2024 AIChE Annual Meeting

(228a) Optimal Design of Hydrogen-Blended Natural Gas Pipeline Network Considering Separation System

Authors

Gu, S. - Presenter, Tufts University
Dong, Y., School of Chemical Engineering, Dalian University of Technology
Du, J., Dalian University of Technology
Important energy challenges that the world is facing now include the security of supply, environmental pollution, greenhouse gas emissions and the consequent issue of climate change. As a relatively clean fuel that can be produced from a range of sustainable sources and gasification of coal, hydrogen is expected to become an alternative energy carrier capable of addressing these challenges, as well as improving the overall energy supply security. To promote hydrogen utilization, establishing economically feasible delivery infrastructure represents a key issue that must be addressed. Rather than constructing specialized pure hydrogen pipeline, blending hydrogen into existing natural gas pipeline is considered the best option for long-distance and large-scale hydrogen transportation for the early market of hydrogen economy.

In this presentation, we propose a mixed-integer nonlinear programming model of injecting hydrogen from multiple sources into natural gas pipelines, and obtain commercial-grade hydrogen gas using separation and purification technologies downstream to satisfy the purity requirement and demand. The model mainly includes two parts, respectively for pipeline network and separation system. (1) In the pipeline network part, several adjustments are applied to the original natural gas pipeline to ensure that our design scheme can satisfy the stability and safety requirements of gas transportation, such as introducing new compressor stations that are compatible with hydrogen environment and reconsidering operation pressure of pipeline to avoid the adverse effects of hydrogen-induced material failure. (2) For the separation part, we develop a universal H2/CH4 separation system based on a stage-wise superstructure that accounts for products with a variety of concentration requirements. In the superstructure, membranes and pressure swing adsorption (PSA) are considered as potential separation units, with the possibility of recovering retentate gas of membranes and off gas of PSA. Based on information such as the pressure, composition, and flow rate of the feed streams, as well as the purity and capacity requirements of the products, the superstructure helps to select appropriate separation units and optimizes the operational parameters of each unit, thereby generating the optimal process scheme. All these separation requirements will be considered together with the pipeline network design to ensure that the important decisions, such as the hydrogen blending ratio, are optimized.

This integrated design model for hydrogen-blended natural gas pipeline network and separation system is a large-scale MINLP model. To address the computational complexity resulting from the large-scale and nonlinear nature of practical design problems, a decomposition algorithm is tailored to the proposed model, with two sub-models determining separation superstructure and pipeline configuration, respectively. Several case studies, including one based on the pipeline of West-East natural gas transmission project in China, are conducted; the results indicate that the proposed model and decomposition algorithm can provide good solutions, solving the issue that the original model is computationally intractable. Hydrogen supply from each source, mixing locations of hydrogen in the pipeline network, layout of compressor stations, and separation schemes for each sink can be obtained by this model. We will also show the quantitative estimation and analysis of the hydrogen contents that are most cost-effective in the transported gas passing through the existing infrastructures of the natural gas transmission. Therefore, this work can provide a decision-support tool for the optimal design of hydrogen-blended natural gas pipeline network and downstream separation processes adaptable to various product demand.