Sparse Non-Congruent Simulation Aggregation for Rare Event Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current stochastic simulation methods require excessive processing resources and time, especially when dealing with large numbers of simulations involving rare events, and lack the ability to combine variables with different probabilities or allow for real-time adjustments.
Innovation Solution
The implementation of sparse and non-congruent stochastic simulation techniques, which store only significant simulation trials and allow for dynamic adjustment of likelihoods, enabling efficient aggregation and real-time modeling of mitigation strategies across multiple assets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If large numbers of simulations are performed to accurately model rare events, then measurement precision is improved, but use of energy and loss of time worsen
Solution Approach 1:
The patent extracts and stores only the significant simulation trials (those representing rare events) separately from the bulk data. By taking out only the critical information needed for accurate rare event modeling and storing it in a condensed format, the system achieves high measurement precision without requiring energy-intensive processing of all simulation trials.
Solution Approach 2:
The simulation data is segmented into two distinct parts: bulk simulation results and significant rare event trials. This segmentation allows the system to process and store data differently based on its importance, enabling accurate rare event analysis with minimal processing resources by focusing computational effort only on the segmented rare event portion.
2Measurement precision
If large numbers of simulations are performed to accurately model rare events, then measurement precision is improved, but loss of time worsens
Solution Approach 1:
The system performs preliminary identification and extraction of significant rare event trials during the simulation process, storing them in advance for later analysis. This preliminary action ensures that when rare event analysis is needed, the precise data is already prepared and stored, eliminating the need for time-consuming re-simulations.
Solution Approach 2:
The patent creates a condensed copy of only the significant rare event trials from the full simulation dataset. This copy contains all the essential information needed for accurate rare event modeling but represents a tiny fraction of the original data volume, enabling fast analysis without processing the complete simulation set.
3Measurement precision
If current simulation methods are used to combine variables with different probabilities, then measurement precision is improved, but device complexity worsens
Solution Approach 1:
The patent transforms the representation of simulation data by changing parameters from storing complete trial sequences to storing condensed significant trial information with associated probability weights. This parameter change enables accurate aggregation of variables with different probabilities while simplifying the system architecture by eliminating complex data processing requirements.
4Adaptability or versatility
If real-time adjustments are implemented in simulation models, then adaptability is improved, but use of energy and loss of time worsen
Solution Approach 1:
The system extracts and stores significant rare event trials in a condensed format that enables rapid re-analysis. When real-time adjustments are needed, the system can quickly re-process only the extracted significant trials rather than re-simulating entire systems, achieving adaptability with minimal additional energy consumption.
Data Source
AI summary
A method of efficiently modeling changes to mitigation of an occurrence of a rare event for a number of simulation trials includes obtaining or generating a number of sparse simulation trials of a simulation including a total number of simulation trials (x) associated with the N occurrences of a rare event, assigning fractions of 1/N to N/N to the sparse simulation trials, filtering the sparse simulation trials by the assigned fractions by a percentage corresponding to y/N to simulate a mitigation of the likelihood of failure, and outputting sparse simulation trials that are less than the percentage to statistically represent the effects of the mitigation on the total number of trials.