Uncertainty-Aware Digital Twin Models Using Random Boundary Conditions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Digital twins fail to effectively account for uncertainties in real-world processes, leading to misinterpreted results and misinformed decisions.
Innovation Solution
Incorporating stochastic boundary conditions and generative machine learning algorithms into digital twins to model epistemic and aleatory uncertainties, allowing for more accurate simulations by separating and managing these uncertainties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If digital twins are used to simulate real-world processes, then simulation capability is improved, but accuracy deteriorates due to unaccounted uncertainties
Solution Approach 1:
The patent segments uncertainty into two distinct types: epistemic uncertainty (related to lack of knowledge) and aleatory uncertainty (inherent randomness). This segmentation allows each type to be modeled and managed separately through different mathematical frameworks, improving the overall accuracy of digital twin simulations while maintaining their adaptability across various applications.
2Measurement precision
If uncertainties are incorporated into digital twins, then simulation accuracy is improved, but system complexity increases
Solution Approach 1:
The patent introduces stochastic boundary conditions as an intermediary mechanism that bridges the gap between deterministic digital twin models and uncertain real-world processes. These boundary conditions serve as a mathematical interface that incorporates both epistemic and aleatory uncertainties without fundamentally altering the underlying deterministic model structure, thus improving accuracy while controlling complexity.
3Reliability
If stochastic boundary conditions are applied, then uncertainty management is improved, but computational requirements increase
Solution Approach 1:
The patent applies preliminary action by pre-defining stochastic boundary conditions that encapsulate uncertainty characteristics before simulations are run. These pre-characterized boundary conditions can be reused across multiple simulation scenarios, improving reliability in uncertainty management while reducing computational overhead compared to generating uncertainty models on-demand for each simulation.
Data Source
AI summary
Accounting for uncertainties in models and digital twins. A database is constructed by sampling from prior distributions of variables including epistemic variables and aleatoric variables. A model, such as a variational auto-encoder, is trained using the data stored in the database. Data from the database is input to an encoder portion of the model and the model is trained to account for uncertainties by concatenating epistemic variables to a sample of a latent space prior to proceeding with the decoder portion of the model. Once trained, a vector that includes a sample from the latent layer space and a sample from a prior distribution (or measured values) are input to the decoder to generate a solution that may be used by a digital twin. Advantageously, the prior distribution of the epistemic variables can be updated over time. The updated prior distribution improves operation of the model without retraining the model.