Runtime Fault Tree Analysis for Critical Component Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional fault tree analysis methods fail to accurately identify critical components at runtime due to the lack of consideration for conditional events and sequential dependencies, particularly with Priority AND gates, making it difficult to analyze system reliability in fault-tolerant systems.
Innovation Solution
A system and method for analyzing system reliability at runtime by inputting fault trees, calculating minimal cut sets, and recording component failures and recoveries, including observations of guard conditions to determine critical components, which enhances the accuracy of minimal cut set calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional fault tree analysis is applied in a bottom-up way to identify critical components at runtime, then the analysis can be performed systematically, but it fails to consider conditional events and sequential dependencies leading to inaccurate identification of critical components
Solution Approach 1:
The patent segments the fault tree analysis into two distinct phases: (1) static analysis phase where the fault tree is parsed and original minimal cut sets are calculated, and (2) runtime analysis phase where the system monitors component states and calculates runtime minimal cut sets based on observed failures and guard conditions. This segmentation allows traditional FTA to be combined with dynamic monitoring to achieve accurate critical component identification while managing complexity through phased processing.
Solution Approach 2:
The patent performs preliminary action by pre-calculating the original minimal cut sets from the fault tree structure before runtime analysis begins. These pre-computed cut sets serve as a foundation for the runtime analysis, where the system only needs to evaluate which cut sets are currently active based on observed component failures and guard conditions. This preliminary computation reduces the complexity of runtime analysis significantly.
2Reliability
If temporal operators are introduced into the minimal cut sets to handle sequential dependency, then the sequential dependency problem can be solved, but the complexity and cost for calculation of runtime MCS increases
Solution Approach 1:
The patent introduces dynamics by making the minimal cut sets adaptive to runtime conditions. Instead of using fixed static cut sets, the system dynamically determines which cut sets are currently active based on observed component failures and guard conditions. The runtime minimal cut sets are computed as subsets of original cut sets, allowing the analysis to adapt to changing system states without requiring complex temporal operators.
Solution Approach 2:
The patent changes parameters by introducing guard conditions that modify the evaluation of minimal cut sets based on runtime states. Rather than using temporal operators to enforce sequence constraints, the system uses parameter changes in the form of guard condition evaluations (true/false) to determine whether a cut set is currently active. This approach handles sequential dependencies through parameter evaluation rather than temporal logic.
3Ease of operation
If traditional FTA is used without considering conditional events, then the analysis is simpler, but it cannot correctly analyze runtime MCS when fault trees comprise conditional events and PAND gates
Solution Approach 1:
The patent introduces an intermediary layer between the static fault tree structure and the runtime analysis. This intermediary consists of guard conditions that are evaluated at runtime to determine whether original minimal cut sets are currently active. The guard conditions act as mediators that translate the static fault tree model into dynamic runtime predictions, enabling accurate analysis of systems with conditional events and PAND gates while maintaining the simplicity of traditional FTA for the static analysis phase.
Data Source
AI summary
The original MCS of a system fault tree includes sufficient conditions required for a top system hazard. If a fault occurs in a component and the component is restored, the current MCS of a system and the critical components can be calculated on the basis of the original MCS by means of several calculation patterns.