基于热泵技术的醇胺法脱硫再生系统多目标优化研究

doi:10.11949/0438-1157.20250394

摘要/Abstract

摘要：

针对天然气净化厂脱硫再生工段现有工艺再生能耗高、塔顶余热利用率低问题，本研究提出了基于机械蒸汽再压缩（MVR）热泵蒸馏改造方案。通过构建吸收-再生模拟系统，创新设计塔顶蒸汽经MVR系统增压升温后供给再沸器热源的闭式循环架构，形成余热品位提升与热能自平衡的新型再生模式。基于能效分析与过程能耗评估，针对压缩机、换热器等建立操作参数优化模型，开发含敏感性分析、响应面法与非支配排序遗传算法（NSGA-Ⅱ）的协同优化框架，构建以节能增效、减排降本为核心的多目标体系，并采用三级递进策略提升能效。结果显示，改造后硫化氢流量波动≤0.13%，再生单元能耗降低39.86%，碳排放强度下降30.66%。该方案为天然气净化低碳改造提供了创新路径，兼具经济与环境效益。

关键词: 吸收, 优化, 蒸汽再压缩, 响应面法, 节能, 遗传算法

Abstract:

To tackle high regeneration energy consumption and underused low-grade waste heat in natural gas desulfurization-regeneration units, this study introduces a mechanical vapor recompression (MVR) heat pump distillation retrofit. A closed-loop system is designed where regenerator overhead vapor, pressurized via MVR, directly heats the reboiler, enabling waste heat upgrading and thermal self-balance. Using sensitivity analysis, response surface methodology, and NSGA-Ⅱ, a multi-objective optimization framework targets energy conservation, efficiency, emissions, and costs. Thermodynamic analysis identifies 320.0 kPa compressor pressure as a critical efficiency inflection point where waste heat recovery outweighs compression energy costs. Optimization results show: H₂S flow fluctuations ≤0.13%, regeneration energy consumption reduced by 39.86%, and carbon emissions down 30.66%, achieve annual operating cost savings of 546000 CNY, with a static investment payback period of 3 a. This scheme provides an innovative path for the low-carbon transformation of natural gas purification, with both economic and environmental benefits.

Key words: absorption, optimization, steam recompression, response surface methodology, energy saving, genetic algorithm

中图分类号:

TE 64

李雪雯, 王治红, 高阳, 吴明鸥, 马文皓, 谭仁敏. 基于热泵技术的醇胺法脱硫再生系统多目标优化研究[J]. 化工学报, 2025, 76(9): 4563-4577.

Xuewen LI, Zhihong WANG, Yang GAO, Ming'ou WU, Wenhao MA, Renmin TAN. Multi-objective optimization of amine-based desulfurization regeneration system integrated with heat pump technology[J]. CIESC Journal, 2025, 76(9): 4563-4577.

图/表 22

表1 原料天然气组成与关键参数

Table 1 Composition and key parameters of feed natural gas

参数		数值
组成	甲烷	0.8692
	乙烷	0.0393
	丙烷	0.0093
	异丁烷	0.0026
	正丁烷	0.0029
	异戊烷	0.0014
	正戊烷	0.0012
	正己烷	0.0018
	H₂O	0.0122
	氮气	0.0016
	H₂S	0.0172
	CO₂	0.0413
温度/℃		25
压力/kPa		600.0
摩尔流量/(kmol/h)		1250.0
质量流量/(kg/h)		23583.8
汽相分率		0.9932
质量密度/(kg/m³)		4.6751

图1 常规再生工艺

Fig.1 Conventional regeneration process

图2 MVR再生塔改造

Fig.2 Retrofitting of MVR regeneration tower

图3 压缩机出口压力-NQ敏感性分析

Fig.3 Sensitivity analysis of compressor outlet pressure-NQ

表2 目标函数及原始值

Table 2 Objective functions and baseline values

序号	目标函数	原始值
1	再生塔顶H₂S质量流量Q_m₁/(kg/d)	17505.7
2	流股换热器基本面积A/m²	—
3	碳排放水平（折合CO₂质量流量）Q_m₂/(kg/d)	17715.0

