Intelligent Constrained Zero-Sum Game Design With Advanced Critic Learning Including a Two-Area Power System...

Area of Science:

• Control Theory
• Game Theory
• Machine Learning

Background:

• Continuous-time multiplayer zero-sum games (MZSGs) with asymmetric constraints present significant control challenges.
• Existing methods often impose restrictive conditions on control matrices.

Purpose of the Study:

• To develop an advanced critic learning technique for MZSGs with asymmetric constraints.
• To introduce a novel algorithm for asymmetric constraints and a new weight tuning rule.

Main Methods:

• A novel asymmetric constraint algorithm is presented, relaxing prior restrictions.
• The Hamilton-Jacobi-Isaacs equation is derived for asymmetric constrained MZSGs (ACMZSGs).
• An advanced critic learning scheme approximates optimal controls and worst disturbances.

Main Results:

• A new weight tuning rule reduces the need for initial admissible controls.
• Stability analysis confirms the control system's robustness.
• Simulations on a two-area power system demonstrate feasibility and efficacy.

Conclusions:

• The proposed critic learning technique effectively addresses ACMZSGs.
• The novel asymmetric constraint algorithm and weight tuning rule enhance control performance and applicability.