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention grasps unevenness of transmitted light of paper as a planar image,
The present invention relates to a formation measurement method that can evaluate the properties and quality of paper and can be used as control means for improving the quality, and a formation control method using the formation measurement method.

[Prior art] The quality of paper formation (small thickness unevenness) means the degree of dispersion of fibers in a sheet. Generally, a sample sheet is placed on an inspection box with a built-in light source. ,
This has been done by visual inspection of the transparency distribution.

Although this method is widely practiced at each factory, it is probably subjective and its judgment requires sufficient knowledge and long experience, resulting in individual differences in test results.

For this reason, conventionally, the ground sum as shown in FIG. 8 has been devised and actually used, and the vertical head is placed at the vertical position of the running paper a for which the formation is to be measured so as to sandwich the paper a. b,
c, a light source d such as a laser connected to a power supply, and a mirror f for irradiating light e from the light source d to the paper a in the lower head c. And a photocell j that receives light e passing through the paper a via a mirror g, a filter h, and a lens i.

Thus, the light e emitted from the light source d is applied to the paper a via the mirror f, and the light e transmitted through the paper a is incident on the photocell j via the mirror g, the filter h, and the lens i, and the voltage e is applied. The voltage value is displayed as a formation index with respect to time as shown in FIG.

In addition, when the formation index is measured, a person judges based on the formation index and appropriately changes the jet wire ratio (J / W ratio) and the like so that the formation is good. .

[Problems to be Solved by the Invention] However, in the above-described total, the light source d
Since the diameter of the light e irradiated from the paper a to the paper a is about 1 mm, and the transmitted light signal is processed one-dimensionally, the fluctuation of the transmitted light level is captured as the floc size. Although formation was quantified, the judgment sample was too small to make an overall judgment as judged by human vision, and had the drawback that formation could not be captured accurately. .

Also, with regard to the formation control, the J / W ratio is only adjusted by trial and error, and the control is based on the measurement result that does not grasp the entirety of the ground total. It has not been possible to easily improve the quality of the product.

In view of such circumstances, the present invention provides a formation measurement method capable of objectively evaluating the formation as a surface rather than as a point, and a paper quality improvement method based on the detection result by the formation measurement method. An object of the present invention is to provide a formation control method that can be performed efficiently.

Means for Solving the Problems According to the present invention, an image of light emitted from a parallel light source having a variable light amount and transmitted through a paper having a required area is captured by a camera, and the image is displayed as a transmitted light image on a display device of an image processing arithmetic unit. The transmitted light image projected on the display device is divided into a required size and a required number of windows, and the average value of the density of each window and the density of each window are calculated from the density of the pixels existing in each window. The secondary variance of the concentration of the entire window is calculated, and the average value of the primary variance of the concentration of the entire window and the secondary variance of the concentration of the entire window are calculated. In addition, based on experimental data obtained in advance, a plurality of levels of quality from the worst to the best of paper are applied to the formation coefficient of paper based on experimental data. Ma Each of the J / W ratio, the foil angle, and the deflector indentation amount corresponding to each of the plurality of membership function curves for specifying from the degree of A plurality of control membership function curves indicating the degree of application to the quantity are obtained in advance, and the formation coefficient measured by the formation measurement method and each membership function curve for specifying the quality of the paper are determined. The degree of matching at each intersection with is determined, and the J / W ratio change amounts are inferred from the control membership function curves indicating the J / W ratio change amounts in accordance with the respective matching degrees. Along with obtaining the result, a final membership function indicating the final inference result of the J / W ratio change amount synthesized by superimposing the respective inference results of the J / W ratio change amount is obtained, and based on the final membership function, Actual J / W ratio J / W ratio is controlled by determining the change amount of the foil, and similarly, the change amount of the foil angle and the deflector pushing amount is sequentially determined and controlled. It depends on the control method.

[Operation] Therefore, in the formation measurement method of the present invention, an image of light emitted from a variable light source and transmitted through a sheet of paper having a required area is captured by a camera and is displayed on a display device of an arithmetic unit for image processing. While being displayed as a transmitted light image, the transmitted light image displayed on the display device is divided into a required size and a required number of windows, and an average value of the density of each window from each density of the pixels present in each window and The primary variance of the concentration of each window is calculated, and further, the average value of the primary variance of the concentration of the entire window and the secondary variance of the concentration of the entire window are calculated. Quantification of formations obtained as is performed more accurately.

