Dynamic System Simulation With Solution-Dependent Time Marks

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing simulation methods for dynamic systems, particularly in the energy industry, struggle to efficiently capture highly dynamic events due to the use of pre-set time marks, leading to inefficiencies in data generation and potential missed events or excessive computational resources.

Innovation Solution

Implementing solution-dependent time marks that are generated in response to highly dynamic events, detected by comparing variable changes to predefined criteria, allowing for targeted data output during simulations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If pre-set time marks are used for data output during simulation, then the simulation can proceed continuously without interruption, but highly dynamic events may be missed or excessive data may be generated

Engineering Contradiction:
Improvedetection accuracy of dynamic eventsVSAvoidsimulation efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent applies the Dynamics principle by transitioning from static, pre-set time marks to dynamic, solution-dependent time marks. The output time marks are no longer fixed beforehand but are determined during simulation based on whether highly dynamic events are detected between consecutive marks. This allows the simulation to adapt its output timing to the actual behavior of the system, capturing events when they occur while maintaining computational efficiency.

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The patent implements Feedback by using the simulation results themselves to control the output process. The detection of highly dynamic events between consecutive output time marks feeds back into the decision of whether to generate an output at the current time mark. This creates a closed-loop system where the simulation output is continuously adjusted based on the detected system behavior, ensuring reliable event capture without excessive data generation.

Inventive Principle:
Principle #23Feedback

2Reliability

If solution-dependent time marks are generated to capture highly dynamic events, then detection accuracy improves, but additional computational resources are required for event detection

Engineering Contradiction:
Improvedetection accuracy of dynamic eventsVSAvoidcomputational resource consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent applies Partial or excessive action by performing event detection only partially - specifically, only between consecutive output time marks rather than continuously throughout the entire simulation. This selective detection approach provides sufficient accuracy to capture highly dynamic events when they occur, while avoiding the excessive computational cost of continuous monitoring. The detection is applied exactly where needed (between output points) and nowhere else.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP3353611B1Solution dependent output time marks for models of dynamic systems
Publication Date: 2025.10.29 BAKER HUGHES CO
  • EP3353611B1 patent drawingFigure 1
  • EP3353611B1 patent drawingFigure 2
  • EP3353611B1 patent drawingFigure 3~4

AI summary

A method of simulating a dynamic system includes receiving input data and applying the input data to a numerical forward model, and executing the forward model to simulate the dynamic system and generate solution data values. The method also includes, during execution of the forward model, comparing the solution data values to a pre-selected criterion. The method further includes, in response to at least one solution data value satisfying the pre-selected criterion, automatically generating a time mark for a time at which the at least one solution data value satisfied the pre-selected criterion and automatically outputting one or more solution data values associated with the time mark.