ESG guidance and artificial intelligence support for power systems analytics in the energy industry.

Area of Science:

• Power Systems Engineering
• Artificial Intelligence
• Environmental Social Governance (ESG)

Background:

• Power system analysis and fault diagnosis require enhanced precision and effectiveness.
• Integrating Environmental Social Governance (ESG) is crucial for assessing power system impacts.
• Existing methods may lack accuracy in load demand forecasting and fault prediction.

Purpose of the Study:

• To assess power systems using AI and ESG for improved analysis and fault diagnosis.
• To develop a CNN-BiLSTM model for accurate power load demand forecasting.
• To implement and optimize a Deep Belief Network (DBN) with Particle Swarm Optimization (PSO) for power grid fault diagnosis.

Main Methods:

• Presentation of an ESG framework to evaluate environmental, social, and governance impacts.
• Development of a Convolutional Neural Network-Bidirectional Long Short-Term Memory (CNN-BiLSTM) model for load demand forecasting.
• Optimization of a Deep Belief Network (DBN) using Particle Swarm Optimization (PSO) for fault diagnosis.

Main Results:

• The CNN-BiLSTM model significantly improved forecasting accuracy with low RMSE (0.054), MAE (0.076), and MAPE (0.102).
• The PSO-optimized DBN achieved 96.22% accuracy in power grid fault diagnosis within 129.94 seconds.
• The proposed models outperformed existing algorithms in both forecasting and fault prediction efficiency.

Conclusions:

• The integrated AI and ESG approach enhances power system analysis and fault diagnosis.
• The study provides a robust framework for sustainable and intelligent growth in the energy industry.
• The developed models offer significant improvements in accuracy and efficiency for power system operations.