2025-10-28
興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊《Journal of Materials Chemistry A》
字體大小
小
中
大
興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊《Journal of Materials Chemistry A》
稿源：中興大學工學院
在全世界努力減少碳排放、發展乾淨能源的潮流中，氫能被認為是很有前途的綠色能源。「電催化水分解」是製造氫氣的主要方法之一，但其中關鍵反應的能量屏障較高，長期成為提升效率的最大挑戰。
為了突破這個瓶頸，國立中興大學研究團隊利用人工智慧機器學習演算法，結合電催化劑的合成技術，成功開發出新的「製程優化模型」，大幅提升水分解產氫的效率。這項研究成果已登上國際頂尖期刊《材料化學雜誌A》（Journal of Materials Chemistry A），並且被選為當期的封面內頁文章。
這項研究為國科會前瞻處人工智慧（AI）專案「主題研究群計畫」的執行成果，由中興大學工學院院長楊明德教授、副院長陳志銘教授、李宏中教授與張厚謙教授等跨領域學者共同參與，並由博士後研究員錢德魯（Chandrasekaran Pitchai）與碩士生羅定榆協助完成。
研究團隊開發出一種新材料：釩摻雜鎳鈷層狀雙氫氧化物（NiCoV LDHs），並利用AI模型分析材料的組成、電解液濃度與合成溫度等條件，預測最佳的合成條件與反應環境。結果顯示，在模型推算出的最佳條件下，這種材料的效率較原本最佳的實驗結果提升 21.4%，而且模型預測結果與實驗驗證的誤差僅 6.1%，表現出模型的高精準性。進一步的材料性質分析也發現，此種材料導電性更好，而且可以長時間穩定反應。
值得一提的是，此AI技術僅需不到五十組的實驗數據，即可預測接近一百萬組的實驗結果，相較於傳統的嘗試錯誤法，節省高達99%以上的時間與材料成本。
陳志銘教授表示，這項技術不只加快新材料的研發，也能讓未來的科學家在設計能源材料時少做許多重複實驗。這代表 AI 不只是能寫文章、下棋或聊天，更能幫助人類在能源科技上找到新突破，推動我們更快邁向綠色能源的未來。
新聞報導彙整
1.工商時報：能源材料新突破！中興大學AI模型優化製程 登上國際期刊
2.大成報：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊《Journal of Materials Chemistry A》
3.警政時報：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊
4.觀傳媒：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊
5.獨家報導：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊
6.yahoo新聞：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊
7.中央社：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊
圖說：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊《Journal of Materials Chemistry A》
圖說：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊《Journal of Materials Chemistry A》
照片圖說：興大工學院楊明德院長（左3）、陳志銘副院長（右3）帶領團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊《Journal of Materials Chemistry A》
轉貼興新聞：https://www2.nchu.edu.tw/news-detail/id/60570
---------------------------------------------------------------------------------------------------------------------
NCHU Team Pioneers AI-Driven Breakthrough in Hydrogen Energy — Featured in Journal of Materials Chemistry A
Source: College of Engineering, National Chung Hsing University
As the global community increasingly focuses on reducing carbon emissions and developing sustainable energy alternatives, hydrogen has become one of the most promising candidates in the transition toward clean energy. Among the various hydrogen production methods, electrocatalytic water splitting stands out for its potential. Yet, the crucial electrochemical steps involved specifically the hydrogen and oxygen evolution reactions, continue to encounter significant energy barriers, which have long hindered improvements in efficiency.
To address this challenge, a research team from National Chung Hsing University (NCHU) has successfully integrated artificial intelligence (AI) and machine learning algorithms with advanced electrocatalyst synthesis techniques. This synergy led to the development of a new AI-guided process optimization model that dramatically enhances hydrogen generation efficiency via water splitting. Their pioneering findings were recently published in the internationally renowned Journal of Materials Chemistry A (Royal Society of Chemistry) and featured as an inside front cover article in the issue.
This groundbreaking study is the part of National Science and Technology Council (NSTC) Frontier AI Research Program, under the “Thematic Research Group Project.” The interdisciplinary team comprises Professor Ming-Der Yang, Dean of the College of Engineering; Professor Chih-Ming Chen, Vice Dean; Professor Hung-Chung Li; and Professor Hou-Chien Chang, with contributions from Dr. Chandrasekaran Pitchai (Postdoctoral Fellow) and Ting-Yu Lo (Master’s student).
This team designed a vanadium-doped nickel–cobalt layered double hydroxide (NiCoV LDH) catalyst and employed an AI model to systematically analyze parameters such as material composition, electrolyte concentration, and synthesis temperature. The model accurately predicted the optimal synthesis and reaction conditions, resulting in a 21.4% increase in hydrogen production efficiency compared to the best previous experimental results with only a 6.1% deviation between AI prediction and experimental verification.
Further electrochemical and structural analyses revealed that the newly developed material exhibited enhanced electrical conductivity and excellent long-term operational stability, making it a promising candidate for scalable hydrogen production technologies.
Remarkably, the AI model required fewer than 50 experimental datasets to explore nearly one million possible outcomes, cutting down both time and material costs by more than 99% compared to conventional trial-and-error approaches.
Professor Chih-Ming Chen highlighted the broader impact of this achievement:
“This research demonstrates that AI can do far more than write essays, play chess, or chat - it can drive breakthroughs in energy science and help humanity transition faster toward a sustainable green energy future.”
