JPH04130358A - Self-diagnostic and self-repairing system for image formation device - Google Patents

Abstract

PURPOSE:To repair a fault only by operating a selected actuator means by deciding whether or not there is the fault and diagnosing the fault state, and executing a repair plan which can remove secondary influence. CONSTITUTION:A digital signal/symbol conversion part 11 converts digital signals from sensors 1a - 1c into symbols according to reference value data which are characteristic to the object machine and stored in an object model storage part 14 and a fault diagnostic part 12 evaluates the symbols to decide whether there is the fault or not, diagnose the fault, and simulate the fault state, thereby determining the fault symptom and fault cause of the object machine. Then a repair planning part 15 reasons a repairing method, develops plural repair plans derived by the reasoning on a mathematical model, and selects the repair plan which exerts the least secondary influence.

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention relates to a self-diagnosis and self-repair system for an image forming apparatus. More specifically, the present invention relates to a device or system that can self-diagnose the operational status of an image forming device and self-repair by utilizing artificial intelligence and knowledge engineering, which have been actively researched in recent years.

<Conventional technology> In the development field of precision machinery and industrial machinery, recent efforts have been made to save labor in maintenance work and extend automatic operation.
Artificial Intelligence
Research on expert systems using CE+ so-called AI) technology is actively being conducted. Some expert systems can self-diagnose whether a device has malfunctioned,
There are also some devices that self-repair any malfunctions that occur.

<Problems to be solved by the invention> However, in the fault diagnosis system using the conventional expert system, (a) the knowledge is not versatile and fault diagnosis for various targets cannot be performed, and (b) unknown faults occur. (c) As the target becomes more complex, the amount of knowledge required for fault diagnosis increases explosively.
Limitations were pointed out, including (d) difficulty in realizing it, and (d) difficulty in acquiring knowledge.

More specifically, conventional automatic adjustment systems and fault diagnosis systems basically operate corresponding actuators based on the outputs of the sensors. In other words, a type of automatic adjustment and failure diagnosis has been performed using a predetermined combination of sensors and actuators. Therefore, basically, each sensor corresponds to a specific actuator, and the relationship between the two is fixed.

Therefore, (1) The relationship between sensor parameters and actuator parameters must be clearly expressed numerically.

(2) For the reason mentioned in (1) above, the relationship between sensor parameters and actuator parameters is strongly dependent on the object and lacks versatility. In other words, it cannot be used for various objects.

(3) The relationship between parameters between each sensor or between parameters between each actuator is irrelevant to control.

Therefore, only simple control can be performed based only on the relationship between the corresponding sensor parameters and actuator parameters, and the types of failures that can be handled are limited in advance.

In other words, it is necessary to predict possible failures at the design stage and include a mechanism for dealing with such failures; unknown failures cannot be handled.

(4) For the reason (3) above, it is not possible to predict the side effects on the parameters of other actuators that may be caused by manipulating the parameters of any actuator.

There were problems such as.

In this way, in conventional automatic adjustment systems and fault diagnosis systems, predicted failure A is caused by sensor A and actuator A.
The predicted failure B is performed based on the set A of the sensor B and the actuator B, and the predicted failure C is performed based on the set B of the sensor B and the actuator B.
Fault diagnosis was performed based on the set C of the sensor C and actuator C, respectively, and the fault repair was performed based on the diagnosis.

The present invention was made against the background of such prior art, and an object of the present invention is to provide a new self-diagnosis and self-repair system for an image forming apparatus that eliminates the drawbacks of the prior art.

Means for Solving the Problems> The present invention is a self-diagnosis and self-repair system for an image forming apparatus that embodies image data and generates a visible image, and includes features that represent the characteristics of the image forming apparatus. Data, a storage means that stores qualitative data representing the image forming device as a combination of multiple elements, qualitatively representing the behavior or attributes of each element, and the connection relationship between each element, and fault diagnosis knowledge, data input means, a plurality of actuator means capable of changing the functional state of the image forming apparatus, and a plurality of actuator means capable of changing the functional state of the image forming apparatus; and a plurality of actuator means capable of changing the functional state of the image forming apparatus; A fault diagnosis means for determining whether the image forming apparatus is normal or abnormal based on the characteristic data, qualitative data, and fault diagnosis knowledge stored in In response to determining that the image forming apparatus is abnormal, the data updating means updates predetermined data among the characteristic data stored in the storage means based on the given state data, and the failure diagnosis means determines that the image forming apparatus is abnormal. In response to determining that
Selection means for selecting the actuator means to be activated based on qualitative data and fault diagnosis knowledge stored in the storage means, side effects that may occur when actuating the actuator means selected by the selection means simulating means for simulating, determining means for determining whether the side effects simulated by the simulating means can be removed by actuating an actuator other than the selected actuator means, and determining means. When it is determined that the side effects cannot be removed, the selection means again selects an actuator means other than the previously selected actuator means based on the qualitative data and fault diagnosis knowledge, and the simulation means The device is characterized by comprising control repetition means for simulating side effects that may occur at that time, and for controlling the determining means to determine whether or not the side effects can be removed. It is.

