System and method for supporting a fault cause analysis

• US 20030135786A1
• Filed: 01/22/2003
• Published: 07/17/2003
• Est. Priority Date: 07/22/2000
• Status: Active Grant

• Current Assignee: Intellectual Ventures II LLC (Intellectual Ventures LLC)
• Original Assignee: ABB Research Ltd. (ABB Limited)
• Inventors: Hu, Zaijun, Vollmar, Gerhard, Kallela, Jari, Greulich, Manuel
• Granted Patent: US 7,200,774 B2
• Time in Patent Office: Days
• US Class Current: 714/25
• CPC Class Codes: G05B 23/0251 Abstraction hierarchy, e.g. "complex systems", i.e. system is divided in subsystems, subsystems are monitored and results are combined to decide on status of whole system
• CPC Class Codes: G05B 23/0278 Qualitative, e.g. if-then rules; Fuzzy logic; Lookup tables; Symptomatic search; FMEA

Abstract

A system and method for supporting a fault cause analysis in a fault event in a plant includes a data processor with memory storing a fault model of XML files accessed by a fault cause navigator and an operating/display device. Each fault model contains an industry-specific process model divided into process steps, with steps and defined fault events needed therefor assigned to plant components/systems, and fault trees assigned to fault events and having fault hypotheses. A checklist with symptoms for verification of the fault hypothesis is assigned to the fault hypotheses. The system enables navigation to the relevant step in the process model by the display and navigator, and presents a fault event list. Following fault event selection, critical components/systems corresponding thereto are found and displayed. Possible symptoms are generated and displayed in a checklist and hypotheses of possible fault causes, contained in the fault trees, are found and displayed.

Claims

1. A system for supporting a fault cause analysis of a fault event in an industrial plant having plant components and plant systems, comprising:

an operating and display device for receiving input data and displaying output data;

a fault cause navigator;

a data processing device having a fault model memory storing at least one hierarchically structured cause/effect model in XML files, said at least one cause/effect model having;

an industry-specific process model divided into process steps, respective ones of said process steps and defined fault events for said process steps being associated with one of the plant components and the plant systems;

fault trees associated with said fault events having fault hypotheses; and

a checklist with symptoms for verification of each of said fault hypotheses associated with a respective one of said fault hypotheses;

said operating and display device and said fault cause navigator connected to said fault model memory and accessing said at least one cause/effect model; and

said operating and display device, said fault cause navigator, and said data processing device programmed;

a) to enable user navigation to a relevant one of said process steps in said at least one cause/effect model through said operating and display device and said fault cause navigator, and to present to a user a fault event list;

b) after selection of a fault event in said fault event list by the user, to find and display critical ones of said plant components and said plant systems corresponding to said selected fault event;

c) after selection of at least one of said critical plant components and said critical plant systems, to generate possible symptoms and display said symptoms in a checklist; and

d) after selection of at least one of said symptoms as a probable symptom, to find and display hypotheses of possible fault causes contained in said fault trees.

2. The system according to claim 1, wherein said operating and display device, said fault cause navigator, and said data processing device are programmed, after displaying said hypotheses of possible fault causes contained in said fault trees, to one of:

enable user confirmation of one of said displayed hypotheses as an actual fault cause; and

mark a fact that none of said displayed hypotheses has been confirmed by the user; and

if a hypothesis is confirmed as relevant, to generate and output a fault path, and, if no hypothesis is confirmed as relevant, to repeat steps b), c), and d) by finding and displaying other ones of said hypotheses in respective lower levels of at least one cause/effect model and respectively enable user confirmation of one of said displayed hypotheses as an actual fault cause.

3. The system according to claim 1, including a web server administering said XML files and a web browser, said data processing device with said fault model memory being part of said web server, said operating and display device and said fault cause navigator being a web client by which a user accesses said at least one cause/effect model through said web browser and said fault cause navigator.