Further, in the formation control method using the formation measurement method of the present invention, the quality of the paper in a plurality of stages from the worst to the best based on experimental data obtained in advance is determined with respect to the formation coefficient of the paper. A plurality of membership function curves for specifying from the degree to which they apply, and the J / W ratio, foil angle,
And a plurality of control membership function curves indicating the degree of application of each change amount of the deflector depression amount are obtained in advance, and the formation coefficient measured by the formation measurement method and the quality of the paper are specified. The degree of matching at each intersection with each membership function curve of
First, each J / W from each control membership function curve indicating the amount of J / W ratio change corresponding to the degree of each matching.
An inference result of the W ratio change amount is obtained, and a final membership function indicating a final inference result of the J / W ratio change amount synthesized by superimposing the respective inference results of the J / W ratio change amount is obtained,
The amount of change in the J / W ratio experiment is determined based on the final membership function, and the J / W ratio is controlled.Following, similarly, the change amount is sequentially determined and controlled for the foil angle and the deflector pushing amount. Is carried out to improve the formation.

[Example] Hereinafter, an example of the present invention will be described with reference to the drawings.

FIGS. 1 to 3 show an embodiment of the present invention. A camera 3 such as a video camera having a uniform light source 2 with a variable light amount and a zoom function on the other side is provided on one side of a paper 1 to be inspected. Place. The camera 3 is connected via a cable 4 to an arithmetic unit for image processing 5 having a display device 6. Further, a separate display device 7 is connected to the camera 3 via a cable 8 in order to always display an actual transmitted light image of the paper 1.

Since the paper 1 to be inspected is actually quite large, the portion to be taken out as a sample (10 mm × 10 mm or more) is the entire paper 1 if only the portion is viewed without missing the characteristics of the paper 1. Area (area that can be regarded as steady).

At the time of measurement, the collimated light source 2 is
The signal input to the camera 3 enters the display device 7 and is always displayed as an image 9a of the actual transmitted light of the paper 1, while the signal input to the camera 3 enters the image processing arithmetic device 5. The signal is displayed on the display device 6 as an image 9 showing an area in which the formation can be determined, that is, the quality of the entire paper 1 can be determined, as shown in FIG. The bright portion) and the heavy portion to which dust or the like is attached are displayed darker (darker) than the average.

Subsequently, windows W ₁ , W ₂ ,..., W _{k are used in the} image 9 with the area having an area of about twice the average size of the voids or the minimum size of the flocs peculiar to paper as a dispersion calculation unit. ,…, W _N
(See FIG. 2 (b)). The size and number of the windows can be appropriately selected depending on the type of paper, the composition, and the like.

Here, as shown in FIG. 3, M = n × m pixels (n = 4, m in the example of the drawing) of the display device 6 in one window.
= 5) When it is included, since the density of each pixel in the i-th row and the j-th column in the k-th window W _k is represented by C _kij , k
The average value Ca _Vk of the concentration in the second window W _k is And a density variance Va _Vk (hereinafter referred to as a first-order variance) in the k-th window W _k .
That is, a value that is a measure of how much the density varies within one window W _k is: (See FIG. 2 (c)).

Furthermore, the window _{W 1, W 2, ...,} W k, ..., the average value a _V of primary variance of the entire W _N , W _k ,..., W _N based on the average value a _V of the overall primary variance, and the variance Va _{V of the} primary variance as a whole of the windows W ₁ , W ₂ ,. Is called secondary variance) And displays the calculation result (see FIG. 2 (d)).

The primary dispersion Va _Vk becomes in local features sampling part of the paper 1, good quality is uniform small primary dispersion Va _Vk in window W _k, primary dispersion according quality deteriorates As the value of Va _Vk increases, 2
Since the secondary variance Va _V also increases, the formation is quantified using the secondary variance Va _V as a formation coefficient.

