Cost-optimized model-based extension of system life

• US 9,251,309 B2
• Filed: 09/21/2011
• Issued: 02/02/2016
• Est. Priority Date: 09/21/2011
• Status: Active Grant

• Current Assignee: GM Global Technology Operations LLC (General Motors Company)
• Original Assignee: GM Global Technology Operations LLC (General Motors Company)
• Inventors: Shin, Kwang-Keun, Salman, Mutasim A.
• Primary Examiner(s): Kundu, Sujoy
• Assistant Examiner(s): Ishizuka, Yoshihisa
• Application Number:
  US13/238,144
• Publication Number:
  US 20130073222A1
• Time in Patent Office: 1,595 Days
• Field of Search: 702/34, 702/35
• US Class Current: 1/1
• CPC Class Codes: G06F 30/00 Computer-aided design [CAD]
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass
• CPC Class Codes: Y02T 10/40 Engine management systems

Abstract

A method includes recording a current performance signature for first and second components in a system, recording a calibrated baseline performance signature for the components, and processing the performance signatures through an aging model to determine a future performance signature for each component. The future performance signatures are processed through a system function model to determine the state of function of the system for each possible repair case. A cost-optimal repair case is then determined from among the possible repair cases, and recorded in memory. An apparatus includes first and second components of a system, and a host machine configured for processing the current and baseline performance signatures through the aging and system function models as noted above. The cost-optimal repair case is determined from among all possible repair cases, and then recorded in memory. An example system may be a cranking system with a starter motor and battery.

Claims

1. A method comprising:

receiving, from a plurality of sensors via a host machine in communication with a system having a plurality n of components, a current performance signature for the plurality n of components, wherein the plurality n includes a first and a second component, and wherein the host machine includes a processor and memory on which is recorded an aging model and a system function model;

recording, via the host machine, a calibrated baseline performance signature for the first and the second components;

processing the current and calibrated baseline performance signatures through the aging model to determine a future performance signature for each of the first and second components, including comparing the current and calibrated baseline performance signatures to a recorded trace of past performance values for additional first and second components in a prior-fielded system over time;

processing the future performance signatures for the first and second components through the system function model to determine a numeric state of function (SOF) of the system for each of 2ⁿ−

1 possible repair cases, wherein the numeric SOF is 0 and 1 for a respective degraded and calibrated new system, such that the SOF degrades from 1 to 0 over time, and each of the 2ⁿ−

1 possible repair cases describes a possible repair combination of the plurality of n components; and

executing a control action when any of the numeric SOFs of the system is 0, including;

calculating a repair cost and remaining useful life (RUL) for each of the 2ⁿ−

1 possible repair cases via the host machine;

calculating a ratio of the calculated repair cost and the calculated RUL for each of the 2ⁿ−

1 possible repair cases via the host machine;

calculating a cost-optimal repair case as the lowest of the calculated ratios from among the 2ⁿ−

1 possible repair cases via the host machine; and

performing the calculated cost-optimal repair case by replacing or repairing at least one of the plurality n of components.

2. The method of claim 1, further comprising recording a desired system life for the system, wherein calculating a cost-optimal repair case as the lowest of the calculated ratios from among the 2ⁿ−

1 possible repair cases includes calculating a repair which is cost-optimal over the desired system life.

3. The method of claim 1, further comprising:

recording the current and calibrated baseline performance signatures via a data recorder; and

processing the current and calibrated baseline performance signatures via a host machine in remote communication with the data recorder.

4. The method of claim 1, wherein the system is a starter system of a vehicle, the first component is a starter motor, the second component is a battery, and the SOF is the cranking function of the starter system which includes the starter motor and the battery.

5. A method comprising:

receiving, from a plurality of sensors via a host machine in communication with a vehicle cranking system having a plurality n of components, including a starter motor and a battery a current performance signature for the plurality n of components, wherein the host machine includes a processor and memory on which is recorded an aging model and a system function model;

recording, via the host machine, a calibrated baseline performance signature for the starter motor and the battery;

processing the current and calibrated baseline performance signatures through the aging model of a host machine to determine a future performance signature for each of the components;

processing the future performance signatures for the starter motor and the battery through the system function model of the host machine to determine a numeric state of function (SOF) of the vehicle cranking system for each of 2ⁿ−

1 possible repair cases, wherein the numeric SOF is 0 and 1 for a respective degraded and calibrated new system, such that the numeric SOF degrades to 0 over time, and wherein each of the possible 2ⁿ−

1 repair cases describes a possible repair combination of the n components; and

executing a control action when any of the numeric SOFs is 0, including;

recording a desired system life for the vehicle cranking system;

calculating, via the host machine, a repair cost and remaining useful life (RUL) for each of the 2ⁿ−

1 possible repair cases;

calculating, via the host machine, a ratio of the calculated repair cost and the calculated RUL for each of the 2ⁿ−

1 possible repair cases;

calculating, via the host machine, a cost-optimal repair case as the lowest of the calculated ratios from among the 2ⁿ−

1 possible repair cases, including calculating, via the host machine, the cost-optimal repair case by calculating a repair which is cost-optimal over the recorded desired system life; and

performing the calculated cost-optimal repair case by replacing or repairing at least one of the plurality n of components.

6. The method of claim 5, further comprising:

recording the current and calibrated baseline performance signatures via a data recorder positioned aboard the vehicle;

transmitting the current and calibrated baseline performance signatures to the host machine; and

processing the current and calibrated baseline performance signatures via a host machine in remote communication with the data recorder.

7. The method of claim 5, wherein processing the current and calibrated baseline performance signatures through an aging model includes comparing the current and calibrated baseline performance signatures to a recorded trace of past performance values for a prior-fielded starter motor and battery.