Machine fault information detection and reporting

• US 7,142,990 B2
• Filed: 04/21/2003
• Issued: 11/28/2006
• Est. Priority Date: 04/22/2002
• Status: Expired due to Term

• Current Assignee: Computational Systems Incorporated (Emerson Electric Company)
• Original Assignee: CSI Technology Incorporated (Emerson Electric Company)
• Inventors: Bouse, Kai, Reeves, Todd
• Primary Examiner(s): Hoff, Marc S.
• Assistant Examiner(s): O MALLEY, MARY CATHERINE
• Application Number:
  US10/419,444
• Publication Number:
  US 20040019461A1
• Time in Patent Office: 1,317 Days
• Field of Search: 702 31- 35, 702/189, 702/188, 700/2, 700/9, 700/19
• US Class Current: 702/35
• CPC Class Codes: G05B 23/0229 knowledge based, e.g. expert systems; genetic algorithms
• CPC Class Codes: G05B 23/0262 Confirmation of fault detection, e.g. extra checks to confirm that a failure has indeed occurred

Abstract

Sensors detect machine parameters such as vibration, turning speed, and temperature, and a local processor performs logic operations based on the parameters and inference rules to produce fault information that is reported to a system processor, which selectively uses the fault information to control the plant machines. The inference rules include rules for determining faults as to balance, alignment, bearing condition, electrical condition, and cavitation. The inference rules are contained in rules code that is separate from the analysis code that performs the logical operations. Thus, the inference rules may be easily changed without changing other code, such as the analysis code.

Claims

1. A fault detection system for detecting faults in a plurality of machines, the system comprising:

a plurality of sensors for detecting parameters from the machines and producing parameter signals corresponding to the detected parameters;

a plurality of local processors, each local processor associated with a subset of the machines and a corresponding subset of the sensors, each local processor receiving the parameter signals from the subset of sensors with which the local processor is associated, each local processor programmed with a plurality of inference rules that define a logic operation to be performed on one or more of the parameter signals to detect a fault in at least one of the machines with which the local processor is associated, and each local processor programmed to perform the logic operation on the one or more parameter signals as defined by the inference rules to detect faults in at least one of the machines with which the local processor is associated;

a system processor in communication with the plurality of local processors; and

each local processor programmed to transmit to the system processor fault information indicating at least whether a fault has been detected in at least one of the machines with which the local processor is associated, whereby the system processor may receive the fault information.

2. The system of claim 1 wherein at least a subset of the inference rules are selected from a group of rules consisting of:

balance rules for determining a balance fault;

alignment rules for determining an alignment of fault;

bearing rules for determining a bearing fault;

electrical rules for determining an electrical fault; and

cavitation rules for determining a cavitation fault.

3. The system of claim 1 wherein one or more of the local processors are programmed with an analysis code for performing the logic operation and a rules code containing the inference rules, the rules code configured such that the inference rules may be changed without changing the analysis code.

4. The system of claim 1 wherein one or more of the local processors are programmed with inference rules that are customized for a particular one of the plurality of machines.

5. The fault detection system of claim 1 wherein the inference rules define on which of the parameter signals the logic operation is to be performed.

6. A fault detection system for detecting faults in a plurality of machines, the system comprising a plurality of local processors, each associated with one or more of the machines, each local processor receiving sensor information indicative of one or more operational characteristics of one or more of the machines, each local processor for performing a logic operation according to an analysis code to detect a fault in one or more of the machines based on the sensor information, the analysis code performing the logic operation to detect a fault according to inference rules contained in a rules code that is not embedded in the analysis code.

7. The fault detection system of claim 6 wherein each local processor has the analysis code and the rules code programmed therein as separate modules.

8. The fault detection system of claim 6 wherein the rules code is operable to be modified to change the inference rules without modification of the analysis code.

9. The fault detection system of claim 6 wherein the analysis code accesses the inference rules in the rules code according to a protocol provided by an interface protocol code.

