JPH0529922B2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] This invention improves the reliability of operation of large-scale plants,
The present invention relates to a plant diagnosis device that can identify abnormal events in a plant online in real time, and can appropriately answer questions about the results in order to improve operation rates. .

[Conventional technology]

A conventional device of this type is shown in FIG. In the figure, 1 is plant data that is process quantities such as temperature and pressure input from the plant, 2 is a plant abnormality diagnosis section, 3 is an abnormality diagnosis result file, and 4 is a message that is displayed to the operator after diagnosing a plant abnormality. 5 is a question input processing unit that processes questions from the operator, 6 is a consultant unit that generates answers to the operator's questions, and 7 is a consultant unit that stores the consultant results. 8 is a message processor for consultants used when responding to operators, 9 is a knowledge database for diagnosis, 10 is a message file for diagnosis, 11 is a knowledge database for consultants, and 12 is a message processor for consultants. A message file for the consultant, 13 and 14 are cathode ray tube display sections, and 15 is a section for inputting questions from the operator.

Next, the operation will be explained. The plant abnormality diagnosis unit 2 inputs information necessary for diagnosis from plant data 1 from an actual plant, uses a knowledge database 9 for diagnosis to identify the cause of the abnormality, and also provides operation guidance, prediction of the next event, etc. is determined and the result is stored in the abnormality diagnosis result file 3. A message processor 4 for diagnosis inputs the prediction of the next event and operation guide information from the abnormality diagnosis result file 3, and uses the message file 10 for diagnosis to input the information on the cathode ray tube display unit 1.
Output and display the message in 3.

On the other hand, when the operator asks a question through the question input section 15, the question input processing section 5 interprets the content of the question, and the consultant section 6 inputs the question from the abnormality diagnosis result file 3. Results and knowledge database for consultants 1
1 to search for answers to the questions and store the results in the file 7 for the consultant. The message processor 8 for the consultant displays the answer to the question on the cathode ray tube display section 14 using the consultant results stored in the file 7 for consultants and the data stored in the message file 12 for consultants. .

[Problem that the invention seeks to solve]

Since the conventional plant diagnosis apparatus is configured as described above, the diagnosis result file obtained only at the time when the question is input is used to provide a consultant to the operator. As a result, messages are inevitably fragmented in time, and there are problems such as the inability to provide information such as how the amount of important processors changes over time after a question is input. Ta.

This invention was made to solve the above-mentioned problems, and when a question from an operator is input, it immediately starts editing the trend of the actual plant process amount necessary to answer the question. Afterwards, the results of quantitative predictions of how the actual plant process amount is changing and how the plant status will change if no operation is performed are displayed on the CRT display at the same time as the consultant results. If requested by the operator, the trend results can be displayed on another screen.
The present invention provides a plant diagnostic device that displays detailed information on a CRT.

[Means for solving problems]

The plant diagnosis device according to the present invention identifies the cause of the abnormality when an abnormality occurs in the plant, and then displays the cause of the abnormality, an operation guide, and the event that will occur next on a cathode ray tube. A function that provides appropriate answers to questions such as predicting the impact on the plant, a function that predicts changes in plant status if no operation is performed after identifying the size of an abnormality, and a function that predicts how the plant status will change after an abnormality occurs. A function to collect process quantity trend data from the predicted results of changes in process quantities, a function to process change trends in process quantities from the collected trend data, The process quantity data and the process quantity change trend results are
It has a function to display on a CRT, and, if trend result display is specified, a function to display trend results on a CRT separate from the CRT that displays answers to questions from operators.

[Effect]

In this invention, a prediction function is provided in addition to the conventional plant diagnosis function, and it is possible to identify the magnitude of plant abnormality by inputting the process quantities necessary for diagnosis from plant data, and to The file information that stores the status is used as the initial status of the prediction model to predict and display the behavior of the plant status after an abnormality occurs. Also, when an operator asks a question, the answer to the operator's question is displayed using the results of interpreting the operator's question and the abnormality diagnosis result file.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings. In FIG. 1, the same parts as in FIG. 2 are denoted by the same reference numerals. In FIG. A trend data processing unit that performs various processes using data files, 21 is an abnormality size identification unit, 22 is a file in which the plant state before the abnormality occurs, 23 is a prediction execution unit, and 24 is a prediction result trend file.

