Today, process engineers are responsible for many project activities, including conceptual design, revamp studies and operational troubleshooting. Increasingly, the process simulator is an essential tool central to these activities. Process simulators are very powerful tools for modeling all or parts of a process. While they are excellent for general-purpose process modeling, it is the process engineers responsibility to understand to what extent these tools can be applied, and how combining their application with more specialized tools might be appropriate. This choice is ultimately based on the business and technical objectives to be achieved.
This article examines three different applications where rigorous heat exchanger models can enhance value derived from process simulation and provide more accurate results. These applications include conceptual designs of new plants, revamps of existing facilities, and operations support.
One of the key responsibilities of the process engineer is related to the conceptual design of processes. With conceptual design, the use of process simulation is central to project activities. The initial stages of conceptual design consider the main process synthesis and separation operations required to convert feedstocks to products. At this early stage, the process flowsheet typically involves simplified models of reactors, distillation columns, and the heating and cooling services required to facilitate the essential parts of the process. At this stage, the type and design of equipment required, for example, to preheat reactants before they enter a reactor, are less important. The traditional functionality of process simulation in providing heat and mass balance over the conceptual process is paramount.
As the conceptual design evolves, it becomes important to take account of the actual equipment involved. The reactors, separators and heat exchangers need to be evaluated to further develop their designs to ensure desired performance; to size them adequately; and to obtain estimates of the capital cost of the process, the heating and cooling utility requirements and the energy cost to operate the process. Heat transfer equipment can typically be up to 30% of the capital cost of process equipment. Therefore, as the process design progresses, it is important to take account of the real design requirements for the major heat transfer equipment items. For any heat exchanger, two main aspects must be considered:
- How much duty does the heat exchanger need to provide?
- How much pressure drop can be consumed?
The first aspect can be modeled by a simple equation:
Q = m (ho hi) (1)
where Q is the rate of heat transfer, m is the mass flow, h is specific enthalpy, and the subscripts o and i refer, respectively, to outlet and inlet. Where there is no phase change, this can be expressed as:
Q = m Cp (To Ti) (2)
where Cp is the specific heat of the fluid, and T is the temperature.
Consider the following simple example, where a water/water exchanger has been modeled with one side heating up from 20°C to 90°C, while the other side is cooling down from 90°C to 20°C (Fig. 1).
| Fig. 1. Water/water exchanger model |
showing one side heating up and one side
However, if the exchanger must be designed to determine how much surface it will require, the basic heat transfer equation (for pure counter-current flow) must be considered:
Q = UA 3 LMTD (3)
where U is the overall heat transfer coefficient, A is the effective area in the heat exchanger, and LMTD is the logarithmic mean temperature difference.
If a generic heat exchanger is assumed to have two ends (here referred to as A and B) at which the hot and cold streams enter or exit on either side, then LMTD is defined by the logarithmic mean, as follows:
LMTD = (ΔTA − ΔTB) ÷ [ln (ΔTA ÷ ΔTB)] (4)
where ΔTA is the temperature difference between the two streams at end A, and ΔTB is the temperature difference between the two streams at end B. In this case, the LMTD will have a limit of 0, so it will need a UA with an infinite limit.
The second aspect to consider for any real equipment is the pressure drop that will be consumed on the hot and cold sides as the respective streams flow through the heat exchanger. It is normal for the process engineer to designate how much pressure drop will be allocated to a particular exchanger. For example, in turbulent ﬂow inside tubes, the local heat transfer coefficient varies approximately with the mass velocity raised to the power 0.8. The pressure drop varies approximately with the mass velocity squared. This means that, if pressure drop is kept low, the heat transfer coefficient will be very low, and a large surface area will be needed for the heat exchanger. A realistic pressure drop must be estimated at this stage to enable the design of the heat exchanger later without having to rework the process design.
A more realistic way to model the exchanger is to assume that one side of the exchanger is between 90°C and 25°C, with the other side heating up from 20°C to 85°C. Both sides have a pressure drop of 0.5 bar (Fig. 2).
| Fig. 2. Heat exchanger model showing one |
side heating up and one side cooling down.
