JPH07113901B2 - Self-diagnosis device for medium transport system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial field of use) The present invention relates to a medium transport system that automatically diagnoses the operating state of the medium transport system, predicts the occurrence of a failure requiring maintenance, and gives a warning in advance. The present invention relates to a self-diagnosis device, especially to a self-diagnosis device suitable for application to an automatic transaction device.

(Prior Art) Conventionally, maintenance of an apparatus is performed by post maintenance.
The post-maintenance is to perform a process such as, when a failure occurs in the device, searching for the failed part and repairing or replacing the part. As an example, the maintenance of the bill depositing / dispensing machine of the automatic transaction device will be described. The banknote depositing / dispensing machine separates the deposited banknotes one by one, and the banknote discriminating section discriminates the authenticity and denomination of the banknotes and stores them in the corresponding banknote storing section, and the operation is almost opposite to this operation. It has a withdrawal function of withdrawing banknotes along a route. In such a banknote deposit / withdrawal machine,
Problems such as skew and double feeding of bills and abnormalities of bill discrimination sensors may cause jams and faults of discrimination. Oblique or double feeding of banknotes is caused by abrasion or misalignment of the separation roller provided on the banknote conveyance path, and abnormality of the discrimination sensor is caused by aging of the sensor or dust attached to the sensor. Often becomes. Therefore, conventionally, when such a failure occurs, the cause of the failure is removed according to the status of the failure.

(Problems to be Solved by the Invention) However, in the above-mentioned conventional post-maintenance, it is natural that the operating rate and the reliability of the apparatus are limited to some extent. For example, in a financial-related automatic transaction device, maintenance is performed by stopping the device each time a failure occurs, but this is not preferable in terms of service to customers, and from the viewpoint of the financial institution that operates the device, The operating rate and reliability are not sufficient.

Therefore, the present invention aims to solve these problems.

(Means for Solving the Problems) According to the first invention of the present application, there is provided a self-diagnosis apparatus for a medium transport system, comprising a roller for moving a medium and a separation roller for separating the medium, A traveling sensor that is provided in the vicinity of each of the roller and the separation roller to read the traveling state of the medium and the sensor information output from the traveling sensor detects a traveling abnormality of the medium and responds to the detected traveling abnormality. A detection unit that outputs the detection information of the operation abnormality of the roller and the separation roller, a data collection unit that temporarily stores the detection information output by the detection unit, and the detection information that is stored in the data collection unit. From the data processing unit that calculates the occurrence ratio of the operation abnormality of the roller and the separation roller as the occurrence frequency information, and the occurrence frequency information calculated by the data processing unit. Memory for accumulating information, a comparison reference value storage unit in which a comparison reference value predetermined as an operation abnormality allowable limit is stored for each diagnosis target portion, the roller stored in the comparison reference value storage unit, and Danger forecast information indicating whether the comparison reference value of the separation roller and the value of the occurrence frequency region of the operation abnormality of the roller and the separation roller are respectively compared, and whether the value of the occurrence frequency information exceeds each operation abnormality allowable limit And a control unit that sets the risk forecast information in association with the occurrence frequency information stored in the memory, and displays the risk forecast of the roller and the separation roller based on the risk forecast information. Provided is a self-diagnosis device for a medium transport system including a pre-alarm display unit.

