JP3438378B2 - Image forming device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention uses an electrophotographic method.
In particular, always keeps images at a predetermined quality.
Control can be performed at low cost and with high accuracy.
Development work related to data collection and optimization design during product development
Image forming apparatus capable of significantly reducing the number
You. 2. Description of the Related Art Conventionally, an image forming method using an electrophotographic method has been known.
In the image forming apparatus, the fee for keeping the image density
Feedback control is very commonly used. This is static
In electrophotography using electricity, the temperature and humidity of the day
Which environmental conditions or changes over time of photoconductors and developers
Changes the image output state of the device itself,
This is because the instability fluctuates. [0003] The contents of the conventional feedback control are further improved.
To put it concretely, the density reproduction status is monitored by the density patch.
To obtain the error from the target density,
Control gain by multiplying
The most common method is to calculate the correction value of the
You. Here, the density patch is a developed image patch.
Often it is. This is a transfer image created on paper,
Developed images are easier to create and erase than fixed images
And the correlation with the fixed image density obtained by the user
Is extremely high. Also, the control actuator
The voltage applied to the charger and the exposure
Amount, term image bias, etc. are often used. For example, Japanese Patent Application Laid-Open No. Sho 63-177176
Bulletin, 63-177177 bulletin, 63-177178
In the technology described in the gazette, by changing the developing potential
The development density is controlled to a desired value. This development potential
The method to change is either one-component or two-component development
But it holds. However, the optimum developing potential is not controlled.
Various possible external factors: temperature, humidity, cumulative copying
It is always affected by the number of sheets, etc.
The setting of the amount and the developing bias always take these conditions into account.
It has the difficulty of having to do it. Only
The amount of state such as temperature and humidity, the amount of charge and exposure,
The relationship with the set value is complicated, and at the current technical level
Has not been fully modeled. Therefore, the control quantified using an approximate expression is
Electrons that are somehow performed but are primarily electrostatic processes
In photographic technology, charging, exposure, and
The relationship between the optimal settings of bias and bias is non-linear.
Therefore, sufficient control accuracy has not been obtained. In these circumstances
More environmental conditions in advance, such as hot and humid conditions
And environmental effects in low-temperature, low-humidity conditions,
Impact must be grasped, aiming at advanced control performance
Must collect detailed data over a wide range of conditions.
Enormous development man-hours were required. [0008] In addition, such huge man-hours are invested in the decision.
The set feedback gain is also different for each
Is always always optimal for the usage conditions of
I couldn't help. In particular, image densities that deteriorate over time
Impact on the degree of deterioration of the components used in each unit
And it depends on how users use it.
Long-term image density control performance from
I couldn't say. In addition, whether the control method is as described above.
In order to obtain control accuracy,
Potential sensor to monitor the electric potential and exposure potential,
Temperature and humidity sensors to monitor environmental conditions
Many control methods are required, raising costs
Was. [0010] Recently, Japanese Patent Laid-Open Publication No.
No. 71, 4-320278, etc.
How to use fuzzy and neural networks
Is being done. These are fuzzy and new
Neural network has a complicated nonlinear relationship between input and output
Using the feature of being able to cope with
It is used as a means for increasing control accuracy. This
As a result, the above-mentioned problem, that is,
Enormous development man-hours and sensors
Increased cost due to heavy use, more likely to go to market
Long-term image density control performance
To solve problems such as lack of security.
Not. Rather, fuzzy and neural networks
If the control accuracy is improved by using
To take advantage of the features that are suitable for computation, multi-input
In many cases, many sensors are used.
Cost is increasing. [0012] Furthermore, in fuzzy, members by engineers
It is necessary to tune the ship function.
Although the learning process itself can be automated,
Technicians must prepare the technician data in advance
In fact, it required considerable development man-hours. [0013] In addition, time-dependent deterioration data is collected in advance and stored.
Fuzzy and neural networks that take into account
Even when used, the relationship between the input and output
May change due to deterioration over time, machine differences, parts replacement, etc.
In the event that the
Was. In other words, each and every one
Image density control performance, even if fuzzy or neural network
Cannot be guaranteed even when using
won. Therefore, it is necessary to memorize the operation amount actually used.
The density value (or the
Or the concentration conversion value) and the actual concentration conversion measurement value.
Then, based on the density difference (or density conversion value difference)
The actual situation depends on which of the existing control rules and how much
Some weights indicate whether they are related. Figure
13 is weighted by the conventional density (or density conversion value)
FIG. 3 is a schematic diagram for explaining an outline of an application method. Also figure
14 is centered on the control example obtained this time (marked by X in FIG. 13)
FIG. 14 is an enlarged schematic view of a part of FIG. Here,
Is one manipulated variable and one concentration for simplicity.