Next, the operation of this invention will be explained. first,
The plant abnormality diagnosis section 2 inputs information necessary for diagnosis from plant data 1 from the actual plant, identifies the cause of the abnormality using the knowledge database 9 for diagnosis, and provides operation guidance, prediction of the next event, etc. is determined and the information is stored in the abnormality diagnosis result file 3. A message process 4 for diagnosis inputs information from the abnormality diagnosis result file 3, and further displays a message on the cathode ray tube display section 13 using a message file 10 necessary for displaying the diagnosis result.

On the other hand, when the operator asks a question through the operator's question input section 15, the question input processing section 5 interprets the content of the question. The consultant section 6 outputs the results from the abnormality diagnosis result file 3 at the time when the question is input to the file 7 for the consultant, and at the same time uses the results and the knowledge database 11 for the consultant to answer the question. Explore. Thereafter, the search results are stored in the file 7 for the consultant, and at the same time, the plant process quantities used when searching for answers to the questions are output to the trend data process quantity determining section 16.

The trend data process quantity determining unit 16 determines, among the input plant process quantities, plant process quantities that are considered important when providing consultant results to operators. The trend data collection unit 17 includes the trend data process amount determination unit 16
Regarding the plant process amount determined in step 1, data is collected sequentially from the plant data 1 and stored in the trend data file 18.

The trend data processing unit 19 processes the rate of change of the plant process amount, whether it is on an upward trend, whether it is on a downward trend, etc. Abnormality size identification unit 21
After inputting a signal indicating success in cause identification from the abnormality diagnosis unit 2, the system identifies the magnitude of the abnormality using the plant process quantities and the model. In addition, the plant abnormality diagnosis unit 2 constantly performs abnormality diagnosis, and also collects information necessary as the initial state of the model when quantitatively predicting how the abnormality will develop. The data is stored in the file 22 at regular intervals from the beginning.

The prediction execution unit 23 inputs the optimal initial state information from the file 22 and the size of the abnormality from the abnormality size identification unit 21, predicts the plant behavior when no operation is performed, and calculates the result. is stored in the prediction result trend file 24. The message processor 8 for the consultant includes a file 7 for the consultant storing diagnosis results at the time when the consultant is requested and answers to questions from the operator, a message file 12 for the consultant, and a trend data processing section. The answer to the question from the operator is displayed on the (first) cathode ray tube display section 14 using the plant data processing result after the operator's question request by No. 19 and the prediction result trend file 24.

The results of the trend data processing section 19 and the results of the prediction execution section 23 are updated at regular intervals (first).
It is displayed on the cathode ray tube display section 14. In addition, if an operator requests detailed trends such as other process quantities, select from the trend data file 18 or the prediction result trend file 24 (second).
It can be displayed on the cathode ray tube display section 20.

〔Effect of the invention〕

As described above, according to the present invention, after an operator inputs a question regarding the diagnosis result, it is possible to determine how the important process quantities of the actual plant are changing over time and the state of the plant when no operation is performed. Since the diagnostic system is configured to freely provide information such as predictions of how the In addition, using this information, the amount of plant operation can be estimated quantitatively,
It also has the effect of reducing the burden on the operator.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a plant diagnosis apparatus showing an embodiment of the present invention, and FIG. 2 is a block diagram of a conventional plant diagnosis apparatus. In the figure, 1...plant data, 6...consulting department, 14, 20...first and second cathode ray tube display sections, 15...question input section from operators, 16...trend data process amount determination section ,
17... Trend data collection unit, 18... Trend data file, 19... Trend data processing unit, 21... Abnormality size identification unit, 22... File storing the plant state before the occurrence of the abnormality, 23 . . . Prediction execution unit, 24 . . . Prediction result trend file.

Claims (1)

[Claims] 1. After identifying the cause of an abnormality when an abnormality occurs in a plant, a plant diagnostic device that displays the cause of the abnormality, operation guide, and next event, etc.
The consultant department inputs the diagnosis reason and questions from the operator from the file that is the diagnosis result of the abnormality diagnosis department, and provides appropriate answers to the questions. a trend data process amount determination section that determines a process amount, a trend data collection section that collects trend data of the process amount, and a trend data processing section that processes a change tendency of the process amount from the collected trend data; an abnormality size identification unit that identifies the size of the abnormality after the abnormality cause is fixed by the abnormality diagnosis unit;
A prediction execution unit that predicts the plant state when no operation is performed, and the answer and trend data of the consultant department.Trend data of the process quantity determined by the process quantity determination unit, and prediction result trend by the prediction execution unit. 1. A plant diagnosis device comprising: a first cathode ray tube display section that displays on the same cathode ray tube; and a second cathode ray tube display section that displays an answer when the result of the trend is specified.