According to the present invention, in response to the status data being given from the data input means, the given status data is compared with, for example, a reference pig as characteristic data, and it is determined whether the given status data is relatively low or relatively low. It is determined that a failure has occurred depending on whether the

Furthermore, the current failure state is diagnosed based on the qualitative data.

If a fault has occurred, then, based on the qualitative data and the fault diagnosis knowledge, it is inferred which actuator means should be operated in order to repair the fault that has occurred, and the actuator means is selected.

Furthermore, it is simulated whether side effects occur when operating the selected actuator means. When simulating side effects, the fault condition diagnosed by the fault diagnosis means is taken into account.

If a side effect occurs as a result of the simulation by the simulating means, whether the generated side effect can be eliminated by operating an actuator means other than the selected actuator means. is determined.

As a result of the determination, if the side effects cannot be removed, the selection means selects an actuator means other than the previously selected actuator means again, and eliminates the side effects that may occur in that case. The control is repeated so that the simulating means simulates the effect, and the determining means determines whether or not the side effects can be removed.

If an actuator means that can eliminate side effects is selected, that actuator means is output as a repair actuator means.

Further, when the image forming apparatus is determined to be normal by the failure diagnosis means, predetermined data of the characteristic data stored in the storage means, such as reference value data, is determined based on the given status data. Characteristic data that is updated and always represents the latest state of the device is stored in the storage means.

Qualitative data is not data specific to the image forming apparatus, but data that is common to many image forming apparatuses and is expressed qualitatively. Therefore, the self-diagnosis and self-repair system according to the present invention It can be easily incorporated into different types of image forming apparatuses.

<Embodiment> Overview of System Configuration FIG. 1 is a block diagram showing the system configuration of an embodiment of the present invention. This system includes multiple sensors la, lb installed on the target machine.

A plurality of actuators 6a, 6b, and 6c are included for changing the functional states of the IC and the target machine.

The plurality of sensors la, lb, and lc are each for detecting a change in an element of the target machine or a state related to the machine elements caused by the operation of the target machine.

Information respectively taken in from the plurality of sensors la and 1b IC is amplified by an amplifier circuit 2, converted from an analog signal to a digital signal by an A/D conversion circuit 3, and provided to a system control circuit 10.

The system control circuit 10 includes a digital signal/symbol conversion section 11, a fault diagnosis section 12, a fault simulation section 13, a target model storage section 14, a repair planning section 15, and a symbol/digital signal conversion section 16.

The digital signal/symbol conversion section 11 is for converting the digital signal provided from the A/D conversion circuit 3 into qualitative information. That is, a digital signal, for example, normal.

A conversion function is provided to convert to any of the three symbols, high and low. Sensor la, lb,
Converting the signal provided from the lc into such qualitative information symbolized facilitates the approach to fault diagnosis. Note that the symbols are not limited to normal, high, and low as in this example, but may be other expressions such as on and off or A, B, C, and D. When the digital signal is converted into a symbol in the conversion unit 11, characteristic data specific to the target machine stored in the target model storage unit 14 is referred to. Details of this feature data and signal conversion will be described later.

The failure diagnosis section 12 and the failure simulation section 13 are
By comparing the symbols converted by the digital signal/symbol conversion unit 11 with the fault diagnosis knowledge stored in the target model storage unit 14, the presence or absence of a fault is determined, and fault diagnosis is performed. This is a component that expresses and outputs the failure state of the machine using qualitative information, that is, symbols.

The repair planning unit 15 is a component for inferring a repair plan for repairing a failure when a failure occurs and deriving a repair work. In inferring the repair plan and deriving the repair work, qualitative data (described in detail later) stored in the target model storage unit 14 is utilized.

The methods of fault diagnosis and fault simulation, inference of a repair plan, and derivation of a repair work in the fault diagnosis section 12, fault simulation section 13, and repair planning section 15 will be described in detail later.

The repair work output from the repair planning unit 15 is represented by the symbol /
In the digital signal conversion section 16, the stored information in the target model storage section 14 is referred to and converted into a digital signal.

