Symbolic Expression Propagation for Model Reconfiguration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional modeling and simulation tools are limited in propagating symbolic expressions across multiple entities in a graphical model, requiring restarts when input values change and involving significant memory and execution overhead for different applications.
Innovation Solution
A symbol propagation engine that identifies and propagates symbolic expressions based on connectivity information and behavioral constraints across entities in a graphical model, generating an updated model while preserving the relationship between symbolic expressions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional propagation methods are used with numeric values only, then the simulation can be completed with finalized attributes, but the system requires restarts when input values change and cannot support reconfiguration without code regeneration
Solution Approach 1:
The patent changes the parameter representation from concrete numeric values to symbolic expressions. This allows the system to maintain propagation relationships in symbolic form, enabling parameter changes without restarting the simulation. The symbolic expressions can be instantiated with different numeric values later, providing adaptability while avoiding repeated propagation computations.
Solution Approach 2:
The patent performs symbolic propagation in advance to establish the structural relationships and propagation rules between entities. By pre-computing the symbolic form of attribute relationships, the system prepares the propagation framework beforehand, allowing rapid reconfiguration when numeric values change without needing to restart the entire propagation process.
2Loss of information
If symbolic expressions are used in conventional tools with single-entity limitation, then symbolic representation is possible, but the expressions cannot be propagated through the graphical model and require significant memory overhead
Solution Approach 1:
The patent segments the symbolic expression propagation into discrete rules associated with each entity type. Each entity has specific propagation rules that define how symbolic expressions are transformed as they pass through that entity. This segmentation allows efficient local processing of symbolic expressions without requiring the entire model to be held in memory simultaneously, reducing overall memory overhead.
Solution Approach 2:
The patent creates a universal symbolic propagation framework that can handle multiple entity types and attribute kinds through a common set of propagation rules. This multi-functional approach allows the same symbolic expression mechanism to propagate various types of attributes (dimensions, sample rates, data types) through different entities, eliminating the need for separate symbolic representation systems for each attribute type and reducing memory requirements.
3Productivity
If numeric values are propagated through the model, then the simulation can proceed with concrete values, but the code must be regenerated when attributes need to be re-targeted to different applications
Solution Approach 1:
The patent maintains symbolic expressions throughout propagation to preserve the structural relationships between attributes. The symbolic form allows the same propagated model to be instantiated with different numeric parameter sets for different applications without regenerating the code structure. Only the numeric instantiation changes, not the underlying propagation logic or code architecture.
Solution Approach 2:
The patent creates a symbolic copy of the propagation relationships that can be instantiated multiple times with different numeric values. Instead of working directly with concrete numeric values that are tied to a specific application, the system creates a symbolic representation that can be copied and instantiated for different applications, enabling code reuse while maintaining simulation productivity.
Data Source
AI summary
A graphical model is received and includes a plurality of entities and connectivity information between the entities. The entities include properties, behavioral descriptions, and optionally behavioral constraints. A symbolic expression is received. The symbolic expression represents a property of a first entity in the graphical model. A second entity is identified. The second entity includes the property represented by the symbolic expression. The second entity is identified based on at least one of the connectivity information, a behavioral description, or a behavioral constraint. The symbolic expression is propagated to the second entity. The second entity is expressed in terms of the propagated symbolic expression. An updated graphical model is generated.