Article Preview
Top1. Introduction
Prins (2005) says that during the oil crisis in the late 1970’s and through into the middle of the 1980’s, coal was regarded as a very important substitute for oil and as result in this period, several coal gasifiers were developed and commercialised. Unfortunately, due to a large drop in the oil price, coal gasification did not gain a much larger share of the energy market, although heavy oil gasification is still commercially practiced at several refineries. In the last few decades, there has once again been a renewed interest in gasification; however, the focus has been less on coal but more on biomass gasification, due to concerns about the greenhouse gas emissions from coal processes.
Knoef (2000) says that from a chemical point of view, the process of biomass gasification is quite complex, and it includes a number of steps:
- •
Thermal decomposition to non-condensable gas, vapours and char (pyrolysis)
- •
Followed by the thermal cracking of vapours to gas and char
- •
The gasification of char by steam or carbon dioxide and the
- •
Partial oxidation of combustible gas, vapours and char
Common among models that predict gasification and gasifier performance is the requirement for some kind of experimental results, kinetic data, access to thermodynamic databases to accurately predict the behaviour of the gasifier and assumptions regarding the reactions that occur in the gasifier (Ngubevana, Hildebrandt and Glasser, 2011). A typical example is the work done by Shabbar and Janajreh (2013), who use experimental data to model the empirical formula for coal.
The Boudouard equilibrium, the heterogeneous water gas shift reaction and the hydrogenating gasification seem to be popular choices as the main gasification reactions, as chosen by authors such as Zainal, Ali, Lean et al. (2001) and Schuster, Loffler, Weigl et al. (2001). This work follows in a similar vein in selecting independent reactions that occur in a biomass gasifier.
In the field of Process Synthesis, which can be best described as the step in design where the chemical engineer selects the component parts and how to interconnect them to create a flowsheet, there are several schools of thought in how to best approach synthesis and represent a design space of alternatives. Grossmann and Daichendt (1996) say that the most common approaches for optimising process-synthesis problems are: hierarchical decomposition, superstructures, and targeting techniques.
The superstructure is assumed to contain all possible alternatives of a potential treatment network, including the optimal solution that is hidden. Saif, Elkamel and Pritzker (2009) have said that a “…common approach for formulating superstructures for process-synthesis problems involving heat and mass exchange has been to use the state-space framework. In this way, unit operations, utility units, and utility streams can be embedded in such a way that all the process synthesis alternatives can be realized.”
A representative example of the heuristic search approach is the hierarchical decomposition method by Douglas (1985), whose major strength lies in the use of engineering knowledge, and the successive refinement and generation of design alternatives. Its weakness however is said to lie in its lack of a systematic framework for process modeling, its inability to perform simultaneous optimisation and its reliance on heuristics which may not always be valid. Westerberg (2004) and Saif et al. (2009) mentioned that despite its considerable value, the hierarchical technique cannot evaluate process alternatives simultaneously, nor can it accommodate multiple objectives. On the other hand, the superstructure approach can handle a wide range of practical synthesis problems. Thus, there seems to be some agreement that the superstructure approach is the most favourable for process-synthesis problems.