The digital signal is then converted from a digital signal to an analog signal by a D/A conversion circuit 4 and is provided to an actuator control circuit 5. The actuator control circuit 5 controls a plurality of actuators 6a, 6b, based on an applied analog signal, that is, an actuator control command.
6c is selectively activated to perform repair work.

FIG. 2 is a flowchart showing the processing of the system control circuit 10 in FIG. Next, with reference to FIG. 2, an overview of the processing of the system control circuit 10 shown in FIG. 1 will be explained.

The detection signal of the sensor la lb or IC is amplified, converted into a digital signal, and read into the system control circuit 10, for example during a predetermined read cycle (step S]).

The read digital signal is converted into symbols in the digital signal/symbol converter 11 (step S2
). This symbolization is performed based on characteristic data preset in the target model storage unit 14, that is, reference value data specific to the target machine. For example, the output ranges of the sensors la, lb, and lc are set in the target model storage unit 14 as reference value data specific to the target machine as follows.

That is, less than sensor 1a - low ~ ka2 = more than normal - less than high - low ~ kb2 - normal: output ka output ka output ka2 : output kb + output kb sensor 1b more than output kb2 - no \ sensor IC = less than output kc1 - Low output kcl ~1 (c2 - exceeds normal output kC2 - set to high. In the digital/symbol conversion unit 11, based on reference value data specific to the target machine set in the target model storage unit 14 Then, sensor 1a~IC
Converts the digital signal from the device into a symbol of ``low,''``normal,'' or ``high.''

Next, the converted symbol is evaluated in the fault diagnosis section 12, and the presence or absence of a fault is determined and the fault symptoms are specified (step S3). The target model storage unit 1 is used to determine the presence or absence of a failure by evaluating symbols and to identify failure symptoms.
The fault diagnosis knowledge stored in 4 is utilized. Fault diagnosis knowledge is, for example, a setting condition that a specific parameter must be normal, for example. If the specific parameter is not normal, there is a failure,
The failure symptom is determined based on the specific parameter. If there is no failure, step Sl
, S2 and S3 are repeated.

If it is determined in step S3 that there is a failure,
The state of the target machine is inferred, that is, fault diagnosis and fault state simulation are performed (step S4). Specifically, the failure diagnosis unit 12 uses the qualitative data stored in the target model storage unit 14 that qualitatively represents the behavior or attributes of each element constituting the device and the coupling relationship between each element. , the parameter causing the failure is searched, and the failure state is simulated in the failure simulation section 13 on the assumption that the searched parameter is the cause of the failure.

Further, in the failure diagnosis section 12, the simulation results are compared with the current parameter values, and the validity of the assumption that the searched parameter is the cause of the failure is determined. The above processing is performed on a plurality of parameters to be searched.

As a result of the failure determination, failure diagnosis, and failure state simulation, the failure symptoms and failure causes of the target machine are determined (step S5). Here, failure symptoms include the output status of the target machine (for example, in the case of a copying machine,
The cause of the failure is,
This is a change in the mechanism or structure of the target machine that causes the symbol to change (for example, in the case of a copying machine, a ``decrease in the light intensity of a halogen lamp'', etc.).

Next, the repair planning unit 15 infers a repair method. In this inference, a simulation of the influence of repair is also performed (step S6).

Then, a repair plan is determined (step S7), and control instructions are developed based on the determined repair plan (step S8). When developing actual control commands, characteristic data unique to the device, such as limit values for actuator operation, is read out from the target model storage unit 14 and utilized.

The developed control command is converted into an analog signal and given to the actuator control circuit 5, and repair control is executed (step S9). Control then ends.

Next, we will explain how to infer failure diagnosis and repair plans.
A detailed explanation will be given with reference to specific examples.

In the following explanation, - as an example, a small ordinary copying machine -
The following describes how to use the peripheral area of a photoreceptor drum as the target machine.

FIG. 13 is an illustrative diagram showing a specific target machine. In FIG. 3, 21 is a photosensitive drum, 22 is a main charger, 23 is a /% logen lamp for illuminating the original, 24 is a developing device, and 25 is a transfer charger.

In this embodiment, for example, there are three sensors la.

lb and lc are provided. That is, sensor 1a is an AE sensor for measuring the amount of light incident on the photoreceptor drum, sensor 1b is a surface potential sensor for measuring the surface potential of the photoreceptor drum, and sensor IC measures the density of the image copied on paper. This is a concentration meter for

Although not shown in FIG. 3, three types of actuators are provided. That is, the main charging volume VR1 is used to change the main charging voltage of the photoreceptor drum, and the lamp volume A is used to control the light amount of the Haloken lamp.
Three volumes are provided as actuators: VR and a transfer volume VR2 for controlling the transfer voltage between the photosensitive drum and copy paper.

