Optimal design of ethylene and propylene coproduction plants with generalized disjunctive programming and state equipment network models

Highlights

• Optimal design of an ethylene and propylene coproduction plant.

• Superstructure optimization with GDP model through custom implementation of Logic-based Outer-Approximation.

• SEN representation for potential distillation trains and acetylene reactors.

• Rigorous models for distillation columns, compressors, turboexpanders, vessels.

• Analysis for four international price scenarios for raw material and utility costs.

Abstract

In this work, we propose a superstructure optimization approach for the optimal design of an ethylene and propylene coproduction plant. We formulate a superstructure that embeds ethane and propane steam cracking technologies, propane dehydrogenation and olefin metathesis processes. We represent the superstructure with a Generalized Disjunctive Programming model, and solve the problem through a custom implementation of the Logic-based Outer Approximation algorithm in GAMS. We propose a state-equipment-network representation to model potential distillation trains, as well as alternative acetylene reactor configurations. Rigorous models are formulated for distillation columns, compressors, turboexpanders, vessels and several process equipment units. The objective function is to maximize the net present value. We analyze four international price scenarios for raw material and utility costs, while considering global ethylene and propylene prices. We obtain the optimal scheme for each case. Numerical results show that the combination of ethane steam cracking, olefin metathesis and ethylene dimerization is the most profitable configuration under low ethane price scenarios, whereas the combination of ethane and propane steam cracking together with propane dehydrogenation is the optimal solution when the propane price is on the order of ethane price.

Introduction

The shale gas revolution has led to the availability of natural gas liquids (NGLs), which represent excellent feedstock for the chemical industry. In particular, there are economic advantages on using NLGs for olefin production instead of naphtha feedstock (Siirola, 2014). For this reason, there is a general trend to modify reactive furnaces to use ethane, instead of naphtha, for ethylene production (Jenkins, 2012), even in countries that import shale gas from other countries (U.S. Energy Information Administration, 2019). These feedstock and technology changes lead to propylene yield reduction, since in the naphtha cracking process, propylene selectivity is higher than in ethane steam cracking. Furthermore, propylene demand continues to rise mainly due to polypropylene consumption (Baker, 2018). The combination of both issues increases the need for special purpose technologies for propylene production (Lavrenov et al., 2015).

Several process alternatives have been proposed to produce propylene from both petrochemical raw materials and chemical intermediates, such as methanol into olefins (Tian et al., 2015), methanol into propylene (Ali et al., 2019; Koempel and Liebner, 2007), olefin metathesis (Mol, 2004), propane dehydrogenation (Nawaz, 2015) and deep catalytic cracking (Akah and Al-Ghrami, 2015). Among these alternatives, both propane dehydrogenation and olefin metathesis are particularly interesting technologies since they can be used synergistically with ethane steam cracking to produce ethylene and propylene more efficiently.

In order to obtain an optimal design of a plant producing olefins, mathematical optimization modeling approaches can be very effective. Diaz and Bandoni (Diaz and Bandoni, 1996) formulated an MINLP problem to make discrete decisions and to optimize the operating conditions in an ethane-based ethylene plant. Lee et al. (Lee et al., 2003) postulated a superstructure based on the state-task-network (Yeomans and Grossmann, 1999a) representation, and determined the optimal scheme that minimizes the cost of separating a given mixture of olefins. Onel, Niziolek, & Floudas (Onel et al., 2016) addressed the olefin production optimal design from natural gas, via methanol. Gong & You (Gong and You, 2018) studied the optimal design of an integrated shale gas separation and chemical manufacturing process by formulating a superstructure which includes steam cracking, oxidative dehydrogenation, and catalytic dehydrogenation as alternative technologies. The simultaneous optimal design of reactor networks and separation systems has been recently addressed (Kong and Maravelias, 2020a; Ryu et al., 2020). Andersen, Diaz, & Grossmann (Andersen et al., 2013) have proposed the optimal design of integrated ethanol and gasoline supply chain. Pedrozo et al.(Pedrozo et al., 2020) have proposed an algorithm for the optimal design of ethylene plants based on multivariable piecewise linear surrogate models.

NGLs-based olefin plants including propane dehydrogenation and olefin metathesis as alternatives have not been addressed in the current literature. Moreover, we should note that several process superstructure optimizations have been reported based on shortcut models for distillation columns (Kong and Maravelias, 2020b; Narváez-García et al., 2017). However, it is worth noting that even though these approaches allow simplifying the optimization model, they can introduce significant errors when compared to rigorous mass, equilibrium, summation and heat (MESH) equation models (Dowling and Biegler, 2015), and consequently, inaccurate estimations regarding economic performance can be obtained.

In this work, we propose a superstructure embedding different process technologies for an olefin plant applying the state equipment network (SEN) representation (Yeomans and Grossmann, 1999a), and rigorous equations (MESH) to model distillation columns taking into account the relevance of the separation scheme, as well as rigorous models for compressors, turboexpanders and several process equipment units. Ethane and propane steam cracking, propane dehydrogenation, and metathesis are considered as potential technologies for ethylene and propylene production. The model is formulated as a Generalized Disjunctive Programming (GDP) problem, which is solved with a custom implementation of the Logic-based Outer Approximation algorithm. Numerical results provide useful insights for integrated olefin plants, as several price scenarios are analyzed considering raw material and utility costs from the United States, the European Union, Russia and Argentina. Furthermore, we perform heat integration for the optimal configuration of each price scenario, and we assess the potential economic improvement.