Is described. In the following description, the acquired control
Classify and store control cases that have the same case status
This is called a “cluster”. Across
For control that can be extracted from several classified control cases
Let the parameters be called control rules. FIG. 13 shows three control rules as control rules.
The control rules A, B, and C are shown, and the target control amount is indicated by a dotted line.
Is indicated by a dotted line. Also, the black dots shown
This is the control example acquired up to this point.
This is a control example. Weighting with conventional technology is obtained this time
Calculated from the control case (actual operation amount) and the control rule
By using the concentration that can be obtained (or the concentration conversion value)
Was. That is, as shown in FIG.
The control amount of your case and the control items acquired this time for each control rule
Control variable calculated by applying the manipulated variable in the example
Is the difference from the calculated value of
The degree of conformity of the control rule in proportion to the
Had been decided. Originally, an image forming apparatus is used due to a change in environment or the like.
When the surrounding state changes, only the control amount (concentration)
Thus, it is impossible to calculate the degree of conformity of the control rule.
This is because, as shown in FIG.
Control amount that can be obtained even with the same operation amount due to deterioration of the
Often differ greatly. As described above, the conventional image
In the image forming apparatus, when calculating the conformity of the control rule,
Assigns conformity to all control rules
Will be assigned. Therefore, the current state is control rule C
Despite the print operation in
Control rule A, which has low relevance to control
There was a problem that it could have an effect. Further, FIG. 13 shows a typical control
Although only rules A, B, and C are shown,
Because many control rules such as
Even if there are many control rules with low probability,
The effect of the book is to double the number of its control rules.
Control variables that should be converged immediately
Bundles take a long time or have a significant effect on the image
There is a problem of performing an unsuitable control. The present invention has been made in view of the above circumstances.
And the effects of shallow control rules on current control
Eliminate or reduce size, necessary for original control
By using control rules, optimal control can be performed for the current situation.
It is an object of the present invention to provide an image forming apparatus which can be used. [0020] To solve the above-mentioned problems.
Therefore, according to the present invention, the quality of the output image depends on the operation amount.
Image quality variable means for changing the quality and control of the output image
Control case storage means for storing an example;
Extract control rules from multiple control cases stored in columns
Control rule extraction means and output image quality
Detection means for outputting the result as a control amount, control amount and target
Quality and compare the result with a preset tolerance
Is exceeded, the control amount is stored in the control case storage device.
Stored in the stage so that it can be used for subsequent control
The comparison means and the comparison result by the comparison means are preset.
After the specified tolerance is exceeded, the control amount corresponds to the target quality.
When the value becomes the value, the operation amount storage means for sequentially storing the subsequent operation amounts
And each control extracted by the control rule extracting means.
Determine the conformity of the control rule to the immediately preceding control case.
After the setting, based on the operation amount stored in the operation amount storage means,
And weights each control rule according to the degree of conformity.
And using the result, output by the detection means.
New operation so that the controlled variable becomes a value corresponding to the target quality.
Operating amount calculating means for obtaining an operation amount,
The amount of operation determined by the output means is
It is characterized by being supplied to a stage. According to the present invention, when the apparatus is in operation,
Control cases are accumulated in the control case storage means.
The control rule extracting means is controlled based on the accumulated control cases.
Extract your rules. On the other hand, the comparison means
Quality and compare the result with a preset tolerance
Is exceeded, the control amount is stored in the control case storage means.
Remember. Further, the manipulated variable storage means stores
After the comparison result by the step exceeds the preset tolerance
When the control amount reaches the value corresponding to the target quality,
The amount of work is stored sequentially. On the other hand, in the operation amount calculation means,
Is a comparison between the control amount detected by the detection means and the target quality.
Operation amount is required so that the control amount approaches the target quality.
You. At this time, for each extracted control rule,
After the suitability for the control case is determined,
Based on the manipulated variable stored in the column, the degree of conformity to each control rule
Is weighted according to the result, and a new
A manipulated variable is required. And this operation amount is
The image quality is manipulated and supplied to the variable means. Embodiments A. Configuration of First Embodiment (1) Basic Configuration A specific embodiment according to the present invention will be described below. Destination
First, the image output unit IOT (i.
The outline of the image output terminal is shown in FIG.
In FIG. 2, the image reading unit and the image processing unit are omitted.
ing. That is, the image output unit IO by electrophotography
Only T is shown. The image forming procedure will be described with reference to FIG.
