Bayesian Failure Diagnostics for Coincident Processing Unit Events

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Complex processing systems face extended downtimes and increased losses due to the difficulty in identifying the root cause of equipment failures, often resulting from less visible failures in interconnected units.

Innovation Solution

The implementation of a Bayesian inference engine to determine the probability of coincident failures across processing units, allowing for prioritization of investigations and remediation based on calculated probabilities, and updating the engine with occurrence data to improve future diagnostics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If traditional failure identification methods are used in complex processing systems, then the system structure remains simple and easy to understand, but the time required to identify root causes of failures increases significantly

Engineering Contradiction:

Improvedowntime for failure identificationVSAvoidcomplexity of failure analysis system

Core Design Contradiction:

Loss of timeVSDevice complexity

Solution Approach 1:

The system implements feedback loops where failure data from processing units is continuously collected, analyzed, and used to update probabilistic models. The Bayesian inference engine processes new failure information and updates the likelihood of potential root causes, creating a dynamic feedback mechanism that improves diagnostic accuracy over time while managing system complexity through iterative learning.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent introduces a Bayesian inference engine as an intermediary between raw failure data and root cause identification. This intermediary component processes complex probabilistic relationships and transforms multiple potential failure causes into ranked likelihoods, simplifying the diagnostic process without requiring direct analysis of all possible failure pathways.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Measurement precision

If comprehensive monitoring of all processing units is implemented to identify root causes, then the precision of failure diagnosis improves, but the complexity and cost of the monitoring system increases

Engineering Contradiction:

Improveprecision of failure root cause identificationVSAvoidcomplexity of monitoring system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The system replaces complex mechanical monitoring infrastructure with a computational approach using Bayesian inference. Instead of implementing elaborate physical monitoring systems for every processing unit, the patent uses software-based probabilistic modeling that processes standard operational data to achieve high diagnostic precision, substituting computational complexity for physical system complexity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent transforms the monitoring approach by changing from direct observation of all possible failure modes to probabilistic estimation based on limited observations. The Bayesian framework allows the system to infer the state of unmonitored parameters (potential root causes) from observed parameters (failure symptoms), achieving comprehensive monitoring coverage without proportionally increasing monitoring infrastructure.

Inventive Principle:

Principle #35Parameter changes

3Reliability

If redundant critical processing units are deployed to prevent facility shutdown, then the reliability of the processing facility improves, but the loss of time and resources for identifying and correcting root causes increases

Engineering Contradiction:

Improvefacility operational reliabilityVSAvoiddowntime for root cause correction

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by continuously analyzing operational data and identifying potential failure risks before they manifest as actual failures. The Bayesian inference engine calculates probabilities of potential root causes in advance, allowing operators to take preventive actions or prepare remediation strategies before failures occur, reducing both the frequency and duration of facility shutdowns.

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11703846B2Equipment failure diagnostics using Bayesian inference

Publication Date: 2023.07.18 CAMERSON INT CORP

AI summary

A method is described herein, comprising registering an event at a first processing unit of a processing facility comprising a plurality of processing units, using a coincidence probability array and an event probability to identify a second processing unit of the plurality of processing units based on the event, determining whether the second processing unit experienced a coincident event, if the second processing unit experienced a coincident event, remediating a condition of the second processing unit that caused the coincident event, and updating the coincidence probability array based on the event.