This type of idealized approach is often used to model an exchanger where a process stream is heated by utility steam in a heat exchanger. Pure fluids, like steam, condense isothermally at constant pressure. If isothermal condensation is present, then EQ. 3 can be applied to good effect; however, in reality, any pressure drop on the steam side will result in a lower saturation temperature, and then the exit temperature will be lower than the inlet temperature.
The main issue with this approach is that it is easy for the process engineer to specify conditions that later make it difficult to achieve a practical exchanger design. This hampers effective collaboration between process engineers and thermal design specialists, resulting in additional cycles of engineering to refine the overall process and equipment designs.
One way to promote better collaboration between disciplines and achieve better designs quickly is to use a rigorous exchanger modeling tool within the process simulation to achieve a preliminary design. This approach enables the process engineer to get a better first approximation for evaluating the feasibility of the process, and to give the thermal specialist a useful starting point for full design optimization. Where this technique is employed, it has been shown to reduce project schedules and eliminate costly rework.
The second type of project where rigorous heat exchanger modeling can improve the engineering workflow is a revamp. Typically, revamp projects have two main aspects. First, there is a check that the actual proposed equipment in the process is accurately simulating the plant performance data. Secondly, what if options can be explored for process and capital improvements, with different equipment geometries and stream sequencings validated against the revamps performance objective.
Modeling an existing exchanger can be easy if plant data is available. The process simulator allows the specification of process conditions for the exchanger. This, in turn, allows simple modeling of an exchanger based on EQ. 3, and it enables the simulator to estimate the exchanger duty. The inherent assumption is that UA will remain constant. The pressure drop will not be recalculated by the simulator, so any variation will need to be estimated with a manual calculation. As mentioned earlier, for single-phase turbulent flow inside tubes, the local heat transfer coefficient will vary according to:
a = f(m0.8) (5)
where a is the local tube-side heat transfer coefficient, and m is the mass velocity in the tubes. This indicates that, as the flow of either stream in an exchanger is varied, the simple modeling of the simulator will result in an error in the estimated duty of an exchanger. Change in steam properties will also be unaccounted for in this simple modeling approach.
In the following example, the first exchanger downstream of the desalter in a crude preheat train is subject to examination in a revamp study where the overall aim is to recover more pumparound energy and increase the throughput of the refinery (Fig. 3). The first step is to model the existing exchanger. The crude on the tube side of this exchanger is focused on in Table 1.
| Fig. 3. Revamp study of an exchanger |
in a crude preheat train.
The first two columns are the values of the pressure drop and the temperature changes on the tube side of the exchanger. The last two columns represent the difference between the simple UA modeling and the rigorous modeling approaches.
In the first set of process conditions, the rigorous model and UA model values are close. This is expected, since the UA model is based on the result of the rigorous calculation performed during the design stage. However, when the process conditions change, the UA model and rigorous model diverge, with the relative difference increasing from less than 1% to more than 3% for the temperature drop, and from less than 2% to more than 20% for the pressure drop. The rigorous modeling shows that the pressure drop increased to a value higher than the limit of 0.6 bar defined in the process. After the revamp and a redesign of the heat exchanger, it is possible to calculate the pressure drop for the rigorous model below the limit of 0.6 bar.
The rigorous modeling of the heat exchanger is needed to check the performance with new process conditions and to properly design a revamped heat exchanger. The integration of rigorous modeling inside the simulator allows the engineer to check the anticipated heat exchanger performance and take any corrective design actions without leaving the simulator environment.
In this case study, an existing exchanger on a gas compression system is water-cooled. The process is modeled with a control operation that simulates the adjustment of the water flow to achieve a specified outlet temperature for the gas being cooled on the tube side of the heat exchanger (Fig. 4).
| Fig. 4. Water-cooled exchanger on a |
gas compression system.
The operator is seeking to reduce the outlet temperature of the heat exchanger to reduce the power consumed by a large compressor. In the process simulator, it is simple to set a lower gas outlet temperature target in the control block, and the coolant flow rate will be increased until the new, higher duty is achieved.
