Zeta converter: Kharitonov polynomials based interval reduced order modeling.

Area of Science:

• Electrical Engineering
• Control Systems Theory
• Computational Mathematics

Background:

• Zeta converters are crucial in power electronics but often require complex models.
• Model order reduction is essential for efficient analysis and controller design.
• Existing methods may not adequately handle interval uncertainties in converter parameters.

Purpose of the Study:

• To develop a continuous interval reduced-order model (ROM) for a fourth-order Zeta converter.
• To utilize Kharitonov polynomials and interval arithmetic for robust model reduction.
• To validate the accuracy and performance of the proposed ROM against higher-order models.

Main Methods:

• Obtaining the fourth-order continuous interval transfer function using interval arithmetic.
• Reducing the model order to first, second, and third orders via Kharitonov polynomials.
• Deriving Kharitonov polynomials for the denominator and constructing a Routh table.
• Matching time-moments (TiMo) and Markov parameters (MaPa) to determine the ROM numerator.

Main Results:

• Successfully generated first, second, and third-order interval reduced-order models for the Zeta converter.
• Demonstrated the efficacy of the proposed method through comparisons with existing models.
• Validated performance using step/impulse responses and Bode/Nichols plots for interval bounds.
• Tabulated time-domain specifications (TDS) and performance error criteria (PEC) for comparative analysis.

Conclusions:

• The proposed Kharitonov polynomial-based method effectively reduces the order of Zeta converter models.
• The reduced-order models maintain high accuracy and performance comparable to higher-order models.
• This approach offers a computationally efficient way to analyze interval systems in power electronics.