Electronics Remaining Useful Life Prediction Under Mission Profiles

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for predicting the remaining useful life (RUL) of electronics, particularly power converters, fail to consider the mission profile and operate under fixed test conditions, leading to conservative RUL estimation, underutilization of parts, or excessive inventory, and require inefficient digital twin implementations susceptible to noise and cyber-attacks.

Innovation Solution

A system and method using probability density functions, measure theory, and machine learning models on FPGAs or GPUs for in-situ health monitoring, enabling component qualification under various operating conditions, and predicting system failure probabilities by combining individual component data, with a scalable approach for digital twins.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If fixed test conditions are used for RUL prediction, then the prediction process is simple, but the RUL estimation is conservative and leads to underutilization of parts

Engineering Contradiction:

Improveprediction process complexityVSAvoidRUL estimation accuracy

Core Design Contradiction:

Device complexityVSReliability

Solution Approach 1:

The patent transitions from static fixed test conditions to dynamic mission profile-based testing. The system continuously updates RUL predictions by incorporating real-time operational data that reflects actual varying conditions, allowing the prediction model to adapt dynamically rather than relying on conservative fixed-condition assumptions

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The patent changes the testing parameters from fixed single-condition values to multiple varying conditions that mirror actual mission profiles. By varying temperature, humidity, voltage, and other environmental parameters during testing to match real operational scenarios, the system achieves more accurate RUL estimates without excessive complexity

Inventive Principle:

Principle #35Parameter changes

2Productivity

If digital twin implementations are used for health monitoring, then real-time monitoring is achieved, but the system becomes susceptible to noise and cyber-attacks

Engineering Contradiction:

Improvehealth monitoring efficiencyVSAvoidnoise and cyber-attack susceptibility

Core Design Contradiction:

ProductivityVSObject-affected harmful factors

Solution Approach 1:

The patent creates a simplified statistical model that copies only the essential reliability characteristics of the physical system rather than a complete digital twin. This statistical representation maintains the core functionality for RUL prediction while reducing the attack surface and noise susceptibility inherent in comprehensive digital twin implementations

Inventive Principle:

Principle #26Copying

Solution Approach 2:

The patent uses lightweight statistical models that are computationally inexpensive and can be rapidly updated or replaced. These simplified models serve the same practical purpose for maintenance planning but with reduced vulnerability to cyber threats and noise compared to complex persistent digital twins

Inventive Principle:

Principle #27Cheap short-living objects (Disposable)

3Productivity

If component qualification is performed at single operating condition, then testing is efficient, but the prediction does not account for mission profile effects

Engineering Contradiction:

Improvetesting efficiencyVSAvoidmission profile adaptability

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent segments the mission profile into discrete operational conditions and stressors. By dividing the complex mission profile into manageable segments (different temperatures, humidities, voltages, vibrations), the system can efficiently test each segment separately and combine the results to achieve comprehensive mission profile-based RUL prediction

Inventive Principle:

Principle #1Segmentation

4Reliability

If conservative RUL estimation is used, then system reliability is maintained, but excessive inventory is required

Engineering Contradiction:

Improvesystem reliabilityVSAvoidinventory quantity

Core Design Contradiction:

ReliabilityVSQuantity of substance

Solution Approach 1:

The patent inverts the traditional conservative approach by using mission profile-based statistical models that predict actual component behavior under real operating conditions. Instead of assuming worst-case scenarios, the system uses inverted logic that leverages actual operational data to demonstrate components can safely operate closer to their true limits, reducing inventory requirements while maintaining reliability

Inventive Principle:

Principle #13The other way round (Inversion)

Data Source

PatentUS12442851B2Systems and methods for remaining useful life prediction in electronics

Publication Date: 2025.10.14 THE GOVERNMENT OF THE UNITED STATES OF AMERICA AS REPRESENTED BY THE SECRETARY DEPARTMENT OF HEALTH & HUMAN SERVICES

AI summary

The systems and methods described herein are for remaining useful life prediction in electronics and include measuring a plurality of circuit parameters for each of a plurality of circuit components at a plurality of different temperatures, determining a probability density function of failure as a function of time for each of the plurality of circuit components and combining the probability density functions for each of the plurality of circuit components as a function of a circuit that contains the plurality of circuit components.