Intelligent demand-side energy management via optimized ANFIS-gene expression programming in hybrid renewable-grid...

Area of Science:

• * Energy Systems Engineering
• * Artificial Intelligence in Sustainable Energy
• * Industrial Power Management

Background:

• * Sustainable industrial operations require efficient energy management, especially with restricted electrical grids.
• * Accurate forecasting and scheduling are crucial for optimizing hybrid renewable energy systems.
• * Existing models may not fully address the complexities of predictive energy management.

Purpose of the Study:

• * To propose a hybrid Gene Expression Programming Adaptive Neuro-Fuzzy Inference System (GEP-ANFIS) for predictive energy management.
• * To evaluate the GEP-ANFIS model's forecasting accuracy for solar PV and industrial loads.
• * To assess the economic benefits and robustness of the proposed GEP-ANFIS controller.

Main Methods:

• * Development and implementation of a hybrid GEP-ANFIS model.
• * Forecasting of solar photovoltaic (PV) power and industrial energy loads.
• * Economic evaluation across different energy system configurations (Grid-only, PV-Battery, Grid-connected PV-Battery).
• * Sensitivity analysis on solar PV power and battery storage capacity.

Main Results:

• * GEP-ANFIS achieved low error rates for solar PV prediction (MAPE < 6% short-term, < 8% long-term) and industrial load forecasting (MAPE < 2.5% short-term, < 3.5% long-term).
• * Significant daily energy cost reductions were observed: 7.4% (Grid-only), 6.5% (PV-Battery), and 6.3% (Grid-connected PV-Battery) compared to ANFIS.
• * Over 20 years, GEP-ANFIS showed a 6.5% reduction vs. ANFIS and a 37.7% improvement vs. HOMER.
• * The model demonstrated robustness, efficiency, and scalability, particularly in PV-dominated microgrids.

Conclusions:

• * The GEP-ANFIS model offers a robust and efficient solution for predictive energy management in hybrid renewable energy systems.
• * It significantly reduces operational costs and improves system performance compared to conventional methods.
• * The controller enhances microgrid sustainability and preserves battery longevity by preventing deep discharge.