Parametric Optimizer for Dynamic Mechanical Assemblies

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional optimization algorithms for mechanical assembly design oversimplify dynamic behavior, leading to inaccurate evaluation of design criteria and difficulty in identifying convergence, and often fail to compute gradients due to computational complexity, limiting the exploration of design space.

Innovation Solution

A computer-implemented method that generates design options by discretizing continuous equations of motion, computing gradients, and updating design variables to minimize objective functions, using a parametric optimizer that simulates time-varying dynamics and performs sensitivity analysis to explore the design space more effectively.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If conventional optimization algorithms oversimplify the definition of design (e.g., converting flexible structures to rigid bodies or disregarding time-dependent behavior), then computational complexity is reduced, but the accuracy of approximating dynamic behavior and determining design criteria compliance deteriorates

Engineering Contradiction:

Improvecomputational complexityVSAvoidaccuracy of dynamic behavior approximation

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent changes the parameter representation from continuous to discrete by formulating design variables, state variables, and constraints in discrete form. This allows the optimization algorithm to work with simplified discrete parameters while still capturing the essential dynamic behavior through the discrete equations of motion, resolving the contradiction between computational simplicity and dynamic accuracy

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent replaces the continuous mechanical system description with a discrete mathematical model. By substituting continuous equations of motion with discrete equivalents and using discrete adjoint variables, the system achieves computational tractability while maintaining the ability to represent complex dynamic behavior accurately

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Loss of information

If conventional optimization algorithms attempt to compute gradients for complex mechanical assembly designs with numerous interconnected parts, then informed exploration of design space is improved, but computational intractability increases

Engineering Contradiction:

Improvegradient information for design optimizationVSAvoidcomputational intractability

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The patent segments the gradient computation problem into manageable discrete components. By discretizing the adjoint variables and computing gradients step-by-step through discrete time steps, the complex gradient calculation for numerous interconnected parts is broken down into sequential, computationally tractable operations

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces discrete adjoint variables as intermediary quantities that facilitate gradient computation. These adjoint variables act as mediators that connect the design variables to the objective function through the discrete equations of motion, enabling efficient gradient calculation without directly differentiating complex continuous equations

Inventive Principle:

Principle #24Intermediary (Mediator)

3Productivity

If conventional optimization algorithms use iterative processes with continuous equations, then design optimization is performed, but convergence identification becomes difficult due to computational complexity

Engineering Contradiction:

Improvedesign optimization capabilityVSAvoidconvergence identification

Core Design Contradiction:

ProductivityVSDifficulty of detecting and measuring

Solution Approach 1:

The patent replaces the continuous iterative optimization process with a discrete formulation. By expressing the optimization problem in discrete form with discrete design variables and discrete equations of motion, the convergence criteria can be clearly defined and detected at discrete steps, making convergence identification straightforward while maintaining optimization capability

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11620418B2Efficient sensitivity analysis for generative parametric design of dynamic mechanical assemblies

Publication Date: 2023.04.04 AUTODESK INC

AI summary

A design engine generates a configuration option that includes a specific arrangement of interconnected mechanical elements adhering to one or more design constraints. Each element within a given configuration option is defined by a set of design variables. The design engine implements a parametric optimizer to optimize the set of design variables associated with each configuration option. For a given configuration option, the parametric optimizer discretizes continuous equations governing the physical dynamics of the configuration. The parametric optimizer then determines the gradient of an objective function based on the discretized equations the gradient of objective and constraint functions based on discrete direct differentiation method or discrete adjoint variable method derived directly from the discretized motion equations. Then, the parametric optimizer traverses a design space where the configuration option resides to reduce improve the objective function, thereby optimizing the design variables.