Elsevier

Energy

Volume 146, 1 March 2018, Pages 82-97
Energy

Cost-effective strategy for heat exchanger network retrofit

https://doi.org/10.1016/j.energy.2017.09.005Get rights and content

Highlights

  • A cost-effective retrofit strategy is proposed.

  • The approach relies on combining structural modifications and enhancement.

  • Pressure drop considerations reduces the energy savings with enhancement.

  • Energy performance is improved by selective use of structural modifications.

Abstract

Cost-effective retrofit of heat exchanger networks (HENs) remains a significant challenge. This paper explores different methods for achieving cost-effective retrofit. The first part of this article presents a novel methodology for the application of heat transfer enhancement in HEN retrofit with a fixed network structure considering pressure drop constraints. Heat transfer enhancement is a low-cost option. However, heat transfer enhancement on its own without changes to the network structure provides a limited scope for energy reduction. The second part of this paper presents a new pinch retrofit method that identifies network structural changes sequentially to meet the retrofit target. However, the high capital cost associated with installing new heat exchangers, relocating existing exchangers, and augmenting the heat transfer area of existing heat exchangers most often leads to uneconomic retrofits. Low-cost retrofit requires few modifications. Therefore, the third part of this paper combines the new pinch retrofit method with the use of heat transfer enhancement to provide low-cost retrofit by combining the merits of both approaches. A case study highlights the benefits of the new approach.

Introduction

Increasing concerns associated with greenhouse gas emissions have led to a rise in interest into the retrofit of heat exchanger networks (HENs). In the process industries, the retrofit of HENs can be a cost-effective method to reduce the energy consumption. Maximising the use of existing equipment increases the profitability of the retrofit process [1]. In existing HENs, operating and physical constraints referred to as bottlenecks restrict the degree of energy savings.

Established techniques and methodologies for the retrofit of HENs focus on modifying the existing HENs to obtain energy savings and overcome network bottlenecks. Modifications to the existing HEN are primarily based on Pinch Analysis, Mathematical Programming, and Hybrid methods. There are several reports in literature dedicated to the retrofit of HENs based on these methods. Sreepathi and Rangaiah [2] presented a detailed review of the different methods for HEN retrofit. The book by Smith [3] has full chapters that introduce the fundamentals of energy targets, capital and total cost targets, and network design of HENs. Pinch Analysis for retrofit makes use of a targeting stage for estimating the maximum energy recovery of a network, and a re-design stage to disconnect and reconnect the cross-pinch exchangers to obey the pinch decomposition. The pioneering work on Pinch Analysis for retrofit was introduced by Tjoe and Linnhoff [4]. The objective of the work is to eliminate heat transfer across the pinch. This concept has since been extended to account for the cost required in HEN retrofit [5]. Most recently, the work by Gadalla [6] presented a new graphical method based on Pinch Analysis. The energy efficiency is evaluated quantitatively in the existing HEN to identify the potential modifications for better energy recovery. The drawbacks of Pinch Analysis are that it requires an expert user for its application and does not highlight the number of modifications required and the appropriate placement for the additional heat transfer area requirement. Also, Pinch Analysis requires too many changes in a single step, which makes it fundamentally not suited to retrofit, as it tries to convert the existing network into an ideal grass root design in a single step instead of accepting the features that already exist.

Mathematical programming methods convert the retrofit problem into an optimisation model and solved. However, the HEN retrofit is a mixed integer non-linear programming (MINLP) problem [7]. Ciric and Floudas [8] presented a superstructure approach for retrofit of HENs. The objective of the approach is to optimise the superstructure which contains all structural features of an existing HEN to remove all unnecessary features and minimise the cost. The difficulty in solving the MINLP problem has led to authors simplifying the retrofit problem to avoid obtaining a local optimum solution. Over the years, several authors developed ways of decomposing the retrofit problem. Ma et al. [9] developed the Constant Approach Temperature model used to linearise the area calculations. Other authors [10] presented a two-stage approach to retrofit. The first stage, the prescreening stage is solved as a mixed integer linear programming (MILP) problem. The second stage, the optimisation stage is solved as a nonlinear programming (NLP) problem. Another way to overcome the difficulty in solving the MINLP problem might be to apply stochastic optimisation [11]. A benefit of mathematical programming methods is that the total cost and environment impact of retrofit can be considered in the optimisation process [12]. Compared to the Pinch Analysis method, mathematical programming methods can consider more variables and identify the optimal HEN. In general, the drawbacks associated with the use of mathematical programming techniques include prolonged computational times, uncertainty in the optimality of the solution due to the assumptions and simplifications made to the model and the lack of user interaction.

