JPS61109111A - Plant supervisory unit - Google Patents

Abstract

PURPOSE:To eliminate the misdiagnosis of an abnormality and to improve the reliability with a plant supervisory unit by giving a delay to the abnormality detection input fed from a plant and changing freely the setting time of said delay to detect the abnormality. CONSTITUTION:The signals sent from detectors 3 and 7 provided to a supervisory unit 2 of a plant 1 are supplied to a limit check device 10 of a computer 8 via each converter 5. The output signal 11 given from the device 10 delivers a start instruction 12 to start a plant diagnosis logic 9. The detectors 3 and 7 decide the faults, etc. with the start of the logic 9 and display them on a display device. While the load state information on the plant 1 is supplied to a function generator 32 via a detector 31 and the converter 5. Then the limit value 10 accordant with said load state information is obtained. Here an on- delay timer 34 is provided as a logical time delay element that can change freely the set time. Thus the matching is freely secured between the abnormality detecting point of the plant 1 and the ON time points of nodes 13 and 14 of the logic 9 respectively. Accordingly, a plant fault supervisory system with no misdiagnosis is attained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to a monitoring device for monitoring abnormal conditions in a plant.

[Conventional technology]

Conventionally, there has been a device as shown in FIG. 1 as a second type of device. In Figure 1, 1 is a plant, 8 is a device that constitutes a part of plant 1, and if an abnormality occurs, the influence will spread to other parts of plan 1, and 3 is a device that occurs in device 2. A detector outputs a signal 4 to detect some kind of abnormality (for example, high temperature, low pressure, etc.). 5 is signal 4
A signal 6t- is outputted by a converter that exchanges the voltage level of .

7 is a detector that detects a process quantity that has some kind of causal relationship with the process quantity detected by the detector 3; 8 is a computer that includes a plant diagnosis logic 9 configured by software and a display device (for example, a CRT) not shown; .

10 inputs the signal 6 of the converter 5, and the signal (2 value signal)ll'i is output. 12 is a start command (signal) for the plant diagnosis logic 9
It is.

In the plant diagnosis logic 9, 18 is an observable point (hereinafter referred to as a node) that is detected as a warning in the plant diagnosis logic 9 when the limit check lO signal 11 is turned on, and 14 is a node 13. Similarly, the node 15 is detected in the plant diagnosis logic 9 and is connected to the detector 7 of the plant I.
gate, 16 is the or gate, 17 is the lower node 13
The minimum time delay among the time delays that occur when reaching the upper node 19 from [8.18] is output after node 13 is detected as a warning and it is determined that the cause is in node 14, and the history is A message 19 consisting of an instruction to prevent the abnormal condition from spreading to the node 13 is transmitted by a time delay τ8. Node detected after disjunction of τ4, 20
is a node in the final state where the abnormality extends, and 21 is a point, which affects the detector 3 when it becomes abnormal, but is a virtual point that cannot be detected or has no detector installed. . B2 is a tentative reception logic, which is provided corresponding to the unobserved point 21, and is a logic that simulates the route that Akichuku takes with friends. 28 is a virtual node corresponding to point 21. A message 24 is additionally provided to correspond to the point 21% provisional acceptance logic 22 and node 23, and instructs to avoid an abnormal state.

25 is one component in device 2-J-9
1. Thank you for your understanding, □□4, 2. A model 29 that simulates the component 25, that is, exhibits characteristics equivalent to this, is an adder that adds the output of the predetermined margin value model 27, and a signal that has a dynamic limit value that dynamically defines the failure of the component. 30t-output.

81 is a detector for detecting the load condition of the plant l; 32 is the output of the detector 31 inputted through the converter 5t-q, and the limit value required by the limit check 10 is determined according to the load condition of the plant. The signal 33
Output.

Next, the operation i will be explained.

The signals I'i outputted from the detectors 8 and 7 of the device 2 are converted by the corresponding 8Y converter 5 and are input to the limit check 10 of the computer 8.

When the limit check 1G outputs the signal 11, i.e., turns on, the plant diagnostic logic 9 is activated by the startup command 12, and the signal 11 is detected at the node 13.
It is known in advance that the abnormality in the detector 3 is an influence from the detector 7 or others, and the logic of the causal relationship is simulated by the AND gate 15 and the OR gate 16.

Therefore, if an abnormality is detected at the time when the plant diagnosis logic 9 is activated and before the node 14 reaches the alarm level, the cause of the abnormality is the node 14.
This means that a large amount of money has been generated that would result in detection of and,
It is determined that the cause of the abnormality in the detector 8 is an abnormality in the detector 7, and a message 18 is output to the operator in order to prevent this abnormality from spreading.

Further, when the plant diagnosis logic 9 is activated and the node 13 is detected, the upper logic t- is examined. As a result, the next possible state of the plant I can be predicted, and the operator can take precautions to avoid reaching the final state of the node 2o. In addition, since the minimum possible time in plant l is set between node 13 and node 19, if node 19 is detected within a time period shorter than time 1 after node 13 is detected. If this occurs, it is determined that there is a failure in the detector 2 of the one example of the plant detected by the node 19i.

