System and method for equipment remaining life estimation
AssigneeGeneral Electric Company
InventorsKai Frank Goebel, Piero Patrone Bonissone, Weizhong Yan, Neil Holger White Eklund, Feng Xue
Life of this patent
Estimating the term…
Abstract
A method to reduce uncertainty bounds of predicting a remaining life of a probe using a set of diverse models is disclosed. The method includes generating an estimated remaining life output by each model of the set of diverse models, aggregating each of the respective estimated remaining life outputs via a fusion model, and in response to the aggregating, predicting the remaining life, the predicting having reduced uncertainty bounds based on the aggregating. The method further includes generating a signal corresponding to the predicted remaining life of the probe.
Claims22
A method to reduce uncertainty bounds of predicting a remaining life of a probe using a plurality of diverse models for predicting the remaining life of the probe, the method comprising:
- generating a plurality of estimated remaining life outputs, wherein each of the estimated remaining life outputs is generated using a respective one of the diverse models;
- aggregating the estimated remaining life outputs via a fusion model;
- in response to the aggregating, predicting the remaining life of the probe, the predicted remaining life having reduced uncertainty bounds based on the aggregating;
- generating a signal corresponding to the predicted remaining life of the probe;
- and
- defining at least one of a parameter and a structure of the fusion model, wherein the defining step comprises:
- defining the parameter of the fusion model, the parameter comprising a weight corresponding to each of the diverse models, wherein the step of defining the parameter comprises:
- making available input conditions describing more than one region of a feature space;
- and
- defining a local weight corresponding to each of the more than one regions of the feature space.
A method to reduce uncertainty bounds of predicting a remaining life of a probe using a plurality of diverse models for predicting the remaining life of the probe, the method comprising:
- generating a plurality of estimated remaining life outputs, wherein each of the estimated remaining life outputs is generated using a respective one of the diverse models;
- aggregating the estimated remaining life outputs via a fusion model;
- in response to the aggregating, predicting the remaining life of the probe, the predicted remaining life having reduced uncertainty bounds based on the aggregating;
- generating a signal corresponding to the predicted remaining life of the probe;
- defining at least one of a parameter and a structure of the fusion model;
- and
- tuning and maintaining at least one of the parameter and the structure via an optimization wrapper, wherein the tuning and maintaining comprises:
- receiving historical data regarding at least one of the parameter and the structure;
- encoding at least one of the parameter and the structure for an evolutionary algorithm;
- creating a performance metric for the estimating;
- tuning at least one of the parameter and the structure using the optimization wrapper to optimize the performance metric;
- and
- storing the tuned at least one of the parameter and the structure for subsequent receiving.
A system to estimate a remaining life of a probe using a plurality of diverse models, the system comprising:
- a processor;
- a computational model application for executing on the processor, the computational model application performing a method, comprising:
- generating a plurality of estimated remaining life outputs, wherein each of the estimated remaining life outputs is generated using a respective one of the diverse models;
- aggregating the respective estimated remaining life outputs via a fusion model;
- in response to the aggregating, predicting the remaining life of the probe, the predicted remaining life having reduced uncertainty bounds based on the aggregating;
- and
- defining at least one of a parameter and a structure of the fusion model, wherein the defining by the computational model application comprises defining the parameter of the fusion model, the parameter comprising a weight corresponding to each respective one of the diverse models, and wherein the defining the parameter by the computational model application comprises:
- making available input conditions describing more than one region of a feature space;
- and
- defining a local weight corresponding to each of the more than one regions of the feature space;
- wherein the processor is responsive to the computational model application to generate a signal corresponding to the predicted remaining life of the probe.
Dependencies are inferred from the claim text.