Multimode Process Mode Identification With Coexistence of Quantitative Information and Qualitative Information

The mode division and identification of the multimode process is to classify the normal production data according to the changes in process information and then to identify the corresponding process mode. The mode division and identification of the multimode process is the primary prerequisite for multimode process monitoring, evaluation, modeling, quality prediction, and process optimization. In the multimode complex industrial process, it usually contains both qualitative information and quantitative information, which makes the mode division and identification much more difficult. To deal with this problem, a new mode division and identification method is proposed in this article. First, the mode of the production process is roughly divided by the mode indicator variables that contain the initial production conditions with more qualitative variables. Then, the mode refinement is performed by using the process variables that contain a large amount of quantitative information. Therefore, the whole process is divided into multiple stable modes and transitional modes. At the end, the efficiency of the proposed method is illustrated with a case of the cyanide leaching process of gold hydrometallurgy. Note to Practitioners-From the practical application point of view, a new method of cross-referencing quantitative information model and qualitative information model is proposed in this article to divide the multimode production process into multiple stable modes and transitional modes. Then, feature models are established for different mode data online for process mode-type identification. The effective mode division and identification of multimode processes is the key to realize multimode process monitoring, evaluation, modeling, quality prediction, and process optimization. Although the validity of the proposed method is verified only in a gold hydrometallurgical cyanide leaching process, the method is equally applicable to other multimode complex industrial processes with the qualitative information and quantitative information coexistence characteristics.