Download Embedded computing systems : applications, optimization, and by Mohamed Khalgui PDF

Show description

2. The Model for the Whole System The PN for the whole system with two storage and charging tanks, respectively, is shown in Figure 8(a). For the reason of simplicity, a simplified one is obtained and shown in Figure 8(b) by omitting the discrete places and its associated arcs, and 10 Discrete transitions for oil flowing in the pipeline The macro-transition for the pipeline the inhibitor arc in a tank PN model. It should be pointed out that, by omitting these places and arcs, it does not mean that the omitted places and arcs are useless in the model.

This results in high set up cost for transporting high fusion point crude oil. Thus, in scheduling crude oil operations, when high fusion point crude oil is transported from storage tanks to charging tanks, it is desired that it can transport as much as possible, or at least an expected amount by a single set up. Often, in determining a refining schedule, a planner wants to know if a given amount of high fusion crude oil in the storage tanks can be transported to charging tanks for processing by a single set up.

Here the segments of different oil types in a pipeline should be identified. Besides, the operation of an oil refinery process is governed by the decisions given in a short-term schedule. In other words, the process is under control. Thus, a model should describe these decisions (commands). A new hybrid Petri net model is developed for this purpose. A reader is referred as to (Zhou and Venkatesh, 1998; and Wang 1998) for the basic knowledge of Petri nets. The PN model used here is a kind of coloredtimed PN (CTPN) defined as CTPN = (P=PD∪PC, T=TD∪TT∪TC, I, O, Φ, M0), where PD and PC are sets of discrete and continuous places; TD, TT, and TC are sets of discrete, timed, and continuous transitions; I and O are input and output functions; Φ(p) and Φ(t) represent the color sets of the places in P and transitions in T; and M0 is the initial marking.