First, a document was read by an image reading unit (not shown).
Or an external computer (not shown).
The image processing unit (Fig.
(Not shown). The resulting input
The image signal is input to the laser output unit 1 and the laser light
Modulate line R. In this way, depending on the input image signal,
The modulated laser beam R illuminates the photoreceptor 2 with raster light.
Fired. On the other hand, the photosensitive member 2 is connected to a scorotron charger 3.
Therefore, when uniformly charged and irradiated with the laser beam R,
An electrostatic latent image corresponding to the input image signal is formed on the surface.
It is. Next, the electrostatic latent image is developed by the developing device 6 with toner.
The developed toner is transferred to a sheet (not shown) by the transfer device 7.
) And is fixed by the fixing device 8. So
After that, the photoconductor 2 is cleaned by the cleaner 11.
Then, one image forming operation is completed. 10 is development
It is a density sensor and a developing package formed outside the image area.
(Hereinafter described). (2) Development Patch Creation Mechanism and its Monitor
Here, the developing patch and its monitor in this embodiment are described.
The mechanism will be described. The development patch is the output image density
To monitor the signal, and as shown in FIG.
(Dot coverage 100%) density patch a1 and hyla
Two types of density patches (dot coverage 20%) density patches a2
Is adopted. Then, these solid density patches a1,
The highlight density patch a2, as shown in FIG.
These are also set to a size of about 2 to 3 cm square, and
It is formed outside the image area. Sand
That is, as shown in FIG. 4, a latent image is formed in the image area 2a.
After that, the solid density patch a1 is
Highlight density patches a2 are sequentially formed.
ing. The density sensor 10 is provided on the surface of the photoreceptor 2.
LED irradiating unit that irradiates light to the surface of photoconductor 2
Sideways from a photo sensor that receives specularly reflected light or diffused light
The line L1 shown in FIG.
4 is a detection line of the sensor 10. Therefore, the solid density
H1 and the highlight density patch a2 correspond to the detection line L1
The developing density sensor 10
Sequentially pass through. FIG. 5 shows the output of the developing density sensor 10.
FIG. 4 is a diagram illustrating an example of a force signal. As shown, first,
A density detection signal corresponding to the original image is obtained.
Density detection of density patch a1 and highlight density patch a2
An outgoing signal is obtained. Solid density patch a1 and highlight density
Since the degree patch a2 is formed outside the image area,
It is not transferred to the paper and the cleaner 11
It is erased when passing through the part. In this embodiment, the development patch
The density is detected only for the fixed image that the user has
High density (final image density)
This is because removal by -11 is possible. Also, the development package
If the timing is other than the time of image formation,
A. (3) Configuration of Control Unit Next, FIG. 1 shows a scorotron charger 3 and a laser output.
FIG. 2 is a block diagram illustrating a configuration of a control unit 20 that controls the force unit 1.
is there. In the figure, reference numeral 21 denotes a density adjustment dial,
The author sets a value according to the desired density. Density adjustment die
The revised value of Al 21 is converted by the converter 22 into the development density
The value converted to the output of the sensor 10 (in the case of this embodiment,
(A value between “0” and “255”). Converter 2
2 is stored in the control amount memory 23.
Is held. In this case, the control amount memory 23
The difference is also stored. On the other hand, the output signal of the development density sensor 10 and the
The output signal of the memory 23 is
Are compared. In this comparison, the memory 23 stores
Is referred to. And the phenomenon concentration
If the difference between the two is within the allowable value,
For example, a control rule searcher is provided via a concentration comparator 24.
30 and if the value is equal to or more than the allowable value, the control case memory 2
5 is supplied. Here, the convergence detector 43 will be described first.
I will tell. The convergence detector 43 is connected via a density comparator.
The output of the developing density sensor 10 output is monitored and
Note that if the difference between the
That is, when the control system converges, a convergence signal (for example, “1”)
Is supplied to the manipulated variable (max, min) memory 44.
On the other hand, the difference between the two ranges from within the allowable value to over the allowable value
And a non-convergence signal (for example, a signal “0”)
(Max, min) is supplied to the memory 44. Operation amount (max, min)
When the memory 44 receives the convergence signal, the memory 44 thereafter performs
The operation amount supplied from the operation amount memory 32 described later is stored.
While receiving a non-convergent signal,
Extract the maximum and minimum values from the previous manipulated variable
Then, the data is supplied to the control rule memory 29. The maximum
The maximum and minimum values are defined as the conformance range of the corresponding control rule.
Box, in other words, indicates the operation amount range of the control rule.
