A split-optimization approach for obtaining multiple solutions in single-objective process parameter optimization

Area of Science:

• Manufacturing Engineering
• Process Optimization
• Computational Intelligence

Background:

• Current process parameter optimization techniques often yield a single solution, which may not be suitable for real-world production or may result in undesirable conditions.
• The challenge lies in identifying diverse, applicable parameter sets that achieve desired performance indicators across different production scenarios.

Purpose of the Study:

• To propose a novel split-optimization approach for single-objective process parameter optimization problems.
• To enable the discovery of multiple optimal process parameter combinations, enhancing applicability in actual production.
• To compare the efficiency of different search space splitting strategies for optimization.

Main Methods:

• A split-optimization approach was developed by dividing the original search space into smaller sub-search spaces.
• Genetic Algorithm (GA) was employed within each sub-search space for parameter optimization.
• Two specific splitting strategies, cluster centers and hill and valley splitting, were evaluated against a baseline method of equal sub-space division.

Main Results:

• The proposed split-optimization approach successfully generated multiple optimal process parameter combinations for electrochemical micro-machining.
• The cluster centers and hill and valley splitting strategies demonstrated superior efficiency in partitioning the search space compared to equal division.
• These advanced splitting methods facilitated the identification of a broader range of viable process parameters.

Conclusions:

• The split-optimization approach offers a significant advancement over traditional methods by providing multiple, potentially more practical, optimal parameter sets.
• The effectiveness of search space partitioning strategies, particularly cluster centers and hill and valley splitting, is crucial for successful multi-solution optimization.
• This methodology enhances the flexibility and applicability of process parameter optimization in industrial settings.