Low voltage ride through enhancement in DFIG based WECS by hybrid method

ABSTRACT

This paper proposes a hybrid technique for the enhancement of Low Voltage Ride Through (LVRT) capacity in a Doubly Fed Induction Generator (DFIG) based Wind Energy Conversion System (WECS). The proposed hybrid technique is the joint execution of S-shaped binary version of Whale Optimisation Algorithm (S-bWOA) and Random Decision Forest (RDF), hence it called S-bWOA-RDF method. The major objective is to ensure the LVRT capacity in the WECS at voltage drop including failure stages. S-bWOA model is used to identify the optimal solutions for offline mode as an available search location under objective function and generates the training dataset. In light of proficient dataset, RDF performs and predicts better feasible machine side with grid side converter control signals. The proposed technique is activated in MATLAB/Simulink site, and the efficiency of the proposed S-bWORDF method is compared with the existing methods, such as Random Decision Forest and Whale Optimisation Algorithm. The IAE and ITAE of the proposed and existing methods are also analysed. The IAE and ITAE of the proposed hybrid technique are identified as 5.0071 and 54.0475. The comparison results demonstrate the efficiency of the S-bWORDF method and confirm its potential for enhancing the LVRT of DFIG based WECS.

KEYWORDS:

• Random decision forest
• S-shaped binary version
• voltage
• active with reactive power
• current

Abbreviations

LVRT: Low Voltage Ride Through; DFIG: Doubly Fed Induction Generator; WECS: Wind Energy Conversion System; S-bWOA: S-shaped binary version of Whale Optimisation Algorithm; RDF: Random Decision Forest; WT: wind turbine; AHC: Auxiliary Hardware Circuit; DVR: Dynamic voltage restorer; AI: artificial intelligence; WOA: whale optimisation algorithm; OOBE: out-of-bag error; SC: stator current; RC: Rotor current

Disclosure statement

No potential conflict of interest was reported by the author(s).