CDHG: A New Partitioning Algorithm based on the Detection of Cycles in Hypergraphs

PONZONI, IGNACIO; SÁNCHEZ, MABEL C.; BRIGNOLE, NÉLIDA B.

LATIN AMERICAN APPLIED RESEARCH

Planta Piloto de Ingeniería Química

Lugar: Bahía Blanca, Argentina; Año: 1998 vol. 28 p. 31 - 36

0327-0793

A graph-based approach for partitioning the occurrence matrix associated with the mathematical model of a chemical plant is presented in this work. The algorithm is based on the detection of cycles in hypergraphs. Some heuristic rules were defined in order to guide the search. The performance of the technique was assessed by applying it to an existing industrial ethane plant of medium size. The degree of partitioning achieved was highly satisfactory and the run-time proved to be  much lower than the one required by a combinatorial approach.