Single Input Interval Type-3 Fuzzy PID Control of a Magnetic Levitation System


Şahin İ., ULU C.

Fuzzy Information and Engineering, vol.18, no.3, pp.339-351, 2026 (ESCI, Scopus)

  • Publication Type: Article / Article
  • Volume: 18 Issue: 3
  • Publication Date: 2026
  • Doi Number: 10.26599/fie.2026.9270016
  • Journal Name: Fuzzy Information and Engineering
  • Journal Indexes: Emerging Sources Citation Index (ESCI), Scopus
  • Page Numbers: pp.339-351
  • Keywords: interval type-3 (IT3) fuzzy, magnetic levitation, robustness, single input interval type-3 fuzzy PID (SIT3-FPID)
  • Yıldız Technical University Affiliated: Yes

Abstract

Single input fuzzy PID (SI-FPID) controllers combine a fuzzy controller with a PID controller, thereby enhancing performance while maintaining a simple and efficient design. In this structure, the fuzzy controller adds a nonlinear characteristic to the PID controller. Interval type-3 (IT3) fuzzy sets offer greater potential in handling disturbances, noise, and higher-order uncertainties than their type-1 and type-2 counterparts, they have not yet been applied in SI-FPID controllers. Therefore, for the first time in the literature, a single-input interval type-3 fuzzy PID (SIT3-FPID) control structure is proposed in this study, in which an interval type-3 fuzzy controller (IT3-FC) is integrated with a conventional PID controller. Three triangular IT3 fuzzy sets and three singletons are used for the input and output variables, resulting in only three rules. The SIT3-FPID controller introduces additional design parameters, such as apex points, alpha-cuts and fuzziness parameters, which enable performance improvements over type-2 controllers. To evaluate the effectiveness of the proposed SIT3-FPID controller, it is applied to a nonlinear magnetic levitation system and compared with SIT2-FPID and conventional PID controllers under a nominal condition, an external disturbance, and parameter uncertainties. The results demonstrate that the proposed SIT3-FPID controller exhibits robustness performance in all scenarios.