JPH0643905B2 - Monitoring / control device - Google Patents

Description

Description: [Industrial field of use] The present invention relates to a monitoring / control device for monitoring the operating state of a system under test such as a machine / plant.

[Prior Art] In a monitoring / control system of this type, a sensor is a detection circuit and detection data supplied from a data converter such as an A / D converter (hereinafter, these are collectively referred to as a detector). Was processed by a computer to monitor the operating state of the system under measurement. Therefore, it is extremely important for the stability of the system to confirm the correctness (credibility) of the detected data.

Therefore, conventionally, disconnection detection of a sensor or wiring has been performed by hardware, or the authenticity of a detector has been determined by checking the physical characteristics of the system under measurement. For example, regarding a detector that detects water temperature, the detected amount is −5 ° C. to 100 ° C.
If it is within the range, it is determined to be normal, otherwise it is determined to be abnormal, and for the exhaust gas temperature detector of the internal combustion engine, the detected amount is much higher than the maximum expected temperature (for example, 550 ° C.), which is a limit value (for example, 880 ° C.). ℃), if it exceeds this, it was judged abnormal, otherwise normal.

[Problems to be Solved by the Invention] However, taking the case of the above water temperature as an example, at 98 ° C. and 0 ° C. even in the same normal range, the meanings are completely different from the viewpoint of monitoring the measured system, and these are uniformly treated. According to the evaluation, it was not possible to carry out sufficient monitoring.

The present invention has been made under such a background, and an object thereof is to provide a monitoring / controlling device capable of comprehensively judging the correctness of a detector while considering the dynamic characteristics of the system under measurement. And

[Means for Solving the Problems] In order to solve the above problems, the present invention provides a plurality of detectors for detecting the state of a system to be measured such as a machine and a plant, and detection data obtained by capturing the detection output of each detector. A data collection unit that outputs as, analyzing the transition state of the measured system from the detection amount of the detector, a position position understanding unit that fuzzy reasoning the position position and the certainty factor of the measured system exists, A phase change understanding unit that analyzes the transition speed of the measured system from a change in the detection amount of the detector, and fuzzy infers the phase change in which the measured system exists and its certainty factor, and the position position understanding unit. The obtained position of the position and its certainty factor,
And a storage unit that stores the phase change and the certainty factor obtained by the phase change understanding unit, and whether or not the operation is normally performed by receiving the output signals of the phase position understanding unit and the phase change understanding unit. According to a system check rule for performing a check, a dynamic state confirmation unit that determines whether the operating state of the system under measurement is correct and outputs the result, and a system check result output from the dynamic state confirmation unit is normal. , The position of the position obtained by the position understanding unit and its certainty factor, and the position change obtained by the position change understanding unit and its certainty factor are stored in the accumulation unit, and if the system check result is abnormal, The present situation is confirmed by applying a higher-order rule that controls the application of the rule and providing feedback to the situation position understanding unit and the situation change understanding unit.

[Operation] By constantly understanding the position of the measured system and the changes in the phase, and by grasping their certainty factors, it is possible to comprehensively judge the correctness of the individual detection data from each detector. become. In addition, when the system check result is abnormal, the present situation state recognition unit can feedback the situation position understanding unit and the situation change understanding unit to make a more correct decision.

Embodiments Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the structure of the monitoring / control system of the present invention, which is based on the one described in Japanese Patent Application No. 1-139700. For details, refer to Japanese Patent Application No. 1-13
As an alternative to the specification of No. 9700, the main points will be described below.

The state of each part of the measured system is m detectors 1-i (i- = 1 to 1
m), and the detection output is detected by the data collection unit 2
-I is sampled, taken in, and output as detection data. These pieces of detection data are supplied to the phase change understanding section 3 and the phase position understanding section 4 to understand the phase change and the phase position.