By the way, the target machine shown in Fig. 3 can be viewed from a physical perspective, and the target machine can be expressed at the physical level as a combination of multiple elements, and the behavior and attributes of each element and the connection relationship between each element can be expressed using parameters. When expressed qualitatively using this method, it is as shown in Table 1. The expression format shown in Table 1 will be referred to as a "substantive model."

Furthermore, the representation shown in FIG. 4 in which the entity model is abstracted and expressed as a connection tree of each parameter will be referred to as a "mathematical model."

Then, by combining the "substantive model" and the "mathematical model",
We will call it the "target model". [Target model] is qualitative data common to image forming apparatuses, which is also utilized for failure repair, which will be described later.

(The following is a blank space) Table 1 "Substantive Model" The contents of the substantive model and the mathematical model as qualitative data are stored in the target model storage unit 14.

The target model storage unit 14 also stores information regarding predetermined parameters among the parameters included in the real model.
For example, reference value data measured at the time of factory shipment is stored. This reference value data is characteristic data specific to this image forming apparatus.

For example, in this machine, as shown in FIG. 5, for the parameters x, v, , o, , v,
Reference value data specifying the normal and high ranges is stored.

In this embodiment, the above-mentioned reference value data can be updated in response to sensing data during subsequent failure diagnosis and failure repair processes, changes in the operating state of the machine, and the like.

Furthermore, the target model storage unit 14 stores evaluation function knowledge as an example of failure diagnosis knowledge that serves as a standard for determining whether or not the target machine is operating normally, based on the converted symbol. ing.

Note that the evaluation function knowledge, in other words, the failure diagnosis knowledge, may be specific to the target device, or may not be specific and may be common to a wide range of image forming apparatuses.

Evaluation function knowledge includes the following knowledge.

Image density O, -normal, fogging degree O1・normal, separation performance S, <normal here, 0. ．． 0.・If S does not meet the above conditions, the target machine is not operating normally.

Now, consider a case where the digitized sensor information of the target machine during normal operation has the following values.

AE sensor value X-30 Surface potential sensor value V, -300 Densitometer value o, -7 Also, densitometer value O1 when using a blank document with optical density D-0 - fog degree 0.1 , the value V of the surface potential sensor with the halogen lamp turned off
, - dark potential V. , and their values are, respectively, fogging degree O1·-50 dark potential V, -700.

The fogging degree O1 and the dark potential V2 may be measured manually, or may be measured by a sensor under certain conditions, for example, each time the target machine is turned on or before copying starts. It may be programmed to be measured automatically. In this embodiment, the latter is adopted.

AE sensor la, surface potential sensor 1b and concentration meter IC
Multi-valued X, V, ,O,,O57,■, obtained by
, are respectively digital signal/symbol converters 11
is converted to a symbol at .

The conversion is performed for each sensor la, lb or I as described above.
The digital value given from C is stored in the target model storage unit 1.
This is done by comparing it with reference value data as feature data stored in 4, and converting it into one of three types of symbols: normal, high, or low.

In this example, each parameter is symbolized as follows.

X-High■, -LowOl-LowVIl-Normal In the fault diagnosis section 12, each of these symbolized parameters is combined with functional evaluation knowledge as an example of fault diagnosis knowledge stored in the target model storage section 14. be compared. As a result, since the image density O1 is not normal, it is determined that there is a failure, and the failure symptom is ``Image density is too low (0
．． − low)”. And next,
0. Fault diagnosis, that is, inference of the cause of the fault, is performed using "-" as a fault symptom.

Fault diagnosis is first performed in the fault simulation section 13 using the mathematical model shown in FIG. 4 to search for parameters that may cause the error.

In the mathematical model in Figure 4, 0. Fig. 6 shows the parameters that have the potential to reduce the . In Figure 6, the parameters marked with upward or downward arrows are parameters that may cause parameter O6-low, and those with upward arrows are parameters that may cause parameter O6-low. If that parameter decreases, 0. -Causes low.

Next, each parameter,ζ,D,that can cause a 0,-rho was explored in the mathematical model.

V+ 70 Vb, V,, V,, X, β, HL
.

Regarding D, the failure diagnosis unit 12 detects the cause of the parameter change.

This detection is performed based on the entity model shown in Table 1, and in this embodiment, the following failure cause candidates are inferred.

