Model Adjuster for Time-Domain Simulation Error Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing simulation techniques introduce errors when solving mathematical models, particularly due to approximations in solving methods like the trapezoidal rule, forward Euler, and backward Euler techniques, leading to inaccuracies in system analysis outputs.
Innovation Solution
A model adjuster adjusts the input model to compensate for these errors by generating an adjusted model that cancels out the approximation errors, thereby improving the accuracy of the solution when applied to the solver.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If standard solving techniques (trapezoidal rule, Euler methods) are used to simulate system outputs, then computational efficiency is maintained, but approximation errors are introduced reducing solution accuracy
Solution Approach 1:
The patent applies preliminary action by pre-computing and storing warping compensation values in lookup tables before the actual simulation process. The model adjuster uses these pre-prepared compensation values to correct the mathematical model parameters, eliminating the need for complex real-time calculations during simulation while maintaining high accuracy
Solution Approach 2:
The patent introduces an intermediary component (model adjuster and warping compensator) that sits between the standard solver and the mathematical model. This intermediary automatically applies corrections based on pre-computed warping compensation values, isolating the complexity from the main simulation process while improving accuracy
2Measurement precision
If higher precision solving methods are used to reduce approximation errors, then solution accuracy improves, but computational complexity and processing time increase
Solution Approach 1:
The patent performs the computationally intensive accuracy improvement work in advance by pre-computing warping compensation values and storing them in lookup tables. During actual simulation, the model adjuster simply retrieves and applies these pre-computed values, achieving high precision without the time cost of complex calculations
Solution Approach 2:
The patent segments the accuracy improvement process into two distinct phases: an offline phase where warping compensation values are pre-computed and stored, and an online phase where these values are efficiently applied during simulation. This segmentation moves the computational burden to when it doesn't impact real-time performance
Data Source
AI summary
A device receives an initial model of a system and information that identifies a solving technique to be used to solve a mathematical problem associated with the initial model. The initial model includes an initial transfer function that describes a relationship between an input to and an output from the system. The device determines an error associated with the solving technique, resulting in an inaccurate solution to the mathematical problem. The device generates an adjusted model, based on the initial model and the error, that includes an adjusted transfer function, based on the initial transfer function, or an adjusted input to the system, based on the input to the system. The device applies the solving technique to the adjusted model, to generate a result that includes a more accurate solution to the mathematical problem than applying the solving technique to the initial model, and outputs or stores the result.