Plant operation support apparatus and method using expert systems

• US 5,315,502 A
• Filed: 06/06/1990
• Issued: 05/24/1994
• Est. Priority Date: 06/09/1989
• Status: Expired due to Term

• Current Assignee: Mitsubishi Denki Kabushiki Kaisha (Mitsubishi Electric Corporation)
• Original Assignee: Mitsubishi Denki Kabushiki Kaisha (Mitsubishi Electric Corporation)
• Inventors: Koyama, Kenji, Uchihara, Shoichi
• Primary Examiner(s): Smith, Jerry
• Assistant Examiner(s): BROWN, THOMAS
• Application Number:
  US07/534,089
• Time in Patent Office: 1,448 Days
• Field of Search: 364/148, 364/149, 364/152, 364/184-187, 364/550, 364/551.01, 395/50, 395/60, 395/68, 395/912
• US Class Current: 700/79
• CPC Class Codes: G05B 23/0229 knowledge based, e.g. expert systems; genetic algorithms
• CPC Class Codes: G05B 23/0267 Fault communication, e.g. human machine interface [HMI]
• CPC Class Codes: Y10S 706/912 Manufacturing or machine, e.g. agricultural machinery, machine tool

Abstract

Disclosed herein is a plant operation supporting method whereby a body of simulation data simulating the data admitted from a plant in error state is used along with a previously prepared body of knowledge to simulate various modes of plant operation and to have appropriate operation guides displayed for verification of the knowledge.

Claims

1. A plant operation support apparatus for monitoring input process data from a plant to detect errors in the plant, and for providing a countermeasure process guide relating to the cause of error for the use of a plant operator or maintenance personnel, said apparatus comprising:

error detecting means for detecting error in the plant based upon process data from the plant;

error cause inference means for determining the cause of the error;

countermeasure determining means for outputting a countermeasure against the cause of the error determined by said error cause inference determining means;

emergency operation determining means for outputting an emergency operation against said cause of error; and

emergency priority determining means for judging the priority level of the detected error, and for signaling said emergency operation determining means when the detected error is of a selected high priority level;

wherein said error detecting means includes a first knowledge base storing knowledge concerning the normal range of each process data and the priority level of each error event, and error detecting means for detecting errors in the plant by comparing process data which is inputted from the plant with stored data on normal ranges of values stored in said first knowledge base, means for determining priority level by using knowledge stored in said first knowledge base and outputting said priority level and an error event name;

said error cause inference means including a second knowledge base storing knowledge on the characteristics of components of the plant in its normal condition and each cause and effect inter-relating each error and the cause of the error, and error cause inference means for inferring the cause of error by using knowledge stored in said second knowledge base;

said countermeasure determining means including a third knowledge base storing knowledge on each relationship between each cause of error and countermeasures against such error, and an emergency operation finding means for obtaining countermeasures for addressing the determined error by using knowledge stored in said third knowledge base; and

said emergency operation finding means including a fourth knowledge base storing knowledge as to each relationship between each cause of error and the emergency operation pertaining thereto, and an emergency operation finding means for obtaining an emergency operation for addressing the determined error by using the knowledge stored in said fourth knowledge base.

2. A plant operation support apparatus for monitoring input process data from the plant to detect errors in the plant and for providing a countermeasure process guide against the cause of error for a plant operator or maintenance personnel, said apparatus comprising:

error detecting means for detecting errors in the plant based on process data from the plant;

alarm determining means for determining whether the detected error justifies an alarm signal;

prediction means for obtaining predicted alarms for determining the trend of the changes in the process data on which the alarms are based; and

error cause inference means for determining the cause of error in the plant

wherein said error detecting means includes a first knowledge base storing knowledge as to the normal range of process data for each process, and error detecting means for detecting errors in the plant by comparing process data inputted from the plant with the normal range stored in said first knowledge base;

said error cause inference means including a second knowledge base storing knowledge as to characteristics of components of the plant in the normal state and each cause and effect between each error and the cause of such error, and including an error cause inference portion for inferring the cause of error by using knowledge stored in said second knowledge base; and

said predicting means including a third knowledge base storing knowledge as to the trend of change in each of the process data, and each cause and effect between said trend of change and each of said process data, and a triggering process finding portion for determining the name of the process data causing the alarm determined by the alarm deciding portion, by using knowledge stored in said third knowledge base, and a spread alarm prediction portion for finding the plant component or components affecting the process data that caused the error and inferring the trend of the changes in the process data of the affecting components and predicting alarms suitable for trends of changes in the found process data.

3. A plant operation support apparatus for monitoring input process data from the plant to detect errors in the plant and provide a countermeasure guide against the cause of the error for a plant operator or maintenance personnel, said apparatus comprising:

error detecting means for detecting error in the plant based on process data from the plant;

error cause inference means for determining cause of the error; and

a checking portion for checking whether current error in the plant is the same as a previous error and activating said cause inference means if the current error and the previous error are the same

wherein said error detecting means includes a first knowledge base storing knowledge as to the normal range of each of the process data, and an error detecting portion for detecting errors in the plant by comparing the process data inputted from the plant with the normal ranges stored in said first knowledge base; and

said error cause inference means including a second knowledge base storing knowledge as to the characteristics of the components of the plant in the normal state and each cause and effect between each error and the cause of the error, and an error cause inference portion for inferring the cause of the error by using knowledge stored in said second knowledge base.