The concentration of organic acids in anaerobic digesters is one of the most critical parameters for monitoring and advanced control of anaerobic digestion processes, making a reliable online-measurement system absolutely necessary. This paper introduces a novel approach to obtaining these measurements indirectly and online using UV/vis spectroscopic probes, in conjunction with powerful pattern recognition methods. An UV/vis spectroscopic probe from S::CAN is used in combination with a custom-built dilution system to monitor the absorption of fully fermented sludge at a spectrum from 200nm to 750nm. Advanced pattern recognition methods, like LDA, Generalized Discriminant Analysis (GerDA) and SVM, are then used to map the measured absorption spectra to laboratory measurements of organic acid concentrations. The validation of the approach at a full-scale 1.3MW industrial biogas plant shows that more than 87% of the measured organic acid concentrations can be detected correctly.
[1]
David G. Stork,et al.
Pattern Classification
,
1973
.
[2]
Yee Whye Teh,et al.
A Fast Learning Algorithm for Deep Belief Nets
,
2006,
Neural Computation.
[3]
A. Puñal,et al.
Automatic control of VFA in anaerobic digestion using a fuzzy logic based approach
,
2002
.
[4]
Chih-Jen Lin,et al.
LIBSVM: A library for support vector machines
,
2011,
TIST.
[5]
Nello Cristianini,et al.
An Introduction to Support Vector Machines and Other Kernel-based Learning Methods
,
2000
.
[6]
André Stuhlsatz,et al.
Discriminative feature extraction with Deep Neural Networks
,
2010,
The 2010 International Joint Conference on Neural Networks (IJCNN).
[7]
André Stuhlsatz,et al.
Feature Extraction for Simple Classification
,
2010,
2010 20th International Conference on Pattern Recognition.