Prediction of Element Component Content Based on Mechanism Analysis and Error Compensation

Abstract

To solve the difficulty of rapid and accurate detection of component content in the rare earth extraction process, a component content modeling method combining mechanism model and error compensation model based on just-in-time learning (JITL) was proposed. Considering the different dynamic characteristics of each section, the extraction section is simplified using the segmented-aggregation method, and the mechanism model of the rare earth extraction process based on material balance is established; in view of the error caused by the simplification of the model and the characteristics of some rare earth solutions with color features, the color features of rare earth solution samples are extracted by machine vision technology, and the error compensation model of the mechanism model is established by the just-in-time learning algorithm. Through the experimental verification of the field sample data of the praseodymium/neodymium (Pr/Nd) extraction process, the results show that the modeling method proposed in this paper is suitable for rapid and accurate detection of elemental component content in the rare earth extraction process with ionic color features.