Performance parameter prediction for sewage sludge digesters using reflectance FT-NIR spectroscopy

Abstract

This study investigates the use of Fourier transform near infrared (FT-NIR) spectroscopy combined with principle components analysis (PCA) and partial least squares regression (PLS-R) as part of a possible process monitoring system for sewage sludge anaerobic digesters. The ability of FT-NIR with PCA to distinguish between different stages of the AD process was investigated, it was found that waste activated sludge (WAS), primary, feed (Primary:WAS 70:30) and digested sludge were distinguishable from each other using this technique. PLS-R was used successfully to track differing proportions of primary:WAS in feedstocks of 5% total solids (Coefficient of Efficiency (CE) = 0.93). The study also looked at the ability of reflectance mode NIR spectroscopy to track process parameters important for stability. Temperature and organic loading rate variations were employed to stress the digesters. Predictive models were produced for volatile fatty acids (VFA), bicarbonate alkalinity (BA) and total and volatile solids (TS and VS) and independently validated for each digester. The models were able to track the relevant process parameters: TS (CE = 0.75), VS (CE = 0.75), BA (CE = 0.71), and VFA (CE = 0.69). This technique could be used to improve the performance of sewage sludge anaerobic digesters.