吉林大学学报(信息科学版) ›› 2020, Vol. 38 ›› Issue (5): 578-587.

• • 上一篇    下一篇

量子教学优化算法及在函数优化中的应用

  

  1. 东北石油大学 计算机与信息技术学院, 黑龙江 大庆 163318
  • 收稿日期:2019-12-17 出版日期:2020-09-24 发布日期:2020-10-22
  • 作者简介:石彤(1995— ), 女, 陕西富平人, 东北石油大学硕士研究生, 主要从事量子优化算法研究, ( Tel) 86-15765986635(E-mail)920436084@ qq. com; 李盼池(1969— ), 男, 河北大城人, 东北石油大学教授, 博士生导师, 主要从事量子智能计算研究, (Tel)86-13936853869(E-mail)lipanchi@ vip. sina. com.
  • 基金资助:
    国家自然科学基金资助项目(61702093)

Quantum Teaching-Learning-Based Optimization Algorithm and Its Application in Function Optimization

  1. School of Computer and Information Technology, Northeast Petroleum University, Daqing 163318, China
  • Received:2019-12-17 Online:2020-09-24 Published:2020-10-22

摘要: 为提升基本教学优化算法的搜索能力, 通过融合量子计算原理, 提出了一种量子教学优化算法。 该方法采用教师自学和学生向教师学两种学习机制搜索全局最优解。 个体采用量子比特编码, 搜索过程在 Bloch 球面上进行, 个体的更新通过量子比特的绕轴旋转实现, 然后将其解码为量子比特的 Bloch 球面坐标。 由于该方法将基本教学算法中每维变量的搜索都扩展到 Bloch 球面进行, 可使搜索过程更为精细, 从而加强了对解空间的遍历性。 不同维度标准函数极值优化的仿真结果表明, 此方法的寻优能力不仅超过基本教学优化算法, 同时也超过其他经典群智能优化算法, 验证了将量子计算的某些机制和智能优化相融合可提升其优化性能。

关键词: 智能优化算法, 量子计算, 教学优化算法, 量子教学优化, 函数优化

Abstract: In order to improve the search ability of traditional teaching-learning optimization algorithm, the corresponding quantum version of the algorithm is proposed by integrating principle of quantum computing. In proposed method, two learning mechanisms, teacher self-study and student learning from teacher, are used to search the global optimal solution. The individual uses qubits coding, and the search process is performed on the Bloch sphere. The individual is updated by the rotation of the qubits about axis, and then is decoded into the Bloch spherical coordinates of the qubits. Because the proposed method extends the search of each dimension in the common teaching-learning algorithm to the Bloch sphere, the search process can be more refined, thus enhancing the traversal effect on the solution space. The experimental results show that the optimization ability of this method is not only better than that of common teaching-learning-based optimization algorithm, but also better than that of other classical swarm intelligence optimization algorithms. The results show that the combination of some mechanisms of quantum computing and intelligent optimization can improve its optimization performance.

Key words: intelligent optimization algorithm, quantum computing, teaching-learning-based optimization algorithm, quantum teaching-learning-based optimization, function optimization

中图分类号: 

  • TP391