An improved immune algorithm with parallel mutation and its application.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Nature-Inspired Computing

Background:

• Immune algorithms (IA) simulate biological immune systems for optimization.
• Standard IA methods face challenges with slow convergence and local optima due to stochastic search.
• Existing improvements include adaptive parameter setting and diversity maintenance.

Purpose of the Study:

• To design a fast and efficient immune algorithm (IIA) for optimization problems.
• To address the limitations of traditional immune algorithms, specifically slow convergence and local minima trapping.
• To apply the proposed algorithm to the Lennard-Jones potential problem (LJPP).

Main Methods:

• Proposed an improved immune algorithm (IIA) incorporating a parallel mutation mechanism (PM).
• Implemented three distinct mutation operators: Cauchy Mutation (CM), Gaussian Mutation (GM), and Lateral Mutation (LM), with conditional selection.
• Evaluated IIA on 2D functions, benchmark functions, and the Lennard-Jones potential problem.

Main Results:

• The IIA demonstrated improved convergence speed compared to traditional methods.
• The algorithm showed effectiveness in balancing exploration and exploitation during the search process.
• Experimental results confirmed the IIA's capability in solving the LJPP with high solution quality.

Conclusions:

• The developed improved immune algorithm (IIA) offers a fast and efficient approach to optimization.
• The parallel mutation mechanism effectively enhances search performance and overcomes common IA drawbacks.
• IIA shows significant potential for real-world applications, including the Lennard-Jones potential problem.