Parameter identification and modeling of the biochemical methane potential of waste activated sludge.

Anaerobic digestion is widely used in waste activated sludge treatment. In this paper, partial least-squares (PLS) is employed to identify the parameters that are determining the biochemical methane potential (BMP) of waste activated sludge. Moreover, a model is developed for the prediction of the BMP. A strong positive correlation is observed between the BMP and volatile fatty acids and carbohydrate concentrations in the sludge. A somewhat weaker correlation with COD is also present. Soluble organics (sCOD, soluble carbohydrates and soluble proteins) were shown not to influence the BMP in the observed region. This finding could be most-valuable in the context of application of sludge pretreatment methods. The obtained model was able to satisfactory predict the BMP.

Efficient tracking of the dominant eigenspace of a normalized kernel matrix.

Various machine learning problems rely on kernel-based methods. The power of these methods resides in the ability to solve highly nonlinear problems by reformulating them in a linear context. The dominant eigenspace of a (normalized) kernel matrix is often required. Unfortunately, the computational requirements of the existing kernel methods are such that the applicability is restricted to relatively small data sets. This letter therefore focuses on a kernel-based method for large data sets. More specifically, a numerically stable tracking algorithm for the dominant eigenspace of a normalized kernel matrix is proposed, which proceeds by an updating (the addition of a new data point) followed by a downdating (the exclusion of an old data point) of the kernel matrix. Testing the algorithm on some representative case studies reveals that a very good approximation of the dominant eigenspace is obtained, while only a minimal amount of operations and memory space per iteration step is required.