In the rigorous exchanger simulation shown in Fig. 5, it is clear that the pressure drop on the water side is below the maximum allowable for the existing operating conditions.
| Fig. 5. Rigorous simulation for a water-|
If a lower gas outlet temperature is prescribed to affect the desired reduced compressor power, the rigorous model in the simulation responds to the increased coolant flow that the adjust mechanism imposes. The exchanger can now achieve the new duty. However, because a rigorous tool is being used, other beneficial calculations can be performed. The results highlighted in Fig. 6 show three issues to consider:
Pressure drop. The increase in water flow has resulted in a pressure drop on the shell side, which exceeds the design allowable. This may mean that sufficient pumping capacity will not be available to achieve the required flow.
Dynamic pressure. The Tubular Exchanger Manufacturers Association (TEMA) defines maximum dynamic pressure as:
q = rho v2 (6)
where rho is the fluid density, and v is the fluid velocity.1
The maximum dynamic pressure will be different based on the exchanger geometry. Exceeding these values brings the risk of excessive erosion and the potential for premature failure of tubes or other pressure parts of the exchanger.
Vibration. The rigorous exchanger model performs a vibration analysis for the exchanger bundle. It can be seen that the increase in the cooling-water flowrate has resulted in a possible risk of flow-induced vibration for this exchanger bundle. This can lead to tube failure, which, in some cases, can be rapid.
| Fig. 6. Results of rigorous simulation |
for a water-cooled exchanger.
The process simulation, coupled with the rigorous heat exchanger analysis, can reveal potential operational problems that go far beyond the simple considerations of heat and mass balance. In this case, the operator can choose to work within limits that avoid the risks of erosion, flow-induced vibration and other operational problems. The simulator and the rigorous exchanger tools can be used to evaluate an alternative control scheme, such as controlling the cooling-water temperature instead of the flowrate.
Best practice in exchanger/process modeling
Today, leading engineering and operating companies in the chemical and energy sectors are exploiting the integration of rigorous exchanger models within process simulation to reduce project schedules, minimize rework, and provide better overall optimization of their processes.
However, traditional organizations often separate process engineering, thermal design and mechanical design functions, which can be a barrier to the adoption of integrated technologies. As companies recognize the benefits provided by closer cooperation between the disciplines involved, many are seeing that they can make much more effective use of specialist skills when process engineers undertake preliminary designs using rigorous models in their simulations. Such simulations can then be fully optimized by the thermal specialist as process activities proceed.
In many smaller engineering organizations, a broader skill base for process engineers allows them to directly exploit the benefits of the integration discussed here. In a case study2 presented at the OPTIMIZE 2011 conference, one chemical company discussed a feasibility study wherein a reduction in capital equipment costs of 15% and an annual energy savings of $200,000 were discovered through the integration of rigorous equipment modeling with process simulation.
Another company obtained an estimated $5.5 million in additional revenue from increased liquefied petroleum gas (LPG) production, while reducing equipment costs by $0.5 million.3 This was achieved through the integration of a plate-fin rigorous modeling tool inside a process simulator. The integration allowed the evaluation of various alternative process solutions and their direct impact on temperature approach in the heat exchanger type selected.
The integration of rigorous modeling tools for heat exchanger modeling inside process simulators allows a faster delivery of projects by shortening the discussion time between different disciplines. Process engineers can be confident with the results of the process modeling by using the real geometry and the most rigorous tool for the heat exchanger calculation. Finally, plant operations are made safer by modeling all aspects of the heat exchanger operation, such as vibration.
1 Standards of the Tubular Exchanger Manufacturers Association, 9th Ed., New York, New York, 2007.
2 Roy, E., presentation at the AspenTech OPTIMIZE 2011 Conference, Washington, DC, May 2011.
3 Venkatesh, L., Petrofac Engineering India Ltd., presentation at Aspentech OPTIMIZE 2011 Conference, Washington, DC, May 2011.
|The author |
Julien Cazenave is an Aspen exchanger design and rating (EDR) business consultant for AspenTech, based at the companys European headquarters in Reading, UK. He has more than 10 years of experience in working with customers of AspenTechs EDR and simulation products across Europe, the Middle East and Africa. Mr. Cazenave ensures that customers derive maximum value from their investment and are regularly updated on new developments in the software and the underlying technology.