Optimal Design of Natural Gas Dryers

Optimal Design of Natural Gas Dryers

Yasser Al Wahedi^a*, Arwa H. Rabie^b, Prodromos Daoutidis^c*

^a Gas Research Center, Abu Dhabi Petroleum Institute, P.O. Box 2533, Abu Dhabi, United Arab Emirates

^b Department of Research & Technology, Abu Dhabi Gas Industries Ltd. (GASCO), Abu Dhabi, P.O. Box 665, Abu Dhabi, United Arab Emirates

^c Department of Chemical Engineering and Materials Science, University of Minnesota, Minneapolis, MN-55455, USA

* Corresponding authors, E-mail addresses: daout001@umn.edu & yaalwahedi@pi.ac.ae

Removal of moisture from gas streams is a commonly applied technology in the fields of natural gas processing, natural gas liquefaction, air separation, and air compression [1].  The degree of “dryness” of the gas stream is commonly expressed in terms of the water dew point [1,2]. In the field of natural gas processing and for shallow Natural Gas Liquid recovery (i.e. recovery of C3+ components) glycol processes can achieve a dew point of -40^oC, which can adequately meet sales gas specifications [2]. When ethane recovery is sought or when liquefied natural gas (LNG) is the ultimate product, glycol based processes fail to deliver the required gas dew point. For such applications, the targeted dew point is achieved by usage of adsorption based systems [2].

Molecular sieves (zeolites) are the adsorbents of choice when very deep dew point depression is sought as in the case of LNG production. In addition, the molecular sieving property of zeolites results in high selectivity towards water molecules and high rejection of hydrocarbons. Commonly used zeolites include Zeolite 3A and Na-X [3]. The current practice in designing natural gas dryers relies on empirical correlations and rules of thumb [2]. Optimality of such designs from a cost perspective is not guaranteed. Minimizing the Net Present Value of the costs (NPVC) ensued due to these dryers requires addressing three main challenges. Firstly, cyclic adsorption systems are by definition dynamic and hence are governed by dynamic Partial Differential Equations (PDEs) [4]. The solution of the governing model is required in order to evaluate the design constraints of absorption time, breakthrough time, and the required regeneration time [4,5]. Secondly, the attainment of the cyclic steady state condition at the optimal point has to be guaranteed [4,5].  Finally, the objective function is nonlinear, contains both integer and continuous decision variables, and may also contain discontinuities [6].

Recently, we have reported a novel approach in addressing similar challenges in the context of optimizing the design of temperature swing adsorption systems for Claus tail gas cleanup [6]. Specifically, we focused on the optimization of a model adsorption cycle. The cycle envisaged adsorption to occur isothermally while regeneration was simulated as an analogue to a pure component phase change. The evaluation of the breakthrough and required regeneration times relied on re-casting the governing PDE model into a semi-empirical form which is analytically solvable.  The approach was shown to lead to substantial savings in computational time. Furthermore, cost estimates of the proposed technology proved its potential in competing with existing commercial technologies in achieving the targeted removal at a substantially lower capital and operating costs.   

In the present work, we report the implementation of the same approach in the design of natural gas dryers. Specifically, this work presents a systematic methodology to design and schedule an optimal natural gas dehydration network with the minimum Net Present Value of Costs (NPVC) while meeting all process constraints. The proposed approach guarantees (local) optimality of the design and through its computational speed allows an effective search for optimal solutions for a large sample of initial guesses.   Furthermore, the high fidelity of the model employed provides higher confidence on the reliability of the design obtained and hence on the cost estimate. A case study of a real industrial dehydration network is presented and solved to illustrate the effectiveness of the devised methodology. Such attributes are critical especially in times were cost minimization is eagerly sought in the face of volatile energy markets.

References:

[1]  Kohl,A.,Nielsen,R., Gas Purification. 1997 Gulf Publishing Company, Houston.

[2] Manning, F. S., Thompson R. E., Oil Field Processing Of Petroleum – Volume One: Natural Gas, 1991, PennWell Publishing Company, Tulsa, Oklahoma  

[3] Tagliabue,M., Farrusseng, D. Valencia, S. Aguado S., Ravon U., Rizzo, C., Corma, A., Mirodatos,C., Natural gas treating by selective adsorption: Material science and chemical engineering interplay, 2009, Chemical Engineering Journal, 155, Issue 3, Pages 553-566

[4] Jiang,L.,Biegler,L.T.,Fox,V.G., Simulation and optimization of pressure- swing adsorption systems for air separation . 2003, AIChEJ. 49, 1140–1157.

[5] Jiang,L.,Biegler,L.T.,Fox,V.G., Design and optimization of pressure swing adsorption systems with parallel implementation . 2005, Comput.Chem.Eng.29, 393–399.

[6] Al Wahedi, Y.,  Torres, A., Al Hashimi, S., Dowling N., Daoutidis P., Tsapatsis, M., Economic assessment of Temperature Swing Adsorption systems as Claus Tail Gas Clean Up Units,  2015, Chemical Engineering Science, 126, pp 186-195.