JPH11207931A - Color control system for printer - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable the control to be performed in consideration of all different operating states by providing an adaptive turner which communicates with a controller for calculating an ink supply amount and adjusts a gain parameter in response to the oration of a printing process. SOLUTION: The control system 200 is equipped with a fuzzy logic parameter turner 212 for performing an adaptive control. In this fuzzy logic parameter turner 212, a gain parameter in a proportion-differential-integral controller (PID controller) 216 is adjusted adaptively to compensate for a dispersion in the behavior of a printer. The method for adjusting the value of parameters of th PID controller 216 using the fuzzy logic parameter tuner 212 is preferably a method for adjusting the PID parameter 216 indirectly or incrementally because of its high stability. Besides, this method is also preferable because it diminishes the severe fluctuation of the parameter value with time.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention generally relates to a web offset printer.
controlling the ink supply in the ting press,
A system and method for obtaining and maintaining a target color value. More specifically, the present invention relates to a fuzzy logic adaptive closed loop controller (fuz).
zy logical adaptive closed
The present invention relates to an ink supply control system using a loop controller.

BACKGROUND OF THE INVENTION Web offset printing presses operate by combining several single-color images to print a multi-color image by overlay printing on a moving substrate or web. Typical four-color printing processes include black, cyan, magenta, and yellow inks. The color quality of the printed image is determined by the degree to which the color of the printed image matches the desired or exemplary reference image. In many cases, the reference image is provided or endorsed by the printing customer. One of the methods for evaluating the color is a visual test of a printed image by a skilled printman. Another method of evaluating color is to measure the optical density of solid color bars printed on a substrate. Generally, the actual color quality is compared to the desired quality and the amount of ink supplied to the substrate is adjusted as needed.

[0002] That is, a printing press has an ink supply structure (inking structure) for each color ink used in the printing process.
assembly). Each ink supply structure includes an ink reservoir and a blade disposed along an outer surface of the ink supply roller. The amount of ink supplied to the roller train of the printing press and ultimately to the substrate, such as paper, is adjusted by varying the spacing between the edge of the blade and the outer surface of the ink supply roller. The blade is
Divided into a plurality of blade segments, the position of each blade segment relative to the ink supply roller is independently adjustable by movement of an adjustment screw or ink key, thereby providing a longitudinally extending substrate. To control the amount of ink supplied to the corresponding strip. A typical printing press includes 24 to 30 ink keys and operates to control ink to an ink key zone having dimensions of about 1.2 to 2.5 inches.

[0003] Ink also spreads laterally from one ink key zone on a substrate to an adjacent zone due to the movement of the vibrating roller. The vibrating roller vibrates laterally with respect to the longitudinal movement of the substrate.

In order to preset the initial positions of the ink keys, a printed copy or proof (p.
It is common for printing press operators to visually inspect the proofs and note the amount of color required for each band of the image to be printed. Based on this visual inspection and experience with the press, ink and substrate type (typically paper), the operator can set the ink keys to approximate the settings required once printing has been performed. It can be set in advance. As an example, low tack yellow inks have low dye strength and require an increase in ink to produce an image with a given optical density. As another example,
Uncoated paper
r) is coated paper (coated paper)
More ink is needed to have an image with a given optical density than r).

[0005] Once the printing press is started, the ink flow from the ink fountain to the web must be controlled by adjusting the ink keys for each ink color. The amount of time spent adjusting the ink keys until the desired solid ink density is obtained for each zone on the printing press.
It is called "makeready". Again, adjustment of the ink keys is typically based on visual inspection and manual adjustment by an experienced press operator. After blotting, continuously monitoring the print output and making appropriate ink key adjustments to ensure proper quality control of the colors in the printed image is not possible for press operators. General. For example, if the color of a zone is too weak, the operator adjusts the corresponding ink key to increase the ink flow in that zone. If the color is too strong, adjust the corresponding ink key to reduce the ink flow. Also, during execution, additional color adjustments may need to be made to compensate for fluctuating printing press conditions or to take into account the personal preferences of the customer.

[0006] Preliminary installation of the above-mentioned ink key, group,
The visual inspection and manual adjustment techniques used in connection with run-time are relatively inaccurate and costly and time consuming. In addition, such techniques require a high degree of operator skill.

[0007] In general, there is a known method for determining an initial estimated value of the amount of ink required for preinstalling an ink key with high accuracy. These methods typically achieve higher accuracy than the results obtained by visual estimation, based on a measurement of plate coverage. The plate coverage is the area of the ink in the total plate area.
This is the ratio of a), which gives a measure of the amount of ink required to print the desired image. By dividing the printing plate into bands corresponding to the ink keys and determining the plate coverage for each band, an initial ink key setting for each band can be determined.

In addition to the visual inspection of a printed image, a method of monitoring the color quality after printing is once performed is also known. These methods typically involve measuring the optical density of the printed image. To measure the optical density of various points on the printed image, the off-line or on-line of the web printing process
Either on a line, using a densitometer or by scanning the densitometer. The optical density can be measured by illuminating the inspection image with a light source and measuring the intensity of light reflected from the image. The optical density (D) is defined by the following equation.

D = -log ₁₀ (R) where R is the reflectance, that is, the ratio of the reflected light intensity to the incident light intensity.

Since the substrate material is discarded until an acceptable color is obtained, a method for accurately and quickly determining ink key settings would require minimal time and material costs. Would. Especially for short print jobs,
Startup waste can account for a large proportion of the overall turnaround time and materials.

Typically, a conventional proportional-integral-differential (PI
D) The controller is the most widely used controller in the industry. Conventional PID controllers were developed in the 1940's based on classical linear time-invariant systems. In theory, such a controller works well in printing applications if the entire printing process is linear and time-invariant, and controls the ink delivery rate. In other words, the color density needs to be proportional to the ink key setting, and the factors affecting the whole printing process must be unchanged.

