Injection Molding Parameter Optimization With Adaptive Gradient Steps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current model-free optimization methods for injection molding process parameters are inefficient due to frequent tests and lack of adaptability to different parameters, leading to high optimization costs and slow convergence rates.
Innovation Solution
An iterative gradient estimation method combined with an adaptive moment estimation algorithm is used to calculate gradient directions and adjust parameter steps adaptively, reducing the number of tests and improving adaptability, thereby optimizing process parameters rapidly and efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If model-based optimization with surrogate models is used, then optimization accuracy can be improved, but the number of tests required increases and optimization cost increases
Solution Approach 1:
The patent extracts only the essential gradient information needed for optimization direction from the complex surrogate model building process. Instead of building complete surrogate models for all parameters, it extracts gradient directions through finite difference methods, obtaining the necessary optimization guidance with minimal testing.
Solution Approach 2:
The patent performs preliminary gradient estimation at each iteration point before full optimization. By calculating gradient directions in advance through perturbation tests and using these pre-computed gradients to guide the next iteration, it avoids the need for extensive testing that would be required to build accurate surrogate models from scratch.
2Adaptability or versatility
If conventional trial-and-error methods are used, then personal experience can be utilized, but optimization efficiency decreases and time consumption increases
Solution Approach 1:
The patent implements a feedback mechanism where the results of each test are immediately used to calculate gradient directions that guide the next parameter adjustment. This closed-loop feedback system replaces trial-and-error with systematic, data-driven optimization, maintaining adaptability while dramatically improving efficiency.
Solution Approach 2:
The patent substitutes the mechanical trial-and-error process with an automated gradient-based optimization system. Instead of relying on manual adjustment based on experience, it uses mathematical gradient calculations to automatically determine optimization directions and update parameters systematically.
3Speed
If gradient-based methods with extra tests for gradient fitting are used, then convergence rate can be improved, but optimization cost increases
Solution Approach 1:
The patent performs only the minimum necessary gradient estimation through perturbation tests rather than comprehensive gradient fitting. By calculating gradients using finite differences with small perturbations and accepting approximate gradient directions, it achieves sufficient convergence speed without the excessive testing cost of precise gradient fitting methods.
4Ease of manufacture
If uniform adjustment steps for all parameters are used, then implementation simplicity is maintained, but adaptability to different parameters decreases
Solution Approach 1:
The patent applies local quality by allowing each parameter to have its own adaptive step size based on its specific characteristics and the local gradient information. Instead of uniform steps, each parameter's adjustment magnitude is tailored to its needs, with steps automatically adjusted based on the gradient magnitude and previous performance for that specific parameter.
Data Source
AI summary
The present invention discloses a model-free optimization method of process parameters of injection molding to solve the problems of frequent tests required and performing adaptive adjustment on different parameters in the existing optimization method. The method need not build a surrogate model between a product quality index and a process parameter to render the process parameter to converge nearby the optimal solution by an on-line iteration method. The present invention calculates the gradient direction of a current point by an iterative gradient estimation method, and uses adaptive moment estimation algorithm to allocate an adaptive step for each parameter. The method can significantly reduce the cost and time required in the process parameter, which greatly helps improving the optimization efficiency of process parameters of injection molding.