JPH0437732B2 - - Google Patents

Description

[Detailed description of the invention]

[Industrial Application Field] The present invention relates to an abnormality diagnosis method for automatically determining the cause of abnormality, that is, the point of failure in chemical plants, various production processing processes, etc. [Prior art] As a method for diagnosing abnormalities in chemical treatment processes, etc.,
A method has been developed that uses a signed digraph that shows the correlation between changes in the physical quantities of the object to be processed between each point in the process, and has been published in the Journal of Operation Research Society Japan.
Operation Research Society Japan.）Vol23.
Detailed information is provided in P295 (1980). In addition, the above-mentioned method can only diagnose abnormalities caused by a single point of failure, that is, a single cause, but a method has been developed that can diagnose abnormalities caused by multiple points of failure, that is, complex causes. Vol. 3, pp. 343-346 (1985). In addition, a method has been put into practical use to make the diagnosis situation more accurate by using data before the diagnosis time, using changes in abnormality patterns over time, and using multilayer graphs. No. 5, No.
609-615 (1984). [Problems to be Solved by the Invention] However, in a method that enables the diagnosis of abnormalities due to multiple failure points (hereinafter, a set of these failure points is referred to as candidates), in some cases, a large number of candidates may be required as a diagnosis result. However, when applied to large-scale processes, the method that utilizes changes in abnormal patterns over time has the problem of making it impossible to identify the point of failure. This method requires a large amount of calculation time, and a problem has arisen that makes it almost impossible to put it into practical use. [Means for Solving the Problems] In order to solve the above-mentioned problems, the present invention is constituted by the following means. In other words, the correlation between changes in the physical quantities of the processed object between each point of the process that processes the processed object is stored, and the physical quantities of the processed object obtained from multiple specific points of the process are positive or negative from the reference value. In an abnormality diagnosis method that detects a change in the direction and determines a failure point in the process based on the direction of change in the detected situation and the correlation between changes in the physical quantity, the first abnormality diagnosis is performed in response to the detection of a change in the physical quantity. , find and memorize failure point candidates corresponding to the cause of the abnormality,
In response to detection of a change in a physical quantity different from the change in the physical quantity, a second abnormality diagnosis is performed in which further candidates are determined from the candidates in the first abnormality diagnosis. [Effect] Therefore, through the first and second abnormality diagnosis, diagnosis is made using the change in the abnormal state over time, and the fault point can be identified by improving the diagnostic accuracy, and the second abnormality diagnosis In the abnormality diagnosis, the final candidate is determined from among the hypothetical candidates determined in the first abnormality diagnosis, thereby reducing the memory capacity and the time required for calculation. [Example] Hereinafter, details of the present invention will be explained with reference to figures showing examples. FIG. 2 is a schematic diagram of the process, in which liquid W is supplied from pipe 1 to tank 2 as the object to be treated, and from there it is sequentially supplied to tanks 4, 6, and 8 via pipes 3, 5, and 7. In addition, after being subjected to predetermined treatment in each tank 4, 6, and 8, it is fed from pipe 9. On the other hand, from tanks 2 and 4, each pipe 1
0 and 11 to tanks 12 and 13, and after being subjected to the prescribed treatment there as well, each pipe is connected to the pipe line 1.
4 and 15. Therefore, the liquid W is transferred from tank 2 to tank 8 and from tank 2 to tank 8.
and 4 to tanks 12 and 13 in a fixed direction, and after being sequentially processed in each tank 4, 6, 8, 12, and 13, it is fed to a portion not shown in the figure. In addition, among the points of each tank 2, 4, 6, 8, 12, and 13, tanks 8, 12, and 13 are selected as specific multiple points, and these are equipped with a liquid level meter using a pressure transmitter or the like.
6 to 18 are provided for each individual, and these determine the liquid volume.
L ₄ , L ₅ , and L ₆ are measured as physical quantities, and the measured values are provided to a computer (hereinafter referred to as CPT) 21. CPT21 has a cathode ray tube display device (hereinafter referred to as
CRT) 22, keyboard (hereinafter referred to as KB) 23, printer (hereinafter referred to as PRT) 24, etc. are included.
Through operations on the KB 23 and input operations using a light pen or the like on the CRT 22, the memory in the CPT 21 stores the configuration of processes for tanks 2 to 13, and the processor in the CPT 21 (hereinafter referred to as
The CPU) executes the program in the memory, performs abnormality diagnosis based on the measured values from the liquid volume meters 16 to 18 and the process configuration, and reports the results.