Here, the phase change indicates the rate of change of the state transition of the measured system. For example, when the system to be measured is an engine, the change in each detected amount of the engine is determined by making a comprehensive determination of each of the detected items such as engine speed, exhaust temperature, turbine speed, etc. Understand (eg, changes in exhaust temperature) and overall phase changes (eg, changes in engine load increase, decrease, etc.). In this embodiment, the following plural phase changes are preset for the engine exhaust temperature and the load condition (however, R3 and R11 are only the load condition).

R1. A sudden descent R2. Rapidly descending R3. Undershoot R4. Transient state after descending R5. R6. Slightly descending R7. Stable R8. Slightly rising R9. R10. Transient state after ascent R11. Overshoot R12. R13. On the other hand, the phase position indicates whether the measured system is in the position of each transition state. In this embodiment, a plurality of phase positions are set for the exhaust temperature and the load state of the engine. . That is, the phase position of the exhaust gas temperature is as follows.

R1.30 to 325 ° C region R2.325 to 350 ° C region R3.350 to 375 ° C region R4.375 to 400 ° C region R5.400 to 425 ° C region R6.425 to 450 ° C region R7. 450 to 475 ° C. region R8.475 to 500 ° C. region R9. 500 ° C. or higher region On the other hand, the position position of the load state is as follows.

R1. Long term suspension R2. Short-term stop R3. Unloaded area R4.25% area R5.50% area R6.75% area R7.100% area R8.110% area R9. Danger Area That is, by understanding the position of the phase and the change of the phase, it is possible to understand what state the engine is currently in and what rate of change the engine is changing. Such information is obtained by integrating the detection data from the individual detectors 1-i. On the contrary, such information determines the correctness of the individual detectors 1-i. It is extremely useful information above. The details of this will be described later.

The above-mentioned aspect change understanding unit 3 understands an aspect change by the following three steps of processing.

Survey the detection data for each detection item along the time axis,
Extract significant changes.

The change level of the significant change obtained above is calculated for each detection item. This change level is obtained by fuzzy reasoning, considering the characteristics of the motion system. Here, the characteristic of the operating system refers to the magnitude of the influence of the change in the detection data on the operation of the system under measurement. For example, taking the exhaust temperature of an engine as an example, it is easy to detect a change of 5 ° C,
Even in the steady state, about 10 ° C changes, so it is meaningless to consider 5 ° C as a change in the state of the engine, that is, a change in the operating system. Taking into account such characteristics of the motion system, the change level is obtained using fuzzy reasoning.

That is, after setting a plurality of change levels D1 to D9 for each detection item in advance, nine fuzzy numbers are set in correspondence with the change levels D1 to D9 while taking the characteristics of the operation system as described above into consideration. Is set. Here, the change levels D1 to D9 are as follows.

D1. Super steep d2. Sudden D3. Down D4. A little descending D5. Stable D6. On the rise D7. Rise D8. Soaring D9. Ultra-rapid rise Then, the level of significant change obtained above is applied to a fuzzy number to perform fuzzy inference, and a change level considering the characteristics of the operating system is obtained for each item (for details, refer to Japanese Patent Application No. 1-139700). See No.).

The change levels for each item obtained by the above fuzzy inference are integrated to make a fuzzy inference as to which of the above-mentioned overall phase changes R1 to R13 the current load state change corresponds to.

That is, for each detection item, the overall phase change R1
~ 13 fuzzy numbers corresponding to R13 are set, and the fuzzy inference is performed by applying the change level obtained by these fuzzy numbers to obtain a fuzzy set for each detection item. In the fuzzy set thus obtained, a partial group (for example, exhaust temperature of each cylinder) is arithmetically averaged to perform in-group addition type fuzzy inference. Then
Additive fuzzy inference is performed by arithmetically averaging the fuzzy sets that have not been subjected to intra-group additive fuzzy inference and the fuzzy sets obtained by intra-group additive fuzzy inference. Finally, the region in the result is subjected to defuzzification that defines one definite value, and the phase change R and its confidence M _R are obtained. The details of the techniques of addition-type fuzzy inference within a group, addition-type fuzzy inference, and defuzzification are described in Japanese Patent Application No. 1-139700.