[0012]

The conventional PID controller does not operate well when the non-linearity of the system to be controlled is very high, that is, when the work environment includes uncertain factors. Since web offset printing is a very complex process, there are many known and unknown factors that affect the measured solid ink density (SID) values, and the entire system is non-linear. Known factors affecting SID values include press operation and model, make and model variations, ink and color variations, reservoir solution pH values, operating temperature variations, differences in paper stock, Includes years of operation and speed of the printing press. As a result, the fixed PID
It is undesirable to control colors using only a PID controller with a parameter set. This is because such controllers cannot take into account all the different operating states.

[0013]

SUMMARY OF THE INVENTION It is an object of the present invention to provide a high precision color control system and method on a web offset press that utilizes adaptive control and overcomes the disadvantages of conventional PID controllers. is there. By adaptive control, gain parameters Kp, K of the PID controller
Adjust i, and Kd to "optimal" values in real time.

It is an additional object of the present invention to achieve adaptive control through the use of fuzzy logic.

[0015] Other features and advantages of the present invention will become apparent to one with skill in the art upon examination of the following detailed description and drawings.

Before describing the embodiments of the present invention in detail, it is to be understood that the invention is limited in its application by the details of structure and arrangement of components set forth in the following description and illustrated in the drawings. That is not to say. The invention is capable of other embodiments and of being practiced or of being carried out in various ways. It will also be understood that the text and terms used herein are for illustrative purposes and should not be considered as limiting.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The following description of the preferred embodiment is given by way of example for a Harris M1000B printing press using 24 ink keys.
printing printing). It should be noted that the invention is applicable to other types of printing presses, printing presses of other manufacturers, and printing presses having a different number of ink keys, such as 22 or 36.

FIG. 1 shows a web offset printing system 10 for printing a multicolor image on a moving web 12. In a preferred embodiment, four printing units 14, 1
6, 18, 20 each print one color of the image on the web 12. The arrangement of the printing units 14, 16, 18, 20 with respect to each other is determined by the printing press and can vary. Each printing unit 1, 16, 18, 20
Includes a printing plate cylinder and a blanket cylinder. This type of printing is generally called web offset printing. That is, each printing unit includes an upper blanket cylinder 22, an upper printing plate cylinder 24, a lower blanket cylinder 26, and a lower printing plate cylinder 28 to allow printing on both sides of the web 12. . In printing system 10, color 3 on printing units 14, 16, 18, 20
1,32,33,34 are typically black (K), cyan (C), magenta (M), and yellow (Y), respectively. Cyan, magenta, and yellow are 3
Subtractive pr inks
(i.m., color link), which is used to reproduce a color image. Black ink is used to add sharpness to features and to reverse overprinting of the three primary inks.

Each printing unit 14, 16, 18, 20
Includes an associated ink supply structure 36. Figure 2
Shown in The ink supply structure 36 operates to supply ink to the row of rollers, including the plate and blanket cylinders, and then to the web 12. That is,
The ink supply structure 36 includes an ink reservoir 38 disposed adjacent an ink reservoir roller 40 (also known as an ink ball) extending transversely across the web.
The blades 42 extend and are sectioned along the ink fountain roller 40 to independently adjust the spacing of each segment with respect to the ink fountain roller 40 to control the ink supplied to each ink key zone on the web 12. It has become possible. As shown in FIGS. 3 (a) and 3 (b), each blade segment 44 has an edge 46 and is adapted to approach an outer surface 48 of the ink fountain roller 40 by adjustment of an associated ink flow control device 50. Move away and away.

More specifically, a part of the ink reservoir roller 40 forms one main wall of the ink storage unit 38. The other major wall of the reservoir 38 is provided by a blade segment 44. The ink is supplied from the ink storage unit 38.
Surface of Ink Roller 40 and Blade Segment 4
4 through the space between the lower edge 46 and the distance from the blade edge 46 to the ink sump roller 40 so as to control the thickness of the ink film supplied to the outer surface 48 of the ink sump roller 40. Act on.

A plurality of ink flow control devices 50 are disposed at equally spaced lateral positions along the ink supply structure 36;
At these locations, pressure is applied to blade segment 44 to form and adjust the width of the gap between roller 40 and blade segment 44.
Each ink flow regulator 50 includes an ink key 54,
It has a screw that engages a thread in a fixed portion of the frame of the ink supply structure 36. The ink key 54 has the tip 5
6, which presses and deflects the corresponding blade segment 44, thereby providing locally adjustable control of the spacing and ink supply.

The ink key 54 is provided for the bidirectional operation motor 5.
8 driven. The bi-directional operating motor 58 operates to move the ink key 54 toward and away from the ink fountain roller 40. Potentiometer 60
Has a movable arm mechanically connected to the ink key 54. Potentiometer 60 has a pair of outer electrical terminals and an inner electrical terminal located between the outer electrical terminals. The inner terminal of the potentiometer is mechanically connected to the movable arm of the potentiometer 60. The position of the movable arm of the potentiometer 60 is determined by the ink key 5
4 depending on the position. Potentiometer 60 is energized at its outer electrical terminal and an electrical signal indicative of the position of the ink key is generated at the inner electrical terminal of the potentiometer. Motor 58 responds to the signal on line 66 to position ink key 54 as desired.

FIG. 4 is a side view of the roller array 96 of the lower printing unit of the Harris M1000B press. The ink is supplied from an ink supply structure 36 to an ink reservoir roller 40.
Through the call roller
r) 98. The call roller 98 continuously moves rearward and forward due to contact with the ink reservoir roller 40 and the roller 100. The ink amount of the ink fountain roller itself can also be adjusted by changing the angle at which the ink fountain roller 40 rotates each stroke. To do this, a conventional ratchet structure (not shown) is adjusted, as is known in the art. The rotation angle, together with the position of the blade segment 44, determines the amount of ink transferred to the pick roller 98. The relationship between the angle of rotation and the amount of ink transferred to the draw roller 98 is assumed to be linear. Ink is supplied from roller 100 to various other rollers 102-124 as shown in FIG. Arrows in FIG. 4 indicate the rotation directions of the rollers 98 to 124. Rollers 100, 104, 114, 118
Is a vibrating roller which operates to spread ink from one ink key zone to an adjacent ink key zone by vibrating back and forth transversely to the longitudinal direction in which the web 12 moves. I do.

