JPH0380331A - Symptom input support type abnormality diagnosis device - Google Patents

Abstract

PURPOSE:To infer the factors based on the accurate symptoms by storing previously the technical knowledge when the abnormal symptoms are narrowed and giving the stored knowledge to an operator at the input of the symptoms. CONSTITUTION:An operator selects one of the highest rank abnormal symptoms (master abnormal symptoms) out of those abnormal symptoms of each group stored in a main storage means. Thus an output means outputs plural abnormal symptoms belonging to the same group as the selected master symptoms in a state of these contents where these abnormal symptoms become fine toward a lower rank from a higher rank. Thus the operator recognizes the abnormal symptoms emerging presently at an abnormality diagnostic subject device in a comparatively rough form and selects the corresponding master abnormal symptoms. Then the operator can naturally input the most detailed contents of the abnormal symptoms emerging at present, i.e., the abnormal symptoms suited to an accurate inference of factors just by selecting successively the output abnormal symptoms toward the lower ranks from the higher ranks via an abnormality symptom selection means.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to an abnormality diagnosis device that diagnoses an abnormality in a device to be diagnosed based on abnormal symptoms of the device to be diagnosed by an operator. In particular, it relates to technology that supports operator input of symptoms.

[Conventional technology]

An abnormality diagnosis device generally includes a main storage means that stores each of a plurality of abnormal symptoms of a device to be abnormally diagnosed and inference information necessary for inferring the cause of each abnormal symptom in association with each other, and abnormal symptoms input by an operator. and an inference means for inferring a cause based on the information and inference information.

[Problem to be solved by the invention]

In this type of abnormality diagnosis device, the accuracy of inference depends on whether the input abnormal symptoms are suitable or not. Therefore, in order to achieve high inference accuracy, it is important to input abnormal symptoms with content suitable for inference. but,
In order to find such qualified abnormal symptoms from the abnormal symptoms exhibited by the abnormality diagnosis target equipment, it is normal that a considerable amount of knowledge about the abnormality diagnosis target equipment or abnormality diagnosis equipment is required. There was a problem in that an operator who did not have one could not effectively use the abnormality diagnosis device.

The present invention has been made with the aim of solving this problem.

[Means to solve the problem]

As shown in FIG. 1, the gist of the present invention is to provide an abnormality diagnosis apparatus including the above-mentioned main storage means and inference means, in which the main storage means stores a plurality of abnormal symptoms of an equipment to be diagnosed. The abnormal symptoms are classified into multiple groups depending on the type of abnormal symptoms, and the multiple abnormal symptoms belonging to each group are placed in multiple positions in the tree structure, and the contents of the abnormal symptoms are detailed from top to bottom. In addition to storing inference information necessary for inferring the cause of the abnormal symptoms in association with the abnormal symptoms, output means outputs a plurality of abnormal symptoms. , an abnormal symptom selection means for selecting one of a plurality of groups and then sequentially selecting one of a plurality of abnormal symptoms belonging to the selected group from the top to the bottom according to an operator's operation based on the output result of the output means; In addition, the inference means infers the cause based on the selection result of the abnormal symptom selection means and the inference information, and outputs it to the output means.

[Effect]

In the device of the present invention, if the operator selects one of the highest abnormal symptoms (hereinafter referred to as parent abnormal symptom) among the abnormal symptoms of each group stored in the main storage means, the selected abnormal symptom is selected. A plurality of abnormal symptoms belonging to the same group as the parent abnormal symptom are output in a state in which the details become more detailed as the abnormal symptoms go from higher to lower.

Therefore, the operator can relatively roughly recognize the abnormal symptoms currently occurring in the equipment to be diagnosed, select the corresponding parent abnormal symptom, and then select the output abnormal symptoms in order from the top to the bottom. Then, the most detailed content of the abnormal symptom currently occurring in the abnormal symptom target device, that is, the abnormal symptom suitable for accurate cause inference will be naturally input. In other words, the device of the present invention supports symptom input by storing in advance specialized knowledge for narrowing down abnormal symptoms and providing that knowledge to the operator when the operator inputs the symptoms.

〔Effect of the invention〕

In this way, according to the present invention, even if the operator does not have sufficient knowledge about the equipment to be diagnosed and the abnormality diagnosis device, qualified abnormal symptoms are input, and the inference means can determine the cause based on the qualified abnormal symptoms. This allows inference to be made, resulting in the effect of improving inference accuracy.

〔Example〕

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment in which the present invention is applied to an abnormality diagnosis device used in an NC machine tool will be described in detail below with reference to the drawings. NC machine tools are equipped with an XY table, a CRT used for inputting machining data, and a memory for storing NC programs.
It is designed to drill a hole at a predetermined position in a material supported on a Y table and then tap the hole.

As shown in FIG. 2, the abnormality diagnosis device includes a keyboard 10, a mouse 12, a CRT 14 (separate from the above-mentioned CRT), and a keyboard 10, mouse 12, and CR.
The main controller 16 is connected to the T14. The main controller 16 is a CPU.