According to a second aspect of the present invention, there is provided a self-diagnosis device for a medium transport system, comprising a roller for moving a medium and a separation roller for separating the medium, the self-diagnosis apparatus being provided near the roller and the separation roller, respectively. A traveling sensor that reads the traveling state of the medium and sensor information output from the traveling sensor detect a traveling abnormality of the medium, and the operation of the roller and the separation roller according to the detected traveling abnormality, respectively. A detection unit that outputs as abnormality detection information, a sensor level detection unit that detects the sensor information output from the traveling sensor at the time of checking the sensor output level and outputs it as sensor level information, and the detection information output from the detection unit and A data collection unit that temporarily stores the sensor level information output from the sensor level detection unit; A data processing unit that calculates the occurrence ratio of the operation abnormality of the roller and the separation roller as the occurrence frequency information from the detection information accumulated in the collection unit, a memory that accumulates the occurrence frequency information calculated by the data processing unit, and an operation. A comparison reference value storage unit in which a comparison reference value predetermined as an abnormality allowable limit is stored for each diagnosis target site, a comparison reference value of the roller and the separation roller stored in the comparison reference value storage unit, and the comparison reference value storage unit. The value of the occurrence frequency information of the operation abnormality of the roller and the separation roller is respectively compared, and the comparison reference value of the sensor level stored in the comparison reference value storage unit is compared with the value of the sensor level information to generate the occurrence. A comparison unit that outputs danger forecast information indicating whether the value of the frequency information or the value of the sensor level information exceeds the respective operation abnormality allowable limit; and the danger forecast information A control unit that sets the occurrence frequency information and the sensor level information stored in the memory in association with each other, and a pre-alarm display unit that displays the danger forecasts of the roller, the separation roller, and the traveling sensor based on the danger forecast information. There is provided a self-diagnosis device for a medium transport system, comprising:

(Operation) In the first invention of the present application, the traveling abnormality of the medium is detected based on the sensor information output from the traveling sensor which is provided near the roller and the separation roller and which reads the traveling state of the medium. Is output as the detection information of the operation abnormality of the roller and the separation roller. The occurrence ratio of the operation abnormality is calculated as the occurrence frequency information from the detection information. As a result, the detection information is standardized as the occurrence ratio of the operation abnormality and stored in the memory. Therefore, by extracting the data from this memory, it becomes possible to know the raw data of the occurrence frequency information. The predetermined comparison reference value of the roller and the separation roller is compared with the value of the occurrence frequency information of the operation abnormality of the roller and the separation roller, respectively, and the danger forecast information is output, and the danger forecast information is displayed on the pre-alarm display unit. Displayed for each separation roller. The developer or quality manager judges the weight of each data based on the data obtained by experiments, etc., and then uses the comparison reference value calculated as the operation abnormality permissible limit, so that the danger forecast information can be centralized. Can be output.

In the second invention of the present application, in addition to the configuration of the first invention,
Since the sensor level detection unit that detects the sensor information output from the travel sensor at the time of checking the sensor output level and outputs the sensor information as the sensor level information is provided, it is possible to detect the information of the output of the travel sensor itself. In this case, as the comparison reference value, the comparison reference value of the sensor level corresponding to the sensor level reference value is also stored for each diagnosis target part, and the comparison reference value of the sensor level and the value of the sensor level information are also stored. By comparison, the danger forecast information based on the value of the sensor level information is also output. Therefore, it is possible to display both the danger forecast based on the operation abnormality and the danger forecast based on the sensor abnormality as the danger forecast displayed on the pre-alarm display unit. It is possible to determine whether the operation is truly abnormal or the travel sensor is abnormal.

(Embodiment) First, the principle of the present invention will be described by taking a bill depositing / dispensing machine incorporated in an automatic transaction device as an example.

As described above, the banknote pay-in / pay-out device includes a roller for running the banknotes, a separation roller for separating the banknotes from a plurality of banknotes one by one, and the like. Due to various causes such as changes over time such as wear of rollers and misalignment, etc., these parts and units may cause problems such as skewing and double feeding of banknotes, resulting in jams and other problems. To do.

Therefore, the present invention constantly diagnoses the operating state of the apparatus by grasping the frequency of occurrence of operation abnormality such as skew feeding and double feeding of bills, and the occurrence of unacceptable failure in the apparatus, in other words, maintenance is required. It is to predict the state and perform maintenance.

