Resilient optimization and control for distributed systems

• US 10,146,611 B2
• Filed: 11/13/2013
• Issued: 12/04/2018
• Est. Priority Date: 11/19/2012
• Status: Active Grant

• Current Assignee: Siemens AG
• Original Assignee: Siemens Corp. (Siemens AG)
• Inventors: Liao, Linxia, Ji, Kun
• Primary Examiner(s): Patel, Kamini B
• Application Number:
  US14/440,906
• Publication Number:
  US 20150301882A1
• Time in Patent Office: 1,847 Days
• Field of Search: 714 473
• CPC Class Codes: G05B 19/4184 characterised by fault tolerance, reliability of production system
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: Y02P 80/10 Efficient use of energy, e.g. using compressed air or pressurized fluid as energy carrier
• CPC Class Codes: Y02P 90/02 Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
• CPC Class Codes: Y02P 90/80 Management or planning

Abstract

A method for controlling a system including a plurality of subsystems, includes receiving operational data from the plurality of subsystems of the system (S21). A future condition of each of the plurality of subsystems is estimated from the received operational data (S22). A control strategy for delaying a need for system maintenance is generated based on the estimated future condition of each of the plurality of subsystems (S23). An operation of the system is controlled based on the generated control strategy (S24).

Claims

1. A method for controlling a system including a plurality of subsystems, comprising:

receiving operational data from the plurality of subsystems of the system;

estimating a future condition of each of the plurality of subsystems from the received operational data, wherein the future condition of each of the plurality of subsystems is collaboratively estimated using prognostic analysis by a plurality of distributed prognostic engines that are disposed locally with respect to the plurality of subsystems, wherein the prognostic analysis comprises;

constructing a health indicator for each subsystem based on combining multiple measurements of the operational data using an evolving self-organizing map algorithm, and

calculating a prediction of remaining useful life wherein;

on a condition the health indicator has no observable trend, the prediction is based on a function of instances when the health indicator exceeds a threshold over a predetermined period of time; and

on a condition that the health indicator has an increasing trend, the prediction is based on a continuous Bayesian update of a fault progression for parameters to a predicted performance deviation reaching a threshold;

generating a control strategy for delaying a need for system maintenance based on the estimated future condition of each of the plurality of subsystems, wherein the control strategy maximizes a period of operational normalcy until the system maintenance is required; and

controlling an operation of the system based on the generated control strategy.

2. The method of claim 1, wherein the operational data includes instructions sent to the subsystems, measurements of subsystem operation, or data output from the subsystems during operation.

3. The method of claim 1, wherein the future condition of each of the plurality of subsystems is estimated by a plurality of distributed prognostic engines that are disposed locally with respect to the plurality of subsystems.

4. The method of claim 1, wherein the future condition of each of the plurality of subsystems is a measure of operational quality of the corresponding subsystem.

5. The method of claim 1, wherein the future condition of each of the plurality of subsystems is a measure of predicted time to failure of the corresponding subsystems.

6. The method of claim 1, wherein the future condition of each of the plurality of subsystems is a measure of predicted operational efficiency of the corresponding subsystem.

7. The method of claim 1, wherein the generated control strategy defines a level of utilization for each subsystem of the plurality of subsystems.

8. The method of claim 7, wherein the generated control strategy dictates that subsystems with a future condition that is estimated to be relatively poor are defined to have a relatively low level of utilization while subsystems with a future condition that is estimated to be relatively strong are defined to have a relatively high level of utilization.

9. The method of claim 1, wherein the generated control strategy maximizes a time until the system maintenance is required.

10. The method of claim 1, wherein the generated control strategy delays the need for the system maintenance until the system maintenance becomes available.

11. The method of claim 1, wherein the generated control strategy minimizes a risk of system failure by reducing utilization to one or more compromised subsystems, wherein a compromised subsystem is a subsystem that has been estimated to have a poor future condition.

12. The method of claim 11, wherein utilization of the compromised subsystems is reduced by diverting operational responsibility to other subsystems of the plurality of subsystems that are not compromised.

18. The method of claim 1, wherein the control strategy extends an amount of time at least one subsystem maintains operational normalcy after a degradation of the at least one subsystem occurs.

19. The method of claim 1, further comprising:

determining, from the operational data, that degradation of one or more of the plurality of subsystems has occurred,

wherein the control strategy maximizes the period of operational normalcy during a period of the degradation of the one or more of the plurality of subsystems.

13. A controlled system, comprising:

a plurality of subsystems, wherein at least one subsystem of the plurality of subsystems comprises a physical component;

an optimization and control engine configured to receive operational data from the plurality of subsystems, generate a control strategy for delaying a need for system maintenance based on an estimated future condition of each of the plurality of subsystems, wherein the control strategy maximizes a period of operational normalcy until the system maintenance is required, and control the operation of the system based on the generated control strategy; and

a plurality of prognostic engines disposed locally with respect to the plurality of subsystems for receiving the operational data from the subsystems and collaboratively estimating the future condition of each of the plurality of subsystems using prognostic analysis, wherein the prognostic analysis comprises;

constructing a health indicator for each subsystem based on combining multiple measurements of the operational data using an evolving self-organizing map algorithm, and

calculating a prediction of remaining useful life wherein;

on a condition the health indicator has no observable trend, the prediction is based on a function of instances when the health indicator exceeds a threshold over a predetermined period of time; and

on a condition that the health indicator has an increasing trend, the prediction is based on a continuous Bayesian update of a fault progression for parameters to a predicted performance deviation reaching a threshold.

14. The system of claim 13, wherein the operational data includes instructions sent to the subsystems, measurements of subsystem operation, or data output from the subsystems during operation.

15. The system of claim 13, wherein the prognostic engine comprises a plurality of distributed processing units that are disposed locally with respect to the plurality of subsystems.

16. The system of claim 13, wherein the future condition of each plurality of subsystems is a measure of operational quality of the corresponding subsystems, a measure of predicted time to failure of the corresponding subsystem, or a measure of predicted operational efficiency of the corresponding subsystem.

17. The system of claim 13, wherein the generated control strategy defines a level of utilization for each subsystem of the plurality of subsystems, dictates that subsystems with a future condition that is estimated to be relatively poor are defined to have a relatively low level of utilization while subsystems with a future condition that is estimated to be relatively strong are defined to have a relatively high level of utilization, maximizes a time until the system maintenance is required, delays the need for the system maintenance until the system maintenance becomes available, or minimizes a risk of system failure by reducing utilization to one or more compromised subsystems, wherein a compromised subsystem is a subsystem that has been estimated to have a poor future condition.