Pyrolytic homogeneity enhancement of municipal solid waste using a clustering-based sorting strategy.

Heterogeneity of pyrolytic parameters in municipal solid waste (MSW) significantly hinders its waste-to-energy efficiency. So far, hardly any light has been shed on current pyrolytic heterogeneity conditions or feasible pyrolytic homogeneity enhancement approaches of MSW. Accordingly, pyrolytic properties (E and logA) of 130 MSW samples in 6 categories were collected from literature. A kinetic parameters clustering-based sorting strategy for MSW was proposed. A so-called C index was established to compare their sorting performance for E and logA against two traditional sorting strategies (substance categorization and density clustering). Results showed that the proposed sorting strategies outperformed the traditional ones in pyrolytic homogeneity enhancement, where the optimal C_E and C_logA reached 1578.30 kJ/mol and 93.11 -log min. Among investigated clustering methods, k-means clustering outperformed hierarchical clustering, which could be attributed to its adaptability to the sample structure. Future perspectives involving data set expansion, model framework development, and downstream technologies matching were also discussed. The index C established in this study can be used to evaluate other clustering models.