The principle of maintenance based on the frequency of occurrence of operation abnormalities such as skew feeding and double feeding of bills will be described below. Regarding the types of operation abnormalities, those that require immediate maintenance by a maintenance person when abnormalities occur (referred to as failures in this description), those that can be easily removed by an operator or personnel, and Depending on the frequency of occurrence such as skew feeding and double feeding, it can be roughly divided into those requiring maintenance. Here, since the failure occurrence is predicted, the latter two operation abnormalities are targeted. Therefore, the occurrence frequency of the same operation abnormality is grasped, and when it exceeds the allowable limit (comparison reference value), a pre-alarm indicating that maintenance is required for the parts and units having these occurrence causes is set. In this case, skewing and double feeding may be due to the cause of the device itself, or to the cause of defective banknotes, etc., although there is no problem with the device itself, so the comparison reference value is set in consideration of this point. To be done. Further, it is preferable that the comparison reference value is a value with some margin. The frequency of occurrence is, for example, 1 for skew.
It can be regarded as the number of skew occurrences with respect to the number of processed bills on a day.

FIG. 1 is a diagram showing the procedure of the self-diagnosis method according to the present invention based on the above-described principle. First, in step 100, data capable of diagnosing the operating state of the device is collected and processed to form diagnostic data, that is, the occurrence frequency information of the same kind of operation abnormality (step 100). Then, the obtained occurrence frequency information is stored in the memory (step 101). Next, the occurrence frequency information is compared with the comparison reference value (step 102). In the following step 103, when the occurrence frequency information exceeds the comparison reference value, a danger forecast (pre-alarm) is set to notify the maintenance personnel of the arrival of the time when maintenance is required (step 10).
Four). This setting is the diagnostic data stored in memory,
That is, this is done by setting a flag in the occurrence frequency information. If there is a display request (step 105),
A pre-alarm is displayed to maintenance personnel (step 106). This display is performed, for example, every day of regular inspection of the device.

