秸秆发酵过程软测量模型关键参数的优化算法研究
Research on Optimization Algorithm of Key Parameters of Soft Measurement Model for Straw Fermentation Process
华天争1
作者信息
- 1. 江苏大学农业工程学院,江苏 镇江 212013
- 折叠
摘要
秸秆发酵制取燃料乙醇过程关键参量的最小二乘支持向量机软测量模型的性能依赖于模型正则化参数和核参数的选择,为此,该文提出了一种基于混沌和三角形游走策略的蜣螂群体优化算法,利用混沌映射初始化种群,提高种群的多样性,位置更新策略引入三角形游走策略.经标准函数集的仿真测试表明,改进优化算法相比于原始的蜣螂群体优化算法,在收敛速度和寻优精度上有较好的改进,可为后续的软测量模型参数寻优提供寻优思路.
Abstract
The performance of the least squares support vector machine soft measurement model for the key parameters in the process of straw fermentation to produce fuel ethanol depends on the selection of the model regularization parameters and kernel parameters.In order to address this problem,this paper proposes an optimization algorithm for the dung beetle population based on the chaotic and triangular wandering strategies,which utilizes chaotic mapping to initialize the population and improve the diversity of the population,and introduces a triangular wandering strategy into the position updating strategy,which can be shown to be better compared to the original optimization algorithm in terms of convergence speed and searching accuracy by simulation tests of the standard function set.The simulation test of the standard function set can show that the improved optimization algorithm has better convergence speed and optimization accuracy than the original dung beetle population optimization algorithm,which can provide optimization ideas for the subsequent parameter optimization of the soft measurement model.
关键词
最小二乘支持向量机/混沌/三角形游走/蜣螂群体优化Key words
least squares support vector machine/chaos/triangle wandering strategy/dung beetle population optimization引用本文复制引用
出版年
2024
NETL
NSTL
万方数据