Referring to FIG. 5, the overall operation of the ink key control system 200 of the present invention is shown. In general, the control system 200 operates to adjust the position of the ink keys 54 and control the position of the blade segments 44 to control the corresponding ink tanks on the web 12 of the printing press.
Controls the amount of ink supplied to the key zone. The control system 200 includes a controller 204 and a color density measurement system 208. Although various methods of measuring SID values can be utilized, preferably, the color density measurement system 208 provides for color bar patches in color bars arranged transversely across the web 12:
Generates a measure of solid ink density (SID). With respect to feedback control, controller 204 operates to maintain the SID value for each color patch within a desired range. The SID measurement value is called a control target variable, and is a final control target of the controller 204.

That is, the measured SID value is compared with the desired or set point SID value and the SID error signal (SID_er
r). The controller 204 includes a fuzzy logic parameter tuner 212, a conventional proportional-integral-derivative (PID) controller 216, and optionally a truncation calculator 220. Fuzzy logic parameter tuner 212 adaptively adjusts gain parameters in PID controller 216.
The PID controller 216 disconnects the signal
To supply. The cutting calculation unit 220 includes the vibration roller 110,
Taking into account the effect of the ink key combination due to the lateral movement of 104, 114, 118, a signal to drive motor 58 is provided to control the position of each ink key 54 independently. In the operation without the cut calculator 220, the signal from the PID controller 216 is supplied directly to the ink key 54.

Optionally, the controller 204
Can also interface with the ratchet structure 224 to control the angle of rotation of the ink reservoir roller 40.

More specifically, the color density measurement system 20
8 includes a color CCD video camera,
The CD video camera is mounted on a transport bar that spans the entire web. System 208 reports the solid ink density values of solid color patches in color bars oriented transversely across web 12. Flash the strobe light at the appropriate time so that the color CCD camera can capture an image of a portion of the color bar on the web 12. An image of a color bar is processed by an algorithm, and an accurate S is calculated for each color patch.
Calculate the ID value. These SID values are supplied to the controller 204. In a series of steps, the camera is moved laterally across the web 12 to capture a continuous image in all ink key zones across the web 12. One of the color density measurement systems 208 that accurately measures the optical density of a printed image while the printing press is operating includes
John C. Seeymour, Jeffrey P .;
Rappette, Frank N.W. Vroman, C
ia-Lin Chu, Bradly S. et al. Moer
sfelder, Michael A .; Gill and Karl R.S. Invented by Voss, "SYST
EM AND METHOD FOR MONITOR
ING COLOR IN A PRINTING P
U.S. patent application Ser. No. 08 / 434,9 entitled "RESS" (system and method for monitoring color in a printing press).
28 is a color measurement system (CMS) described in No. 28. The contents thereof are also included in the present application by this reference.

The controller 204 performs several functions. First, the controller 204 receives the SID measurements from the color density measurement system 208 and performs the following calculations.

SID_err (j, k, t) = SID_
set_point (j, k, t) -Measured
_SID (j, k, t) where: j: color index (j = C, M, Y, or K) k: ink key index on the web (k = 1,
..., 24) t: sampling time index (t = 1, 2, ...)
The controller 204 also determines that the SID_err has an increasing tendency, that is, the current SID_err at time t and the time (t
The difference from the previous sample ISD error in -1) is calculated as follows.

SID_derr (j, k, t) = SID
_Err (j, k, t) -SID_err (j, k, t
-1) Here, j, k, and t are defined above.

FIG. 6 shows the relationship between the PID controller 216 and the fuzzy logic parameter tuner 212. As shown, two calculation signals SID_e
rr (j, k, t) (ie, SID error signal for color j and ink key k at time t) and SID
_Derr (j, k, t) (ie, color j at time t)
And change of SID error signal for ink key t)
To the fuzzy logic parameter tuner 212
And the PID controller 216. PI
The D controller 216 calculates the ink key settings and obtains the desired set point SID values for each ink key band and for each ink color without considering ink key combinations. Fuzzy logic parameter tuner 21
The function of 2 adaptively adjusts the three gain parameters in the PID controller 216 to determine the behavior of the printing press (pe
to compensate for variations in R.f. To adjust the value of the PID parameter, the following 2
There are two ways. 1) direct output of the current PID gain parameter value by a fuzzy logic parameter tuner, or 2) indirect or incremental adjustment of the PID parameter. The second method is preferable because it has higher stability and reduces a sharp fluctuation of the parameter value over time.

Hereinafter, the following notation will be used. K
p (j, k, t), Ki (j, k, t), and Kd
(J, k, t) are the proportional, integral, and differential gain parameters used by PID controller 216 for color j and ink key k at time t, respectively. These gain parameters are based on fuzzy logic
It is adjusted in real time by the parameter tuner 212.

D_Kp (j, k, t), d_Ki (j,
k, t) and d_Kd (j, k, t) are Kp, K for color j and ink key k at time t.
It is the incremental adjustment of the i, and Kd parameters.

Ink_key (j, k, t) is the command ink key setting for color j and ink key k at time t, without taking into account the ink key combination.

The entire output of the PID controller 216 is
Unadjusted command ink key settings. Since the color density measurement system 208 sequentially reports the SID values, the PI
The D controller 208 can be implemented sequentially. P
The entire output of the ID controller 216 is as follows:
It is a linear combination of the proportional, integral, and derivative terms.

[0036]

(Equation 1)

Here, K _INT = int_key (j,
k, 0). In the adaptive control, the PID controller 2
The sixteen parameters Kp, Ki, and Kd are adjusted in real time to certain "optimum values". One way to achieve adaptive control is to use fuzzy logic.

The fuzzy logic operates based on fuzzy set theory to map the input space to the output space. When used with adaptive control, fuzzy logic incorporates the human expert's operational knowledge into the control loop.
Fuzzy logic is also useful in modeling nonlinear functions of arbitrary complexity. Fuzzy logic can be mixed with conventional control techniques, such as conventional PID control. The embodiment described here is an example of an indirect fuzzy logic control system. Indirect fuzzy logic control systems are, for example, conventional PI
Used with a D-controller, it has the advantage of separating the control design from the adaptive mechanism. In contrast, direct fuzzy logic controllers typically use a static incremental process model to relate the calculated error in the control action to the deviation in the desired behavior.