It is something. When there is only one operation amount,
, The one operation amount has a maximum value and a minimum value. Also,
Can we extract control rules with few acquired control cases?
In this case, the above-mentioned operation amount is not stored. the above
The maximum and minimum values of the operation amount are supplied to the control rule memory 29.
And stored in association with the control rules. The control case memory 25 includes a density comparator
When the output signal from the H.24 is supplied, the subsequent control
It is a memory that stores examples, and stores state quantities, operation quantities, and control quantities.
The three quantities are stored as a set. Thus, the control case
Is stored in the past in the present embodiment.
To perform various controls based on the control cases
This is a control method based on a method called example-based reasoning. Here, it is stored in the control case memory 25.
State quantities have a dominant effect on the electrophotographic process.
Temperature, humidity, or deterioration over time.
These state quantities are considered to be almost constant within a limited time.
In the case of this embodiment,
The occurrence time (date, hour, minute, second) is used. However, occurrence
The time is in a predetermined time unit (3 minutes, 5 minutes, 10 minutes, etc.)
Within the predetermined time unit), the state quantity
I treat it as new. This is the time of occurrence
Are close to each other, they are almost the same temperature
Degree of deterioration over time under humidity
Because it can be expected. Time indicating the occurrence time
The data is, in this embodiment, the clock shown in FIG.
It is supplied from a timer 40. Next, the manipulated variable is the output value of the controlled object.
The amount of adjustment of the parameter to be changed.
In this case, the grid voltage setting value of the scorotron charger 3 (0
~ 255, hereinafter abbreviated as "scoro set value") and laser
Power setting value (0 to 255, hereinafter abbreviated as LP setting value)
). The reason why these two quantities are manipulated variables is that
The final image density to be controlled is
It must be two points of the light density part, and the scoro set value
And LP revision values show high correlation between solid density and highlight density
This is because Further, the scoro revision value and the LP revision value are as follows:
Each of them is stored in the operation amount memory 32, and the operation amount correction
The value corresponding to the output signal of the arithmetic unit 31 is appropriately read.
Swelling. Then, read from the operation amount memory 32
The revised scoro values are supplied to the grid power supply 15 and
As a result, the grid power supply 15
Pressure is applied to the scorotron charger 3. In addition, the operation
The LP setting value read from the memory 32 is
To the light amount controller 1.
6 outputs the laser power according to the LP set value.
Give to force part 1. Next, the control supplied to the control case memory 25 will be described.
The control amount is the output signal of the developing density sensor 10 and
As a result, in the control memory 25, for example, as shown in the following table,
The control case is stored. [Table 1] In this table, for example, Case 1 is a
) Was revised on April 1, 1994 at 12:00:10, LP
Value is “83”, the scoro revision value is “130”, and the control amount (
Sensor output value) is “185” for the solid part,
In the case 4, the state quantity 19
At 9:05:00 on April 2, 1994, the LP set value was "14
8 ", the scoring set value is" 115 ", and the control amount is
Is “185” and the highlight is “30”.
You. Next, the state quantity comparator 2 shown in FIG.
6. Cluster memory 27 and control rule calculator 28
Refers to the control case stored in the control case memory 25.
A control rule. Note that this
The operation of these blocks will be described later in detail. The control rule memory 29 stores the control rules.
For storing a plurality of control rules calculated by the computing unit 28
Each control rule includes the operation amount (max, mi
n) The maximum value, minimum value, that is, the operation amount from the memory 44
The operation amount range is stored in association with the
When there is a request from the rule searcher 30, the system responds to the request.
Return the rule and its operation amount range. in this case,
The control rule searcher 30 controls the control rule memory 29
Request each rule to be returned in turn, and
And the corresponding manipulated variable range (maximum
Value, minimum value), supplied from the density comparator 24.
Density difference (control amount) and the operation amount memory 32
MV supplied (ie LP set value, SCORO revision)
Value), which control rule best matches the control rule.
To determine the fitness of each control rule
Is weighted for each control rule, and
Search for the most frequent control rules, and ultimately
Request to the control rule memory 29.
You. Next, the manipulated variable correction value calculator 31 controls the control
The current control case searched by the rule searcher 30
Using the control rule with the highest degree of conformity to
The correction value is obtained, and the obtained correction value is stored in the operation amount memory 32.
Supply. As a result, the operation amount memory 32 stores the operation amount
The manipulated variable corresponding to the positive value, that is, the LP
The roller set value is stored in the grid power supply 15 and the light amount control.