It is mainly a computer that includes ROM, RAM, and a bus. A knowledge base memory 20 and a program memory 22 are provided in the ROM.

The knowledge base memory 20 stores a knowledge base in which symptom data representing each of a plurality of abnormal symptoms of the NC machine tool is associated with cause data representing the cause of each abnormal symptom. The knowledge base memory 20 also stores the plurality of symptom data in rA, CR, as partially shown in a table in FIG.
There is an abnormality on the T screen (the CRT screen of the machine tool; hereinafter, the CRT without a code indicates the CRT of the NC machine tool) 1, r13. Poor machining accuracy1. Each of the most general abnormal symptoms is considered a parent abnormal symptom, and below that parent abnormal symptom there are multiple abnormal symptoms of the same type, with details increasing from the top to the bottom. It is stored in a hierarchical manner as follows.

The program memory 22 stores a control program represented by a flowchart in FIG.

The operation of the abnormality diagnosis device will be explained below based on its flowchart, but this time it will be assumed that an abnormal symptom occurs in the NC machine tool such that the position of the hole drilled in the r material sometimes deviates in the X-axis direction. and explain.

When the power is turned on, first step 3100 (hereinafter referred to as
(Simply represented by 5100, the same applies to other steps), the highest level of the abnormal symptoms of the plurality of groups stored in the knowledge base memory 20, that is, the plurality of parent abnormal symptoms are displayed on the screen of the CRT 14 all at once. be done. Based on the displayed result, the operator moves the mouse 12 to the same parent abnormal symptom as the abnormal symptom currently displayed by the NC machine tool, and then operates a specific switch provided on the mouse 12. As a result, the abnormal symptom that matches the mouse 12 is displayed in a different state from other abnormal symptoms, and the operator is informed of the fact that the specific parent abnormal symptom is currently selected by the mouse 12. If the abnormal symptom currently selected by the mouse 12 is desired by the operator, the operator operates the execution key on the keyboard 10.

If the execution key is operated, in 5110, the mouse 1
2, the symptom data corresponding to the currently selected parent abnormal symptom is imported into the computer. This time, based on the above assumption, as is clear from FIG. 3, symptom data corresponding to the parent abnormal symptom of rB and poor machining accuracy of 1 will be imported into the computer.

Thereafter, in 5120, the abnormal symptom (currently the parent abnormal symptom) that is lower than the abnormal symptom selected in 5110 (currently the parent abnormal symptom)
This time, it is determined whether abnormal child symptoms) are present. This time, the parent abnormal symptom of ``rB, machining accuracy is poor'' was selected, so as is clear from Figure 3, the child abnormal symptoms of ``b+, tap machining accuracy is poor'' and 'bz, 1st class misalignment. exists, so the determination result this time is YES.

At 5130, the plurality of child abnormal symptoms are displayed on the CRT 14 all at once instead of the plurality of parent abnormal symptoms. Thereafter, the operator determines whether or not the corresponding child abnormal symptom exists, and if it is determined that it does not exist, and inputs specific data using the keyboard 10 and mouse 12, 51
The determination result of 40 is YES, and the process moves to 5150.

This time, based on the above assumption, rb! Since the child abnormal symptom J, in which the 0 position is shifted, corresponds to the child abnormal symptom, the operator inputs data for selecting the child abnormal symptom, not the above-mentioned specific data, so the judgment result of 3140 is NO, and
Return to 0.

At 5110, symptom data representative of the selected child abnormal symptom is captured into the computer.

Then, in 5120, as in the previous case, it is determined whether there is an abnormal symptom lower than the current abnormal symptom, that is, whether a grandchild abnormal symptom exists this time. This time, based on the above assumption, rb! I.

Since there are grandchild abnormal symptoms such as J, 'tlzz, which shifts the position in the X-axis direction, and 1st class, which shifts the position in the Y-axis direction, the current judgment result of 5120 is also YES, and in 5130, the grandchild abnormal symptom is replaced with the child abnormal symptom. Displayed on CRT14. 5
At 140, it is determined whether any of the displayed grandchild abnormal symptoms are applicable at this time. This time, from the above assumption, 'btu, the grandchild abnormality symptom of 1, where the position in the X-axis direction is shifted, corresponds to this, so the current judgment result of 5140 is also No.
So, we return to 5ILO again.

Thereafter, in 5110, symptom data representing the current grandchild abnormal symptom is imported into the computer, and then in 5120, it is determined whether there is an abnormal symptom lower than the current abnormal symptom. This time, rbz++, 1 and ' which sometimes deviate.
Since there is a great-grandchild abnormal symptom with bttt and j where the origin position shifts, the current judgment result for 5120 is also YES,
In 5130, instead of grandchild abnormal symptoms, great-grandchild abnormal symptoms are C.
Displayed on RT14. At 5140, it is determined whether any of the displayed great-grandchildren's abnormal symptoms are applicable this time. This time, based on the above assumption, the great-grandchild abnormal symptom of ``b□30 sometimes deviates from 1'' corresponds to this, so the current judgment result of 5140 is also No, and it returns to 5110 again.