10. A fault detection system for detecting faults in a plurality of machines, the system comprising:

one or more local processors, each associated with one or more of the machines, each local processor receiving sensor information indicative of one or more operational characteristics of one or more of the machines, each local processor for performing one or more logic operations to detect a fault in one or more of the machines based on the sensor information;

analysis code for execution in the one or more local processors to perform the one or more logic operations to detect a fault according to one or more inference rules; and

rules code including the one or more inference rules, the rules code accessible by the analysis code and reprogrammable independently of the analysis code.

11. The fault detection system of claim 10 wherein the rules code includes one or more inference rules selected from a group of inference rules comprising:

balance rules for determining a balance fault;

alignment rules for determining an alignment fault;

bearing rules for determining a bearing fault;

electrical rules for determining an electrical fault; and

cavitation rules for determining a cavitation fault.

12. The fault detection system of claim 10 farther comprising an interface protocol code for providing a protocol by which the analysis code accesses the inference rules in the rules code.

13. A fault detection system for detecting faults in a plurality of machines, the system comprising a plurality of local processors, each associated with one or more of the machines, each local processor receiving sensor information indicative of one or more operational characteristics of one or more of the machines, each local processor for performing a logic operation according to an analysis code to detect a fault in one or more of the machines based on the sensor information, the analysis code performing the logic operation according to inference rules contained in a rules code that is not embedded in the analysis code, wherein the analysis code accesses the inference rules in the rules code according to a protocol provided by an interface protocol code, and the interface protocol code is part of the analysis code.

14. A fault detection system for detecting faults in a plurality of machines, the system comprising:

one or more local processors, each associated with one or more of the machines, each local processor receiving sensor information indicative of one or more operational characteristics of one or more of the machines, each local processor for performing one or more logic operations to detect a fault in one or more of the machines based on the sensor information;

analysis code for execution in the one or more of the local processors to perform the one or more logic operations according to one or more inference rules;

rules code including the one or more inference rules, the rules code accessible by the analysis code and reprogrammable independently of the analysis code; and

an interface protocol code for providing a protocol by which the analysis code accesses the inference rules in the rules code, wherein the interface protocol code is part of the analysis code.

15. A fault detection system for detecting faults in a plurality of machines, the system comprising a plurality of local processors, each associated with one or more of the machines, each local processor for performing a logic operation according to an analysis code to detect a fault in one or more of the machines, the analysis code performing the logic operation to detect a fault according to inference rules contained in a rules code, the analysis code accessing the inference rules in the rules code according to an interface protocol code that establishes a protocol whereby the analysis code accesses the rules code, whereby the rules code may be changed to provide a new rules code that conforms to the interface protocol established by the interface protocol code.

16. The fault detection system of claim 15 wherein the rules code may be revised without changing either of the analysis code and the interface protocol code.

17. A fault detection system for detecting faults in a plurality of machines, the system comprising a plurality of local processors, each associated with one or more of the machines, each local processor for performing a logic operation according to an analysis code to detect a fault in one or more of the machines, the analysis code performing the logic operation to detect a fault according to inference rules contained in a rules code, the analysis code accessing the inference rules in the rules code according to an interface protocol code that establishes a protocol whereby the analysis code accesses the rules code, whereby the rules code may be changed to provide a new rules code that conforms to the interface protocol established by the interface protocol code, and wherein the interface protocol code is part of the analysis code.

18. A fault detection system for detecting faults in a plurality of machines, the system comprising a plurality of local processors, each associated with one or more of the machines, each local processor for performing a logic operation according to an analysis code to detect a fault in one or more of the machines, the analysis code performing the logic operation to detect a fault according to inference rules contained in a rules code, the analysis code accessing the inference rules in the rules code according to an interface protocol code that establishes a protocol whereby the analysis code accesses the rules code, wherein one or more of the local processors are programmed with inference rules that are customized for a particular one of the plurality of machines.