Method for Monitoring Hydraulic Coke Removal Using Cylindrical Array Fiber Acoustic Sensors

We demonstrated an on-line monitoring method for hydraulic coke removal when refining petroleum based on the feature extraction of collision sound using cylindrical array fiber acoustic sensors, which are installed around the oil refinery coke drum. Twenty optical fiber Fabry-Perot (F-P) acoustic sensors with high sensitivity and anti-electromagnetic interference can simultaneously monitor the vibration signal in the coke drum. The subsequent signal processing circuit analyses the change regulation of the signal intensity and the characteristics of the fast Fourier transform (FFT) spectrum. When the signal intensity of the closed acoustic sensor reaches the maximum and the main frequency of the FFT spectrum suddenly changes in order of priority, it can be estimated that the coke has been completely removed. The experimental results show that the cylindrical array fiber sensor group can satisfy the harsh working environment in which petroleum is refined and simplifies the signal processing for calculating the thickness of coke.