Fuzzy logic includes the concept of fuzzy sets. A fuzzy set is one that has no clear and sharp boundaries, but instead describes a somewhat ambiguous concept. The following are examples of fuzzy sets.

A set of old people, a set of tall people, a set of hot, a set of good drivers, a set of bad restaurant services, and a set of hot climates.

The degree to which an item belongs to a fuzzy set is measured by its membership function. A membership function is a function in which each point in the input space is 0 and 1
Is a curve that defines how to map to membership values (or membership degrees) that are values between. As an example, the membership value of a 69-year-old who belongs to a fuzzy set of "old" is. 8 (degree of belonging)
It is considered to be. The membership function can be represented by curves of various shapes including, for example, triangles, Gaussian, bell-shaped, S-shaped, polynomial-based curves, and others.

Another feature of fuzzy sets is that they do not follow the "mutually exclusive" rules. An item can belong to two or more different fuzzy sets at the same time. Using the same example as above, 69
An old person can belong to the fuzzy set "young people", but its membership function value is small. 2 and
At the same time, it belongs to the fuzzy set "old man" and its membership function value is. 8

The fuzzy inference system as shown in FIG. 7 can realize a non-linear mapping from the input space to the output space. Mapping is performed according to a number of fuzzy if-then rules. Each if-th
The en rule describes the partial behavior of the mapping and reflects some knowledge of the human expert's decision process. For example, the following rule is an example of a method for determining a tip amount in a restaurant.

Rule 1: If the quality of the food is excellent and the quality of the service is average, the tip is medium.

Rule 2: If the quality of the food is poor and the quality of the service is below average, the tip is set to the lowest.

The rule establishes a simple input-output inference system, where "food quality" and "service quality" are input fuzzy variables and a single output fuzzy variable is "tip amount". is there. The premise of the rule defines a fuzzy region in the input space, while the required part (con
(sequence) specifies a fuzzy region in the output space.

The fuzzy inference system basically consists of fuzzification, inference, aggregation, and defuzzification (defuz
(function). One way to perform the above steps is to use the Mandani fuzzy inference system.
ce system).
This is known in the art. Figure 8 illustrates some of the processing steps involved in the Mandani fuzzy inference system.
Shown in The Mandani inference system includes output membership functions (denoted by C1 and C2), which are also fuzzy sets.

Since the inputs to the fuzzy inference system are general crisp values, they are subjected to a fuzzification process to obtain a fuzzy if-value.
It is necessary to apply the then rule. Similarly, the results of multiple fuzzy if-then rules are aggregated and then defuzzified to produce a crisp output.

Fuzzification is performed using a plurality of input membership functions, and the membership value of each membership function is determined for a given input variable. The next step is to determine which if-then rule to activate for a given input variable. if-the
If the membership value of the fuzzy variable included in the premise of the n rule is non-zero, the if-then rule is activated. Interpreting the if-then rule involves evaluating the antecedent (fuzzifying the input and applying any necessary fuzzy operators),
And applying the results to the required parts. If the premise of a rule has more than one fuzzy variable, a fuzzy operator must be applied. For example, A AND where A and B are in the range (0,1)
The output of the statement B is min (A, B)
(Ie, the minimum of the two values). This is shown in the left part of the figure. Similarly, the output of the statement A OR B, where A and B are in the range (0,1), can be determined by max (A, B)) (ie, the maximum of the two values).

The outputs of the activated rules are aggregated. To aggregate the output fuzzy sets, they are combined into a single output fuzzy set, typically using a maximum operator, as shown in the right part of FIG. The resulting set is defuzzified, ie, solved into a single numerical value.

Various defuzzification methods are known.
Defuzzification is the conversion of fuzzy quantities to exact quantities. Timothy J., copyrighted by McGraw-Hill in 1995. "Fu" by Ross
zzy Logic with Engineering
Applications "(Fuzzy Logic and Design Applications) describes four known methods of defuzzification. Preferably, using the centroid method, also known as the area center or centroid method, Perform defuzzification.

The design and implementation of the fuzzy logic parameter tuner 212 for ink key control is performed as follows. As described above, the basic principle is to construct a fuzzy inference system for parameter adjustment of PID parameters. The two fuzzy input variables are SID_err (j, k, t) and S
ID_derr (j, k, t). Each input variable is
Fuzzy into multiple membership functions. For example, each input variable can be fuzzified into five membership functions, as shown in FIG. In addition, 4, 6 or 7
Note that a different number of membership functions can be employed, such as

In the control system 200 described here,
The membership functions are selected to be triangular, such that one input has a non-zero value for at most two simultaneous membership functions. The membership functions are as follows.

NL (large in the negative direction) NM (medium in the negative direction) ZE (zero) PM (medium in the positive direction) PL (large in the positive direction) There are two fuzzy output variables FOp and FOi. The output set in the preferred embodiment also includes five membership functions, as shown in FIG.