Next, an example in which the present invention is applied to a bill depositing / dispensing machine of the automatic transaction device will be described. FIG. 2 is a block diagram showing the hardware configuration of this example. In the figure, 10 is a travel sensor. A large number of light receiving sensors are provided in addition to a light source in the vicinity of a bill traveling path and rollers of the bill depositing / dispensing machine. Here, these light receiving sensors are collectively shown as a traveling sensor as one block. Reference numeral 12 is a sensor level detection unit. The sensor level detection unit 12 detects the output level of each traveling sensor (light receiving sensor).
Reference numeral 14 is a running number detecting unit. The traveling sheet number detection unit 14 detects an output signal from a light receiving sensor located in the bill separating / feeding unit, specifically, a signal that is turned off when a bill passes therethrough. Reference numeral 16 is a skewed sheet number detection unit. The skewed sheet number detecting unit 16 detects skewing based on an output signal from a light receiving sensor provided in the vicinity of a traveling path for detecting skewing and a separation roller for separating one banknote. In particular,
Oblique detection is performed by detecting a difference in generation time of output signals from two light receiving sensors that are spaced apart in a direction perpendicular to the traveling direction of bills. Reference numeral 18 is a double feed detection unit. The double-feed detection unit 18 detects the double-feed of the bill based on the output signal from the light receiving sensor that detects the thickness of the bill when the bill travels. 20 is a discrimination sensor, which outputs a signal for discriminating the authenticity and denomination of a bill. An identification unit 22 identifies the authenticity of banknotes, unfit banknotes (contaminated or damaged banknotes), and denomination based on the output signal from the discrimination sensor 20. 24 is a timekeeping section, which measures time. 26 is a data collection unit. The data collection unit 26 collects output signals from the sensor level detection unit 12, the traveling number detection unit 14, the skew detection unit 16, the double feed detection unit 18, the identification unit 22 and the time counting unit 24, and temporarily stores them. That is, the data collection unit 26 accumulates the output level of the light receiving sensor, the number of traveling sheets (the number of separated sheets), the number of skewed sheets, the number of double feeds, the number of discrimination rejects, and the like. Reference numeral 28 denotes a data processing unit, which receives signals from the skew detection unit 16, the double feed detection unit 18, and the identification unit 22 via the data collection unit 26 via the data collection unit 26 from the running number detection unit 14. The generated output signals are used to obtain the respective occurrence frequencies. For example, focusing on skew, the data collection unit
When the number of skews per day accumulated in 26 is N _s and the number of runs per day (number of separated sheets) is N _a , N _s / N _a × 100 is calculated and the skew occurrence frequency S Calculate _x (skew rate). The frequency of double feed (double feed rate) and the frequency of identification reject (reject rate) are calculated in the same manner. 30 is a memory for storing the obtained diagnostic data. Specifically, the memory 30 stores the output level and the number of traveling sheets of the light receiving sensor collected by the data collecting section 26, and the skew feeding rate, the double feeding rate, and the identification reject rate created by the data processing section 28. FIG. 4 shows a writing format of diagnostic data (occurrence frequency information) in the memory 30. As will be described later, FIG. 10A shows a format in which the occurrence frequency information exceeds the comparison reference value as a result of the comparison between the occurrence frequency information and the comparison reference value, and a flag is added to the occurrence frequency information as a dangerous forecast. FIG. 6B shows a format when the diagnostic data does not exceed the comparison reference value. Reference numeral 32 is a comparison reference value storage unit. The comparison reference value storage unit 32 stores the comparison reference value of the output level of the light receiving sensor, the comparison reference value of the skew feeding rate, the comparison reference value of the double feed rate, and the comparison reference value of the identification reject rate. As described above, these comparison reference values are set to values with a margin so that normal operation can be performed within a certain range (time) even if the diagnostic data exceeds that value. 34 is a pre-alarm display section that displays a pre-alarm corresponding to the part or unit being diagnosed. Specifically, the pre-alarm display unit 34 displays a pre-alarm for all the light receiving sensors, a pre-alarm for a unit including rollers that cause skew and double feeding, and a pre-alarm for the discrimination sensor 20. And so on. 36 is a keyboard. The keyboard 36 performs two controls on the diagnostic data and flags stored in the memory 30. Specifically, it specifies the batch clear control that clears only the diagnostic data with a flag and the data without a flag, and the single instruction control that clears one diagnostic data. . Collective clear control is used when maintenance is completed at the time of regular inspection, and single command control is used when parts or units are replaced for reasons other than pre-alarm information. Also, the keyboard 36
Is also used to call up the pre-alarm display and change the comparison reference value. Reference numeral 38 denotes a comparison unit which stores the occurrence frequency information obtained from the collection unit 26 or the data processing unit 28 via the control unit 40 or the average of these and the occurrence frequency information read from the memory 30 and the comparison reference value 32. If the occurrence frequency information does not exceed the comparison reference value, the fourth comparison is made.
The occurrence frequency information compared in the format of FIG. 4B is stored in the memory, and when it is equal to or exceeds the comparison reference value, the occurrence frequency information compared in the format of FIG. 4A is stored in the memory. To do. A control unit 40 controls each of the above-mentioned units.

As described above, by grasping the skew feeding, the double feeding, the identification rejection rate, and the like, the operating state of the device depending on the occurrence frequency of the operation abnormality is known, and the preliminary maintenance is performed to eliminate the cause of the failure before the failure actually occurs.