The image processing operation device 5 is provided with a ₁ to an at the b _{i -th} stage of the pixels constituting the image 9 projected on the display device 6.
A data logger (not shown) that accumulates the density data of the pixels in the column, an analyzer (not shown) that extracts the data accumulated in the data logger, captures the data as a change with time, and frequency analyzes the data; Or by processing the data stored in the data logger by analysis software, and determining whether or not a pulsating component exists in the paper 1 based on the frequency analysis result.

In this way, the formation can be quantified as a surface rather than as a point as in the past, and the formation can be objectively evaluated with higher accuracy. It can be used for control and the like.

The formation measurement method described above can be applied to both online and offline. However, in the case of online, it is indispensable to adopt a high-speed shuttering as the camera 3, while a strobe that emits light in conjunction with the high-speed shuttering can be used as the parallel light source 2 instead of the one that constantly emits light. is there.

Next, based on the formation coefficient obtained by the formation measurement method described above, the membership function is considered (Retsu Yamakawa, "The Idea of FUZZY Computers", November 10, 1988, third print, Kodansha Co., Ltd.) A method of controlling formation by applying is described.

FIG. 4 is a side view showing a wire part of the paper machine,
10 is a head box, 11 is a breast roll, 12 is a bottom wire, 13 is a forming board, 14 is a foil, 15 is a wet suction box, 16 is a top wire, 17,1
8 is a deflector, 19 is a suction box, 20 and 21 are showers for cleaning dirt attached to the bottom wire 12 and the top wire 16, and 22 and 23 are automatic valves for adjusting the amount of water injected from the showers 20 and 21. , 24 indicates a felt, and a J / W ratio between a jet velocity jetted from the head box 10 and the velocity of the bottom wire 12 and the top wire 16;
The foil 14 angle in the initial dewatering region 25, and the deflector 17, 18 indentation amount in the bottom middle dehydration region 26 and the top middle dehydration region 27, respectively, based on the formation coefficient obtained by the formation measurement method. The suction amount of the suction box 19 and the opening degrees of the automatic valves 22 and 23 of the showers 20 and 21 are changed as necessary.

FIG. 5 (i)-
As shown in (v), the five levels of quality of the paper 1 including “worst”, “bad”, “normal”, “good”, and “best” are determined with respect to the formation coefficient (horizontal axis) of the paper 1. Applicable degree (vertical axis)
From five membership function curves M _{1 to} M ₅
And a change amount (horizontal axis) of the J / W ratio (here, only the wire speed is increased / decreased) as an example corresponding to each of the membership function curves M _{1 to} M ₅ for specifying the quality of the paper 1. obtained in advance and the true extent for five control shown (vertical axis) membership function curve M ₁ a to M ₅ a and M ₁ b to M ₅ b against.

The formation coefficient of paper 1 obtained by the formation measurement method and each membership function curve for specifying the quality of paper 1
The degree of matching at each intersection with M _{1 to} M ₅ is determined, and each control membership function curve M ₁ a to M ₅ a indicating the amount of change in the J / W ratio corresponding to the degree of matching. And M ₁ b to M ₅ b
And the inference result of each J / W ratio change amount from
The inference results of the J / W ratio change amount are superimposed and synthesized to form a fifth
Figure final membership function showing the final inference result in as shown in (vi) J / W ratio change amount M _L a, obtains an M _L b, the final membership functions M _L a, one of the M _L b centroid and (final membership membership function M _L a or position to halve the area surrounded by the M _L b and aspect shaft) actual wire speed of increase or decrease the horizontal axis component of the actual the J / W ratio J / W
Control the ratio.