The CRT 22 and PRT 24 are used for display and printout. FIG. 3 is a diagram showing a signed directed graph used when storing the process structure in the memory in the CPT 21.
2 and 13 represents the correlation of changes in the liquid amounts L ₁ to _{L 6} at each point. That is, the liquid volumes L ₁ to _{L 6} are transferred to the pipes 3, 5, 7, and
10 and 11 are connected by corresponding arrows, and a + sign is added when the downstream fluid volume change occurs in the same direction as the fluid volume change on the upstream side, and - when the relationship between the two is in the opposite direction. In the example of FIG. 2, all the directions are in the same direction, so the symbols in FIG. 3 are all +. In addition, the CPT 21 determines whether there is a change in the positive or negative direction by comparing the measured values of the liquid level meters 16 to 18 with the reference value, and this situation is as shown in Figure 4. . That is, a reference value 0 is set for each measured value L, and tolerance ranges are set twice in the positive and negative directions, and judgment levels α ⁺ , α ⁻ ,
β ⁺ , β ^- are defined, and according to the following relationships, a positive direction change “+”, a negative direction change “−”, an ambiguous positive direction change “+?”, and a similar negative direction change “−?”
is being detected. α ^- ≦L≦α ⁺ ...“0” (no change) L>β ⁺ ...“+” L<β ^- ...“−” β ⁺ ≧L>α ⁺ ...“+?” α ^- > L≧β ^- ...“-?” Figure 5 is a comprehensive flowchart of the abnormality diagnosis situation by the program in CPT21. Depending on the detection, the system will move to execution of an abnormality diagnosis program. That is, by "sign measurement of measurement points" 101, the measurement values from the liquid volume meters 16 to 18 are determined to be "0",
“+”, “-”, “+?”, and “-?” are judged respectively, and if any of them is “+” or “-”, the point “+” or “-” is determined. Yes?” 102 is Y
(YES), and the process moves to "Abnormality Diagnosis Processing" 111. Figure 1 is a flowchart showing the details of the "abnormality diagnosis process." In order to count the number n of failures that occur simultaneously, a counter configured in the CPU is set with "n=1" 201. By first assuming that the number of failures is single and searching for candidates, 202,
Based on the change detected in step 102 of FIG. 5, a search is made to see if there is a consistent relationship between the relationship in FIG. It is determined whether all the combinations of points, that is, all the candidates have been calculated, by "All combinations completed?" 203, and this is N.
(NO), repeat steps 202 and after, and if step 203 is Y, select “Candidates?” 211
, set the counter in step 201 to "n=n+1"
In order to change the number of failures and perform diagnosis assuming a complex failure, the number of causes of failure is increased by one, and steps 202 and subsequent steps are repeated. If the result in step 211 is Y, the set of candidates obtained by the above first abnormality diagnosis "C = (C ₁ , C ₂ . . .
C _r ) Store" 213 to store the data in memory. Next, in the same way as steps 101 and 102 in FIG. ?” 223, the steps in Figure 1
Change detection point and step 222 when 101 becomes Y
It is determined whether the change detection point when becomes Y is the same or not, and whether the detection situation has changed even at the same point, and if this is Y, steps 221 and subsequent steps are repeated. If step 223 is N,
Since a fluid volume change different from the previous fluid volume change is detected, a second abnormality diagnosis is started accordingly. That is, this time, in order to count the ring i that sequentially selects the hypothetical candidates C ₁ to _{C r} from the set C of candidates obtained by the first diagnosis stored in step 213, the counter configured in the CPU is set to ``i''. =
1" 231, and the corresponding hypothetical candidate "search for candidates based on Ci" 232 is performed in the same manner as step 202, and through N of "i=r?" 233, "i
= i + 1'' 234, the counter in step 231 is incremented, the next hypothetical candidate is sequentially used, and step 232 is added.
After repeating the following steps, if step 233 becomes Y,
In response to the "Candidates available?" 241, "Display candidates" 242 is performed by the CRT 22. Also, if step 241 is N, step 231 is
The counter in step 201 is again set to ``1'' by ``i=1'' 251, and the counter in step 201 is further incremented by ``n=n+1'' 252. 253 is the same as step 202, but multiple failure points are virtualized at the same time, and while ``i=r?'' 261 is N, ``i=i+1'' 262 is performed as in step 234.
After performing the addition, steps 253 and subsequent steps are repeated, and as step 261 becomes Y, ``Candidates exist?'' 263 is checked, and while this is N, steps 252 and subsequent steps are repeated. However, since steps 231 to 233 normally result in step 241 being Y, steps 251 and subsequent steps can be omitted. Also, in general, after step 242, the steps
Return to 213 and repeat the diagnosis. Figures 6 and 7 are specific examples showing the status of each of the above abnormality diagnosis, and Figure 6 shows step 201.
-212, and FIG. 7 corresponds to steps 231-241. In this example, the detection of time 1 in step 102 of FIG. 5 and the detection of time 1 in step 223 of FIG.
In FIG. 2, it is assumed that the detection of time 2 was carried out as shown in the following table.

