本文選題：全局優(yōu)化算法　切入點：理論結(jié)構(gòu)預(yù)測　出處：《物理》2017年09期 　論文類型：期刊論文

【摘要】：凝聚態(tài)物質(zhì)內(nèi)部的原子堆垛方式,即微觀原子結(jié)構(gòu),是深入理解其各種宏觀物理和化學(xué)性質(zhì)的基礎(chǔ)。近年來,隨著基于群智理論的全局優(yōu)化算法和第一性原理計算方法的發(fā)展,只根據(jù)物質(zhì)的化學(xué)組分和外界條件,通過理論計算來確定或預(yù)測物質(zhì)的微觀原子結(jié)構(gòu)成為可能。文章將對目前國內(nèi)外主要理論結(jié)構(gòu)預(yù)測方法進(jìn)行簡要的概述,重點介紹基于群智算法的卡里普索(CALYPSO)結(jié)構(gòu)預(yù)測方法的基本原理及其在凝聚態(tài)物質(zhì)結(jié)構(gòu)研究中的一些典型應(yīng)用。
[Abstract]:In recent years, with the development of the global optimization algorithm based on group intelligence theory and the development of first-principle calculation methods, the atomic stacking method in condensed matter, that is, the microscopic atomic structure, is the basis of deep understanding of various macroscopic physical and chemical properties of condensed matter. It is possible to determine or predict the microscopic atomic structure of matter by theoretical calculation only according to the chemical composition and external conditions of matter. In this paper, the main theoretical structure prediction methods at home and abroad are briefly summarized. This paper mainly introduces the basic principle of the structure prediction method of Carripsos Carriposo (CALYPSO) based on group intelligence algorithm and its typical applications in the study of condensed matter structure.
【作者單位】： 吉林大學(xué)物理學(xué)院超硬材料國家重點實驗室;
【分類號】：O469
，

本文編號：1639320