Next, aspects changing understanding unit 4, the detection data supplied from the data acquisition unit 2-i, obtaining the a phase position P and its confidence M _P and subjected to the same fuzzy inference and fuzzy inference as described above . In this case, the above-mentioned n position positions P
The n fuzzy numbers corresponding to 1 to Pn are set for each detection item, and the fuzzy inference is executed.

The thus obtained phase change R and its certainty factor M _R , and the position position P and its certainty factor M _P are stored in the storage unit 7. Here, the belief M _P can correspond to the following language surface according to the score P, and thus the human sense can be expressed.

100 ≧ p> 80 …… Surely 80 ≧ p> 60 …… Probably 60 ≧ p> 40 …… Maybe 40 ≧ p …… Maybe it's not clear, but next, I'll explain the phase state transition history pattern understanding unit 5. . The understanding unit 5 is a part that understands the transition history of the phase state, and includes a near past understanding unit 51, a current understanding unit 52, and a global understanding unit 53.

The near and past understanding unit 51 traces back from the present to the past within the set range, and the phase state data, that is, the result data R of the phase change and the certainty factor M _R , and the certainty M of the result data P of the position of the position. Check _P and see, for example, if there was a sudden change in the situation. The phase change suddenly rises (R12, 13 described above) or suddenly falls (R1, R
2) If yes, check how many previous samplings it happened. Next, if the phase position P changes within a certain sampling number width (set value) after that point, it is interpreted that the sudden change is caused by the transition of the position of the phase, and the conclusion is that the phase is "what before sampling". The output of the content is that “the position of the phase has transitioned from“ Px to Py ”. On the other hand, if not, it is interpreted that "a large change within the same position Px" has occurred. In the same manner as for the gradual change, the phase changes R5, R6, R7, R8 and R9 are investigated. As a result of understanding such transition, Px
It is sufficient to save a linguistic expression such as "a sharp change from Py to Py". That is, unlike the conventional system in which a huge amount of all data is stored, the storage capacity can be reduced because the logically organized language code is stored. Moreover, since the linguistic result is easily adapted to the logic, the subsequent knowledge engineering logic can be strongly developed.

The current situation comprehension unit 52 can compare the result obtained by the near past understanding unit 51 with the current state to grasp whether or not a situation transition has occurred. Also, the current phase change data R
Then, the current change state can be surfaced by the certainty factor M _R.

In the global understanding unit 53, the phase change R is on the descending side (R1 to R
In the case of 6), -1 is set for stable (R7), 0 is set for stable (R8 to R13), and +1 is set for the rising side (R8 to R13), and the next calculation is performed from a preset point in the past.

Count when it is not zero. That is, the number of fluctuations is obtained.

Perform simple addition. That is, it is determined whether it is the falling side, the rising side, or the vertical fluctuation.

With this, it is possible to express (a) whether the phase change such as load fluctuation is large or small, (b) whether it is constant, or (c) when the phase change such as load fluctuation is going down or rising. The expression that a phase change is large, small, large, or small is
This is a powerful information for engine maintenance.

The current situation state confirmation unit 6 confirms the situation position P and the situation change R, and the dynamic state confirmation unit 6 confirms the dynamic state.
1 and the present situation state confirmation unit 62.

The dynamic state confirmation unit 61 is a unit that confirms the dynamic state of the system according to the system check rule. Here, the system check rule means a rule for checking whether the operation is normally performed,
There are the following.

The temperature does not rise from a sudden fall to a sudden rise and vice versa. That is, in the temperature item, a transition from a change level D1 (super-deep fall) or D2 (rapid fall) to D8 (rapid rise) or D9 (super-rapid rise) or vice versa occurs during one sampling. There is no.