After that, in 5110, symptom data representing the current great-grandchild abnormal symptom is imported into the computer, and then in 5120, it is determined whether there is an abnormal symptom lower than the current abnormal symptom. This time, there are no abnormal symptoms below the great-grandchild abnormal symptoms, so the current judgment result of 3120 is No.
Then, the process moves to 5150. As above, NC
The most detailed contents of the abnormal symptoms currently occurring in the machine tool are induced.

Thereafter, at 5150, cause data corresponding to the most detailed abnormal symptom is read from the knowledge base memory 20. This time, as is clear from Figure 3, rX
Poor contact of the axis servo motor connector 3. The cause of rXX-axis mode driver failure 1, etc. is inferred as a candidate cause.

The X servo motor is driven by the X-axis motor driver to
The Y table is moved in the X-axis direction, and the connector connects the X-axis servo motor side and the X-axis motor driver side. At 5150, further interviews are repeated with the operator in order to identify the true cause from among the multiple cause candidates, and the operator uses the keyboard IO
and input the response result by operating the mouse 12,
The cause will be identified based on the response results.

Thereafter, at 5160, the identified cause is CR.
Displayed on T14. In this way, the current execution of the control program ends.

As is clear from the above explanation, in this example,
The knowledge base memory 20 constitutes the main storage means, and the CRT 1
4 and 5i of FIG. 4 of the computer of the main controller 16.
The keyboard 10, the mouse 12, and the part that executes 5110, 5120, and 5140 of the computer constitute abnormal symptom selection means, and the part that executes 5150 and 3160 of the computer constitutes an output means. The executing part constitutes the inference means.

Additionally, in this embodiment, all of a plurality of abnormal symptoms of the NC machine tool are expressed as phenomena. Since the phenomenon is a fact that can be recognized intuitively by the operator, even if the operator does not have sufficient technical knowledge dedicated to NC machine tools, it is possible to recognize and select appropriate abnormal symptoms, and the utilization rate of the abnormality diagnosis device can be improved. This has the effect of improving.

Furthermore, in this embodiment, the cause of "incorrect parameter setting 1" shown in FIG. 3 is not caused by the component parts of the NC machine tool itself, but by the operator's operation Alternatively, it is also possible to infer the cause due to the set symptom, which has the effect of expanding the range of abnormal symptoms that can be diagnosed.

Furthermore, in this embodiment, after a parent abnormal symptom is selected, abnormal symptoms of the same type as the parent abnormal symptom are displayed on the CRT 14 for each person. If j with an abnormality on the screen is selected, ra.

J,'at, the CRT screen is pitch black...
ra,, the child abnormal symptoms of 1st grade where the letters are tilted are displayed,
For example, an abnormal symptom that is different in type from the current abnormal symptom, such as 1゜r where the r tool cannot be replaced and 1゜r where the origin cannot be returned, that is, an abnormal symptom that is completely unrelated to the selection of the abnormal symptom, is CRT.
14 will not be displayed together. In this way, in this embodiment, the operator can select the abnormal symptom that he or she considers most suitable from among the abnormal symptoms belonging to the same type and at the same level, making it possible to select accurately and quickly. This effect can also be obtained.

Although one embodiment of the present invention has been described above in detail based on the drawings, the present invention can also be used in various equipment such as spring sinks, casting machines, plastic working machines, welding machines, cutting machines, etc. Application to the abnormality diagnosis device and various modifications based on the knowledge of those skilled in the art.

The invention can be practiced in modified forms.

[Brief explanation of drawings]

FIG. 1 is a block diagram conceptually showing the configuration of the present invention. FIG. 2 is a block diagram showing an abnormality diagnosis device for NC machine tools which is an embodiment of the present invention, FIG. 3 is a diagram showing some abnormal symptoms stored in the computer of the abnormality diagnosis device, and FIG. FIG. 4 is a flowchart showing the control program stored in the computer. 10: Keyboard 12: Mouse 14: CRT 16: Main controller 20: Knowledge base memory

Claims (1)

[Claims] A plurality of abnormal symptoms of a device to be diagnosed as an abnormality are classified into a plurality of groups depending on the type of the abnormal symptoms,
In addition, multiple abnormal symptoms belonging to each group are stored in multiple positions in a tree structure, and the contents of the abnormal symptoms become more detailed from the top to the bottom. a main storage means which also stores inference information necessary for inferring the cause of the abnormal symptoms in association with the above; an output means for outputting the plurality of abnormal symptoms; abnormal symptom selection means for selecting one of a plurality of groups and then sequentially selecting one of a plurality of abnormal symptoms belonging to the selected group from top to bottom; and a selection result of the abnormal symptom selection means and the inference information. and inference means for inferring the cause based on the above and outputting it to the output means.