Next, the operation of the configuration of FIG. 2 will be described with reference to FIG. FIG. 3 is an operation flowchart of the configuration of FIG. In the operation described below, a case will be described as an example where the occurrence frequency information related to the operation of the apparatus for one day is obtained and compared with the comparison reference value. In FIG. 3, steps 200 to 2
09 is an operation regarding each banknote in one transaction. First, the bills are separated and fed out, and one bill is sent to the traveling path (step 200). Then, the number of separated (running) counts of one transaction is counted by +1 by the output of the running number detection unit 14, and the result n _a is temporarily stored in the data collection unit 26. Next, the skew detection unit 16 detects skew (step 202). When a skew is detected, it is counted and set to +1. As a result, n _s is temporarily stored in the data collection unit 26 (step 203) and the process proceeds to step 204. If skew is not detected, the process proceeds to step 204. Next, the double feed detection unit 18 detects a double feed (step 20).
Four). When double feed is detected, +1 is counted, and as a result, n
_d is temporarily stored in the data collection unit 26 (step 205), and the process proceeds to step 206. If the double feed is not detected, the process proceeds to step 206. Next, the identification unit 22 determines the authenticity of the bill and the denomination (step 206). When normal, denominations are counted according to the discrimination result (step 208). When it is not normal, that is, when it is a reject bill, the number of identification rejected sheets is counted by +1 and the result n _b is temporarily stored in the data collection unit 26 (step 209). Then, the process proceeds to step 210, and the process for the next banknote is similarly performed.

When the separation end of one transaction is detected in step 210, the separation count number n _a is first added to the total separation number on the day, and the result N _a is stored in the data collection unit 26 (step 21
1). Also, the skew feed number n _s is added to the total skew feed number on the day, and the result N _s is stored in the data collection unit 26 (step 21
2). Further, the number of multi-fed sheets n _d is added to the total number of multi-fed sheets on the day, and the result N _d is stored in the data collecting unit 26 (step
3). Further, the number of identification rejected sheets n _b is added to the total number of rejected sheets on that day, and the result N _b is stored (step 21
Four).

Then, in step 215, the transaction end operation for the current day (one day) is performed, and when the transaction ends for one day, the data processing unit 28 is stored in the data collection unit 26. Read N _a and N _s and calculate the skew rate S _x (%) = N _a / N _s × 100 (step 21
6). Next, the comparison unit 38 compares the calculated skew feeding ratio S _x with the skew feeding reference ratio S _s which is the comparison reference value stored in the comparison reference value storage unit 32 (step 217). When S _s ≤S _x is satisfied in step 218, the process proceeds to step 219, and the date is written in the memory 30 as a flag in the flag area according to the format of FIG. 4A, and the process proceeds to step 220. When S _s ≦ S _x is not satisfied, the process proceeds to step 220. In step 220, according to the format of FIG. 4 (b),
The skew feeding ratio S _x is written in the diagnostic data area and stored in the memory 30. Similarly, the double feed rate D _X is calculated using N _d and N _a , this is compared with the double feed reference value D _s, and the double feed rate is stored in the diagnostic data area according to the comparison result (step 221). ~ 225). Furthermore, similarly, the identification reject rate B _x is calculated using N _b and N _a , this is compared with the identification rejection reference value B _s, and the identification reject rate is stored in the diagnostic data area according to the comparison result ( Step 230).

Further, the output level of the light receiving sensor is diagnosed as follows. In FIG. 3 (c), first, at the output level check timing (step 231), the light emission amounts of the light sources PD _{1 to} PD _n of the light receiving sensor are decreased (step 23).
2). Next, the sensor level detector 12 detects the light receiving sensors PH _{1 to} PH.
_{The n} output levels V _{1 to} V _n are sequentially collected and stored in the data collecting unit 26 (step 233). Next, the comparison unit 38 compares the detected output levels V _{1 to} V _n with the comparison reference value V _{s of the} light receiving sensor stored in the comparison reference value storage unit 32 (step
234). In step 235, if there is nothing less than the comparison reference value V _s , the process proceeds to step 238, and the light emission amounts of the light sources PD _{1 to} PD _n of the light receiving sensor are returned to the original values. On the other hand, if there is a value equal to or smaller than the comparison reference value V _s , the optical sensor PH _k , PH _m is determined to be a pre-alarm target (step 236), and the generation date and optical sensor name PH _k , PH _k , PH _m is written in the flag area of the memory 30, and the output levels V _k , V _m are stored in the diagnostic data area of the memory 30 (step 237). Then, it proceeds to step 238. The above-mentioned comparison operations are performed, for example, when the apparatus is started up on the next day.