That is, V, = low, - defective transfer transformer ζ - deterioration of low knee paper ■, - high: → defective developing bias γ. - Low 21 toner deterioration V, = Low - Defective main charging voltage H5 - Defective high knee Haloken lamp setting D - Port -:
→The original is thin. Among the parameters, β is the sensitivity of the photoreceptor, and since it does not increase, it is excluded.

Since D, , V, and X are represented by other parameters, they are also excluded.

Then, in response to the above inference made in the fault diagnosis section 12, a fault state simulation is performed in the fault simulation section 13.

Simulating a fault state means inferring the state of the target machine when the above-mentioned inferred fault occurs. More specifically, it is assumed that the cause of 0-low, that is, the cause of failure is, for example, a defective transfer transformer, and Vl-low is set for a model in a normal state. Then, the influence on each parameter in that state is examined using a mathematical model. ■、−
If low is set, 0. -low and SP-low, and all other parameters are normal, so this contradicts the X=high and V,-low obtained from the sensor. Therefore, the result is that the inference regarding the cause of the failure is incorrect.

Similarly, ζ-rho is set on the normal state mathematical model and the result is compared with the symbols obtained from the sensor. In this case as well, on the mathematical model, for X-normal,
Since the symbol from the sensor is X-high, there is a contradiction and the inference of the cause of the failure is determined to be incorrect.

In this way, a simulation of the failure state is performed for all failure cause candidates, and it is confirmed whether or not the inference of the failure cause is correct.

As a result, in this example, when the cause of the failure is determined to be "incorrect halogen lamp settings (HL-high)", a result that matches the actual condition of the target machine can be obtained, and other causes of the failure can be determined. It is concluded that all candidates are inconsistent with the actual state of the device.

Therefore, it can be concluded that the cause of the failure in this case is improper setting of the halogen lamp. Table 2 shows the status of each parameter of the target machine at that time.

Table 2: Halogen setting failure If the states of the parameters shown in Table 2 are traced onto a mathematical model, FIG. 7 is obtained. In FIG. 7, the downward arrow attached to the right side of each parameter indicates low, the upward arrow indicates high, and N indicates normal.

Next, we will explain the reasoning behind the restoration plan.

As a result of fault determination, "Image density is too low (0°-low)"
” was taken up as a failure symptom, the goal of repair was to
○, is to increase.

Therefore, from the relationship on the mathematical model shown in Figure 4, it is possible to increase the repair target 0 by increasing D, ■, or ζ. It is inferred.

Next, when inference is made with the goal of increasing D, ■
Raise , or lower ■, or γ.

Either rise or get a conclusion. in this way,
By reasoning or iterating on the basis of a mathematical model,
Candidates for repair operations can be obtained on the mathematical model. The results obtained are shown in Table 3.

(Margin below) Table 3 By the way, among the repair candidates obtained based on the mathematical model, there are some that can be realized and some that cannot be realized. For example, D: The optical density of the original cannot be changed, and β: The sensitivity of the photoreceptor is also difficult to change.

γ. :Toner sensitivity cannot be changed, ζ・Paper sensitivity is also unchangeable.

Also, in this specific example, ■,: Bias voltage It is also impossible to change because there is no actuator.

Of course, by adding an actuator, ■,
can be made changeable.

Furthermore, X: Logarithm of the amount of light reflected from the original, ■: The surface potential of the drum after exposure, and the toner density on the double drum cannot be changed themselves, but can be changed indirectly by changing other parameters. Since it can only be changed, it is excluded from the repair candidates here.

Although not directly related to this specific example, A1. :The amplitude of the separation AC voltage can also be changed by adding an actuator.

Depending on the above, in this specific example, as a repair candidate, ■・：Transfer voltage vo: Surface potential after main charging HL: Logarithm of halogen lamp output light intensity will be taken up.

On the other hand, the target model storage unit 14 stores in advance the following knowledge as repair plan knowledge. That is, (a) Increase the control voltage of the transfer transformer to increase V. (b) Decrease V → Decrease the control voltage of the transfer transformer. (c) Increase the control voltage of the lifetime charging transformer to increase V. (d) Decrease V, lower the control voltage of the lifetime charging transformer (e) Increase HL - shift the halogen lamp control signal to the high voltage side (f) Decrease HL - shift the halogen lamp control signal to a low voltage Shift to the side.

It is. The repair plan knowledge stored in the target model storage unit 14 is characteristic data specific to this device. By applying the repair plan knowledge to the repair candidates obtained based on the mathematical model, 0. As a repair operation to increase, (a) Increase the control voltage of the transfer transformer to increase V. (c) Increase the control voltage of the charging transformer to increase Vfi. (f) Decrease HL → Halogen lamp. Three methods of shifting the control signal to the lower voltage side are available.

