遗传规划算法在化合物设计、筛选研究中的应用

采用计算机科学中新兴的遗传规划算法思想，结合化学物质的本质特点，运用进化操作来实现化合物的合成设计和筛选。文中针对算法运用讨论了函数集、终止集问题，通过计算元素组成的字符串的化合价的结果来确定适应度函数，既符合化学学科的本质规律，又满足了算法的要求。通过复制、交换和突变操作，经过多代次的进化终止，取得了满意的结果。文章还针对其实用性，从化学本质出发，提出了建议和研究方向。可以说本文是遗传规划在化学化合物合成筛选中运用的成功探索，同时也为进一步研究打下了坚实的基础。

Application of genetic programming in the study of design and sieve of the compounds

To adopt the developing ideas of genetic programming, integrate the essential matters character of chemical , and handle the evolutional operation to realizethe synthesis and sieve of compound. Aiming to the algorithm operation to discuss the function set, to terminate the question set. Through calculating the valence of character string made up of by elements to confirm the adapted function, which not only being suited to the essential rules in chemistry, but also to meet the requirement of algorithm. By copying, exchanging and sieving, after themulti-generation evolutional termination, to get a satisfying result. In this thesis, still aiming to the practicability, and begin with the chemical essence to put forward the suggestions and theresearching direction. It is a thesis that adopted the genetic programming to probe in the compound of synthesis and sieve. Meanwhile, it formed the strong base for further study.