Organic Rankine Cycle (ORC) technology is proven in the field of waste heat recovery (WHR). Its ability to recover low grade heat from combustion flue gas, kiln and furnace exhausts makes it an ideal technology choice for WHR from industry. However recovering the waste heat from industrial process comes with a number of technical challenges from heat exchanger tube fouling, erosion, corrosion and an increased system pressure drop.
Tube fouling, the concentration and characteristics of particles in the exhaust gas attaching themselves to the heat exchange surfaces reduce the efficiency of energy transfer. These particles can also be abrasive and or corrosive in nature depending on the combustion, smelting or calcination process, thus reducing the heat exchanger life expectancy. The build-up of particles around the heat exchanger tubes also increases the pressure drop in the system resulting in a reduced flow rate or requiring increased electrical fan power to maintain the same flow rate. The operational risk of integrating the WHR exhaust gas heat exchanger into the process can be minimised at the design stage by using well developed Computational Fluid Dynamic (CFD) modelling.
The use of CFD a computer aided engineering software is widely used throughout engineering design and is applied to model the thermo-fluid dynamics of the exhaust gas entering the WHR system. CFD can accurately predict the performance of fluid flows, heat transfers, mass transfers and chemical reactions. It works on the principle of discretizing Navier-Stokes equations to describe the flow-related characteristics with a system of non-linear partial differential equation . CFD analysis for the design of WHR system with single phase method, can be solved with less computational time and resources. A single phase method of modelling exhaust gas characteristics can determine the velocity, pressure drop and flow pattern. This helps to modify the installation of ductwork around the heat exchanger and to achieve equal distribution of fluid flow around the heat exchanger tubes, which will result in a lower pressure drop for the system.
To model a complex system analysing the characteristics of exhaust gas and particle loads at the same time, a dual phase approach is required. Modelling dual phase flow of exhaust gas and particle loads, improves the fluidic boundary conditions of the heat exchanger and particles can be removed with additional equipment such as cyclones or dedusters. CFD modelling for cyclones or dedusters helps to track the particles, determines particle dropout efficiency, pressure drop, velocity and temperature. To run a multiphase simulation with a CFD model, there are two distinct approaches available. The Eulerian approach treats the dispersed phases as interpenetrating continua while the Lagrangian approach tears the dispersed phases as individual entities . The advantage of Lagrangian over the Eulerian approach is the tracking of particle injected into the domain. The average values of velocity, temperature and mass of particle phase can be modelled by Eulerian approach, whereas Lagrangian approach calculates these values for each representative particle .
Figure 1: CFD analysis on WHR system with single phase method
In conclusion the modelling of WHR system using CFD reduces the risks of particle fouling and minimises pressure drops resulting in an optimised performance and improved life expectancy. When the concentration of particle loads is low, exhaust gas characteristics are almost unaffected, in such cases a single phase CFD method of modelling is sufficient. When there is a high concentration of particle loading present it is recommended that a dual CFD method of modelling is applied to produce the optimum performance design.
|||M. Giavazzi, “A study on dual phase CFD for the dimensioning of flue gas treatment equipment,” World Cement, pp. 59-68, 1 March 2015.|
|||A. Majid, Y. Changqi, S. Zhongning, W. Jianjun and G. Haifeng, “CFD simulation of dust particle removal efficiency of a venturi scrubber in CFX,” Nuclear Engineering and Design, vol. I, no. 256, pp. 169-177, 2013.|