Combined expert system/neural networks method for process fault diagnosis

• US 5,442,555 A
• Filed: 10/07/1993
• Issued: 08/15/1995
• Est. Priority Date: 05/18/1992
• Status: Expired due to Fees

• Current Assignee: University of Chicago
• Original Assignee: Argonne National Laboratory (Government of the United States of America)
• Inventors: Reifman, Jaques, Wei, Thomas Y. C.
• Primary Examiner(s): Voeltz, Emanual T.
• Assistant Examiner(s): Tran, Alan
• Application Number:
  US08/132,888
• Time in Patent Office: 677 Days
• Field of Search: 364/551.01, 364/495, 376/215, 376/267, 376/433, 376/245, 376/217, 376/255, 395/915
• US Class Current: 701/99
• CPC Class Codes: G21D 3/04 Safety arrangements emergency protection of reactor G21C9/00
• CPC Class Codes: Y02E 30/00 Energy generation of nuclear origin
• CPC Class Codes: Y02E 30/30 Nuclear fission reactors
• CPC Class Codes: Y10S 706/915 Power plant

Abstract

A two-level hierarchical approach for process fault diagnosis is an operating system employs a function-oriented approach at a first level and a component characteristic-oriented approach at a second level, where the decision-making procedure is structured in order of decreasing intelligence with increasing precision. At the first level, the diagnostic method is general and has knowledge of the overall process including a wide variety of plant transients and the functional behavior of the process components. An expert system classifies malfunctions by function to narrow the diagnostic focus to a particular set of possible faulty components that could be responsible for the detected functional misbehavior of the operating system. At the second level, the diagnostic method limits its scope to component malfunctions, using more detailed knowledge of component characteristics. Trained artificial neural networks are used to further narrow the diagnosis and to uniquely identify the faulty component by classifying the abnormal condition data as a failure of one of the hypothesized components through component characteristics. Once an anomaly is detected, the hierarchical structure is used to successively narrow the diagnostic focus from a function misbehavior, i.e., a function oriented approach, until the fault can be determined, i.e., a component characteristic-oriented approach.

Claims

1. A method for diagnosing failures in the operation of a thermal-hydraulic system including detection of a malfunctioning component in said thermal-hydraulic system, said method comprising the steps of:

classifying a malfunction of the component by a function performed by the component as a mass source/sink, or a momentum source/sink, or an energy source/sink imbalance including the steps of;

assigning to one or more of the components of the thermal-hydraulic system a thermal-hydraulic control volume which characterizes the total mass, momentum, and energy inventories of said one or more components during normal and off-normal operations;

assigning to each thermal-hydraulic control volume its associated mass, momentum, and energy conservation equations;

assigning to one or more components a functional classification as a source or sink of mass, momentum, or energy;

assigning components to generic component classes by said functional classification;

monitoring operation of the thermal-hydraulic control volumes for detecting an imbalance in the total mass, momentum, or energy inventories in any of the thermal-hydraulic control volumes; and

comparing a detected imbalance in mass, momentum, or energy inventories of the thermal-hydraulic control volumes with the conservation equations and the functional classification for each of the components and identifying a given component as faulty when a detected imbalance in mass, momentum, or energy of the thermal-hydraulic control volume matches the classification of the component as a source or sink of mass, momentum, or energy;

classifying the function as one of a plurality of generic component classes for each of said mass, momentum and energy source/sink imbalances; and

classifying a specific component within one of said generic component classes as said malfunctioning component.

2. The method of claim 1 wherein the control volume imbalances in mass, momentum and energy inventories are determined from trends in the associated thermal-hydraulic and thermodynamic properties such as temperature and pressure through equations of state.

3. The method of claim 1 wherein the step of classifying the malfunction of the component by a function performed by the component includes associating each system component with a designated control volume characterized by associated thermal-hydraulic and thermodynamic properties such as temperature and pressure.

4. The method of claim 3 wherein a component malfunction causes properties of associated and surrounding control volumes to vary causing imbalances of mass, momentum and energy inventories.

5. The method of claim 4 wherein the step of classifying the malfunction of a component by a function performed by the component further includes determining imbalances in mass, momentum and energy sources or sinks from imbalances in mass, momentum and energy inventories through application of the macroscopic conservation equations of mass, momentum, and energy.

6. The method of claim 5 further comprising the step of designating faulty components by relating said imbalances of mass, momentum and energy sources or sinks in said conservation equations with a function of a plurality of generic components.

7. The method of claim 1 wherein the step of classifying the function as one of a plurality of generic component classes for each of said mass, momentum and energy source/sink imbalances includes classifying the function by generic component thermal-hydraulic characteristics.

8. The method of claim 7 wherein the step of classifying the function further as one of a plurality of generic component classes for said momentum source/sink imbalance includes the step of classifying the function as active such as that of pumps, or passive such as that of open valves.

9. The method of claim 8 wherein the step of classifying the function further as one of a plurality of generic component classes for said momentum source/sink imbalances for a pump includes determining and using pump head flow characteristics.

10. The method of claim 8 wherein the step of classifying the function further as one of a plurality of generic component classes for said momentum source/sink imbalance for a valve includes determining and using valve flow head loss characteristics.

11. The method of claim 1 wherein the step of classifying the specific component within one of said generic component classes as said malfunctioning component includes classifying the specific component by specific component thermal hydraulic characteristics.

12. The method of claim 11 wherein the step of classifying a specific power operated relief valve within one of said generic component classes for mass source/sink imbalances as said malfunctioning component includes determining and using power operated relief valve pressure flow characteristics.

13. The method of claim 1 wherein the step of classifying the malfunction of the component by a function performed by the component as a mass source/sink, or a momentum source/sink, or an energy source/sink imbalance includes utilizing an expert system in identifying the imbalance with a plurality of generic components.

14. The method of claim 1 wherein the steps of classifying the function as one of a plurality of generic component classes and classifying a specific component within one of a plurality of generic component classes includes utilizing artificial neural networks in differentiating between classes of generic components and between individual components within a given class.