〔Effect of the invention〕

As is clear from the above description, according to the present invention, the calculation time and memory capacity required for process abnormality diagnosis are significantly reduced, which is particularly advantageous when applied to large-scale processes, and the present invention automatically detects abnormalities in various processes. Significant effects can be obtained in diagnosis.

[Brief explanation of drawings]

The figures show an embodiment of the present invention, in which Fig. 1 is a flowchart of abnormality diagnosis processing, Fig. 2 is a schematic diagram of the process, Fig. 3 is a signed directed graph representing the correlation between the liquid volume changes in Fig. 2, Figure 4 is a diagram of the judgment levels used to detect fluid volume changes, Figure 5 is a comprehensive flowchart of abnormality diagnosis, Figures 6 and 7 are illustrations of specific examples showing the status of abnormality diagnosis, and Figure 8 This figure is a diagram of a specific example of a conventional method corresponding to FIG. 7. 1, 3, 5, 7, 9, 10, 11, 14, 15
... Pipeline, 2, 4, 6, 8, 12, 13 ... Tank,
16-18...Liquid level meter, 21...CPT (electronic computer), 22...CRT (cathode ray tube display), 23
...KB (keyboard), 24...PRT (printer), W...liquid, _L1 to _L6 ...liquid volume.

Claims (1)

[Claims] 1 Memorize the correlation of changes in physical quantities of the object to be processed between each point of a process in which the object to be processed is processed, and make sure that the physical quantities of the object to be processed obtained from a plurality of specific points of the process are correct from a reference value. Alternatively, in an abnormality diagnosis method that detects a change in the negative direction and determines a failure point in the process based on the direction of change in the detected situation and the correlation between changes in the physical quantity,
A first abnormality diagnosis is performed in response to the detection of a change in the physical quantity, and a candidate for a failure point corresponding to the cause of the abnormality is determined and stored. 1. A method for diagnosing an abnormality in a process, characterized by performing a second abnormality diagnosis in which candidates are further determined from among the candidates in the abnormality diagnosis.