The following is an example of the if-then rule of the five membership function inference system. 1. If (sid_erris NL) and (si
d_derris NL) then (FOPis NL) (FOiis P
L) 2. If (sid_erris NL) and (si
d_derris NM) then (FOPis NL) (FOiis P
L) 3. If (sid_erris NL) and (si
d_derris is ZE) then (FOpis PM) (FOi is P
L) 4. If (sid_erris NL) and (si
d_derris ism) then (FOpis PM) (FOi is P
M) 5. If (sid_erris NL) and (si
d_derris PL) then (FOpis ZE) (FOiis Z
E) 6. If (sid_erris NM) and (si
d_derris NL) then (FOPis NL) (FOiis P
L) 7. If (sid_erris NM) and (si
d_derris NM) then (FOPis NM) (FOiis P
M) 8. If (sid_erris NM) and (si
d_derris is ZE) then (FOpis PM) (FOi is P
M) 9. If (sid_erris NM) and (si
d_derris is PM) then (FOp is ZE) (FOi is Z
E) 10. If (sid_error NM) and (s
id_derr isPL) then (FOp is ZE) (FOi is Z
E) 11. If (sid_erris ZE) and (s
id_derr isNL) then (FOp is NM) (FOi is N
L) 12. If (sid_erris ZE) and (s
id_derr isNM) then (FOp is NM) (FOi is N
M) 13. If (sid_erris ZE) and (s
id_derr isZE) then (FOp is ZE) (FOi is Z
E) 14. If (sid_erris ZE) and (s
id_derr isPM) then (FOPis NM) (FOiisN
M) 15. If (sid_erris ZE) and (s
id_derr isPL) then (FOp is NM) (FOi is N
L) 16. If (sid_error PM) and (s
id_derr isNL) then (FOp is ZE) (FOi is N
M) 17. If (sid_error PM) and (s
id_derr isNM) then (FOp is ZE) (FOi is Z
E) 18. If (sid_error PM) and (s
id_derr isZE) then (FOp is PM) (FOi is P
M) 19. If (sid_error PM) and (s
id_derr isPM) then (FOPis NM) (FOiisP
M) 20. If (sid_error PM) and (s
id_derr isPL) then (FOp is NL) (FOi is P
L) 21. If (sid_erris PL) and (s
id_derr isNL) then (FOp is ZE) (FOi is Z
E) 22. If (sid_erris PL) and (s
id_derr isNM) then (FOp is PM) (FOi is P
M) 23. If (sid_erris PL) and (s
id_derr isZE) then (FOp is PM) (FOi is P
L) 24. If (sid_err is PL) and
(Sid_derr isPM) then (FOP is NL) (FOi is P
L) 25. If (sid_erris PL) and (s
id_derr isPL) then (FOp is NL) (FOi is P
L) Next, the fuzzy output variables are used in the following equation.

FAp (j, k, t) = FAp (j, k, t)
t-1) + alphaP (j) * FOp (j, k, t) where FAp is a fuzzy accumulator output for the proportional term, and FOp is a fuzzy tuner output for the proportional term.

FAi (j, k, t) = FAi (j, k, t)
t-1) + alphaI (j) * FOi (j, k, t) where FAi is a fuzzy accumulator output for the integral term, and FOi is a fuzzy tuner output for the integral term.

The alphaP and alphaI terms each harmonize the fuzzy tuner output associated with it and add it to the fuzzy accumulator. This step is to further stabilize the adjustment process.

The formula used to update the PID parameter is as follows: Kp (j, k, t) = FAp (j, k, t) * MaxP
gain (j) * COVERAGE (j, k) Ki (j, k, t) = FAi (j, k, t) * MaxI
gain (j) * COVERAGE (j, k) Kd (j, k, t) = FAi (j, k, t) * MaxD
gain (j) * COVERAGE (j, k) where Kp (j, k, t), Ki (j, k, t), K
d (j, k, t) is a gain parameter of the PID controller, and MaxPgain (j), MaxIGai
n (j) and MaxDGain (j) are constants empirically determined for each color of each ink, and COVERA
GE (j, k) is the plate coverage value for color j for each ink key k.

In a preferred embodiment, COVERAGE
(J, k) for all keys for all colors,. Set to 20. However, if obtained, it is also possible to use the actual value of the plate coverage for each ink key zone to speed up the convergence. Also note that the FAi term is used to calculate Kd. However, the FOd term can be used as the output of an inference rule to determine a separate FAd term.

From the above, an incremental adjustment of the gain parameter is obtained as follows. d_Kp (j, k, t) = alphaP (j) * FOp
(J, k, t) * MaxPGain (j) * COVER
AGE (j, k) d_Ki (j, k, t) = alphaI (j) * FOi
(J, k, t) * MaxIGain (j) * COVER
AGE (j, k) d_Kd (j, k, t) = alphaI (j) * FOi
(J, k, t) * MaxDGain (j) * COVER
AGE (j, k) The following list summarizes examples of values associated with the Harris M100B press used in the above equations.

BLACK (j = K) MaxPGain = 15 MaxIGain = 45 MaxDGain = 20 AlphaP =. 05 AlphaI =. 10 CYAN (j = C) MaxPGain = 20 MaxIGain = 30 MaxDGain = 25 AlphaP =. 08 AlphaI =. 10 MAGENTA (j = M) MaxPGain = 20 MaxIGain = 35 MaxDGain = 25 AlphaP =. 08 AlphaI =. 15 YELLOW (j = Y) MaxPGain = 20 MaxIGain = 60 MaxDGain = 25 AlphaP =. 15 AlphaI =. The initial values of 30 Kp, Ki, and Kd are known Ziegler-
It can be determined by the Nichols (Ziegler-Nichols) method.

