A Survey on Biomimetic and Intelligent Algorithms with Applications

Area of Science:

• Computational intelligence
• Optimization algorithms
• Neural networks

Background:

• Intelligence algorithms are inspired by natural phenomena to solve complex optimization problems.
• Zeroing Neural Networks (ZNNs) are a specialized class of neural networks designed for dynamic optimization tasks.
• Bio-inspired algorithms, such as genetic algorithms and particle swarm optimization, offer alternative approaches to problem-solving.

Purpose of the Study:

• To provide a comprehensive guide for researchers interested in applying intelligence algorithms to optimization problems.
• To detail the principles, variants, and applications of Zeroing Neural Networks (ZNNs).
• To outline classic bio-inspired algorithms and their practical uses.

Main Methods:

• Comprehensive discussion of Zeroing Neural Networks (ZNNs), including their origin, principles, and mechanisms.
• Introduction of a novel classification method for ZNNs based on performance indices.
• Overview of Genetic Algorithms (GAs) and Particle Swarm Optimization (PSO), covering their design and applications.

Main Results:

• ZNNs are presented as effective tools for solving time-varying optimization problems.
• A new classification scheme enhances the understanding and selection of appropriate ZNN models.
• Demonstration of the applicability of intelligence algorithms in diverse fields.

Conclusions:

• Intelligence algorithms, including ZNNs and bio-inspired methods, offer powerful solutions for various optimization challenges.
• The presented classification method aids in selecting suitable ZNNs for specific problems.
• The study highlights the broad applicability of these algorithms in scientific and engineering domains.