(1. 上海大學 計算機工程與科學學院，上海 200444；
2. 上海大學 材料基因組工程研究院，上海 200444)
摘 要: 由于鈣鈦礦型材料具有廣泛的應用前景，因此對其結(jié)構(gòu)及物理、化學性質(zhì)的研究一直是材料研究領(lǐng)域的熱點之一。其中，利用容許因子(Tolerance factor)來預測鈣鈦礦型材料的結(jié)構(gòu)穩(wěn)定性可以幫助研究者發(fā)現(xiàn)更多的新型功能材料，而傳統(tǒng)的基于離子半徑定義的容許因子tIR存在一定的局限性。本文基于SISSO(Sure independence screening and sparsifying operator)方法和鍵價模型提出一種新型的容許因子τBV，其可以有效地避免由離子半徑帶來的局限性。本工作使用機器學習中的決策樹算法建立容許因子驗證模型，實驗結(jié)果表明，新型容許因子τBV可以很好地預測ABO3型化合物是否具有鈣鈦礦結(jié)構(gòu)，并大大提高了預測精度。
關(guān)鍵字: 鈣鈦礦；結(jié)構(gòu)穩(wěn)定性；SISSO；新型容許因子
（1. School of Computer Engineering and Science, Shanghai University, Shanghai 200444, China;
2. Materials Genome Institute, Shanghai University, Shanghai 200444, China）
Abstract:Due to the wide application prospects of perovskite materials, research on their structures and physical and chemical properties of perovskite materials has been one of the hot topics in the field of materials research. Among them, predicting the stability of perovskite structure with the help of tolerance factor can help researchers discover more new functional materials. The conventional tolerance factor tIR for determining the stability of the perovskite structure based on ion radius has certain shortcomings and limitation. In view of this, this work proposes a new type of tolerance factor τBV based on the bond valence model using the SISSO (sure independence screening and sparsifying operator) method which can effectively avoid the defect limitation caused by the ionic radius. This work uses the decision tree algorithm in machine learning to establish the new tolerance factor verification model and the results show that the new tolerance factor τBV can excellently predict whether the ABO3 compound is perovskite or non-perovskite, which greatly improves the prediction accuracy.
Key words: perovskites; structural stability; sure independence screening sparsifying operator (SISSO); new tolerance factor