Asante and Zhu [13] pioneered the hybrid method for the retrofit of HENs. The method proposed, referred to as the Network Pinch Approach, consists of a diagnosis and an optimisation stage. In the diagnosis stage, the MILP model is used to identify the possible structural modifications that can provide maximum energy recovery subject to an assumed minimum temperature approach. The optimisation stage makes use of an NLP model to optimise the capital-energy trade-off of the structural modifications determined in the first stage. The sequential approach enables the automation of the design procedure while maintaining user interaction. Smith et al. [14] modified the Network Pinch Approach by converting the process from sequential to simultaneous (considering structural modifications and capital-energy optimisation in a single step). The Network Pinch Approach by Asante and Zhu [13] was later extended to handle more complex networks considering some practical features to increase the possibility of identifying cost-effective design solutions [15]. The Network Pinch Approach provides energy savings by manipulating the existing degrees of freedom (utility paths, loops and stream splits) in an existing network [3]. However, the only way of overcoming the Network Pinch for energy savings is by performing structural modifications. Examples of modifications considered to overcome the Network Pinch are adding a new heat exchanger (new match), relocating exchangers (resequencing), adding stream splits. The Network Pinch Approach is an automated sequential method restricted to one change at a time. The drawback with the Network Pinch Approach is the lack of insights into the decision-making process of identifying the best series of modifications that can be applied to a given HEN. As such, there is a possibility of selecting a retrofit option early in the procedure that prevents obtaining the optimal solution in subsequent steps.

In recent years, there has been a rise in the research into the use of heat transfer enhancement in retrofit. Heat transfer enhancement not only increases energy recovery but can be a low-cost option for retrofit as existing network structure can be maintained. This makes the implementation of enhancement devices relatively simple and can be performed during the normal shutdown period. Optimisation methods presented by Pan et al. [16] applied detailed models of different heat transfer enhancement techniques for the application of heat transfer enhancement. The authors also provided a systematic design method for the application of conventional retrofit strategies [17]. Optimisation methods provide no insights into the identification of the best heat exchangers to enhance. Also, the simplifications in the retrofit method lead to uncertainty in the retrofit solution. Heuristic based methods have been developed to tackle the issues presented by optimisation methods for the application of heat transfer enhancement. Wang et al. [18] presented a method based on sensitivity analysis. However, this approach did not consider the impact of enhancement on the network. Also, the degree of enhancement is assumed. Jiang et al. [19] extended the approach to consider accurate modelling of the chosen enhancement technique to ensure accurate representation of proposed energy savings. The work by Akpomiemie and Smith [20] extended both methodologies to account for the downstream effects on the network after the application of heat transfer enhancement. The drawbacks with the use of sensitivity analysis for identifying the best heat exchanger to enhance were highlighted by Akpomiemie and Smith [21]. The authors [21] presented an alternative method known as the area ratio approach for the identification of the best heat exchangers to enhance. With this method, the decision on the best heat exchanger is not dependent on a key utility exchanger, as in the case of sensitivity analysis, but on the degree of enhancement that a heat exchanger can provide relative to its base case value. As such, this method is more suited to HENs with multiple utilities. A drawback of the heuristic based methods considering enhancement is the lack of pressure drop considerations with heat transfer enhancement. Ignoring the effects of pressure drop in retrofit might present a retrofit result that cannot be realised industrially as the existing pumps/compressors might not be able to cope with the increased pressure drop requirements.

Polley et al. [22] first considered the effects of pressure drop in retrofit. However, the work only considers a targeting stage based on area efficiency and does not present a systematic way of applying heat transfer enhancement in retrofit. Nie and Zhu [23] presented practical methods for mitigating the effects of pressure drop in a network after structural modifications are made. Techniques considered include decreasing the number of tube passes and modifying shell arrangements from series to parallel. Both methods have an impact on the tube-side velocity of the exchangers, which dictates not only the heat transfer coefficient but also the pressure drop requirement.

Recent methods used for the retrofit of HENs with pressure drop considerations have either been based on an iterative MILP optimisation approach [24] or a combination of a set of heuristic rules and NLP optimisation [25]. The drawbacks of the MILP optimisation approach are that there is a lack of insights into identifying the best exchangers to enhance and the use of the MILP model to solve the retrofit problem. To address these drawbacks, the work by Akpomiemie and Smith [25] present a novel sequential approach to retrofit with heat transfer enhancement considering pressure drop constraints. The new approach provides insights into identifying the best exchangers to enhance. The new method also defines a ranking criterion for selecting the best pressure drop mitigation technique. An NLP model is used to ensure the feasibility of the retrofit solution based on a set of constraints. However, the energy savings that can be obtained for fixed network structure with heat transfer enhancement is limited [26]. To obtain high energy recovery, structural modifications to the existing HEN are required.

From the reviewed methods for performing structural modifications, the Network Pinch Approach presents the right balance between providing optimal solutions and user interaction. Studies based on the Network Pinch approach [14], [15] do not provide insights into the decision-making process for selecting the best series of structural modifications in retrofit. To overcome this drawback, this work presents a new pinch retrofit method that provides insights into the fundamental interactions and features of an existing HEN. The Network Pinch Approach identifies the bottleneck for heat recovery in an existing system. However, it provides no guidance as to how to overcome the Network Pinch. Overcoming Network Pinch is left to the use of NLP optimisation. This provides no insights into the solution and no insights as to whether other options might provide an equivalent answer. User intervention is particularly important in industrial retrofit. The problem is not the purchase of new equipment, but the many other issues of pipework and civil engineering requirements and in many situations simply the ability to accommodate new equipment. These issues cannot be included in optimisation, but require user insights. The paper presents an approach that allows the designer to include insights into the decision-making.

The main objective of this work focuses on providing low-cost retrofit methods for HENs. This work presents the benefits of combining the use of heat transfer enhancement and structural changes to existing HENs. The new method is applied sequentially with an objective of maximising energy recovery with minimum investment. A case study is used to illustrate the proposed methods and highlight their benefits by making a comparative analysis of the various options considered in this work for HEN retrofit.

Section snippets

Heat transfer enhancement for a fixed network structure

This section of the paper summarises the method presented by Akpomiemie and Smith [25] for the application of heat transfer enhancement for a fixed network structure with pressure drop considerations. The approach combines a set of heuristic rules and optimisation to meet the retrofit target. Fig. 1 shows the retrofit methodology. Initially, the retrofit profit (RP) is 0.

Structural modifications (The pinch retrofit method)

The pinch retrofit method is based on the Network Pinch Approach presented by Smith et al. [14]. However, the primary purpose of the pinch retrofit method is to present guidelines for identifying the best location to apply a series of modifications based on fundamental insights in a step-by-step approach from the existing network. The aim is to identify the best series of modifications that can achieve maximum energy recovery with the minimum number of modifications to the existing network. The

Combination of structural modifications with enhancement

This section presents a methodology for considering structural modifications based on the pinch retrofit method alongside enhancement to achieve cost-effective energy savings. By combining structural modifications with enhancement, the level of energy savings obtained with structural modifications can be maintained but at a reduced retrofit cost. Increasing the heat transfer area of existing exchangers is capital intensive. However, a benefit of heat transfer enhancement is that an enhanced

Case study

A simplified crude-oil preheat train is used to illustrate the application of the retrofit methodologies in this work. Fig. 10 shows the existing HEN structure. Table 3, Table 4, Table 5 shows the stream, exchanger and cost data respectively.

Conclusions

New retrofit methodologies for HENs based on the application of heat transfer enhancement considering pressure drop constraints, structural modifications and a combination of both methods have been presented in this work. The benefits of these methodologies have been demonstrated with a case study. The results show that the application of enhancement alone can bring about cost-effective energy savings, but the degree of energy savings is limited, i.e. achieving only 5.5% of the HEN initial

References (31)

Cited by (32)

View all citing articles on Scopus
View full text