To each other. On the other hand, the reference patch signal generator 42
Specify creation of density patch al and highlight density patch a2
This is the circuit shown and calibrated at the patch creation timing
The reference patch signal is output to the image output unit IOT. this
The solid density patch a1 shown in FIG.
The density patch a2 is created. In this case, the operation of the reference patch signal generator 42
The operation timing is performed by the 〓 / 0 adjustment unit 41.
The 〓 / ○ adjusting unit 41 is a timer output from the clock timer 40.
Monitor the image signal, and check the solid density patch a1 and highlight density.
So that the reference patch a2 is formed at a predetermined position.
The operation timing signal is supplied to the H signal generator 42. B. Next, the operation of the first embodiment will be described.
Here, FIGS. 6 to 9 illustrate the operation of the first embodiment.
FIG. FIG. 6 shows the first embodiment.
The transition of control cases acquired so far and the cluster
It is a schematic diagram which shows formation. In the figure, the horizontal axis
And the vertical axis indicates the density (control amount). Show
As described above, in this system, a predetermined width
Has an allowable value of If the control rules are met
Indicates the control amount (control example) indicated by a black dot
Although it is within the box, the control rule does not conform
It indicates that the value exceeds the allowable value. In FIG. 6, when printing is started, the temperature
Changes in humidity and humidity, or deterioration over time,
-Since the charge amount and the charging characteristics of the photoconductor change, the laser
Power and scorotron grid voltage at the same setting
Even so, the concentration will be significantly different. For example, high temperature
When the humidity is high, the concentration shifts to the higher side.
It shifts to the lower side. That is, the temperature and humidity at the time of control
And the degree of deterioration over time have already been collected and stored.
Different from those in your case, existing control rules
It does not fit the actual control case. Thus, the system
When the status of the program becomes different from the
Control rules used in the application
And the control amount becomes a value exceeding the target quality.
Lint number 1, 2, 3). Control amount exceeds allowable value
And the system acquires this control case, and the control case memory
25 is additionally stored. This additional storage is performed as follows.
You. First, the density comparator 24 shown in FIG.
Is determined to be above, and the output of the developing density sensor 10 at that time is determined.
The signal is transferred to the control case memory 25. Control case memory
25 is the state variable and the newly supplied control variable.
And the operation amount are stored as a set. And the state quantity comparison
Data 26 is newly written to the control case memory 25.
Based on the case, compare the time with the latest cluster and resemble the status
Is determined. That is, it is a control case group
Time information of control cases in the latest cluster and control case notes
Compare the time information of the control case newly written in
Comparison, within a predetermined time, it is determined that the state is similar
If they are separated by more than a predetermined time, it is determined that the states are not similar.
Refuse. If it is determined that the states are similar,
To add control cases for the latest cluster,
Write to the memory 27. At this time, the control rule calculator 2
8 is a control that includes the newly added control case.
A rule is calculated, and a coefficient indicating the control rule is determined by the control rule.
To the memory 29. And the control rules are extracted
Have created the most appropriate control rules for the current situation.
Therefore, the control amount after that converges to the target quality.
(Number of prints shown in FIG. 6, 4, 5, 6, 7 or less)
See descent). When the control amount converges, the convergence detector 43
A convergence signal is sent to the operation amount (max, min) memory 44. The manipulated variable (max, min) memory 44 stores the convergence signal.
Is supplied to the manipulated variable memory 32 thereafter.
The amount of operation to be performed, in the example shown in FIG.
7 are stored. And again, cis
The status of the system changes and new control cases are acquired
(See 8, 11 of the number of prints in FIG. 6)
Until then, the manipulated variable (max, min) has been stored in the memory 44
In addition, the previous operation amount (the number of prints in FIG.
Operation value), maximum value (max), minimum value (min)
And associate it with the control rules from the previous state,
It is stored in the control rule memory 29. This is an example
In a certain state, the
The maximum and minimum values can be selected from the
If it is associated with the control rules extracted from your case,
Any other method may be used. New control
Not only when it comes to the situation of acquiring cases, but also the amount of operation
Sequentially select if maximum value and minimum value can be determined by memorizing
May be. Note that if the manipulated variable that converged the control is one,
If not, the values are the maximum and minimum. Also,
The number of acquired control cases is too small to extract control rules
In this case, the operation amount storage according to the present embodiment is not performed. FIG. 7 shows the control rules and the control rules.
FIG. 7 is a schematic diagram illustrating an operation amount range associated with a rule.
In FIG. 7, the control rule A is in a state where it can be created.
When the control amount achieves (convergence) the target quality,
The operation amount range consisting of the maximum and minimum values of the operation amount
(Dotted line) A1. Control rule B,
C is also associated with the manipulated variable ranges B1 and C1, respectively.
