Modal Partition And Identification Of Hydrometallurgical Leaching Process Based On Interval PCA And Interval RBFNN

With the economic development of our country,the rapid increasing in the demand for gold resource by society has accelerated the exploitation of resources,which leads to the depletion of high-grade gold resources.After investigation,China is rich in low grade gold resources.How to use the mineral resources economically and effectively is of great significance to the sustainable development strategy of China.As one of the important technologies for refining metallurgy,hydrometallurgy technology is more suitable for recycling of mineral resources because of its advantages of dealing with low-grade ore and less harm to the environment.Therefore,it has been widely used in mineral industry.As a core part of the hydrometallurgical process,leaching process's operation conditions directly affect the quality of the whole gold hydrometallurgy process,so scholars have gradually realized the importance of research on the leaching process,the process of monitoring,fault diagnosis and operation state evaluation for the leaching process has gradually become a research hotspot.However,in the leaching process,the most prominent feature of the process is the multi-mode production process,that is,there are many stable working points in leaching process.The classification and identification of the mode of leaching process has become the basis for the study of process monitoring and fault diagnosis in leaching process.In more than one mode of the leaching process,due to the precision of the sensor limitations or external disturbance causes widespread the problem of inaccurate data,so for the interval data,multi modal division and recognition of the leaching process has become the prerequisite for studying the wet metallurgical leaching process.This thesis is aimed at a typical example of hydrometallurgy of complex industrial processes,which is hydrometallurgy of gold,based on the detailed analysis of the process characteristics of the leaching process,a model of modal division and recognition for the hydrometallurgical leaching process based on interval data is established.The main research work is summarized as follows(1)The process and operation principle of leaching process with sodium cyanide as the leaching agent were analyzed,which lay the foundation for the multi-modal partition and identification model of leaching process.(2)According to interval data,the interval PCA method and interval RBF neural network combination method for classification and recognition of multi modal model of leaching process,namely using the interval PCA method is used to reduce the dimensionality of interval sample data,using interval RBF neural network for interval data after dimensionality reduction to achieve modal classification and identification.(3)This thesis studies on hydrometallurgical leaching process,is based on the theory of Interval PCA method and Interval RBF neural network,a multi-modal partition and recognition model for leaching process based on interval data is established,and multi-modal partition and recognition for hydrometallurgical leaching process based on interval data is simulated.