Sangmyung University Undergraduates Improve Solar Power Prediction with AI... Published in International SCIE Journal

System Semiconductor Engineering Research Team Presents Low-Cost, High-Accuracy Renewable Energy Forecasting Technology
Through Comparison of LSTM and ESN Models

Undergraduate students from the Department of System Semiconductor Engineering at Sangmyung University have gained international recognition for their research using artificial intelligence (AI) to precisely predict solar power generation.

A research paper authored by the undergraduate team was published in the prestigious SCIE-indexed international journal Sustainability, opening up new possibilities for next-generation forecasting technologies in the field of renewable energy.

Sangmyung University announced on October 14 that the paper titled "Performance Comparison of LSTM and ESN Models in Time-Series Prediction of Solar Power Generation," co-authored by fourth-year students Joo Ye-han, Kim Do-gyun, Noh Young-min, and Choi Jae-won from the Department of System Semiconductor Engineering, was recently published in the international journal Sustainability.

This research conducted a performance comparison to more accurately predict solar power generation, a core element of renewable energy, using AI-based time-series analysis techniques.

By optimizing the structure of the Echo State Network (ESN) model compared to the existing Long Short-Term Memory (LSTM) model, the study found that the ESN outperformed in both prediction accuracy and computational efficiency.

The research team systematically designed the key variables that determine the performance of the ESN model-spectral radius, input noise, and leakage rate-achieving high performance without complex structures.

As a result, the team was highly praised for presenting a practical model that reduces computational costs while improving prediction accuracy compared to conventional deep learning-based models.

Joo Ye-han, the first author, stated, "It is a great honor to have our paper published in an SCIE-indexed international journal as undergraduates. This achievement was made possible thanks to the meticulous guidance of our supervising professor and the cooperation of our team members. This meaningful research allowed us to directly confirm the potential of AI technology to enhance the efficiency of renewable energy."

Professor Lee Jonghwan of the Department of System Semiconductor Engineering at Sangmyung University commented, "It is very rare for student research to be published in an international journal. This is a result of passion and perseverance. I hope this experience will serve as a foundation for the students' academic confidence and future research capabilities."

This content was produced with the assistance of AI translation services.