Repost Link:https://www2.nchu.edu.tw/en-news-detail/id/1168
DOI:10.1039/D5TA03069B
稿源：中興大學工學院
在全世界努力減少碳排放、發展乾淨能源的潮流中，氫能被認為是很有前途的綠色能源。「電催化水分解」是製造氫氣的主要方法之一，但其中關鍵反應的能量屏障較高，長期成為提升效率的最大挑戰。
為了突破這個瓶頸，國立中興大學研究團隊利用人工智慧機器學習演算法，結合電催化劑的合成技術，成功開發出新的「製程優化模型」，大幅提升水分解產氫的效率。這項研究成果已登上國際頂尖期刊《材料化學雜誌A》（Journal of Materials Chemistry A），並且被選為當期的封面內頁文章。
這項研究為國科會前瞻處人工智慧（AI）專案「主題研究群計畫」的執行成果，由中興大學工學院院長楊明德教授、副院長陳志銘教授、李宏中教授與張厚謙教授等跨領域學者共同參與，並由博士後研究員錢德魯（Chandrasekaran Pitchai）與碩士生羅定榆協助完成。
研究團隊開發出一種新材料：釩摻雜鎳鈷層狀雙氫氧化物（NiCoV LDHs），並利用AI模型分析材料的組成、電解液濃度與合成溫度等條件，預測最佳的合成條件與反應環境。結果顯示，在模型推算出的最佳條件下，這種材料的效率較原本最佳的實驗結果提升 21.4%，而且模型預測結果與實驗驗證的誤差僅 6.1%，表現出模型的高精準性。進一步的材料性質分析也發現，此種材料導電性更好，而且可以長時間穩定反應。
值得一提的是，此AI技術僅需不到五十組的實驗數據，即可預測接近一百萬組的實驗結果，相較於傳統的嘗試錯誤法，節省高達99%以上的時間與材料成本。
陳志銘教授表示，這項技術不只加快新材料的研發，也能讓未來的科學家在設計能源材料時少做許多重複實驗。這代表 AI 不只是能寫文章、下棋或聊天，更能幫助人類在能源科技上找到新突破，推動我們更快邁向綠色能源的未來。
新聞報導彙整
1.工商時報：能源材料新突破！中興大學AI模型優化製程 登上國際期刊
2.大成報：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊《Journal of Materials Chemistry A》
3.警政時報：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊
4.觀傳媒：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊
5.獨家報導：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊
6.yahoo新聞：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊
7.中央社：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊
圖說：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊《Journal of Materials Chemistry A》
圖說：興大團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊《Journal of Materials Chemistry A》
照片圖說：興大工學院楊明德院長（左3）、陳志銘副院長（右3）帶領團隊以人工智慧推動氫能源研究 成果登上國際頂尖期刊《Journal of Materials Chemistry A》
轉貼興新聞：https://www2.nchu.edu.tw/news-detail/id/60570
---------------------------------------------------------------------------------------------------------------------
NCHU Team Pioneers AI-Driven Breakthrough in Hydrogen Energy — Featured in Journal of Materials Chemistry A
Source: College of Engineering, National Chung Hsing University
As the global community increasingly focuses on reducing carbon emissions and developing sustainable energy alternatives, hydrogen has become one of the most promising candidates in the transition toward clean energy. Among the various hydrogen production methods, electrocatalytic water splitting stands out for its potential. Yet, the crucial electrochemical steps involved specifically the hydrogen and oxygen evolution reactions, continue to encounter significant energy barriers, which have long hindered improvements in efficiency.
To address this challenge, a research team from National Chung Hsing University (NCHU) has successfully integrated artificial intelligence (AI) and machine learning algorithms with advanced electrocatalyst synthesis techniques. This synergy led to the development of a new AI-guided process optimization model that dramatically enhances hydrogen generation efficiency via water splitting. Their pioneering findings were recently published in the internationally renowned Journal of Materials Chemistry A (Royal Society of Chemistry) and featured as an inside front cover article in the issue.
This groundbreaking study is the part of National Science and Technology Council (NSTC) Frontier AI Research Program, under the “Thematic Research Group Project.” The interdisciplinary team comprises Professor Ming-Der Yang, Dean of the College of Engineering; Professor Chih-Ming Chen, Vice Dean; Professor Hung-Chung Li; and Professor Hou-Chien Chang, with contributions from Dr. Chandrasekaran Pitchai (Postdoctoral Fellow) and Ting-Yu Lo (Master’s student).
This team designed a vanadium-doped nickel–cobalt layered double hydroxide (NiCoV LDH) catalyst and employed an AI model to systematically analyze parameters such as material composition, electrolyte concentration, and synthesis temperature. The model accurately predicted the optimal synthesis and reaction conditions, resulting in a 21.4% increase in hydrogen production efficiency compared to the best previous experimental results with only a 6.1% deviation between AI prediction and experimental verification.
Further electrochemical and structural analyses revealed that the newly developed material exhibited enhanced electrical conductivity and excellent long-term operational stability, making it a promising candidate for scalable hydrogen production technologies.
Remarkably, the AI model required fewer than 50 experimental datasets to explore nearly one million possible outcomes, cutting down both time and material costs by more than 99% compared to conventional trial-and-error approaches.
Professor Chih-Ming Chen highlighted the broader impact of this achievement:
“This research demonstrates that AI can do far more than write essays, play chess, or chat - it can drive breakthroughs in energy science and help humanity transition faster toward a sustainable green energy future.”
Repost Link:https://www2.nchu.edu.tw/en-news-detail/id/1168
DOI:10.1039/D5TA03069B