You. Next, referring to FIG.
The manner in which the degree of conformity of a file is calculated will be described. In FIG.
In the first embodiment, the control rule and its control
The control amount reaches the target quality when the rule is applied (convergence
Weighting based on the manipulated variable range at the time of
This shows how the final control rule conformance is calculated.
Is shown. First, the control rules are already 3 for A, B, and C.
There are For each, the manipulated variable ranges A1, B
1, C1 are associated. Now, do the following control
Amount of operation to determine, but the state of the system has changed
Therefore, a control example of the operation amount and the control amount as indicated by the x mark
Suppose you can. In the conventional technology, only existing control rules are used.
To which control rule the current control case conforms
Was calculated, the current control case and each control rule
Weights W1 and W shown in FIG.
2. The degree of conformity of each control rule is determined by the weight W3.
Was. On the other hand, in the first embodiment, the operation amount range
By weighting A1, B1, and C1 as well,
Therefore, the conformity of existing control rules is improved as follows.
You. First, the weighting for each control rule is as follows:
This is performed in the same manner as in the related art. That is, each control rule
The weights for the
Weights W1 and W
2. Determine the weight W3. Next, the current manipulated variable is
Control amount ranges A1 and B associated with files A, B and C
1, it is determined whether or not it is included in C1. As shown in FIG.
In this example, the manipulated variable of the current control case is
It is included only in the associated operation amount range B1. But
Therefore, the weighting based on the operation amount range is performed according to the control rule B.
Weighting is performed only in the production range B1. this
The weight is defined as weight Wa. In this example, the other control
The weights of the manipulated variable ranges A1 and C1 of the rules A and C become zero.
You. The total weight for each control rule A, B, C is
It is calculated by weight of control rule × weight of operation amount range. I
Therefore, the total weight of control rule A and control rule C
Is zero, and the overall weight of control rule B is
W1 × weight Wa. As a result, control rule B conforms
The degree is the highest. Here, an extreme example is shown.
However, actually, as shown in FIG. 9, it is within the operation amount range.
In this case, the weight is set to “1” and even if the weight is outside the operation amount range,
If the weight is within the specified range, set the weight to “0.7”
It does not matter. In addition, the fuzzy control
Fuzzy determination of weight by replacing with barship function
It is also possible. For example, if the current system state is controlled
If it is rule A, the next control case (operation amount)
The operation amount range of the control rule A indicated by the symbol "+"
It will come. The weighting by the manipulated variable range at this time is
As shown, the weight of the control rule B in the manipulated variable range B1
Becomes the weight Wb, and the operation amount range A1 of the control rule A
Becomes the weight Wc. Here, the control rule C
The manipulated variable range is no longer relevant, which is the next control
No longer affect or could be harvested
It decreases depending on the operation amount range. As described above, in the first embodiment, the control amount
Is converged to the target quality,
Lively, the system status changes again and new control
When it comes to acquiring cases, I remembered it until then,
Extract the maximum and minimum values from the previous manipulated variable, and
It is stored as the operation amount range of the control rule in the
Control rules that change the status of the system and match the control case
When selecting, in addition to the existing methods, each of the above control rules
Weighting based on the manipulated variable range of
And the effect on the control rules shallow in the current control
Can be eliminated or reduced, which is necessary for the original control.
Control rules to determine the current system status.
Appropriate control can be performed. C. Second Embodiment Hereinafter, a second embodiment according to the present invention will be described. First,
Image output unit IOT of the image forming apparatus according to the second embodiment
(Image Output Terminal)
This is the same as the first embodiment, and has the configuration shown in FIG.
Therefore, the description is omitted. (1) Configuration of Control Unit FIG. 10 shows a control unit of the image forming apparatus according to the second embodiment.
FIG. 3 is a block diagram showing a configuration of the embodiment. In addition, it corresponds to FIG.
The same reference numerals are given to the same parts and the description is omitted. In the figure
Here, 61 is an operation amount (average) memory, which converges
When the convergence signal from the detector 43 is received,
The operation amount supplied from the operation amount memory 32 described below is stored.
On the other hand, when a non-convergent signal is received,
Calculates the average value (average) of the previous
It is supplied to the control rule memory 62. Note that the previous operation amount is 1
If there is only one, the one manipulated variable becomes the average value.
You. Also, few control cases have been acquired, and control rules have been extracted.
If not, the operation amount is stored.
Absent. The average value of the operation amount is stored in the control rule memory 62.
Provided and stored in association with the control rules. D. Next, the operation of the above-described second embodiment will be described.
