Multi-Resolution Model Parameter Mapping for Faster High-Resolution Tuning

Solution Overview

Problem

Optimizing parameters of large-scale, high-resolution numerical models is a costly and time-consuming process due to the need for multiple executions with different parameter sets, which is inefficient and computationally expensive.

Innovation Solution

A system and method using machine-learning techniques to learn a mapping between low-resolution and high-resolution model parameters through a deep neural network (PMN) and a parameter mapping transform (PMT), allowing parameter tuning at low resolution and conversion to high-resolution equivalents, thereby reducing computational costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If high-resolution model runs are performed with multiple parameter sets to optimize model accuracy, then model parameter optimization quality is improved, but computational cost and time consumption increase significantly

Engineering Contradiction:

Improvemodel parameter optimization qualityVSAvoidtime consumption

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs parameter optimization at low resolution first as a preliminary step, obtaining initial parameter estimates that serve as starting points for subsequent high-resolution optimization. This preliminary action at lower computational cost guides the more expensive high-resolution runs, reducing the number of iterations needed and overall time consumption while maintaining optimization quality.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The optimization process is segmented into multiple resolution levels. The patent divides the parameter optimization task into coarse-resolution stages and fine-resolution stages, where each stage operates at appropriate computational granularity. This segmentation allows efficient exploration of parameter space at low resolution followed by precise refinement at high resolution, balancing optimization quality with time efficiency.

Inventive Principle:

Principle #1Segmentation

2Measurement precision

If high-resolution model runs are performed with multiple parameter sets to optimize model accuracy, then model parameter optimization quality is improved, but computational cost increases significantly

Engineering Contradiction:

Improvemodel parameter optimization qualityVSAvoidcomputational cost

Core Design Contradiction:

Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent performs parameter optimization at low resolution first as a preliminary step, obtaining initial parameter estimates that serve as starting points for subsequent high-resolution optimization. This preliminary action at lower computational cost guides the more expensive high-resolution runs, reducing the number of iterations needed and overall time consumption while maintaining optimization quality.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The optimization process is segmented into multiple resolution levels. The patent divides the parameter optimization task into coarse-resolution stages and fine-resolution stages, where each stage operates at appropriate computational granularity. This segmentation allows efficient exploration of parameter space at low resolution followed by precise refinement at high resolution, balancing optimization quality with time efficiency.

Inventive Principle:

Principle #1Segmentation

3Reliability

If multiple executions of the model with different parameter sets are performed at high resolution, then model accuracy is improved, but the process becomes expensive and time-consuming

Engineering Contradiction:

Improvemodel accuracyVSAvoidtuning process efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent performs parameter optimization at low resolution first as a preliminary step, obtaining initial parameter estimates that serve as starting points for subsequent high-resolution optimization. This preliminary action at lower computational cost guides the more expensive high-resolution runs, reducing the number of iterations needed and overall time consumption while maintaining optimization quality.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The optimization process is segmented into multiple resolution levels. The patent divides the parameter optimization task into coarse-resolution stages and fine-resolution stages, where each stage operates at appropriate computational granularity. This segmentation allows efficient exploration of parameter space at low resolution followed by precise refinement at high resolution, balancing optimization quality with time efficiency.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS20250005360A1System and method for model parameter optimization

Publication Date: 2025.01.02 HEWLETT PACKARD ENTERPRISE DEV LP

AI summary

One embodiment can provide a method and system for tuning parameters of a numerical model of a physical system. During operation, the system can obtain, using a machine-learning technique, a parameter-transform model for mapping parameters of the numerical model at a first resolution to parameters of the numerical model at a second resolution, the second resolution being higher than the first resolution. The system can perform a parameter-tuning operation on the numerical model at a first resolution to obtain a first set of tuned parameters and apply the parameter-transform model on the first set of tuned parameters to obtain a second set of tuned parameters at a second resolution. The system can then generate behavior information associated with the physical system by running the numerical model at the second resolution using the second set of tuned parameters.