Dynamic performance enhancement of adjustable blade pitch angle for wind generation system applications based on...

Area of Science:

• Electrical Engineering
• Control Systems
• Renewable Energy Systems

Background:

• Growing reliance on renewable energy sources like wind power presents challenges to power system stability.
• Large-scale Wind Energy Conversion Systems (WECS) based on Doubly-Fed Induction Generators (DFIGs) require advanced control strategies.
• Conventional controllers (PID, FPID) face limitations with system nonlinearity and model accuracy.

Purpose of the Study:

• To propose an improved pitch-angle control strategy for a 1.5 MW DFIG-based WECS.
• To enhance the control performance by integrating Neural Network (NN) architectures with conventional controllers.
• To address the limitations of existing control methods in handling system nonlinearity and model uncertainties.

Main Methods:

• Investigated Proportional-Integral-Derivative (PID) and Fractional PID (FPID) control strategies.
• Integrated PID and FPID controllers with Multilayer Feedforward (MLFFNN), Cascade Forward (CFNN), and Elman NN architectures.
• Utilized MATLAB/Simulink for simulation and performance evaluation under various wind speed conditions.

Main Results:

• The MLFFNN architecture demonstrated superior performance compared to PID and FPID controllers.
• Achieved a minimum Mean Square Error (MSE) of 0.0027024.
• Reached a power performance efficiency of 98.9% under step, ramp, and random wind speed variations.

Conclusions:

• The proposed NN-based pitch-angle control strategy offers a robust solution for large-scale wind energy applications.
• NN controllers provide significantly improved stability and efficiency over traditional methods.
• The MLFFNN controller is identified as the most effective for enhancing WECS performance.