Optimized CNN-BiLSTM framework for reactive power management and voltage profile improvement in renewable energy...

Area of Science:

• Electrical Engineering
• Artificial Intelligence
• Power Systems

Background:

• Unpredictable renewable energy sources (RES) like wind and solar cause voltage fluctuations in smart grids.
• Maintaining stable voltage profiles is crucial for grid reliability and efficient power delivery.

Purpose of the Study:

• To improve power grid voltage profiles by effectively regulating reactive power.
• To enhance the performance of Distribution Static Synchronous Compensators (DSTATCOM) using advanced AI techniques.

Main Methods:

• Integration of hybrid renewable energy systems (HRES) with DSTATCOM for reactive power control.
• Development of a hybrid Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) model for dynamic reactive power management.
• Optimization of the CNN-BiLSTM network using an Adaptive Parrot Optimizer (APO).

Main Results:

• Achieved up to 33.4% reduction in power loss.
• Improved voltage stability index (VSI) to 1.02 p.u.
• Minimized total harmonic distortion (THD) below 1.7% and reduced settling time to 0.075 s.
• Attained high prediction accuracy (R²=0.9672, RMSE=3.0094).

Conclusions:

• The proposed AI-driven method effectively regulates reactive power, enhancing smart grid voltage stability and performance.
• Integration of HRES with optimized DSTATCOM control significantly improves grid reliability and reduces energy losses.
• The CNN-BiLSTM-APO model demonstrates robust capabilities for real-time voltage regulation in smart grids.