Process Monitoring And Fault Diagnosis For Thickener Decantation Process In Hydrometallurgy

Hydrometallurgy is the process to extract metal from ore by recovering the metal from an aqueous solution of the metal. Solid-liquid separation is one of the most important operations of hydrometallurgy. Thickener is the equipment for solid-liquid separation due to gravitational settling, and decantation refers to the separation process using thickener. In practice, the sever environment and many influencing factors make the working condition very complicated, sometimes even cause the’hold scrapping-slime raker’ or ’abnormal suspended solids’. Thickener’s accident could lead to huge economic losses, even may threaten the workers’ security. Therefore, online monitoring and fault diagnosis on hydrometallurgy thickener decantation process have great significance.PCA (Principal Component Analysis) is a typical process monitoring technique based on statistical information. PCA has attracted extensive attention as it is a simple mathematical method without using experiential knowledge. FTA (Fault Tree Analysis) is a top down, deductive failure analysis in which an undesired state of a system is analyzed using boolean logic combining the possible reasons. It is a simple effective method for fault diagnosis which is usually used in system fault analysis, prediction and diagnosis.This thesis presents a PCA&FTA method for monitoring and fault diagnosis of thickener decantation process based on Shandong gold mining’s hydrometallurgy. Firstly, we explain the thickener decantation process thoroughly, adopting PCA as the technique to conduct monitor process, detect unusual operating conditions and find out the abnormal variables. Then we establish the fault tree of decantation by analyzing the process deeply and trace back to the source causes using a fault inference method. Giving out reasonable suggestions to resolve the fault provides the security and efficiency of production process. In the end, a thickener decantation monitoring and fault diagnosis system is designed using C#and SQL Server. The fault diagnosis’s validity is tested and verified by simulation experiment.