摘要
针对离散制造车间多目标性、多约束性的特点,以车间总耗能最小为优化目标,构建离散制造车间能效优化模型;针对上述模型提出了一种基于改进教与学算法的离散车间能效优化方法,该改进算法在教阶段引入自适应参数,提高算法的学习效率和适应性,同时在教和学两个阶段加入二次离散过程,在保证算法的收敛性快速和寻优能力强的特点的前提下,使该算法能够应用于离散制造车间优化中。对具体实例进行测试,将基本教与学算法、粒子群算法、鸡群算法与改进教与学算法的结果进行比较,通过分析该改进算法优化的能效明显优于另外两种算法,验证了算法的有效性。
Because the discrete manufacturing workshop is multi-objective and multi-constraint, an energy efficiency optimization model whose optimization objective was to minimize the total energy consumption was built for discrete manufacturing workshop. Besides, an improved teaching-learning- based （TLBO） optimization algorithm for discrete energy efficiency workshop optimization was proposed. This improved algorithm introduced adaptive parameter in training phase to improve the learning efficiency and adaptability of the algorithm. In addition, second discrete process was introduced in the teaching stage and learning stage, respectively. This algorithm could be applied to the optimization of discrete manufacturing workshop under the precondition of ensuring the convergence of the algorithm is fast and strong searching ability of the characteristics. The experimental result was compared with basic teaching-learning-based optimization algorithm （TLBO）, particle swarm optimization algorithm （PSO）, chicken swarm optimization （CSO）. Based on the analysis, optimization of this improved algorithm is superior to the other two algorithms, which indicates that the proposed algorithm is effective.
出处
《系统仿真学报》
CAS
CSCD
北大核心
2016年第12期3019-3026,共8页
Journal of System Simulation
基金
国家自然科学基金(61572238)
国家高技术研究发展计划(2014AA041505)
江苏省杰出青年基金(BK20160001)
关键词
离散制造车间
能效优化
教与学优化算法
自适应参数
二次离散
discrete manufacturing workshop
energy efficiency optimization
teaching-learning-based optimization algorithm
adaptive parameter
second discrete process