Image density 0. If you just want to increase
Repair is possible by performing any of the methods.

However, the target machine has an image density of 0. It is conceivable that increasing the amount of water will have various side effects. Therefore, in this embodiment, as explained below, the inference of secondary effects is performed based on a mathematical model.

When the three repair plans derived in the inference of the side effect repair plan are developed on a mathematical model, Figures 8 to 13 are obtained. In other words, (a) Figures 8 and 9 show the case where V is increased (Figure 9 shows 0.. when D-0 is expressed on the mathematical model), (c) When V is increased, as shown in Figs. 10 and 11 (Fig. 11 shows 05 when D-0 is represented on the mathematical model), (f
) When HL is lowered, the results are shown in FIGS. 12 and 13 (in FIG. 13, 05 when set to D-0 is represented on the mathematical model).

Then, when a functional evaluation is performed based on the mathematical model, the following state is inferred.

That is, (1) When V is increased (Figures 8 and 9) (a)
The output image density increases.

(b) When D-00, there is a case where Ol.>Normal.

In other words, fogging may occur.

(c) S becomes > normal, and there is a possibility that a separation failure will occur.

(2) When V is increased (Figures 10 and 11) (a) Output image density increases.

(b) When D-00, O6.>Normal, and fogging may occur.

(3) When HL is lowered (FIGS. 12 and 13) (a) Only the output image density increases, and there are no other side effects.

Therefore, the repair planning unit 15 selects a repair plan that has the least side effects, that is, a repair plan that lowers the HL. This repair plan is consistent with the operation to eliminate the cause of the failure found in the failure diagnosis.

In other words, looking at it from a different perspective, inferring the cause of a failure in fault diagnosis involves tracing the actual state of the device when it fails on a mathematical model and understanding the state of each component when the device fails. In contrast to the previous method of estimating the cause of a failure, estimating a repair plan is based on the assumption that the device is not in a faulty state, but rather that the device is normal, tracing the state of the device on a mathematical model, and performing repairs based on that. Reasoning plans.

In the above-mentioned specific example, the same failure cause and repair plan were obtained both in the inference in the failure diagnosis and in the inference in the repair plan.

However, in some cases, inferences for fault diagnosis are based on devices in a faulty state, while inferences for repair plans are based on devices in a normal state, so the results obtained may differ between the two. .

In such a case, if only those conclusions that do not contradict the conclusions obtained in the inference process of failure diagnosis are selected when inferring a repair plan, the inference process of the repair plan can be performed in a shorter time.

In the above case, if the repair plan of lowering HL cannot be selected, for example, if the AVH volume for shifting the halogen lamp control signal to the lower voltage side is already at the lower limit, then V of (2) has little impact. A restoration plan is selected to increase the

However, if the repair plan is selected to increase ■, , it is predicted that there will be a side effect of the possibility of fogging, so in the mathematical model shown in Figure 11, in order to decrease Ol Which parameters should be manipulated is examined based on the mathematical model of the 11th test, and the manipulation is selected based on the knowledge of the repair plan. As a result, it is selected whether to raise HL, lower Vo, or lower (2), and a repair plan including prevention of fogging is carried out.

That is, as shown in FIG. 14, the reasoning for the repair operation is developed assuming side effects. In the inference development of repair operations as shown in Figure 14, (a) branches that contradict the previous repair plan on the mathematical model are not selected (b) branches with the least side effects are selected (c ) Unrolling is performed based on the knowledge that the loop that formed it will stop unrolling at that point.

In Figure 14, (1) ■. Two repair plans remain: the loop of ↑−H1↑−V, ↑, and the loop of (2:)V, ↑−v1 ↓→V, ↑.

Now, suppose that when loop (1) is performed as a restoration plan, the image density becomes an appropriate density, that is, 0, or normal. In such a case,■. Since HL and HL have been increased, the value of the surface potential measured by the sensor 1b changes to a much higher value than the value initially measured in the post-repair state where the image density returns to O1 or normal. It should be there. However, since the repair operation was successful, the state of the parameter V after repair must be symbolized normally. Therefore, in such a case, when the repair is completed, the reference data for symbolizing the parameter V shown in FIG. 5 is changed based on the measurement values determined by the sensor 1b, for example, the data shown in FIG. can be rewritten as

In this way, the reference data is updated as needed after the repair work is completed.