The FIG. 5, the texture coefficient indicates the case where 35, the vertical axis component of the point of intersection between the membership function curve M ₁ indicating "worst" of FIG. 5 (i) (matching as an example Degree) is about 0.25, so extend the line horizontally from the intersection to "significantly increase" or "significantly decrease" the wire speed.
The head of the control membership function curve M ₁ a or M ₁ b scraping the degree of about 0.25 matching that employs only hatched portion as the inference result of the wire speed decrease amount.
Further, the vertical axis component of the point of intersection between the membership function curve M ₂ showing a "bad" of FIG. 5 (ii) (degree of matching) is about 0.75
Therefore, a line is horizontally extended from the intersection, and the head of the control membership function curve M ₂ a or M ₂ b for “increase” or “decrease” the wire speed is cut off at a matching degree of about 0.75. Only the hatched portion is adopted as the inference result of the wire speed increase / decrease amount. Furthermore, FIG. 5 (iii)
Since the vertical axis component intersection of the membership function M ₃ indicating "normal" in the (degree of matching) is about 0.7, to extend the horizontal line from the intersection point of the wire speed "slightly increased", or "slightly Control function curve M ₃ a
Or cut off the head of M ₃ b with a matching degree of about 0.7,
Only the hatched portion is adopted as the inference result of the wire speed change amount. In addition, since the vertical axis component of the point of intersection between the membership function curve M ₄ indicating "good" of FIG. 5 (iv) (degree of matching) is about 0.1, wire speed to extend the horizontal line from the intersection point the "increased slightly", or "slightly reduce" scraped by the control membership function curve M ₄ a or M ₄ b degree of about 0.1 head matching that, adopting only hatched portion as the inference result of the wire speed decrease amount I do. Moreover, since formation factor for the membership function curve M ₅ indicating the "best" of FIG. 5 (v) is not an intersection case 35, the wire speed "Masa not" or "reduced without the degree of matching for the control membership function curve M ₅ a or M ₅ b of "is 0, that is, that the wire speed has to be increased or decreased. Therefore, when the formation coefficient is about 35, when the above inference results are superimposed and synthesized, as shown in FIG. 5 (vi), the degree of matching is increased from "slightly increased" to "increased". During or “slightly reduce”
The center of gravity is created around between "reduce" and the amount of increase or decrease in wire speed. Here, when determining whether to increase or decrease the wire speed, when changing the wire speed for the first time, only one of the centers of gravity (for example, the center of gravity for decreasing) is adopted from the final inference result, and the ground is determined. If the improvement of the formation is recognized, then control is performed in the direction of decreasing the wire speed.If the improvement of formation is not recognized, control is further performed in the direction of increasing the wire speed thereafter. To

As described above, by using the phage (FUZZY) theory using the membership function together, it is possible to perform not the on-off control but the gentle control that does not suddenly change the paper 1 as the product. Becomes possible.

In this manner, the J / W is changed, and as shown in FIG. 6, when the change in the J / W ratio is not repeated and the formation cannot be improved anymore as shown in FIG. 6, it is shown in FIGS. 5 (i) to (vi). A membership function is used in the same manner as in the case of changing the J / W ratio, and control is sequentially performed to change the foil angle and the deflector pushing amount, thereby improving the formation.

FIGS. 7 (a) to 7 (e) show a modification of the way of displaying an image in the formation measurement method of the present invention. In order to further clarify No. 9, the image processing is performed by expressing the density in three stages, for example, “dark”, “medium”, and “light”.
'Display, the image 9' Fig image 9 as shown in (b) the ratio i.e. void fraction V _theta to the entire area S of the total area [sigma] s _v of at transmitted light up to a region 30 in And the ratio of the total area ΣS _k of the area 31 of the minimum transmitted light in the image 9 ′ to the total area S, that is, the overweight ratio K, By calculating and displaying the results (see FIG. 7 (c)), the formation of the paper 1 is quantified, and if necessary, a part of the image 9 can be clearly identified as a defect or can be clearly identified. 7 (d) is displayed as an enlarged image 9 "as shown in FIG. 7 (d), and image processing is performed on the enlarged image 9" to represent the density in three stages, as shown in FIG. 7 (e). Such an enlarged image 9 is displayed and used as data for investigating the cause of formation.

In the other embodiments described above, plane information can be grasped by emphasizing the characteristics of the paper 1 by image processing, and particularly, holes, dust, etc., which could not be evaluated conventionally, can be initially detected, and quality improvement and productivity can be improved. It helps to improve the quality of the data, and the best and worst parts can be picked up and analyzed through the screen.

The formation measurement method and the formation control method using the formation measurement method of the present invention are not limited to the above-described embodiment, and various changes may be made without departing from the gist of the present invention. Of course, it can be added.

[Effects of the Invention] As described above, according to the formation measurement method of the present invention and the formation control method using the formation measurement method, the following various excellent effects can be obtained.

