论文部分内容阅读
针对不相关并行机调度问题,提出一种基于信息熵的自适应分布估计算法.根据问题特性,设计了面向工件机器分配的概率模型及其基于增量学习的更新方式,学习速率基于信息熵进行调整.为了增强算法局部寻优能力,采用基于关键机器的邻域结构进行局部搜索;同时讨论了信息熵与学习速率的关系,并探讨了关键参数对算法性能的影响.基于标准算例的测试结果与算法比较,验证了学习速率的自适应调整机制以及所提出算法的有效性.
In order to solve the problem of irrelevant parallel machine scheduling, an adaptive distribution estimation algorithm based on information entropy is proposed. According to the problem characteristics, a probability model for job machine allocation and an update method based on incremental learning are designed. The learning rate is based on information entropy In order to enhance the local optimization ability of the algorithm, a local search based on the key machine’s neighborhood structure was used. The relationship between the information entropy and the learning rate was also discussed and the influence of the key parameters on the performance of the algorithm was discussed. The result is compared with the algorithm to verify the adaptive adjustment mechanism of learning rate and the effectiveness of the proposed algorithm.