As yet another example, an input set can be defined that includes six membership functions instead of five. In this case, the six input membership functions are ZE
Into the two functions named PZE (positive zero) and NZE (negative zero), except that
It can be the same as two membership functions.
When using five output sets, the fuzzy logic adaptive controller 216 can use the following inference rules. 1. If (sid_erris NL) and (si
d_derris is NL) then (pgain is NL) (igain i
1. s PL) (dgain is PL) If (sid_erris NL) and (si
d_derris is NM) then (pgain is NL) (igain i
2. s PL) (dgain is PM) If (sid_erris NL) and (si
d_derr is NZE) then (pgain is PM) (igain i
s PM) (dgain is ZE) If (sid_erris NL) and (si
d_derr is PZE) then (pgain is PM) (igain i
s PM) (dgain is ZE) 5. If (sid_erris NL) and (si
d_derris is PM) then (pgain is PM) (igain i
s PM) (dgain is NM) If (sid_erris NL) and (si
d_derris is PL) then (pgain is ZE) (igain i
s ZE) (dgain is NL) 7. If (sid_erris NM) and (si
d_derris is NL) then (pgain is NL) (igain i
7. s PL) (dgain is PL) If (sid_erris NM) and (si
d_derris is NM) then (pgain is NM) (igain i
8. s PM) (dgain is PM) If (sid_erris NM) and (si
d_derr is NZE) then (pgain is PM) (igain i
s PM) (dgain is ZE) If (sid_error NM) and (s
id_derr isPZE) then (pgain is PM) (igain i
s ZE) (dgain is ZE) If (sid_error NM) and (s
id_derr isPM) then (pgain is ZE) (igain i
s ZE) (dgain is ZE) If (sid_error NM) and (s
id_derr isPL) then (pgain is ZE) (igain i
sNM) (dgain is NM) 13. If (sid_error is NZE) and
(Sid_derr isNL) then (pgain is NM) (igain i
s PL) (dgain is PL) If (sid_error is NZE) and
(Sid_derr isNM) then (pgain is NM) (igain i
s PM) (dgain is PM) If (sid_error is NZE) and
(Sid_derr isNZE) then (pga
in is ZE) (igain is ZE) (dg
ain is ZE) 16. If (sid_error is NZE) and
(Sid_derr isPZE) then (pga
in is ZE) (igain is ZE) (dg
ain is ZE) 17. If (sid_error is NZE) and
(Sid_derr isPM) then (pgain is NM) (igain i
sNL) (dgain is PM) 18. If (sid_error is NZE) and
(Sid_derr isPL) then (pgain is NM) (igain i
sNL) (dgain is PL) 19. If (sid_err is PZE) and
(Sid_derr isNL) then (pgain is NM) (igain i
sNL) (dgain is PL) 20. If (sid_err is PZE) and
(Sid_derr isNM) then (pgain is NM) (igain i
sNM) (dgain is PM) 21. If (sid_err is PZE) and
(Sid_derr isNZE) then (pga
in is ZE) (igain is ZE) (dg
ain is ZE) 22. If (sid_err is PZE) and
(Sid_derr isPZE) then (pga
in is ZE) (igain is ZE) (dg
ain is ZE) 23. If (sid_err is PZE) and
(Sid_derr isPM) then (pgain is NM) (igain i
s PM) (dgain is PM) 24. If (sid_err is PZE) and
(Sid_derr isPL) then (pgain is NM) (igain i
s PL) (dgain is PL) 25. If (sid_error PM) and (s
id_derr isNL) then (pgain is ZE) (igain i
sNM) (dgain is NM) 26. If (sid_error PM) and (s
id_derr isNM) then (pgain is ZE) (igain i
s ZE) (dgain is ZE) 27. If (sid_error PM) and (s
id_derr isNZE) then (pgain is PM) (igain i
s ZE) (dgain is ZE) 28. If (sid_error PM) and (s
id_derr isPZE) then (pgain is PM) (igain i
s PM) (dgain is ZE) 29. If (sid_error PM) and (s
id_derr isPM) then (pgain is NM) (igain i
s PM) (dgain is PM) 30. If (sid_error PM) and (s
id_derr isPL) then (pgain is NM) (igain i
s PL) (dgain is PL) 31. If (sid_erris PL) and (s
id_derr isNL) then (pgain is ZE) (igain i
s ZE) (dgain is NL) 32. If (sid_erris PL) and (s
id_derr isNM) then (pgain is PM) (igain i
s PM) (dgain is NM) 33. If (sid_erris PL) and (s
id_derr isNZE) then (pgain is PM) (igain i
s PM) (dgain is ZE) 34. If (sid_erris PL) and (s
id_derr isPZE) then (pgain is PM) (igain i
s PM) (dgain is ZE) 35. If (sid_erris PL) and (s
id_derr isPM) then (pgain is NL) (igain i
s PL) (dgain is PM) 36. If (sid_erris PL) and (s
id_derr isPL) then (pgain is NL) (igain i
s PL) (dgain is PL) As described above, the ink keys can be controlled directly using the effective ink key settings from the PID controller 212, or further processed by the cut calculator 220. However, it is also possible to generate adjusted ink key settings, ie actual ink key settings.

The problem of ink-key coupling is due to ink diffusion due to the movement of the vibrating roller. If the controller 204 determines that the ink flow to a particular ink key zone should be increased, then increasing the ink flow to a zone will increase the amount of ink spread to adjacent ink key zones. , The flow of ink to adjacent zones will also increase. To compensate for this, the ink flow to adjacent keys must be reduced. this is,
Adjacent ink keys are similarly effective.

Before describing one method of compensating for ink diffusion, the color control system may be used to determine the sequence of color patches.
Several different methods need to be described that allow the operation of controlling the ink key using the ID reading. One side of the web 24 has 24 keybands, which correspond to 24 SID measurements. One way to implement the system is to wait until 24 SID measurements are obtained, and then change the 24 ink key readings at once. However, this method is time consuming. Another way to implement the system is to change the ink key immediately after the corresponding SID measurement is obtained, without taking into account the effect of the adjacent ink key. In this case, the method eventually stabilizes, but does not take into account the effects of adjacent ink keys.

An ink key distribution function, or ink key diffusion function, can be determined that represents the diffusion of ink from the ink source, which is the width of the ink key zone. ink·
The key diffusion function can be represented by a vector, whose elements represent the amount of ink in the corresponding band. One way to determine the ink key diffusion vector is to release one ink key and determine how the ink
Observe if it spreads to the key band. For example, as a result of one such inspection, the following vector V was obtained.

V = [. 007. 009. 016.
043. 196. 460. 196. 043.
016. 009. 007] The vector V is obtained by averaging the empirically obtained data with the width corresponding to each ink key band, and then scaling all the vector elements to add up to 1. Thus, the elements of the vector V are the fraction of the ink (fra
ction). Each ink
The key has its own ink distribution,
It is proportional to the key opening. On the Harris M1000B press, 46% of the ink supplied from a given ink key
Goes directly to the corresponding ink key zone,
% Moves to the adjacent band, 4% moves to the next closest set, and so on.

The effect of the vibrating roller is shown in FIG.
20. Mathematically, this is a deconvolution.
Given an ink key distribution function and an effective ink key setting, attempt to obtain an ink key setting. In the preferred embodiment of the color control system 200, the SID measurements for each ink key band arrive at the PID loop serially in time, rather than all at once.

It is possible to describe the determinant and relate the actual and effective ink key openings.

E = SA where E is a vector representing the effective ink key opening, A is a vector representing the actual ink key opening, and S is an ink key diffusion matrix determined from the vector V. is there. E and A are both 24x1 element matrices. S is a 24 × 24 element matrix. (This size is
Determined by the fact that there are 24 ink keys on the Harris M1000B printing press). If the ink diffusion is invariant across the ink keys, the matrix S is a Toeplitzmatrix
x), ie, a matrix where each row is a shifted version of the row above it. Each row contains the elements of the vector V. Matrix S
Is shown in FIG.

