Design of an adaptive neural network based power system stabilizer

Area of Science:

• Electrical Engineering
• Control Systems
• Artificial Intelligence

Background:

• Low-frequency power system oscillations pose a significant challenge to grid stability.
• Conventional power system stabilizers (CPSS) have limitations that necessitate advanced solutions.
• Existing techniques for power system stabilization require comprehensive analysis.

Purpose of the Study:

• To design and present an indirect adaptive neural network based power system stabilizer (IDNC).
• To address the drawbacks of conventional power system stabilizers.
• To improve the damping of low-frequency power system oscillations.

Main Methods:

• Developed an IDNC comprising a neuro-controller and a neuro-identifier.
• The neuro-controller generates supplementary control signals for the excitation system.
• The neuro-identifier models power system dynamics and adapts controller parameters.

Main Results:

• The proposed IDNC demonstrated a simple structure, adaptivity, and fast response.
• Evaluated on a single-machine infinite-bus system under various conditions.
• Effectiveness and robustness were confirmed through simulations.

Conclusions:

• The indirect adaptive neural network based power system stabilizer (IDNC) is a viable and effective solution.
• The IDNC offers superior performance compared to conventional methods.
• The proposed stabilizer enhances power system stability and robustness.