DE-based tuning of PI(λ)D(μ) controllers

Area of Science:

• * Control Engineering
• * Computational Intelligence
• * Fractional Calculus

Background:

• * Fractional order controllers offer enhanced flexibility by increasing tunable parameters.
• * Traditional controller tuning methods may not fully exploit the potential of fractional order systems.
• * Optimizing fractional order controllers requires advanced computational techniques.

Purpose of the Study:

• * To propose a new method for tuning fractional order PI(λ)D(μ) controllers using evolutionary computation.
• * To leverage the increased design flexibility of fractional order controllers to meet multiple control specifications.
• * To demonstrate the effectiveness of the proposed method in both simulated and real-world applications.

Main Methods:

• * A Differential Evolution (DE) algorithm is employed for parameter optimization.
• * The optimization process minimizes a fitness function to satisfy design specifications.
• * Fractional orders (λ and μ) for integral and derivative parts are tuned.
• * The method is validated on a DC motor platform and in simulations.

Main Results:

• * The proposed DE algorithm successfully tunes fractional order PI(λ)D(μ) controller parameters.
• * The method achieves performance improvements in both time and frequency domains.
• * Experimental results on a DC motor platform confirm the effectiveness of the approach.
• * The enhanced control capabilities of fractional order controllers are demonstrated.

Conclusions:

• * Evolutionary computation, specifically DE, provides an effective approach for tuning fractional order controllers.
• * The proposed method enhances system performance by meeting diverse control specifications.
• * The study validates the practical applicability of fractional order control tuning via DE.