(I) Since the conventional method of capturing information as a point is regarded as a surface, it approaches a method of evaluating formation by human eyes and the formation can be quantified, thereby increasing objectivity. or,
The stiffness of the control can be increased by characterizing the dispersion.

(Ii) Quantification of formation and FU using membership function
By using the ZZY theory together, it is possible to perform not only on / off control but also gentle control without abrupt change to the product paper, which can greatly contribute to improving the quality of paper.

(Iii) Applicable to both online and offline.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an example of an apparatus for implementing the formation measurement method of the present invention, FIG. 2 (a), (b), (c), and (d) are illustrations of image processing in a display device, and FIG. FIG. 4 is a schematic view showing the relationship between a window and a pixel in a display device, FIG. 4 is a side view showing a wire part of a paper machine, and FIGS. 5 (i) (ii) (ii)
i), (iv), (v), and (vi) are explanatory diagrams of a procedure for determining a J / W ratio change amount based on a formation coefficient, and FIG. 6 is a formation control method using the formation measurement method of the present invention. FIG. 7 (a), (b), (c), (d), and (e) are image processing explanatory diagrams showing a modification of the image display method in the formation measurement method of the present invention. FIG. 8 is a block diagram showing an example of the conventional ground sum, and FIG. 9 is a diagram showing the relationship between the time obtained by the ground sum shown in FIG. 8 and the formation index. 1 is paper, 2 is a parallel light source, 3 is a camera, 5 is an image processing arithmetic unit, 6 is a display device, 9 is an image, 10 is a head box, 12 is a bottom wire, 14 is a foil, 16 is a top wire, 17 , 18 is a deflector, 19 is a suction box, 20,
21 is a shower, 22 and 23 are automatic valves, 25 is an initial dehydration region, 26 is a bottom middle dehydration region, 27 is a top mid dehydration region, W ₁ , W ₂ ,…, W _k ,…, W _N are windows and Ca _V1 , Ca _V2 ,…, C
a _Vk , ..., Ca _VN is the average of the concentration of each window, Va _V1 , V
a _V2 , ..., Va _Vk , ..., Va _VN is the primary variance of the concentration of each window, a _V is the average value of the primary variance of the concentration of the entire window, and Va _V is the secondary of the concentration of the entire window Variance, M _{1 to} M
₅ shows the membership function _{curve, M 1 a~M 5 a, M} 1 b~M 1 b is the control membership function curve, M _L a, M _L b is the final membership function.

Claims (2)

(57) [Claims] An image of light emitted from a variable light source and transmitted through a sheet of paper having a required area is captured by a camera and displayed on a display device of an image processing arithmetic device as a transmitted light image, and is also displayed on the display device. The transmitted light image is divided into a required size and a required number of windows, and an average value of the densities of the respective windows and a primary variance of the densities of the respective windows are calculated from the respective densities of the pixels existing in the respective windows. Calculating the average value of the primary variance of the concentration of the entire window and the secondary variance of the concentration of the entire window, and using the secondary variance of the concentration of the entire window as a formation coefficient. Total measurement method. 2. A plurality of membership function curves for specifying a plurality of levels of worst to best paper quality from the degree to which the paper formation coefficient is applied based on experimental data obtained in advance. A plurality of membership functions for control indicating the degree to which each of the J / W ratio, the foil angle, and the deflector indentation amount is applied in correspondence with each of the membership function curves for specifying the paper quality. A curve is obtained in advance, and the degree of matching at each intersection between the formation coefficient measured by the formation measurement method according to claim 1 and each membership function curve for specifying the quality of the paper is determined. First, an inference result of each J / W ratio change amount is obtained from each control membership function curve indicating the J / W ratio change amount in accordance with each matching degree, and the J / W ratio change amount is obtained.
A final membership function indicating the final inference result of the J / W ratio change amount synthesized by superimposing the respective inference results of the / W ratio change amount is obtained, and the actual J / W ratio of the J / W ratio is determined based on the final membership function. A formation control using a formation measurement method characterized in that the amount of change is determined and the J / W ratio is controlled, and then the amount of change in the foil angle and the amount of deflector depression is determined and controlled in the same manner. Method.