Method for Modelica-based system fault analysis at the design stage

Patent metadata

Field	Value
Publication number	US-10558766-B2
Application number	US-201615392196-A
Country	US
Kind code	B2
Filing date	Dec 28, 2016
Priority date	Dec 31, 2015
Publication date	Feb 11, 2020
Grant date	Feb 11, 2020

Abstract

A new and/or improved method, apparatus and/or system is disclosed which aids in extending correct behavioral models to include fault modes and in fault mode analysis of components and/or systems in simulated model environments, including, e.g., FMEA and FMECA and diagnostic fault tree generation.

First claim

The invention claimed is:

1. A method to predict failure of a system, the method comprising:
analyzing the system to identify fault susceptible components of the system;
augmenting component models of the of the fault susceptible components with fault modes;
using the augmented component models to simulate faults and to determine a system-level severity;
applying parameterized physics-of-failure models corresponding to a root cause of the simulated faults to predict a fault likelihood;
combining the system-level severity with the predicted fault likelihood to predict component degradations over time;
aggregating the predicted component degradations to predict when the system will fail to meet performance requirements; and
deriving a fault tree by simulating fault modes with a varying fault amount.

2. The method of claim 1 , wherein the prediction of when the system will fail to meet the performance requirements further comprises using a set of initial conditions, faults and component ages to determine a conditional probability that the system meets a requirement.

3. The method of claim 1 , wherein the prediction of when the system will fail to meet the performance requirements is used to optimize a design of the system or components, or to optimally select among possible subsystem/component options from vendors.

4. The method of claim 3 , wherein the design is optimized by an explicit quantitative Fault Modes, Effects and Criticality Analysis (FMECA) obtained by combining the system-level severity of faults with the predicted fault likelihood.

5. The method of claim 1 , wherein the prediction of when the system will fail to meet the performance requirements includes a mean time to failure.

6. The method of claim 1 , wherein the prediction of when the system will fail to meet the performance requirements includes a full probability distribution.

7. The method of claim 1 , wherein the prediction of when the system will fail to meet the performance requirements includes a full probabilistic trajectory of system dynamics.

8. The method of claim 1 , wherein a fault mode of the fault modes is a short circuit.

9. The method of claim 1 , wherein a fault mode of the fault modes is that a shaft is harder to turn than normal.

10. The method of claim 1 , wherein the prediction of when the system will fail to meet the performance requirements is used to select among possible subsystem/component options.

11. The method of claim 1 , wherein the fault amount is linked stochastically to system usage.

12. The method of claim 1 , wherein the fault modes include a worn clutch, and the method further comprises creating a feedback loop that shows how the worn clutch increases a load on an engine to maintain a same velocity for a vehicle.

13. The method of claim 1 , wherein the method further comprises creating a feedback loop showing how a fault mode of the fault modes affects a variable.

14. An apparatus to predict failure of a system, the apparatus comprising one or more processors configured for:
analyzing the system to identify fault susceptible components of the system;
augmenting component models of the of the fault susceptible components with fault modes;
using the augmented component models to simulate faults and to determine a system-level severity;
applying parameterized physics-of-failure models to the simulated faults to predict a fault likelihood;
combining the system-level severity with the predicted fault likelihood to predict component degradations over time; and
aggregating the predicted component degradations to predict when the system will fail to meet performance requirements; and
deriving a fault tree by simulating fault modes with a varying fault amount.

Assignees

• Palo Alto Res Ct Inc

Inventors

• Saha Bhaskar
• Honda Tomonori
• Matei Ion
• Bobrow Daniel G
• Dekleer Johan
• Janssen William C
• Kurtoglu Tolga

Classifications

• G05B23/0243

  model based detection method, e.g. first-principles knowledge model · CPC title

• G06F17/18

  for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)} · CPC title

• G06F11/008 Primary

  Reliability or availability analysis · CPC title

• G06N7/023

  Learning or tuning the parameters of a fuzzy system · CPC title

• G06F30/20 Primary

  Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30) · CPC title