The device is self-diagnosed as described above. Then, for example, on a regular inspection day basis, the control unit 40 refers to the contents of the memory 30 automatically or in accordance with an instruction from the keyboard 36, and displays a pre-alarm on the parts and units related to the occurrence frequency information to which the flag is added. The part 34 displays a pre-alarm. The maintenance staff sees the display of the pre-alarm and knows the place to be maintained, performs all the maintenance, and when the maintenance is completed, the collective clear control is performed via the keyboard 36. The occurrence frequency information to which the pre-alarm flag is not added is stored and retained as it is.

The example in which the present invention is applied to the automatic transaction device has been described above. In this example, the average of the comparison (step 217) is oblique rate within the period (e.g., 1 week) with the skew standard rate S _x the skew reference rate S _s and Hasugyoritsu S _x (e.g., moving average ) May be used. The same applies to the double feed rate and the identification reject rate. Further, the present invention can be similarly applied to devices other than the automatic transaction device. Further, the information stored in the memory 30 can be transmitted to a monitoring device or a host computer at a remote place via a communication line to obtain information for preventing the occurrence of a failure at the remote place.

(Effects of the Invention) As described above, according to the present invention, the occurrence of a failure requiring maintenance is predicted based on the frequency of occurrence of malfunctions. Therefore, before the device actually causes a failure and goes down. In addition, it is possible to eliminate the cause of failure, and improve the operating rate and reliability of the device.

In particular, according to the present invention, the traveling information of the medium is detected by the sensor information output from the traveling sensor that is provided near the roller and the separation roller and reads the traveling state of the medium. It is output as detection information of the operation abnormality of the roller and the separation roller. The occurrence ratio of the operation abnormality is calculated as the occurrence frequency information from the detection information. The predetermined comparison reference value of the roller and the separation roller is compared with the value of the occurrence frequency information of the operation abnormality of the roller and the separation roller, respectively, and the danger forecast information is output, and the danger forecast information is displayed on the pre-alarm display unit. Displayed for each separation roller. in this way,
In the medium transport system, the frequency of running abnormalities such as skewing and double feeding of the medium is detected to know the degree of abnormality such as wear of the roller and the separation roller, and the extent of the abnormality is within the limits determined for each roller and separation roller. When it exceeds, the pre-alarm is displayed, so it is not really a failure, but there is often a situation where it is extremely difficult to judge that it is a condition requiring maintenance in advance. It is possible to perform accurate and prompt maintenance depending on the situation.

Further, since the detection information is standardized as the occurrence rate of operation abnormalities and stored in the memory, the maintenance personnel can access the memory to obtain the standardized and easy-to-see raw data of operation abnormalities as needed. Obtainable. Since the data to be compared is standardized, it becomes easy to compare in a comparison section including a digital processor or the like.
As a comparison reference value, the developer or quality manager uses the data obtained from experiments and other factors to judge the weight of each data, and uses the operation abnormality allowable limit value calculated based on the data. For individual parts of
Danger forecast information can be output centrally. As a result, even if the maintenance staff is inexperienced, he / she will not hesitate to judge the danger forecast. In addition to this, even a maintenance person who has sufficient experience can easily find out a place where a failure is likely to occur but that is likely to be overlooked.

In addition to the above effects, it is possible to detect and display both operational abnormalities and running sensor abnormalities. It is easy to determine whether the sensor is abnormal, and more accurate maintenance can be performed.

[Brief description of drawings]

FIG. 1 is a flowchart showing the procedure of the present invention, FIG. 2 is a block diagram of a hardware configuration when the present invention is applied to a banknote depositing / dispensing machine of an automatic transaction device, and FIG. 3 is an operation of the configuration of FIG. A flow chart and FIG. 4 are diagrams showing a write format of diagnostic data in the memory. 10 …… Running sensor, 12 …… Sensor level detecting unit, 14 …… Running number detecting unit, 13 …… Skew detecting unit, 18 …… Double feed detecting unit, 20 …… Identification sensor, 22 …… Identifying unit, 24 …… Timekeeping section, 26 …… Data collection section, 28 …… Data processing section, 30 …… Memory, 32 …… Comparison reference value storage section, 34 …… Pre-alarm display section, 36 …… Keyboard, 38 …… Comparison unit, 40 ... Control unit.