In this embodiment, the above-mentioned (1) in FIG.
Specifically, when the main charging volume VRI is operated to increase the surface potential of the photoreceptor drum 21, and if a rash occurs on the resulting copy, the lamp volume AVR is operated to increase the halogen charge voltage. The light intensity of the lamp is increased and the image density of the copy is diluted.

Then, while increasing the main charging volume VRI and the lamp volume AVR alternately and appropriately, when the image density became normal, that is, when the parameter O1 became normal, it was obtained from the detection output of the densitometer, which is the sensor IC. At this point, the repair process is terminated.

Furthermore, if the above two repair plans are infeasible, the repair plan (3) of increasing vl is selected, and the two side effects of this are fogging and poor separation. A failure diagnosis is performed based on the assumption that a repair plan is selected.

The selected repair plan is then executed, and in the case of loop processing, failure is determined when the operation of the parameters on the loop reaches its limit.

In addition, in the case of this embodiment, as explained in the specific example, the completion of repair is determined when the state becomes O or normal.
In this state, the repair is stopped.

In the inference of the side effects described above, the three repair plans derived in the inference of the repair plan are sequentially developed on a mathematical model, and the side effects are collectively inferred for each of the three repair plans. ing.

Instead of this method of inferring side effects, the following processing may be performed.

That is, suppose that, for example, three repair plans are derived in the repair plan inference. In that case, take only one repair plan out of the three, simulate the side effects that may occur if the actuator means is actuated based on that repair plan, and It is determined whether the negative influence can be removed by actuating an actuator means other than the actuator means selected by the repair plan.

When it is determined that the side effects can be eliminated, the actuators selected by the repair plan are actually operated to carry out the repair, and the side effects are eliminated by activating other actuator means. It removes it.

As a result, there is no need to simulate the side effects based on the other two plans derived from the repair plan, and the overall repair operation time can be shortened.

In the above case, if for the first selected repair plan, a side effect is simulated and it is determined that the simulated side effect cannot be removed by actuating another actuator means. If it is done,
The first repair plan is abandoned and a second repair plan is taken up, and the side effects that may occur if the actuator means selected based on the second repair plan are actuated. simulating secondary effects, determining whether the simulated secondary effects can be removed by actuating an actuator means other than the actuator means; and when the secondary effects can be removed; , performs repair work based on the second repair plan.

In this way, if the first repair plan is extracted from among the multiple repair plans derived in the inference of the repair plan, the side effects in that case are inferred, and the side effects can be removed. Then, immediately execute the repair based on the first repair plan.

If that repair plan has too many side effects, it is abandoned, the next repair plan is selected, and the side effects in that case are simulated.

In such a case, it is preferable to select which repair plan first among the plurality of repair plans derived in the inference of the repair plan, taking into account the cause of the failure obtained in the failure diagnosis, for example.

In the above embodiments, the repair itself is considerably limited due to the small number of actuator parameters, but the flexibility and possibility of repair can be further improved by increasing the number of actuator parameters.

In the specific example explained above, if any repair work is successful, the state of the equipment after the success is determined to be normal, so it is determined based on the digital data values provided from each sensor. It is preferable that the reference value data of each parameter (the reference values shown in FIG. 5) be updated and the parameters be symbolized based on the new reference value data.

Further, in the above-mentioned specific example, although the operating range of each actuator was not particularly mentioned, the target model storage unit 14
If the operating range data that sets the actuator's operating range is included in the device-specific characteristic data stored in the device, the actuator can be operated when the actuator's output state is within the stored operating range. When the output state of the actuator reaches the upper or lower limit of the stored operating range, it can be determined that the actuator cannot be operated, and this can be used to determine whether the repair work is correct or not.

Furthermore, in the above-mentioned specific example, a system that automatically performs self-diagnosis and self-repair based on changes in sensor output was introduced, but if the image forming apparatus is provided with a self-diagnosis mode setting key, etc. Self-diagnosis and/or self-repair may be performed only when the self-diagnosis mode setting key is operated.

In the above specific example, the device is completely autonomous, that is, the device itself automatically performs self-diagnosis and self-repairs in the event of a failure, without any operation by service personnel or users. The system was explained. However, in the present invention, the sensor is removed from the configuration requirements, and a service person or the like can measure functional status data at predetermined locations of the image forming apparatus and input the measured data into the apparatus. By doing so, it is possible to configure an image forming apparatus that can perform non-autonomous self-diagnosis and perform autonomous repair based on the self-diagnosis.

In addition, if you configure a system that only selects the actuator to repair a failure based on the results of the self-diagnosis performed by the device, but displays the actuator that should be operated without actually operating the actuator, service The image forming apparatus may include a non-autonomous repair system that only requires the operator to operate the displayed actuator.

