Hybrid System Modeling for Accurate Complex Behavior Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing modeling techniques struggle to accurately describe and predict the behavior of complex systems, particularly when dealing with systems that have multiple causes and results, as they often rely on single models that are insufficient for capturing the intricacies of real-world systems.
Innovation Solution
A hybrid system model that combines the strengths of physical-driven and data-driven models, using sub-models to infer and predict system behavior based on structural information, allowing for robust analysis and prediction across a broader range of applications.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single model is used to describe complex systems, then the model structure is simple, but the accuracy of predicting system behavior deteriorates
Solution Approach 1:
The patent divides a complex system model into multiple sub-models, each responsible for specific aspects of system behavior. This segmentation allows each sub-model to focus on particular relationships or components, improving overall prediction accuracy while maintaining manageable complexity through modular architecture
Solution Approach 2:
The patent combines multiple sub-models into an integrated hybrid model that leverages both physics-driven and data-driven approaches. By merging complementary modeling techniques, the system achieves higher prediction accuracy for complex behaviors that cannot be captured by a single modeling approach alone
2Loss of information
If physics-driven models are used to describe causal relationships, then interpretability is improved, but applicability to systems with insufficient physical knowledge deteriorates
Solution Approach 1:
The patent applies different modeling approaches to different parts of the system based on local characteristics. Physics-driven models are used where causal relationships are well-understood and interpretability is crucial, while data-driven models are applied where physical knowledge is insufficient or systems exhibit complex emergent behavior
Solution Approach 2:
The patent creates a composite modeling approach by integrating physics-driven and data-driven models into a hybrid framework. This composite model structure combines the interpretability advantages of physics-based approaches with the adaptability and pattern recognition capabilities of data-driven methods
3Adaptability or versatility
If data-driven models are used to capture complex patterns, then adaptability to diverse systems is improved, but interpretability of causal relationships deteriorates
Solution Approach 1:
The patent introduces physics-driven models as intermediary components that bridge the gap between data-driven predictions and causal understanding. These physics-based sub-models serve as mediators that provide interpretability for specific causal relationships while allowing the overall system to leverage the adaptability of data-driven approaches
4Measurement precision
If multiple sub-models are integrated to describe complex systems, then prediction accuracy is improved, but model complexity increases
Solution Approach 1:
The patent segments the complex modeling task into distinct sub-models, each handling specific aspects of system behavior. This segmentation improves prediction accuracy by allowing specialized models for different components while managing integration complexity through clear modular boundaries and defined interfaces
Data Source
AI summary
Disclosed herein is a computing system for implementing and operating a system model describing a target system, the computer system comprises a processor configured to select first data as input data of a first sub-module, based on structural information, from among new input data for the target system, provide second data as input data of a second sub-module, based on the structural information, wherein the second sub-module is defined to receive output data of the first sub-module as input data thereof by the structural information, control the second sub-model to infer a behavior of the target system based on the second data, and provide third data based on output data of the second sub-module as an output of the system model describing the behavior of the target system.