Operating State Evaluation For The Multi-mode Industrial Process

With the deepening integration of informationization and industrialization,the process industry,as the pillar industry of national economy,is developing in the direction of high efficiency,green and integration,and the pursuit of production efficiency and economic efficiency is increasing day by day.Therefore,it is significant to judge the optimal degree of process operating state under normal working conditions,so that the managers and operators can grasp the process operating state,find the non-optimal operating state timely and propose reference suggestions based on the analysis results of non-optimal cause for further operating adjustment and process state improvement.Aiming at the characteristics of long process,multi-process,multi-mode,the coexistence of qualitative and quantitative information of process industrial process,a new operating state evaluation method based on multiple fuzzy probabilistic rough sets is proposed in this paper.In the offline modeling stage,aiming at the long process and multi-process characteristics,the whole process is divided into several sub-processes firstly according to the analysis of process mechanism.Then,in view of its multi-mode characteristic,the method of mode division and identification which combines rough division with fine division proposed in this paper is used to divide the historical modeling data,and each sub-process is divide into several stable modes and transition modes.Meanwhile,extract and express the mode features in order to realize the mode identification of online data.On this basis,aiming at the characteristic of coexistence of qualitative and quantitative information,a new method for evaluating the operating state of stable mode and transition mode of sub-process based on fuzzy probability rough set is proposed,and according to the historical data,the operating state evaluation rules of the whole process are extracted.When the operating state is non-optimal,the non-optimal cause identification method based on rule matching equivalence degree is performed to quickly locate the non-optimal reasons and provide guidance for operation and maintenance.Finally,a gold hydrometallurgy process which is the typical multi-mode process industrial process is taken as the background.Combining with actual production,two different production situations are simulated and analyzed respectively for the stable mode and the transition mode,which verified the effectiveness of the method proposed in this paper.