Design and development of PI controller for DFIG grid integration using neural tuning method ensembled with dense...

Area of Science:

• Electrical Engineering
• Control Systems
• Artificial Intelligence

Background:

• Doubly-fed induction generators (DFIGs) are crucial for renewable energy integration, but their grid interconnection presents control challenges.
• Effective control of real and reactive power is essential for stable DFIG operation and grid voltage regulation.
• Conventional PI controllers often require fine-tuning for optimal performance in dynamic grid conditions.

Purpose of the Study:

• To propose and evaluate a Neural Tuning Machine (NTM) for enhanced real and reactive power control in DFIG grid-interconnected systems.
• To leverage recurrent neural networks and dense connections for adaptive and robust control parameter tuning.
• To demonstrate the superiority of the NTM approach over traditional PI controllers for DFIGs.

Main Methods:

• A Neural Tuning Machine (NTM) based on a recurrent neural network was developed to fine-tune PI controller parameters.
• Dense plexus terminals (dense connections) were integrated into the neural network for improved adaptability and nonlinear dynamics.
• Control strategies focused on reactive power regulation via the rotor-side converter and grid voltage control based on bus voltage profiles.
• Simulations were performed using MATLAB Simulink, validated against three benchmarking models adhering to IEEE and IEC standards.

Main Results:

• The NTM method demonstrated enhanced adaptability and robustness in controlling DFIG systems.
• The NTM effectively optimized PI controller parameters, leading to improved real and reactive power flow regulation.
• Comparative analysis showed advantages of the NTM over conventional PI controllers in managing inner current loop parameters.
• The control algorithm proved robust across different simulation scenarios and benchmarking models.

Conclusions:

• The proposed NTM offers an advanced control methodology for DFIG grid integration, enhancing system stability and performance.
• NTM provides a robust and adaptive solution for real and reactive power control, addressing grid integration challenges.
• This research paves the way for further applications of neural tuning methods in power system control and optimization.