摘要
针对重油热裂解模型的参数估计问题呈高维、高度非线性的特征,提出一种基于新型蚁群算法优化的重油热裂解模型。通过新型蚁群算法优化确定模型参数,获得具有良好预测精度的模型。新型蚁群算法通过将解空间划分成若干子域,并引入遗传操作,实现连续优化问题的寻优。仿真结果表明它具有良好的性能,且优于传统的遗传算法。
Due to the character of multi-dimension and nonlinearity in estimation of heavy oil thermal cracking model parameters, a novel ant colony algorithm was proposed to obtain the optimal model parameters, the model with favorable prediction ability was developed. Further, the proposed ant colony algorithm was succeeded in the continuous space optimization problem through dividing the space into sub-domains and adopting the genetic operation. The simulation results prove that the proposed ant colony algorithm has favorable capability of search, and its performance is better than that of traditional genetic algorithm.
出处
《化工自动化及仪表》
CAS
2007年第6期20-23,27,共5页
Control and Instruments in Chemical Industry
基金
国家自然科学基金资助项目(20506003)
教育部科学技术研究重点项目(106073)
国家"863"计划项目(2007AA04Z164
2007AA04Z171)
关键词
重油热裂解模型
蚁群算法
空间分割
遗传操作
heavy oil thermal cracking model
ant colony algorithm
space dividing
genetic operation
