Development of intelligent controller for high performance electric drives with hybrid CSA and RERNN technique

Area of Science:

• Electrical Engineering
• Control Systems
• Artificial Intelligence

Background:

• Electric drives are vital in industry, demanding precise control under variable conditions.
• Existing control methods face challenges with complex mechanical configurations and parameter fluctuations.

Purpose of the Study:

• To develop an intelligent controller for precise speed control of electric drives.
• To enhance adaptability and stability in electric drive systems with varying parameters.

Main Methods:

• A hybrid approach combining Circle Search Algorithm (CSA) and Recalling-Enhanced Recurrent Neural Network (RERNN) was developed (CSA-RERNN).
• CSA optimizes voltage source inverter control signals, while RERNN predicts optimal signals.
• The system accounts for variables like inertia, load torque, and rotor shaft angle.

Main Results:

• CSA-RERNN demonstrated superior performance over SSA, ANN, and BCO in MATLAB/Simulink simulations.
• Experimental results showed lower error rates and improved speed control accuracy.
• The system maintained stable response characteristics despite inertia fluctuations (1.2-2.3 kgm²).

Conclusions:

• The CSA-RERNN technique offers a robust, stable, and cost-effective solution for electric drive control.
• It ensures long-term performance stability without disabling adaptation mechanisms.
• This method is suitable for diverse industrial applications requiring intelligent electric drive control.