In order to inform the operators of these 1@ causes and future predictions,
The entire plant diagnosis logic 9 is displayed on a display device in the format shown, and the abnormal state quantities and messages 18 of the detectors 3 and 7 are additionally displayed on the displayed nodes 13 and 14, respectively. This allows operators to accurately grasp the status of the plant and take appropriate actions to avoid abnormal situations.
If the nodes below 8 are checked, but no abnormality is detected in any node (cause identification failure), the cause of the abnormality is caused by the plant, and the unobserved abnormality at point 21, that is, node 28, is determined to be abnormal. to decide.

Note that this message from 24 companies states that the cause of the abnormality is in a non-observable node, such as node 28, and also includes instructions for operations in accordance with this message.

A case will be described in which the detector 7 is a detector that detects an output signal of one component 25 constituting the device 2, that is, a case where it is assumed that the cause of the abnormality is a failure of the component 25. The input signal to component 25 is detected by detector 26 . The output of the detector 26 is input via the converter 5 to a model 27 exhibiting characteristics similar to those of the component 25. The output of model 27 is addition,
9 adds to the margin value and dynamically prescribes it in a manner consistent with the encouragement of component 25.
A signal 30i with a limit value is obtained. As a result, the limit check IO issues an accurate startup command 11 even during plant transients.

Additionally, information regarding the load status of the plant, such as the outputs of generators, turbines, boilers, nuclear reactors, etc., is input from the plant 1 to the function generator 32 via the detector 31 and the converter 5, and the output is limited. Check 10 to obtain limit values that are well suited to the plant loading conditions. This allows all the nodes in the plant diagnosis logic 9, including nodes 3 and 14t, to provide appropriate information to the operator even during plant transitions, making it possible to avoid plant abnormalities.

The conventional plant monitoring device has the above-mentioned configuration, and the abnormality detection timing at nodes 13 and 14 is determined by whether the detection points (a, 7, s5) of the plant are abnormal? : The introduction was not set at the appropriate timing or position, but it was too early, and all nodes were affected, resulting in misdiagnoses of plant abnormalities and failures. There was a drawback that a message of accurate abnormal situation avoidance operation was issued.

This invention was made after eliminating the drawbacks of the conventional ones as described above. By providing a time delay for abnormality detection input from the plant and freely changing the set time, it is possible to perform optimal abnormality detection. The purpose of this invention is to provide a plant monitoring device that can output highly reliable messages without erroneously diagnosing plant abnormalities by transmitting the four reports to the plant diagnosis logic.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Second
In the figure, numerals 1 to 88 represent the same functions as in FIG. 3
4 represents a logical time delay element, ie, an on-delay timer (or off-delay timer) t-, whose set time can be freely set.

Next, the operation will be explained. Since the operations 1 to 33 are the same as those of the conventional plant abnormality diagnosis logic device, they will be omitted here.
A logical on-delay timer (Moshikake off-delay timer) 34t that can be freely changed and set in the output from M10 is installed to match the time of plant abnormality detection and the ON time of No.1''13.14 in the plant diagnosis logic. To enable plant matching to be freely set and to accurately report plant abnormalities to a plant diagnosis logic 9 to perform plant abnormality monitoring without causing erroneous diagnosis.

Although this invention was realized with 8/W, it is also possible to assemble it with a hard wired one.

As described above, according to the present invention, it is possible to match the timing of detecting a plant abnormality with the timing of diagnosing a plant abnormality, thereby eliminating the possibility of erroneous diagnosis due to timing discrepancies, thereby increasing the There is a feature that allows reliable information to be provided to operators.

[Brief explanation of the drawing]

Is Figure 1 a conventional plant monitoring device? FIG. 2 is a block diagram of an embodiment of the plant monitoring device of the present invention. 1---Plant, 3,7,26.31---One detector,
5-m-transducer, 8-m-computer, 9-m-plant diagnostic logic, 10--limit check, 1
3, 14, 19, 20, 23---node, 1
5-11 and gate, 16----or gate,
21-go-point, 25-m-component, 27
---Model, 29-m-adder, 32-m-function generator, 34-m-o7 delay timer (or on-delay timer). Note that the same reference numerals indicate the same parts 2 in the drawings.

Claims (1)

[Claims] When the output of a detector that detects the process amount input to a specific part of the plant indicates a second abnormality level that is set to be safer than a predetermined first abnormality level, the output of the detector is transmitted to the plant. In a plant monitoring device that performs the above diagnosis as an input condition of a logical decision tree for diagnosis, a function generator that receives the load state of the plant as input and responds according to a predetermined function, and an output of this function generator that performs the above diagnosis. A plant monitoring device comprising a first limit check for correcting a second abnormality level and an abnormality detection timing adjustment device.