Front Page Continuation (72) Inventor Shinichi Sudo 1-7-12 Toranomon, Minato-ku, Tokyo Oki Electric Industry Co., Ltd. (56) Reference JP-A-53-83651 (JP, A) JP-A-56-137455 (JP, A) JP 58-182763 (JP, A) JP 59-79363 (JP, A) JP 53-42590 (JP, B2)

Claims (2)

[Claims] 1. A self-diagnosis apparatus for a medium transport system, comprising a roller for moving a medium and a separation roller for separating the medium, the self-diagnosis apparatus being provided in the vicinity of the roller and the separation roller, respectively. A traveling sensor that reads the traveling state of the medium, and sensor information output from the traveling sensor, detects a traveling abnormality of the medium, and detects the operating abnormality of the roller and the separation roller according to the detected traveling abnormality, respectively. A detection unit that outputs the data, a data collection unit that temporarily stores the detection information output by the detection unit, and the occurrence ratio of the operation abnormality of the roller and the separation roller is generated from the detection information stored in the data collection unit. A data processing unit that is calculated as frequency information, a memory that stores the occurrence frequency information calculated by the data processing unit, and an operation error A comparison reference value storage unit in which a comparison reference value predetermined as a limit is stored for each site to be diagnosed, a comparison reference value of the roller and the separation roller stored in the comparison reference value storage unit, and the roller and Comparing with the value of the occurrence frequency information of the operation abnormality of the separation roller, respectively, a comparison unit that outputs danger forecast information indicating whether the value of the occurrence frequency information exceeds the respective operation abnormality allowable limit, and the hazard prediction information And a pre-alarm display unit for respectively displaying the risk forecasts of the roller and the separation roller based on the risk forecast information. Self-diagnosis device for medium transport system. 2. A self-diagnosis apparatus for a medium carrying system, comprising a roller for moving a medium and a separation roller for separating the medium, the self-diagnosis apparatus being provided in the vicinity of the roller and the separation roller, respectively. A traveling sensor that reads the traveling state of the medium, and sensor information output from the traveling sensor, detects a traveling abnormality of the medium, and detects the operating abnormality of the roller and the separation roller according to the detected traveling abnormality, respectively. A detection unit that outputs, a sensor level detection unit that detects the sensor information output from the traveling sensor when the sensor output level is checked, and outputs the sensor information as sensor level information, and the detection information and the sensor level detection unit that are output from the detection unit A data collecting unit that temporarily stores the sensor level information output from the A data processing unit that calculates the occurrence ratio of the operation abnormality of the roller and the separation roller as the occurrence frequency information from the accumulated detection information, a memory that stores the occurrence frequency information calculated by the data processing unit, and an operation abnormality allowable limit A comparison reference value storage unit in which a predetermined comparison reference value is stored for each diagnosis target region, a comparison reference value of the roller and the separation roller, and the roller and the roller stored in the comparison reference value storage unit. The value of the occurrence frequency information of the separation roller operation abnormality is compared with each other, and the comparison reference value of the sensor level stored in the comparison reference value storage unit and the value of the sensor level information are compared to determine the occurrence frequency information. A comparison unit that outputs danger forecast information indicating whether or not the value or the value of the sensor level information exceeds the allowable limit of the operation abnormality, The occurrence frequency information accumulated in the memory, the control unit which is set in correspondence with the sensor level information, and the pre-alarm display unit which respectively displays the danger forecast of the roller, the separation roller and the traveling sensor based on the danger forecast information, A self-diagnosis device for a medium transport system, comprising: