Oct-2013

Applying pinch technology to 
energy recovery

A pinch technology-based study of heating and cooling of material streams in a large-scale olefins plant identifies major opportunities for energy savings

FARBOD RIKHTEGAR, Iranian Fuel Conservation Company
SEPEHR SADIGHI, Research Institute of Petroleum Industry

Article Summary

Energy saving is one of the most important issues associated with cost, regulations and environmental performance in the petroleum and petrochemical industries. Most of the available methods for energy targeting, retrofitting and design of heat exchanger networks are based on the pinch method.

The term “pinch technology” was introduced by Bodo Linnhoff in 1991 to represent a thermodynamically based methodology that guarantees minimum energy levels in the design of heat exchanger networks (HEN); therefore, this approach has been used to save energy in processes and across complete sites. Wherever heating and cooling of material streams take place, there is a potential opportunity to save energy. The design philosophy starts at the heart of the onion model, the reactor, and moves out to the separation system (see Figure 1). Heating and cooling duties for the next layer of the onion are the heat recovery systems. Consequently, targets can be set for the HEN to evaluate the performance of the process design, and it can enable both the energy and capital costs of the HEN to be assessed. It is obvious that, without a screening approach, selection between many design options cannot be easily afforded in terms of the time and effort required.

In this article, the discussion covers the basic principles and capabilities of pinch technology, and how the technology can be utilised to determine scope for reducing energy consumption and costs. In particular, the article demonstrates how the technology’s design methodology can be used for improving the heat recovery networks of an olefin plant, as a case study.

Pinch technology Composite curve
For analysing a heat exchanger network, sources of hot and cold streams (source and sink) should be first identified using material and energy balances. For instance, the current typical flow sheet of a chemical process is shown in Figure 2. The supply and target temperature and enthalpy changes of four process streams are also given in Table 1.

Consider steam at 200°C and cooling water at 20° for heating and cooling utilities, respectively. It is preferable to recover as much heat as possible between process streams. The scope for heat recovery can be determined by plotting all streams on a T-H diagram (see Figure 3). This figure shows DTmin=10°C for the proposed flow sheet; therefore, the hot and cold utility recoveries are 960 and 120 units, respectively (see Figure 4). It can be concluded that DTmin determines the relative location of the hot and cold streams, so it is an important variable for setting the amount of heat recovery.
 
Heat recovery pinch
To achieve the lowest DTmin, the type of heat exchanger and fluid regime are important. In Table 2, the minimum approach temperature for several industries is shown. As a rule of thumb, rating with a DTmin less than 10°C should be avoided. The correct setting of composite curves is defined by an economic trade-off between energy and capital cost. In Figure 5, the trade-off between energy, capital cost and economic amount of energy recovery is illustrated; thus, the trade-off can be carried out using energy and capital cost targets.

Utility selection
After recovering the heat using process-process heat exchangers, the remaining required heat for the plant should be obtained by the utility system. In pinch analysis, the grand composite curve (GCC) is an appropriate tool to show the interface between the process and the utility system (see Figure 6).
In this method, our aim is to use the specified utility at an appropriate level. Thus, for the hot utility, we should use the lowest temperature and generate the highest temperature. In contrast, for the cold utility, we should use the highest temperature and generate the lowest temperature.

Heat exchanger area target
It is possible to predict the required surface area for the whole problem by using vertical enthalpy intervals. The area calculated with this model is minimised when the heat transfer coefficients of all streams are equal. For each enthalpy interval, we can predict the required area from the composite curves. The duty and heat transfer coefficients of the streams are obtained from the stream data, and the log-mean temperature difference (DTLM) is derived from the composite curves (see Figure 7).

Capital cost target
The capital cost of a heat exchanger network is mainly dependent on the surface area of each heat exchanger, the number of shells, the material of construction, the heat exchanger type and the pressure rating. However, the key in capital cost targeting is the surface area required for exchangers that are included in the network. For the total cost targeting procedure, lots of appropriate software has been introduced, but SuperTarget from KBC is particularly regarded for handling energy targeting projects. SuperTarget is a suite of programs to optimise the energy consumption and utilities of a plant by applying pinch analysis. Furthermore, it enables the user to determine the absolute maximum potential for heat recovery, estimate utility and capital costs for a given heat recovery, determine the minimum approach temperature (DTmin), examine capital/energy trade-off, and set the basis for the heat exchanger network’s design. Additionally, the software can be integrated with other process simulators, such as PRO II and Hysys.
 
Case study
As a case study, an olefin plant located in the south of Iran at Bandar Imam was considered. The capacity of the plant is 411 000 t/y of ethylene product. Process data was taken from design process flow diagrams and from discussions with operating engineers. Due to the significant differences in temperatures and economics of operation, the olefin unit was separated into cold and the hot sections. A DTmin of 10°C was chosen to identify future potential savings. The stream data and composite curves of the process obtained using SuperTarget software are shown in Table 3 and Figure 8, respectively.