Research On Modeling And Correcting Methods Of Leaching Rate In Leaching Process

Establishing a relationship model associated with some important production index, such as operating variables and quality, is the basis of the index-related prediction and process optimization. But in the process of industrial production, according to the production requirements, production often need to be done in a variety of working conditions. Meanwhile, the slowly time-varying of the production equipment and production process will also be able to affect the model performance. So, in the case of working condition or slowly time-varying of production equipment and production process, research on how to effectively to correcting model has important theoretical and practical significance, condition and slowly time-varying circumstances.In this thesis, we make acid leaching process of cobalt as the research background. First, we introduced the leaching process and the mechanism model of cobalt leaching process, and analyzed the main influencing factors of the cobalt leaching rate. In the base of analyzing of partial least-squares(PLS) regression analysis, we made the mechanism model of cobalt leaching process as the actual leaching process and established the prediction mode of the leaching rate by using the method of PLS. We monitored the performance parameters of the prediction mode of the leaching rate by using Principal component analysis(PCA) method, in order to realize monitoring and identifying the slowly time-varying or sharp change of working conditions in the leaching process. The simulation results verify the effectiveness of the method. Secondly, we introduced the leaching process modeling based on model migration and correction methods. On this basis, according to the monitoring and identifying result of slowly time-varying or sharp change of working conditions in the leaching process, we put forward different model correction method according to different situation of leaching process. The overall idea:when sharp change of working conditions occurs in the leaching process, we will correct current model by using the model correction method based on model database and model migration; When slowly time-varying occurs in the leaching process, we will correct current model by using increment of PLS method. Specifically, when sharp change of working conditions was monitored by using model monitor, we match existing models in the model database. If existing model is able to be used, we use directly the existing model to predict the rate of leaching. If we can’t find the matching model, we will quickly build a new model by model migration method to realize the correction and the add the new model to the model database. When slowly time-varying was monitored by using model monitor, we will correct current model by using increment of PLS method. Finally, according to the simulation research, it was proved to be a effective method to correct current model by using model correction method based on model monitoring and model database.