生物氧化冶金预处理过程氧化率的软测量模型

针对生物氧化预处理过程中的矿物氧化率不能直接测量的问题,提出一种基于遗传算法和BP神经网络相结合的软测量方法。该方法在对生物氧化预处理过程影响因素分析的基础上,选取温度、pH值、进风量和氧化还原电位(ORP)作为辅助变量,氧化率作为主导变量,利用遗传算法对BP神经网络进行优化,然后采用BP神经网络算法实现对氧化率的预测。为了突出表现GA-BP模型的准确性,与PSO-BP的软测量模型进行了对比。结果表明,基于GA-BP的软测量方法更利于实现氧化率的精确预测,有助于对氧化率的优化控制提供指导。

Soft-Sensing Model for Oxidation Rate of Biooxidation Metallurgy Pretreatment

Aiming at the problem that the mineral oxidation rate cannot be measured directly during the biological oxidation pretreatment process, a soft measurement method based on the combination of genetic algorithm and BP neural network is proposed. Based on the analysis of the influencing factors in the biological oxidation pretreatment process, this method first selects temperature, pH values, air intake and redox potential (ORP) as auxiliary variables, and oxidation rate as the dominant variable, and then uses genetic algorithm to the neural network are optimized, and finally the BP neural network algorithm is used to predict the oxidation rate. In order to highlight the accuracy of the GA-BP model, it is compared with the PSO-BP soft measurement model. The experimental results show that the soft measurement method based on GA-BP is more conducive to the accurate prediction of the oxidation rate and helps to provide guidance for the optimal control of the oxidation rate.