Agglomeration Monitoring of Fluidized Bed Reactor Based on Voiceprint Feature Extraction
Shuai Shi, Zhong Zhao · 2018
I201 order to detect the polymer agglomeration in fluidized bed reactor (FBR), an agglomeration monitoring method is proposed based on voiceprint feature extraction and extreme learning machine (ELM). First, the acoustic emission (AE) detection technology is applied to collect the acoustic signals generated by the polymer impacting on the inner wall of FBR. Then the voiceprint features LP-MFCCs of the acoustic signals are extracted with the Mel Frequency Cepstrum Coefficients and the Linear Prediction Cepstrum Coefficients. Finally, extreme learning machine is applied to classify the extracted voiceprint features. The application results to a pilot plant have verified the feasibility and effectiveness of the proposed method.