The above equation can be rewritten as follows to solve for A. A = S ^-1 E If the matrix S is non-invertible, the solution of the actual ink key opening can be obtained. However, the inversion matrix S ^-1 may have some poor conditions. This means that the matrix can amplify noise. In addition, the inversion matrix includes an entry in each of the 24 columns. Therefore, in order to multiply E by a row of ^-1 , all 24 entries need to be used. One approach is to approximate S ^-1 using a matrix M ^-1 that approximates what S ^-1 does. That is, M ^-1 approximates the despreading function. M ^-1
Is shown in FIG. The matrix M ^-1 is a symmetric matrix, and the numerical values used to derive this matrix are. 518,. 1
96 and. 045. In other words, for any ink key zone, 51.8% of the ink remains in that zone and 19.6% moves to the adjacent zone,
Assume that 4.5% reaches a band two bands apart. S ^-1
The use of M ^-1 instead of, since only the row of M ^-1 is at most five entries, if you get most five SID measurements at a given time, subsequent changes to the ink keys Can be implemented. Numeric value. 518,. 19
6 and. Numeral 045 is a specific diffusion coefficient set that causes the control loop to converge.

When the matrix M ^-1 is used, there is a possibility that an edge effect is mixed in the ink key set value calculated for the ink key at each end.
The edge effect is due to the fact that at the edges, the increased amount of ink in the ink key affects the amount of ink supplied to the adjacent key on one side only. With regard to the ink key at the end, the ink
One way to calculate the key setting can be obtained by changing the value of an element in the matrix M ^-1 . For example, to the amount of ink supplied to the ink key zone at the end,
The latter is taken into account by theoretically including the amount of ink supplied to one side of the web. In other words, [(.196 + 0.045) /. 518] is incremented in the first row, first column of M ⁻¹ . Similarly, by adding (0.045 / .518), the element in the second row and the first column of M ⁻¹ is increased. The ink key settings for the affected ink key settings on the other side of the web are considered by changing the elements in the last column of the first and second or last rows. [(.196
+. 045) /. 518] to increase the element in the last row, last column. Also, by adding (0.045 / .518),
Increment elements in second to last row, last column. Various other fine adjustments are possible to account for edge effects.

In the preferred embodiment, the control loop operates with the following constraints. The SID measurement is the .SID of the desired SID value. If it is within 1, the PID controller 216
Operate without using the fuzzy logic parameter tuner 212 to adjust the PID gain parameter. This is because there is a concern that the rule set is not optimized within that range. In addition, there is a dead zone. For example, if the SID value is. If not, the PID controller 216 does not take any further action to adjust the ink key settings.

Since both the ink key setting and the rotation angle of the ratchet structure control the amount of ink supplied to each ink key zone, the ink key setting and / or ratchet setting in the ratchet structure. It is possible to change R. In theory, any ratchet setting is acceptable. However,
In practice, ratchet settings are limited. If the ratchet value is too low, it may be necessary to open the ink key beyond the physical limits of the ink key. on the other hand,
If the ratchet is set too high, the opening of the ink key will be very small and the sensitivity of the ink film pressure to changes in the ink key opening will be high. This results in reduced accuracy of the ink key opening.

Optimal conditions are obtained when the ratchet setting is as low as possible without causing the ink key opening to exceed a predetermined percentage of the physical limit. This ratio is necessary to provide room for subsequent adjustments.

One complication that can occur is that control algorithms can require ink key settings that exceed the physical limits of the ink keys. For example, if the requested ink key setting is greater than 100% aperture,
Or it may be a negative set value. In the simplest embodiment, requests for out-of-range ink key openings are simply excluded, so as not to exceed an excessive value.

In the preferred embodiment, there are separate operations for ink keys required to move more than 100% and ink keys required to move less than zero. In the former case, it is still possible to achieve the indicated density by increasing the ratchet setting. To do this, it is necessary to move the required ink key setting within the physical limits by an amount
Increase the ratchet setting.

The correction is simple because the ratchet settings and the ink key openings are multiplicative. For example, the required ink
If the key opening is 120%, the current ratchet setting value is
It must be increased to at least 1.2 times its current value. In this case, the new ink key opening is 100%
Is set to Alternatively, set the ratchet value to 10%
In some cases, it may be preferable to only increase and reserve extra ranges for adjustment.

When the set value of the ratchet is changed, all the ink key openings must be compensated together. If the ratchet setting is increased by multiplying by Q, all ink key openings must be reduced by dividing by Q.

It will be understood that the invention is not limited to the specific structure and arrangement of the components shown and described herein, but encompasses any modifications that may fall within the scope of the following claims. . Obviously, many modifications and variations are possible in light of the above teaching. Therefore,
It is to be understood that, other than as specifically described, the invention may be practiced within the scope of the appended claims. Alternative embodiments and variations of the methods taught herein will occur to those of ordinary skill in the art upon reading the foregoing description.

For example, the above color control system is
It was designed to work with a color type II ink fountain. However, such systems are useful when used with other fountains.

[Brief description of the drawings]

FIG. 1 is a block diagram of a web offset printing press according to the present invention.

FIG. 2 is an illustration of an ink supply structure including an ink fountain roller, an ink reservoir, and an ink key.

FIG. 3 (a) is a side view of the ink supply structure of FIG. 2 along line 3-3 with the ink key partially open, and FIG. 3 (b). FIG. 3 is a side view of the ink supply structure of FIG. 2 with the ink key closed.

FIG. 4 is a schematic view of a roller array of a lower printing unit of the Harris M1000B printing machine.

FIG. 5 is a schematic diagram of an ink key control system according to the present invention.

FIG. 6 is a schematic diagram of the relationship between a PID controller and a fuzzy logic parameter tuner.

FIG. 7 is a block diagram of a general fuzzy inference system.

FIG. 8 is a diagram of a Mandani fuzzy inference system.

FIG. 9 is a diagram of a five-input membership function.

