This paper presents new network simulation algorithms that:
(1) drastically affect the formulation of network problems,
(2) permit the simulation of very large networks,
(3) afford large savings in computing cost. These algorithms may be used as computational aids to the design, evaluation, and operation of natural gas gathering, transmission, and distribution systems.
The first algorithm applies to networks without flow circuits. Its designation "CINS - I", is an acronym for Converging Information Network Simulator. The second algorithm is a modified version of CINS - I, designed to simulate networks with few flow circuits. Its acronym is "CINS - II". The third algorithm is a combination of CINS -I and the Newton-Raphson Method, which is used for very large multi-circuit networks. It is called "GINS", which is the acronym for General Information Network Simulator. The power of the CINS algorithms stem from the circumvention of matrix manipulations normally encountered in network simulation. GINS is designed to greatly reduce the size of the matrices required in the Newton-Raphson Method and to exploit the sparseness of the reduced matrices. Several applications of the algorithms to typical gas networks are also given.
The gratification of a modern society is deeply imbedded in a maze of networks which, among other things, collect and disseminate information, transport people, and distribute goods and energy. The enormous investment in and dependence on these networks demand that existing ones be rationally used and the new ones intelligently planned.
"A typical approach to a current decision making problem (i.e. a network design and/or operating decision) is to allow intuition, experience and current knowledge to dictate the selection of factors to be considered. There are some indications that good intuition is a manifestation of the correct mix of factors".