For example, considering the number of revolutions, 500 rpm easily rises when starting from a long-term stop, but 1
There is no increase of 500 rpm from 00% load. In this way, in a certain position P, the range of sudden rise and descent is restricted, and in the position P7, the change level D9 does not occur.

There are also certain restrictions on response relationships. For example, when the engine starts to rotate, the turbine rotates and air pressure is generated.

When the detection item is temperature and the above-described change level D changes abruptly (D1, D9), there is a possibility that an abnormality has occurred in the detector.

For example, if the engine speed is 0 while the load phase position is P5 (50%), there is a possibility that an abnormality in the rotation speed detector has occurred.

Whether or not the operating state is correct is determined using such a system check rule, and the result is supplied to the present situation state confirmation unit 62.

The current phase state confirmation unit 62 stores the obtained data R, M _R , P, M _P in the storage unit 7 when the system check result is normal.

On the other hand, in the case of abnormality, methanol is applied, and feedback is performed to the addition type fuzzy inference section and the defuzzification section of the phase change understanding unit 3 and the phase position understanding unit 4, respectively. Here, the meta-rule means a higher-level rule and controls application of the rule itself. For example, the addition type fuzzy inference part is instructed to change the fuzzy number, and the addition type fuzzy inference is instructed to be redone by changing the set value (excluding the abnormal detector). Instruct to perform defuzzification except some detectors. This may reduce the degree of certainty, but it enables the correct decision to be made.

As can be seen from the above description, in this embodiment, the abnormality of the engine itself and the abnormality of the detector can be separated.

Next, the case where the above embodiment is applied to the detection of the exhaust temperature of the engine will be described.

In the case of a four-cylinder engine, temperature detectors 1-1 to 1-4 are provided at each cylinder outlet of the engine to detect the engine exhaust temperature. These exhaust gas temperatures are collected by the data collection units 2-1 and 2-2.
-4 is sampled, captured as detection data, and supplied to the phase change understanding unit 3 and the phase understanding unit 4.

FIG. 2 is a diagram showing the fluctuation of the detection data thus fetched, in which the horizontal axis shows the elapsed time and the vertical axis shows the exhaust gas temperature. In FIG. 5A, it seems that the exhaust temperature of the cylinder 4 is maintained at a low temperature by abruptly dropping while the exhaust temperatures of the cylinders 1 to 3 remain unchanged. On the other hand, in the same figure (b), it seems that the exhaust temperatures of the cylinders 1 to 3 have risen slightly and the exhaust temperature of the cylinder 4 has dropped sharply.

Actually, the figure (a) shows the case where the exhaust temperature detector is abnormal, and the figure (b) shows the case where the fuel valve of the cylinder is abnormal.
That is, in FIG. 4A, since the exhaust temperatures of the cylinders 1 to 3 have not changed, it is considered that there is no abnormality in the engine and the cylinder 4 is operating normally. Nonetheless, the fact that the exhaust temperature detection data drops drastically is presumed that an abnormality has occurred in the exhaust temperature detector of the cylinder 4. On the other hand, in the same figure (b), since the exhaust temperatures of the cylinders 1 to 3 are rising, it is presumed that the operation of the cylinder 4 is abnormal and the temperature of the cylinder 4 is rapidly decreasing. . That is, the other cylinders 1 to 3 take over the output decrease due to the deterioration of combustion in the cylinder 4,
As a result, it is estimated that the exhaust temperatures of the cylinders 1 to 3 have risen. As a cause of such deterioration of combustion, clogging of the combustion valve is considered.

These failure diagnoses are performed as follows. First, the second
In the case of FIG. 3A, the point RT is related to the exhaust temperature of the cylinder 4.
The change at 1 is D5 (stable), and at the point RT2 is D2
(Rapid descent), and at point PT3, D3 (descent). Also,
The overall phase change is R7 (stable). Such a result is output from the phase change understanding unit 3.

