基于集成模型与遍历搜索算法的铅锌烧结透气性优化  

Permeability optimization based on integrated model and traversing search algorithm for lead zinc sintering process

作　　者：孙良君[1] 吴敏[1] 王春生[1] 徐辰华[1] 

机构地区：[1]中南大学信息科学与工程学院,湖南长沙410083

出　　处：《计算机与应用化学》2008年第7期897-901,共5页Computers and Applied Chemistry

基　　金：国家863计划资助项目(2008AA04Z128);国家杰出青年科学基金(60425310).

摘　　要：以铅锌烧结过程为研究对象，针对烧结过程透气性的控制问题，提出了基于集成预测模型与遍历优化搜索算法的铅锌烧结透气性优化控制方法。首先采用优化组合集成技术将神经网络预测模型和灰色系统预测模型有机结合，建立烧结综合透气性集成预测模型，然后结合14^#风箱温度和烧穿点温度建立透气性状况综合评判模型，最后通过遍历优化搜索算法，获得二配配比和混合料水分设定值，并进行跟踪控制，从而实现烧结过程透气性的优化控制。仿真结果表明：该方法能有效改善烧结过程的透气性，稳定烧结过程。To deal with the problem of the permeability control for the lead zinc sintering process, a permeability optimization control method for the sintering process based on integrated predictive model and traversing search algorithm is presented in this paper. Firstly, a neural network predictive model and a gray forecast model are established. By using optimal combination algorithm, a synthetic permeability integrated predictive model of the lead zinc sintering process is designed through combining two models. Then, Permeability of comprehensive evaluation model is established by combining with the temperature of 14^# bellows and burning through point. Finally, the permeability optimization control for the sintering process is realized by controlling values of the second proportion and the water content of mixture, which is gained through traversing search algorithm. The result of simulation shows that optimization control method can improve the permeability of the sintering process and stabilize sintering process.

关 键 词：铅锌烧结过程 透气性 神经网络模型 灰色模型 集成预测模型 遍历搜索算法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]