Of course, it is also possible to exclude the configuration requirements of the self-repair system and configure an image forming apparatus having only the self-diagnosis system.

That is, according to the present invention, (1) an image forming apparatus having a fully autonomous self-diagnosis and self-healing system; (2) an image forming apparatus having an autonomous self-diagnosis system and a non-autonomous self-healing system; (3) (4) an image forming device having a non-autonomous self-diagnosis system and a non-autonomous self-healing system; (4) an image forming device having a non-autonomous self-diagnosis system and an autonomous self-healing system; or (5) an image forming device having only an autonomous self-diagnosis system. The forming device can be configured as desired.

<Effects of the Invention> According to the present invention, when the image forming apparatus is in an abnormal state, it is possible to determine whether or not there is a failure by simply measuring the status data and inputting it from the data input means. If so, the fault condition is diagnosed, the actuator means to be operated to repair the fault is inferred and output, and the side effects that may occur when the actuator means is operated are determined. simulated.

If a side effect occurs, it is determined whether the generated side effect can be removed by operating another actuator means.

As a result of the determination, if the side effect cannot be removed, another actuator means is selected by the selection means again, and it is simulated whether or not the side effect occurs in that case, and If side effects occur, it is determined whether they can be eliminated.

Finally, if there is a repair plan that can eliminate the side effects, actuator means for executing the repair plan is output.

Therefore, a service person or the like can easily perform repairs simply by operating the selected actuator means that are output.

Therefore, it is possible to provide an image forming apparatus that can perform self-diagnosis and autonomous repair reasoning using the input of status data as a trigger to perform repair.

Further, according to the present invention, since the cause of failure is determined based on qualitative data common to image forming apparatuses, a self-diagnosis and self-repair system that can handle unknown failures that are not explicitly described is provided. The image forming apparatus may have the following features.

Furthermore, the self-diagnosis and self-repair system according to the present invention can be commonly applied to many types of image forming apparatuses, rather than to a specific image forming apparatus, and as a result, is inexpensive. An imaging device having a self-diagnosis and self-repair system can be provided.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the system configuration of an embodiment of the present invention. FIG. 2 is a flowchart showing the operation of the control circuit in FIG. FIG. 3 is a diagram showing a schematic configuration of the present invention when used in a plain paper copying machine. FIG. 4 is a diagram representing a mathematical model of this embodiment. FIG. 5 is a diagram showing reference value data of each parameter necessary when each parameter is symbolized. FIGS. 6 and 7 are diagrams showing development for fault diagnosis on a mathematical model. FIGS. 8 to 13 are diagrams showing development for inferring side effects on a mathematical model. FIG. 14 is a diagram showing operations when selecting a repair plan. FIG. 15 is a diagram showing updated reference value data. In the figure, la, lb, lc-sensor, 6a6b,
6c... Actuator, 10... System control circuit, 11... Digital signal/symbol converter, 12
. . . Fault diagnosis section, 13 . . . Fault simulation section, 14 . . . Target model storage section, 15 .

Claims (1)

[Claims] 1. A self-diagnosis and self-repair system for an image forming apparatus that embodies image data to generate a visible image, comprising: characteristic data representing characteristics of the image forming apparatus; storage means, data input means, and image forming apparatus in which qualitative data and fault diagnosis knowledge are stored that represent the behavior or attributes of each element and the connection relationship between the elements, expressed as a combination of a plurality of elements; a plurality of actuator means capable of changing the functional state of the image forming apparatus, in response to being provided with state data representative of the functional state of the image forming apparatus from the data input means, the state data and the characteristics stored in the storage means; Failure diagnosis means for determining whether the image forming apparatus is normal or abnormal and diagnosing its condition based on data, qualitative data, and failure diagnosis knowledge, and that the failure diagnosis means has determined that the image forming apparatus is normal. data updating means for updating predetermined data of the characteristic data stored in the storage means based on given status data; and a failure diagnosis means determining that the image forming apparatus is abnormal. a selection means for selecting the actuator means to be actuated based on the qualitative data and fault diagnosis knowledge stored in the storage means; simulating means for simulating side effects that are not present; determining whether the side effects simulated by the simulating means can be removed by actuating an actuator other than the selected actuator means; When the determination means and the determination means determine that the side effects cannot be removed, the selection means again selects an actuator means other than the previously selected actuator means based on qualitative data and fault diagnosis knowledge. control repetition means for causing the simulating means to simulate side effects that may occur at that time, and controlling the determining means to determine whether or not the side effects can be removed. A self-diagnosis and self-repair system for an image forming apparatus, characterized in that: