Abstract

Pipeline simulation tools have traditionally been used for modeling large pipeline networks; however, their quick solve times have made these tools attractive for many other applications ranging from process plant piping to valve manifolds. The governing equations and solver used by pipeline simulation tools are indeed applicable to any flow network, although many software packages offer a variety of choices that need to be properly adopted for each application. The first half of this paper will use examples to show in detail the impact of including thermal effects (adiabatic vs. full heat transfer), selecting appropriate fluid properties (single phase and simplified emulsion models), including viscosity effects on pump performance curves for a variety of systems, and model tuning and validation.

Another important consideration for modeling systems ranging in size is how to represent the flow network. For example, the impact of pressure drops caused by a bend, tee or reduction is much greater on a valve manifold than on a transcontinental pipeline, thus it is important to adopt the modeling methodology to the system scale. In addition, to accurately represent a liquid system it is necessary to represent the piping from all primary flow paths; however, to achieve quick model run times piping networks often need to be simplified. The second half of this paper will compare methods for simplifying piping network ranging in size such that pressure drops of all pipes, bends, and tees are represented and discuss the effect of these methods on steady state and transient modeling. A case study will be presented to show the impact of the modeling considerations.

Introduction and Background

Pipeline modeling tools for predicting the performance of liquid systems are well established in the industry. With the success of these tools, they are beginning to be applied to systems not originally envisioned such as cooling loops for power generation and fuel injection manifolds. A variety of tutorials exist in the literature such as those in references [1] and [2] on what to consider when modeling traditional liquid systems. It is, however, important to address several modeling approaches and assumptions that must be adapted to the scale and type of liquid system. The purpose of this paper is to fill the void by describing key concepts to consider when modeling non-traditional systems and to demonstrate the concepts by analyzing two case studies. These concepts can be applied to any modeling tool.

This content is only available via PDF.
You can access this article if you purchase or spend a download.