11 and 12 illustrate the operation of the second embodiment.
FIG. In addition, the system
The transition of your case and the cluster formation are described above.
This will be described with reference to FIG. When creating a cluster,
For details on creating new control rules for
Since this is the same as the first embodiment, a brief description will be given. After the printing is started, each printing
For control cases, follow the control rules with the highest fitness.
Thus, the operation amount is determined, and the system is controlled. Soshi
Therefore, if a control case out of the allowable value range occurs (FIG. 6)
Number of prints, 1-3) and create a new cluster
Along with new control according to these cluster control cases.
Create your rules. Thereafter, the print sheets shown in FIG.
For the control cases of numbers 4, 5, 6, 7 this new
The operation amount is determined according to the control rule. in this way,
When the control amount converges, the convergence detector 43 sets the operation amount (average
age) Sends a convergence signal to the memory 61. The operation amount (average) memory 61 stores the convergence signal.
Is supplied to the manipulated variable memory 32 thereafter.
The amount of operation to be performed is stored. And again, the system
The status of the system changes and new control cases are acquired
Then, the operation amount (average) is stored in the memory 61 until then.
Operating amount (the number of prints in FIG.
Calculate the average value of the
Associated with the rule and stored in the control rule memory 29.
You. This is an example, and under certain conditions,
The average value can be selected from the bundled operation amount,
Control rules extracted from the obtained control cases
If you can, you can use any other method.
No. Also, when it becomes a situation to acquire a new control case
Not only when the amount of operation is stored and the average value can be determined.
If so, they may be selected one after another. FIG. 11 shows the control rules and the control rules.
A schematic diagram showing the average value of the operation amount associated with the rule.
is there. In FIG. 11, control rule A is created by
The controlled variable achieves (converges) the target quality when
The average is associated with the average value A2 of the operation amount during the operation.
Control rules B and C are also associated with average values B2 and C2, respectively.
Have been killed. Next, referring to FIG.
A description will be given of how to calculate the rule adaptability. FIG.
In the second embodiment, the control rules and the
When the control quantity reaches the target quality when the
Weighting based on the average
Shows how to calculate the final control rule conformance
are doing. First, there are already three control rules, A, B, and C.
Suppose there is. For each, the average value A2, B2, C
2 are associated. Well, to perform the following control
We decided the amount of operation, but because the state of the system changed,
A control example of the operation amount and control amount as shown by the X mark was created.
And In the conventional technique, only the existing control rules are used.
To which control rule the current control case conforms
Was calculated, the current control case and each control rule
Weights W1 and W shown in FIG.
2. The degree of conformity of each control rule is determined by the weight W3.
Was. On the other hand, in the first embodiment, the operation amount range
By weighting A1, B1, and C1 as well,
Therefore, the conformity of existing control rules is improved as follows.
You. First, the weighting for each control rule is as follows:
This is performed in the same manner as in the related art. That is, each control rule
Is weighted by the control cases indicated by x and each control rule.
A weight W1, a weight W2, a weight W
Only decide W3. Next, the current operation amount and the control rule
Average value A2 of the manipulated variables associated with A, B, and C
Calculate the absolute value of the difference between B2 and C2, and calculate the minimum value
Divide each absolute value and calculate its reciprocal. Example shown in FIG.
Then, the distance LA is the difference between the current operation amount and the average value A2.
The distance LB is an absolute value, and the current operation amount and the average value B2
Is the absolute value of the difference from the distance, the distance LC is the current manipulated variable and the average value C
The absolute value of the difference from 2. Therefore, in the illustrated example,
The distance LB becomes minimum, and the distance LB / the distance LA,
Distance LB / distance LB = 1, distance LB / distance LC. Next, based on the above-mentioned values, the flatness of the operation amount is calculated.
Calculate the weight by the average value. Weight by the average value of the operation amount
Are distance LB / distance LA (average value A2), 1
(Average value B2), distance LB / distance LC (average value C2)
Is divided by 1 + distance LB / distance LA + distance LB / distance LC.
Is calculated. Therefore, the current manipulated variable and
The greater the difference from the average value of the crop, the greater the distance
The weight decreases as the weight increases. For each control rule A, B, C
The overall weight is calculated by the weight of control rule x average value.
Is calculated by As a result, the conformity of control rule B
Is the highest, and finally the fitness is the highest. As described above, in the second embodiment, the control amount
Is converged to the target quality,
Lively, the system status changes again and new control
When it comes to acquiring cases, I remembered it until then,
The average value is calculated from the previous manipulated variables, and the control
Memorize them in association with your rules so that the system status
Select control rules that change and match the control case
At the time, in addition to the existing methods, operation of each of the above control rules
Since weighting is based on the average value of the amount,
What is the effect on control rules that are not involved in the current control?
