改进蚁群算法在铁矿石价格指数预测中的应用  

Application of Improved Ant Colony Optimization in the Prediction of Iron Ore Price Index

作　　者：张智峰[1] 王涛[1] 兰洁[1] ZHANG Zhi-feng;WANG Tao;LAN Jie(College of Science, Liaoning University of Technology,Jinzhou 121001 ,China)

机构地区：[1]辽宁工业大学理学院,辽宁锦州121001

出　　处：《模糊系统与数学》2018年第2期185-190,共6页Fuzzy Systems and Mathematics

基　　金：辽宁省高校基本科研业务项目(JL201615410)

摘　　要：钢铁作为国民产业的重要组成部分,其价格对国民经济产生巨大的影响,而钢铁的一重要来源就是铁矿石。因此,对铁矿石价格指数的精准预测对国民经济有着重要的调节作用。论文运用具有全局信息素迹更新的改进蚁群算法(IACO)来进行模糊逻辑系统规则的筛选,以此来提高模糊逻辑系统(FLS)的精确性。将选择得到的规则作为规则基,融入神经网络中设计相应的模糊逻辑系统。将设计的智能模糊逻辑系统用于中国铁矿石价格指数(CIOPI)的预测,仿真结果表明,提出的方法是可行和有效的。与无规则筛选的和原始蚁群算法(ACO)的模糊逻辑系统相比,都具有优越性,以此也证实了进行规则筛选的必要性。Iron as an important part of the national industry,the price has a huge impact to the national economy, and an important source of iron is the iron ore. Therefore, the accurate prediction of the iron ore price index has an important regulatory effect on the national economy. This paper proposes an improved ant colony optimization （IACO）for global pheromone trail updating, and uses this algorithm to select rules of the fuzzy logical system to improve the accuracy of fuzzy logical system （FLS）.Regarded those selected rules as rule bases, and integrated into the neural network to design the corresponding fuzzy logic system. In order to test the performance of the system, the system is designed for the prediction of international petroleum price, and the simulation results show that the proposed method is effective. Compared with the fuzzy logic system without rules selected and the ant colony algorithm （ACO） without improved, the algorithm can obtain a better result.

关 键 词：改进蚁群算法 模糊逻辑系统 神经网络 蚁群算法 BP算法 

分 类 号：F062[经济管理—政治经济学]