Zhang neural network versus gradient neural network for solving time-varying linear inequalities

Area of Science:

• Artificial Intelligence
• Machine Learning
• Neural Networks

Background:

• Solving time-varying linear inequalities is crucial in various dynamic systems.
• Existing methods may face challenges with real-time adaptation and convergence.

Purpose of the Study:

• To introduce and analyze a new Zhang neural network (ZNN) for online solutions.
• To compare the ZNN model's performance against conventional gradient neural networks.

Main Methods:

• Developing a novel recurrent neural network based on the Zhang design method.
• Conducting theoretical analysis on the convergence properties of the ZNN model.
• Implementing and simulating both ZNN and gradient neural networks for comparative evaluation.

Main Results:

• The proposed ZNN model shows effective online solutions for time-varying linear inequalities.
• Theoretical analysis confirms the convergence properties of the ZNN.
• Computer simulations demonstrate the ZNN's superior efficacy and novelty over gradient methods.

Conclusions:

• The Zhang neural network (ZNN) offers a novel and superior approach for solving time-varying linear inequalities.
• The ZNN model exhibits robust convergence and practical advantages in dynamic system applications.