The control required for the original control can be made smaller or smaller.
The best rule according to the current system state
Control can be performed. Also, the first implementation
In the example, when the operation amount is out of the range, the weight is set in advance.
However, in the second embodiment, the current operation
Based on the difference (distance) between the yield and the average value of each control rule
Since the weight is calculated, there is no need to set it in advance. As described above, according to the present invention,
Eliminates the effects of control rules that have little bearing on current control
Control that can be made smaller or smaller and is necessary for the original control
The most suitable control can be performed using the rules
The advantage is obtained.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram illustrating a configuration of a control unit according to a first embodiment of the present invention. FIG. 2 is a configuration diagram schematically illustrating an image output unit according to the first embodiment. FIG. 3 is a schematic diagram showing a density patch in the first embodiment. FIG. 4 is a schematic diagram showing a position where a density patch is created in the first embodiment. FIG. 5 is a waveform chart showing an example of an output signal of a developing density sensor in the first embodiment. FIG. 6 is a schematic diagram showing transition of control cases and cluster formation obtained so far in the first embodiment. FIG. 7 is a schematic diagram showing a control rule and an operation amount range associated with the control rule in the first embodiment. FIG. 8 is a schematic diagram showing a manner of calculating the degree of conformity of the weighting method and the final control rule in the first embodiment. FIG. 9 is a schematic diagram showing a modification of the weighting method in the first embodiment. FIG. 10 is a block diagram illustrating a configuration of a control unit according to a second embodiment of the present invention. FIG. 11 is a schematic diagram showing a control rule and an average value of an operation amount associated with the control rule in the second embodiment. FIG. 12 is a schematic diagram showing a manner of calculating a degree of conformity of a weighting method and a final control rule in the second embodiment. FIG. 13 is a schematic diagram for explaining an outline of a conventional weighting method based on density (or density conversion value). FIG. 14 is a schematic diagram in which a part of FIG. 13 is enlarged in a conventional weighting method based on density (or density conversion value). [Description of Signs] 1 Laser output unit (image quality variable means) 2 Charger 10 Developing density sensor (detection means) 15 Grid power supply 16 Light quantity controller 24 Density comparator 25 Control case memory (Control case storage unit) 26 State quantity comparator ( Comparison means) 27 cluster memory 28 control rule calculator (control rule extraction means) 29 control rule memory (control amount calculation means) 30 control rule searcher (control amount calculation means) 31 control amount correction calculator (control amount calculation means) 32 operation amount memory (operation amount calculation means) 43 convergence detector 44 operation amount (max, min) memory (operation amount storage means) 60 control unit 61 operation amount (average) memory (operation amount storage means) 62 control rule memory

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-8-137146 (JP, A) JP-A-3-89373 (JP, A) JP-A-4-42172 (JP, A) JP-A-5-137 336318 (JP, A) JP-A-4-319971 (JP, A) JP-A-4-324870 (JP, A) JP-A-4-320278 (JP, A) JP-A-6-324557 (JP, A) JP-A-5-27565 (JP, A) JP-A-4-130471 (JP, A) JP-A-4-130472 (JP, A) (58) Fields studied (Int. Cl. ⁷ , DB name) G03G 15/00 303 G03G 21/00 370-512 G06F 15/18 550 H04N 1/00

Claims (1)

(57) [Claims] 1. An image quality changing means for changing the quality of an output image according to an operation amount, a control case storage means for storing a control case of an output image, and the control case storage Control rule extracting means for extracting a control rule from a plurality of control cases stored in the means, detecting means for detecting an output image quality and outputting a detection result as a control amount, comparing the control amount with a target quality, When the comparison result exceeds a preset allowable value, the control amount is stored in the control case storage means so that the control amount can be used for the next and subsequent controls. After the result exceeds the preset allowable value, when the control amount becomes a value corresponding to the target quality,
An operation amount storage unit for sequentially storing subsequent operation amounts; and an operation amount stored in the operation amount storage unit after judging the degree of conformity to the immediately preceding control case for each control rule extracted by the control rule extraction unit. On the basis of the,
An operation amount calculation unit that weights each control rule according to the degree of conformity and uses the result to obtain a new operation amount so that the control amount output by the detection unit becomes a value corresponding to the target quality. An image forming apparatus, comprising: supplying an operation amount obtained by the operation amount calculation unit to the image quality variable unit.