Research And Implementation Of Key Variable Detection And Anormaly And Fault Monitoring Methods In Thick Process

With the rapid development of our social economy and the advancement of industrialization,we faces the problem of serious shortage of mineral resources.Mineral processing is an important production process to solve the problem of mineral resources in our country.As a key process in the beneficiation process,the key processes are difficult to detect,abnormal and fault monitoring.Therefore,it is of great theoretical and practical significance to study the soft-measurement of critical variables in critical processes and the problems of abnormal and fault monitoring.This thesis first introduces the basic principle of thickening process of mineral processing and the equipment process of thickener.Taking the actual thickening process of a mineral processing enterprise as the research object,the key variables detection and abnormality and fault monitoring problems in the thickening process are analyzed.In response to the detection of the feed concentration,the project team proposed and implemented a new concentration detection device for the feed concentration detection,the field operation effect is good,and its accuracy meets the production requirements;On this basis,this thesis proposes and implements a soft measurement method of feed concentration based on clustering and least squares modeling method based on the information of on-site thickener feeding concentrate pump,the experimental results show that the method achieves soft measurement of feed concentration without increasing the field equipment,and its accuracy basically meets the practical application requirements.Aiming at the measurement of the underflow concentration,a soft-measurement method of the underflow concentration based on the internal pressure value of the thickener is proposed,the software and hardware are implemented in the field.The on-site operation is good and the measurement accuracy meets the production requirements;at the same time,for the problem that the underflow concentration detection result cannot be provided when a pressure sensor is damaged,a soft measurement method of the underflow concentration under the condition of a certain pressure sensor damage is proposed,method for softly measuring the value of damaged pressure sensor by undamaged pressure sensor value,soft measurement of underflow concentration under the condition of damage of a pressure sensor is realized,and can automatically switch the model.The experimental results verify that the method can provide an underflow concentration measurement that basically meets the production requirements when the damaged pressure sensor is not replaced.For the problem of feeding abnormality and discharge failure in the dense process,a method of monitoring the feeding anomaly and discharge failure based on wavelet decomposition is proposed and the software implementation is completed,the experimental results show that the method can effectively realize the faults of the flotation column,the flotation column emptying,the underflow blockage and other faults and the underflow abnormality monitoring.Finally,using historical data analysis,the abnormality of the ore and the safety monitoring of the thickener are realized.