Fault Diagnosis System Research Of Cement Rotary Kiln

The coal consumption amount of cement industry accounted for about 15 percent of total coal consumption.So the cement industry is the main energy consumption industry.The comprehensive energy consumption of cement in China is around 20 percent higher than international level and the NOx emissions accounts for 10 percent of the total emissions.The saving energy and reducing consumption of cement industry play a very important role to achieve the goal of saving energy.Mechanism of cement production process is complex.It takes bad influence on the yield and quality of cement production because the abnormal running conditions often appear.Even danger may be taken to the field devices and the staff. Monitoring and fault diagnosis are not only the premise to realize energy saving target,but also the guarantee to realize safe production.Based on energy optimization project of Ping Yi China United cement industry,combine fractal theory with Support Vector Machine,and realize working condition recognition and fault diagnosis.There has been a tight correlation between the change of rotary kiln current and operation conditions,such as the kiln thermal condition,the thickness of the crust,the kiln tyre collapse.Considering different fractal characteristics that cement rotary kiln current signal exhibit in different conditions,we study on the generalized dimension of kiln current signal under different conditions as the characteristic parameter. The levels and trends of rotary kiln current and condition will be identified by computing the correlation coefficient according to the generalized dimension of detected signal and cement rotary kiln current under different working condition.It needs high requirements of data samples to distinguish the faults by using Support Vector Machine(SVM) and the classification effect is not ideal when characteristics of some attributes in the data sample is not distinguished obvious.This paper tries to introduce fractal theory into the processing of the data sample and replace the original data without strong attributes by the generalized fractal dimension characteristic.Then realize the classification by SVM. According to the 6 kinds of faults in the cement production,classification accuracy rate were 88.67% and 99.67%.The results show that we can realize the classification of fault diagnosis more effectively by combining the fractal theory and Support Vector Machine.