@Article{ee.2025.070668,
AUTHOR = {Mir Majid Etghani, Homayoun Boodaghi},
TITLE = {Artificial Neural Networks and Taguchi Methods for Energy Systems Optimization: A Comprehensive Review},
JOURNAL = {Energy Engineering},
VOLUME = {122},
YEAR = {2025},
NUMBER = {11},
PAGES = {4385--4474},
URL = {http://www.techscience.com/energy/v122n11/64228},
ISSN = {1546-0118},
ABSTRACT = {Energy system optimization has become crucial for enhancing efficiency and environmental sustainability. This comprehensive review examines the synergistic application of Artificial Neural Networks (ANN) and Taguchi methods in optimizing diverse energy systems. While previous reviews have focused on these methods separately, this paper presents the first integrated analysis of both approaches across multiple energy applications. We systematically analyze their implementation in: Internal combustion engines, Thermal energy storage systems, Solar energy systems, Wind and tidal turbines, Heat exchangers, and hybrid energy systems. Our findings reveal that ANN models consistently achieve prediction accuracies exceeding 90% when compared to experimental data, while Taguchi-based methods combined with Grey Relational Analysis (GRA) or TOPSIS can improve system performance by up to 20%–30% in multi-objective optimization scenarios. The review introduces novel frameworks for combining these methods and provides critical insights into their complementary strengths. Key statistical metrics, including determination coefficients and error analyses, validate the superior performance of integrated approaches. This work serves as a foundational reference for researchers and practitioners in energy system optimization, offering structured methodologies and future research directions.},
DOI = {10.32604/ee.2025.070668}
}