Method for the analysis of process data of an industrial plant

• US 5,999,894 A
• Filed: 05/02/1997
• Issued: 12/07/1999
• Est. Priority Date: 11/02/1994
• Status: Expired due to Term

• Current Assignee: Siemens AG
• Original Assignee: Siemens AG
• Inventors: Michelis, Rainer, Panyr, Jiri, Fuhring, Thorsten, Mederer, Hans-Gerd, Jacoby, Konstantin
• Primary Examiner(s): SHAH, KAMINI S
• Application Number:
  US08/850,857
• Time in Patent Office: 949 Days
• Field of Search: 364/184, 364/188, 364/492, 702/62, 702/182, 395/914, 395/915, 395/907, 73/865.9, 376/215, 376/245, 376/256, 206/915
• US Class Current: 702/182
• CPC Class Codes: G05B 23/0221 Preprocessing measurements, e.g. data collection rate adjustment; Standardization of measurements; Time series or signal analysis, e.g. frequency analysis or wavelets; Trustworthiness of measurements; Indexes therefor; Measurements using easily measured parameters to estimate parameters difficult to measure; Virtual sensor creation; De-noising; Sensor fusion; Unconventional preprocessing inherently present in specific fault detection methods like PCA-based methods
• 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
• CPC Class Codes: Y10S 706/915 Power plant

Abstract

In a method for the analysis of process data of an industrial plant, in particular a power station plant, parts of the plant are automatically controlled for the purpose of information compression, filtering and diagnosing faults in good time. Features characterizing the plant process are first prescribed and parameters relevant to the plant process are provided. Subsequently, the presence of each feature is checked for each plant part by using the parameters. Correlations between combinations of plant parts or features are determined by using features common to various plant parts and plant parts common to various features. Next, the plant parts and/or the features are represented as information elements positioned in such a way that the distance between two information elements in each case represents the degree of their correlation.

Claims

1. A method for the analysis of process data of an industrial plant having automatically controlled plant parts, which comprises:

determining which of the plant parts are represented for determining correlation by using a prescribable criterion;

prescribing a time window as the criterion;

prescribing features characterizing a plant process and providing parameters relevant to the plant process;

checking the presence of each feature for each plant part using the parameters;

determining correlations between combinations of plant parts or features using features common to various plant parts and plant parts common to various features; and

representing at least one of the plant parts and the features as information elements positioned in relation to one another, for representing a degree of correlation of two of the information elements in each case by a distance between the two information elements.

2. The method according to claim 1, which comprises combining a specific identification of plant parts and the parameters relating to respective operating states of the plant parts to form event messages.

3. The method according to claim 1, which comprises determining the positioning of the information elements representing the plant parts and of the information elements representing the features, for preventing the distance between an information element representing a feature and an information element representing a plant part from exceeding a prescribable first limiting value if that plant part exhibits that feature, and for preventing that distance from falling below a prescribable second limiting value if that plant part does not exhibit that feature.

4. The method according to claim 1, which comprises prescribing a time window as a feature.

5. The method according to claim 2, which comprises representing the information elements of successive event messages together as a state complex.

6. The method according to claim 1, which comprises forecasting a system state from a common representation of the information elements.

7. The method according to claim 5, which comprises comparing the state complex with a reference complex derived from plant-specific knowledge.

8. The method according to claim 1, which comprises determining three spatial coordinates to fix the position of each information element.