On the other hand, the position P of the stage corresponds to the temperature, and the position PT1
The position of the stage is 100% area, the point PT2 is 75% area, and the point PT3 is 50% area. Also, the total position is 10
It becomes 0%. These results are output from the phase position understanding unit 4. Further, regarding the other cylinders 1 to 3, the change is D5 (stable), and the position of the stage is the 100% region.
Therefore, the current situation is judged to be stable in the 100% area. By taking in such a phase change and the position of the phase, the dynamic state confirmation unit 61 determines the abnormality of the detector of the cylinder 4. That is, the change of a specific cylinder (in this case, cylinder 4) is a sudden drop, a sudden drop, or a drop, and the changes of other cylinders (in this case, cylinders 1 to 3) are stable, and the overall phase change If it is stable, the detector is determined to be abnormal according to the check rule that it is determined that the exhaust temperature detector of the specific cylinder (cylinder 4) is abnormal.

Next, in the case of FIG. 2B, the change in the cylinders 1 to 3 is D
At 6 (upward), the position of the stage is 100% from 110
% Over the area. On the other hand, the change of cylinder 4 is D2
(Sudden) and the position of the stage is 50% from the 100% area
Are moving to the realm. In addition, the overall situation is stable at 100%. Such a phase change and a phase position are supplied to the dynamic state confirmation unit 61, and it is determined that the combustion valve is clogged. That is, the phase change of a specific cylinder (in this case, cylinder 4) is a sudden drop, or a sudden drop, or a drop, and the phase change of another cylinder (in this case, cylinders 1 to 3) tends to rise or rise. When the overall phase change is stable, the abnormality of the combustion valve is determined by the check rule of determining that the combustion valve of a specific cylinder (cylinder 4) is clogged.

In this way, the detector is checked in consideration of the phase change and the position.

[Effects of the Invention] As described above, the present invention can bring the following effects.

The dynamic characteristics of the system under measurement are integrated, the position and change of the phase of this system under measurement are determined, and the correctness of the detector is comprehensively determined while considering these, so that an appropriate detector can be checked. Became.

As a result, the reliability of the monitoring / control system and the accuracy of monitoring diagnosis can be improved.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the monitoring / control system of the present invention, and FIG. 2 is a diagram showing the variation of exhaust temperature at each cylinder outlet of the engine.
FIG. 7B is a diagram when the exhaust temperature detector is abnormal, and FIG. 8B is a diagram when the combustion valve of the cylinder is abnormal. 1 ... Detector, 2 ... Data collection section, 3 ... Phase change understanding section, 4 ... Phase position understanding section, 5 ... Phase state transition history pattern understanding section, 6 ... Current phase state confirmation section, 7 …… Accumulation section.

Claims (1)

[Claims] 1. A plurality of detectors for detecting a state of a system to be measured such as a machine or a plant, a data collecting unit for taking in detection output of each detector and outputting as detection data, and a detection amount of the detector. A phase position understanding unit that analyzes the transition state of the measured system and fuzzyly infers the position of the measured system and its confidence level, and the transition of the measured system from the change in the detection amount of the detector. A speed change, a phase change understanding unit that fuzzy infers a phase change in which the measured system exists and its certainty factor, and a position of the position and its certainty factor obtained by the position understanding unit,
And a storage unit that stores the phase change and the certainty factor obtained by the phase change understanding unit, and whether or not the operation is normally performed by receiving the output signals of the phase position understanding unit and the phase change understanding unit. According to a system check rule for performing a check, a dynamic state confirmation unit that determines whether the operating state of the system under measurement is correct and outputs the result, and a system check result output from the dynamic state confirmation unit is normal. , The position of the position obtained by the position understanding unit and its certainty factor, and the position change obtained by the position change understanding unit and its certainty factor are stored in the accumulation unit, and if the system check result is abnormal, A monitoring / control device comprising: a current position state confirmation unit that applies a higher-level rule for controlling the application of rules and feeds back to the position position understanding unit and the position change understanding unit. .