FIG. 10 is a diagram of a five-output membership function.

FIG. 11 is an example of an ink key diffusion matrix.

FIG. 12 is an example of a pseudo despread matrix.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Web offset printing system 12 Web 14, 16, 18, 20 Printing unit 22 Upper blanket cylinder 24 Upper printing plate cylinder 26 Lower type blanket cylinder 28 Lower type printing plate cylinder 31, 32, 33, 34 Color 36 Ink supply structure 38 Ink reservoir 40 Ink reservoir roller 42 Blade 44 Blade segment 46 Edge 48 Outer surface 50 Ink flow regulator 54 Ink key 56 Tip 58 Bidirectional operation motor 60 Potentiometer 96 Roller row 98 Recall roller 100-124 Roller 200 Ink Key Control System 204 Controller 208 Color Density Measurement System 212 Fuzzy Logic Parameter Tuner 216 Proportional-Integral-Derivative (PID) Controller 2 0 cutting calculator

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Douglas J. Sykola, Green Meadow Drive, Waukesha, Wisconsin, United States of America 53, 719 (72) Inventor Stephen Jay Ratchajizki, Wisconsin 53072, Wisconsin, United States of America・ Street West 268 ・ North 2710 (72) Inventor: Bradley S. Mesfelder 53188, Wisconsin, USA Magnolia Drive, Waukesha 817

Claims (30)

[Claims] An adaptive control system for controlling an amount of ink supplied to a substrate for use with a printing process, the controller calculating an ink supply amount based on a measured ink color value and a target ink color value, At least one
An adaptive control system, comprising: a controller using two gain parameters; and an adaptive tuner communicating with the controller and adjusting the at least one gain parameter in response to operation of the printing process. 2. The adaptive control system according to claim 1, wherein said adaptive tuner is a fuzzy logic adaptive tuner. 3. The adaptive control system according to claim 1, wherein said controller implements fuzzy logic. 4. The adaptive control system according to claim 1, wherein said controller includes a PID controller. 5. The adaptive control system according to claim 4, wherein said adaptive tuner is a fuzzy logic adaptive tuner. 6. The adaptive control system according to claim 4, wherein said at least one gain parameter is equal to said PI.
An adaptive control system comprising at least one of the P, I and D gain parameters of a D controller. 7. The adaptive control system according to claim 1, wherein said ink color value is a density value. 8. A method for controlling the amount of ink supplied to a substrate of a printing press by an ink measuring device, the method comprising: selecting at least one controller gain parameter; and adaptively adjusting the at least one controller gain parameter. Adjusting and compensating for variations in the behavior of the printing press; using the adjusted at least one controller gain parameter to calculate a set value for the ink measuring device; and transmitting the set value to the ink measuring device. Performing the method. 9. The method of claim 8, wherein said at least one controller gain parameter comprises P, I, and D gain parameters of a PID controller. 10. The method of claim 8, wherein said at least one controller gain parameter is adjusted using fuzzy logic. 11. The method according to claim 8, wherein the set value is calculated based on a measured ink color value and a target ink color value. 12. The method according to claim 11, wherein said ink color value is a density value. 13. An adaptive tuner for use with a controller for controlling an amount of ink supplied to a substrate of a printing press by an ink measuring device, wherein the at least one gain of the controller is responsive to a change in the behavior of the printing press. An adaptive tuner including an adjustment mechanism for determining an adjustment value for a parameter. 14. The adaptive tuner according to claim 13, wherein the adjusting mechanism uses a measured ink density value and a target ink density value. 15. The adaptive tuner of claim 13, wherein said at least one gain parameter comprises at least one of a PID controller P, I, and D gain parameter. 16. The adaptive tuner according to claim 13, wherein said adjustment mechanism implements fuzzy logic. 17. In a printing press, wherein an ink control device controls the amount of ink supplied to a zone of a first roller and transfers ink from said first roller to a row of rollers and then to a substrate for printing an image. A method for adaptively controlling a set value for at least one ink control device, comprising: (a) measuring an ink color value in a band; and (b) converting the measured ink color value into at least one gain parameter. And (c) calculating a new set value of the ink control device corresponding to the band using the controller and the adaptive tuner, wherein the adaptive tuner calculates the new setting value of the printing press. Calculating an adjustment value for the at least one gain parameter based on behavior; and (d) setting the ink control device. And changing the new setting value, (f) until the measured ink color values is within a predetermined range,
Repeating steps (b) through (e). 18. The method according to claim 17, wherein said ink control device is an ink key. 19. The method of claim 17, wherein said ink color value is a density value. 20. The method of claim 17, wherein said ink control device includes a ratchet structure. 21. The method according to claim 17, wherein said adaptive tuner implements fuzzy logic. 22. The method of claim 17, wherein said controller comprises a PID controller. 23. The method of claim 17, wherein said at least one gain parameter comprises at least one of a PID controller's P, I, and D gain parameters. 24. An adaptive control system for controlling ink supplied to a substrate in a printing press, comprising: an ink color measurement device operatively adjacent to the substrate; and an ink supply measurement controlling an amount of ink supplied to the substrate. And a unit for communicating with the measuring device and receiving ink color values therefrom, and communicating with the measuring unit to communicate changes to be made to the amount of ink to be supplied to the substrate, and for measuring the measured ink density value to the target ink density An adaptive controller that converges to a value, the adaptive controller calculating a change to be added to an amount of ink supplied to the substrate based on an operation of the ink supply measuring unit. 25. The adaptive control system according to claim 24, wherein said ink value measuring device is an ink density measuring device. 26. The adaptive control system according to claim 24, wherein said adaptive controller is a fuzzy logic adaptive controller. 27. The adaptive control system according to claim 24, wherein said adaptive controller includes a controller and a parameter tuner. 28. The adaptive control system according to claim 27, wherein said parameter tuner is a fuzzy logic parameter tuner. 29. The adaptive control system according to claim 27, wherein said parameter tuner adjusts at least one gain parameter of said controller, wherein said at least one gain parameter is a PID controller P, I and D gain. An adaptive control system comprising at least one of the parameters. 30. An adaptive control system according to claim 24, wherein said ink supply measuring section includes an ink key.