摘要
多尺度量子谐振子算法(MQHOA)的能级稳定过程是算法的核心部分,对于避免算法陷入局部最优、提高算法求解精度具有重要作用.对算法能级稳定过程进行研究,发现不同的能级稳定判据,会造成算法在同一能级下不同的表现.相对宽松的判据使算法在能级稳定过程中迭代不充分,容易陷入早熟.而更严格的判据能使波函数在同一能级下达到稳定状态,提高算法的全局搜索能力,但会增加算法的计算代价.通过实验证明,相对宽松的能级稳定判据对单峰简单函数具有良好的求解效果,严格的能级稳定判据适用于算法对多峰复杂函数的求解.算法在资源优化、自适应控制及能耗优化管理等方面已取得有效应用.
The energy level stabilization process of the multi-scale quantum harmonic oscillator algorithm (MQHOA) is the core part of the algorithm,which plays an important role in avoiding the algorithm falling into local optimum and improving the accuracy of the algorithm.In the studying of the energy level stabilization process of the algorithm,it is found that different energy level stabilization criteria will result in different performance of the algorithm at the same energy level.The relatively loose criterion makes the iteration of the algorithm inadequate in the process of energy level stabilization and easy to fall into premature.The more stringent criterion can make the wave function reach a stable state at the same energy level,improve the global search ability of the algorithm,but meanwhile the computing cost will also rise.Experiments show that loose energy level stability criterion of the algorithm has good effect on solving unimodal simple functions,and strict energy level stability criterion of it is suitable for solving multimodal complex functions.The algorithm has been effectively applied in resource optimization,adaptive control and energy consumption optimization management.
作者
周岩
王鹏
辛罡
李波
王德志
ZHOU Yan;WANG Peng;XIN Gang;LI Bo;WANG De-zhi(School of Computer Science and Technology,Southwest Minzu University,Chengdu,Sichuan 610225,China;Chengdu Institute of Computer Application, Chinese Academy of Sciences, Chengdu,Sichuan 610041, China;University of Chinese Academy of Sciences, Beijing 100049, China)
出处
《电子学报》
EI
CAS
CSCD
北大核心
2019年第6期1337-1343,共7页
Acta Electronica Sinica
基金
国家自然科学基金(No.60702075)
国家自然科学基金面上项目(No.71673032)
四川省教育厅2018一般项目(No.18ZB0623)
西南民族大学中央高校基本科研业务费专项资金资助(No.2019NYB22)
关键词
优化算法
多尺度量子谐振子算法
能级稳定
量子计算
量子谐振子
波函数
量子退火
optimization algorithm
multi-scale quantum harmonic oscillator algorithm
energy level stability
quantum computation
quantum harmonic oscillator
wave function
quantum annealing