5. The system according to claim 1, which further comprises:

enabling the user, in a query step following the fourth step, one of;

to confirm one of the displayed hypotheses as an actual fault cause; and

to mark a fact that none of the hypotheses has been confirmed;

if a hypothesis is confirmed as relevant, generating and outputting a fault path; and

if no hypothesis is confirmed as relevant, performing a repetition of the second to fourth steps to find and display other hypotheses in respective lower levels of at least one cause/effect model and in each case to connect the other hypotheses to the query step.

4. A method for supporting a fault cause analysis of a fault event in an industrial plant having plant components and plant systems, which comprises:

providing a data processing device with a fault model memory;

storing in the fault model memory at least one hierarchically structured cause/effect model in the form of XML files, the at least one cause/effect model having;

an industry-specific process model divided into process steps, respective ones of the process steps and defined fault events for the process steps being associated with one of the plant components and the plant systems;

fault trees associated with the fault events, the fault trees having fault hypotheses; and

a checklist with symptoms for verification of each of the fault hypotheses associated with a respective one of the fault hypotheses;

accessing the at least one cause/effect model through a fault cause navigator and an operating and display device;

navigating, by a user, to a relevant one of the process steps in the at least one cause/effect model in a first search step through the operating and display device and the fault cause navigator, and presenting the user with a fault event list;

in a second search step following selection of a fault event in the fault event list by the user, finding and displaying critical ones of the plant components and the plant systems corresponding to the selected fault event;

in a third search step following selection of at least one of the critical plant components and plant systems, generating possible symptoms and displaying the possible symptoms in a checklist; and

in a fourth step following selection of at least one of the symptoms as a probable symptom, finding and displaying hypotheses of possible fault causes, the hypotheses contained in the fault trees.

6. The system according to claim 4, which further comprises:

providing the data processing device with the fault model memory as part of a web server;

administering the XML files with the web server;

configuring the operating and display device and the fault cause navigator as a web client; and

accessing, through the web client, the at least one cause/effect model through a web browser and the navigator.

7. A method for supporting a fault cause analysis of a fault event in an industrial plant having plant components and plant systems, which comprises:

providing a data processing device with a fault model memory;

storing in the fault model memory at least one hierarchically structured cause/effect model in the form of XML files, the at least one cause/effect model having;

an industry-specific process model divided into process steps, respective ones of the process steps and defined fault events for the process steps being associated with one of the plant components and the plant systems;

fault trees associated with the fault events, the fault trees having fault hypotheses; and

a checklist with symptoms for verification of each of the fault hypotheses associated with a respective one of the fault hypotheses;

accessing the at least one cause/effect model through a fault cause navigator and an operating and display device;

navigating, by a user, to a relevant one of the process steps in the at least one cause/effect model through the operating and display device and the fault cause navigator, and presenting the user with a fault event list;

finding and displaying critical ones of the plant components and the plant systems corresponding to the selected fault event;

generating possible symptoms and displaying the possible symptoms in a checklist; and

finding and displaying hypotheses of possible fault causes, the hypotheses contained in the fault trees.

8. The system according to claim 7, which further comprises:

enabling the user, in a query step following the step of finding and displaying hypotheses of possible fault causes, one of;

to confirm one of the displayed hypotheses as an actual fault cause; and

to mark a fact that none of the hypotheses has been confirmed;

if a hypothesis is confirmed as relevant, generating and outputting a fault path; and

if no hypothesis is confirmed as relevant, performing a repetition of the first finding and displaying step, the generating step, and the second finding and displaying step to find and display other hypotheses in respective lower levels of at least one cause/effect model and in each case to connect the other hypotheses to the query step.

9. The system according to claim 8, which further comprises:

providing the data processing device with the fault model memory as part of a web server;

administering the XML files with the web server;

configuring the operating and display device and the fault cause navigator as a web client; and

accessing, through the web client, the at least one cause/effect model through a web browser and the navigator.