铅锌烧结配料过程的智能集成建模与综合优化方法

Intelligent Integrated Modeling and Synthetic Optimization for Blending Process in Lead-Zinc Sintering

以铅锌烧结配料过程为背景,针对传统配料方法中存在的成本高和准确率低的问题,提出一种智能集成建模与综合优化方法.首先,在建立过程神经网络模型和改进灰色系统预测模型的基础上,利用信息论中熵值的概念,提出一种既可保证预测精度又能满足配料计算对数据完备性要求的烧结块成分集成预测模型;其次,以成本最小为目标建立烧结配料优化模型,采用基于专家推理策略和改进免疫遗传算法的定性定量综合集成方法,实现烧结配料的优化.仿真结果验证了该方法的有效性.

To deal with the problem of high cost and low accuracy existing in conventional methods for the blending process in lead-zinc sintering, a kind of methodology for intelligent integrated modeling and synthetic optimization is proposed in this paper. First, based on the process neural network model and improved grey system prediction model, an intelligent integrated model is presented using the concept of entropy to not only guarantee the composition prediction precision of Pb-Zn agglomerate but also meet the requirements of the data completeness by blending computation. Then, a blending optimization model is established for the purpose of minimizing the costs. Finally, the mixture ra- tios are optimized by using a qualitative and quantitative meta-synthesis methodology based on the expert reasoning strategies and improved immune genetic algorithm. The simulation results demonstrate the validity of the proposed methodology.