Method for analyzing process signals of a technical plant, in particular a power plant

• US 6,108,615 A
• Filed: 02/09/1998
• Issued: 08/22/2000
• Est. Priority Date: 08/09/1995
• Status: Expired due to Fees

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Panyr, Jiri, Fuhring, Thorsten, Politiadis-Behrens, Alexander, Mederer, Hans-Gerd, Preiser, Ulrich
• Primary Examiner(s): Callahan, Timothy P.
• Assistant Examiner(s): NGUYEN, LINH M
• Application Number:
  US09/020,604
• Time in Patent Office: 925 Days
• Field of Search: 702/188, 702/182, 702/183, 702/184, 702/185, 345/440
• US Class Current: 702/183
• CPC Class Codes: G05B 23/024 Quantitative history assessment, e.g. mathematical relationships between available data; Functions therefor; Principal component analysis [PCA]; Partial least square [PLS]; Statistical classifiers, e.g. Bayesian networks, linear regression or correlation analysis; Neural networks
• CPC Class Codes: G05B 23/0272 Presentation of monitored results, e.g. selection of status reports to be displayed; Filtering information to the user
• CPC Class Codes: Y04S 10/52 Outage or fault management, e.g. fault detection or location

Abstract

A method for analyzing process signals of a technical plant, in particular a power station plant, having a system process subdivided into functional units, includes performing simplified signal tracing for a diagnosis or a forecast of a system state by checking if a process signal is a component of a functional unit, for each process signal and each functional unit. Thereafter, a degree of correlation between each two process signals is determined with the aid of the number of those functional units in which they are jointly a component. Subsequently, the process signals correlated with a prescribed process signal are represented as information elements that are positioned relative to one another in such a way that the distance between each two information elements represents the degree of their correlation.

Claims

1. In a method for analyzing process signals of a technical plant performing a plant process subdivided into functional units, the improvement which comprises:

checking if a process signal is a component of a functional unit, for each process signal and each functional unit;

determining a degree of correlation of each two process signals from the number of the functional units in which they are jointly a component; and

displaying the process signals correlated with a predetermined process signal as information elements positioned relative to one another for representing the degree of their correlation by a spacing between each two information elements, and positioning the information elements representing the most highly correlated process signals at the least spacing from one another.

2. The method according to claim 1, which comprises diagnosing a plant status from a process signal generated from measurement values or report signals detected on-line.

3. The method according to claim 1, which comprises forming a prognosis of a plant status from a simulated process signal.

4. The method according to claim 1, which comprises displaying at least one of a chronological and a logical sequence of process signals.

5. The method according to claim 1, which comprises displaying a number of the functional units in the form of information elements on the basis of their correlations with one another, and displaying the number of the functional units with the process signals.

6. The method according to claim 1, which comprises displaying process signals of linked-together operators in the form of information elements.

7. The method according to claim 5, which comprises displaying process signals of linked-together operators in the form of information elements.