Enhanced Philoponella Prominens Optimization (EESPPO) Algorithm Integrated with Experience Exchange Strategy for...

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Bio-inspired Computing

Background:

• Traditional meta-heuristic algorithms struggle with high-dimensional, nonlinear, and constrained engineering design problems.
• Common issues include loss of population diversity and premature convergence, limiting their effectiveness.

Purpose of the Study:

• To propose a novel bio-inspired optimization algorithm, Experience Exchange Strategy-Enhanced Philoponella Prominens Optimization (EESPPO).
• To enhance the performance of meta-heuristic algorithms in solving complex engineering design problems by addressing premature convergence and diversity loss.

Main Methods:

• Inspired by the social behaviors of *P. prominens* (jumping spiders), EESPPO utilizes an Experience Exchange Strategy framework.
• The algorithm incorporates three progressive evolutionary stages: Experience Scarcity (ESC), Experience Crossover (ECR), and Experience Sharing (ESH).
• These stages dynamically balance global exploration and local exploitation through experience library maintenance, enhanced diversity mechanisms, and adaptive information sharing.

Main Results:

• EESPPO demonstrated superior performance compared to 12 advanced algorithms on the CEC2017 benchmark functions.
• The algorithm achieved higher convergence accuracy and robustness.
• EESPPO proved effective and precise when applied to four challenging constrained engineering design problems.

Conclusions:

• EESPPO successfully overcomes the limitations of traditional meta-heuristics, particularly premature convergence.
• The proposed algorithm offers a robust and high-precision solution for complex, constrained engineering optimization tasks.
• The bio-inspired approach effectively balances exploration and exploitation for improved optimization outcomes.