A Modified Slime Mould Algorithm for Global Optimization.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Metaheuristics

Background:

• The Slime Mould Algorithm (SMA) is a population-based metaheuristic inspired by slime mould behavior.
• SMA faces challenges with unbalanced exploration-exploitation and susceptibility to local optima.

Purpose of the Study:

• To propose an improved Slime Mould Algorithm (MSMA) addressing SMA's limitations.
• To enhance population diversity, balance exploration-exploitation, and escape local optima.

Main Methods:

• Implemented a chaotic opposition-based learning strategy for enhanced population diversity.
• Introduced two adaptive parameter control strategies to balance exploitation and exploration.
• Incorporated a spiral search strategy to mitigate local optima entrapment.

Main Results:

• MSMA demonstrated superior performance on 13 multidimensional and 10 fixed-dimensional test functions.
• The algorithm showed effectiveness in solving two real-world engineering optimization problems.
• Simulation results confirmed MSMA's advantages in convergence accuracy, speed, and stability over comparative algorithms.

Conclusions:

• The proposed MSMA effectively overcomes the limitations of the standard SMA.
• MSMA offers a robust and efficient approach for complex optimization tasks.
• The enhanced algorithm shows significant potential for practical engineering applications.