图4 再生塔进料温度敏感性分析

Fig.4 Sensitivity analysis of feed temperature in the regeneration tower

图5 再生塔塔板数敏感性分析

Fig.5 Sensitivity analysis of the number of trays in the regeneration tower

图6 再生塔进料位置敏感性分析

Fig.6 Sensitivity analysis of feed location in the regeneration tower

表3 决策变量及范围

Table 3 Decision variables and ranges

序号	决策变量	决策范围
1	再生塔进料温度T_in/℃	85~95
2	再生塔塔板数N_DC	18~22
3	再生塔进料位置F_DC	-2~+2

表4 三因素水平

Table 4 Three-factor levels

因素水平	决策变量
因素水平	T_in/℃	N_DC	F_DC
-1	85	18	-2
0	90	20	0
1	95	22	2

表5 回归模型的ANOVA摘要

Table 5 Summary of analysis of variance （ANOVA） for the regression models

项目	Q_m₁/(kg/d)		A/m²		Q_m₂/(kg/d)
	R²= 0.9768		R²= 0.9983		R²= 0.9997
	F值	p值	F值	p值	F值	p值
模型	32.81	<0.0001	459.19	<0.0001	2248.33	<0.0001
a	0.03	0.87	3485.95	<0.0001	18964.47	<0.0001
b	43.04	0.0003	41.39	0.0004	2.99	0.13
c	191.62	<0.0001	135.84	<0.0001	6.97	0.03
ab	0.04	0.85	0.16	0.70	4.65	0.07
ac	0.0002	0.99	0.69	0.43	11.83	0.01
bc	20.08	0.0030	5.03	0.06	1.50	0.26
a²	0.14	0.72	451.12	<0.0001	1230.85	<0.0001
b²	6.73	0.04	0.25	0.63	0.32	0.59
c²	31.41	0.0008	3.70	0.10	0.19	0.67
差异性	0.3800	0.7800	70.9800	0.0006	34.6600	0.0030

图7 Qm1(a)、A(b)、Qm2(c)预测值与实际值的比较

Fig.7 Model predictions vs actual of Qm1 (a), A (b), Qm2 (c)

图8 H2S质量流量Qm1的3D响应面图

Fig.8 3D response surface plot of H2S mass flow rate Qm1

图9 流股换热器基本面积A的3D响应面图

Fig.9 3D response surface plot of the basic area A of the heat exchanger between streams

图10 碳排放Qm2的3D响应面图

Fig.10 3D response surface plot of carbon emissions Qm2

图11 函数gamultiobj运行逻辑图

Fig.11 Logic diagram of the gamultiobj function

图12 函数gamultiobj语法结构图

Fig.12 Syntax structure diagram of the gamultiobj function

表6 gamultiobj的参数设置

Table 6 The parameter configuration for gamultiobj

参数	数值
种群大小	100
代数	200
停滞代数限制	200
函数容忍度	10^-100
交叉比例	0.9
迁移比例	0.03

图13 (a)Pareto最优解决策变量的分布；(b)Pareto最优解的分布

Fig.13 (a) Distribution of decision variables for Pareto optimal solutions; (b) Distribution of Pareto optimal solutions

表7 随机组平均误差分析

Table 7 Random group average error analysis

方案序号	Q_m₁/(kg/d)		A/m²		Q_m₂/(kg/d)		平均误差/%
方案序号	预测值	实际值	预测值	实际值	预测值	实际值	平均误差/%
1	17488.6	17493.7	200.9	194.3	12828.2	12939.7	1.43
2	17507.1	17506.8	238.2	236.2	12452.2	12436.3	0.33
3	17505.7	17493.2	202.5	207.5	12941.6	12751.3	1.32
4	17485.5	17483.6	218.9	219.6	12271.2	12282.8	0.14
5	17494.4	17500.2	150.2	152.8	13088.3	13091.7	0.59

图14 无热泵/优化热泵对比雷达图

Fig.14 Radar chart comparing no heat pump vs optimized heat pump

表8 经济效益分析

Table 8 Economic benefit analysis

项目	节约效益/（×10⁴ CNY/a）	新增成本/(×10⁴ CNY)					静态投资回收期/a
项目	节约效益/（×10⁴ CNY/a）	电费	热泵压缩机	流股换热器	设备安装费	合计	静态投资回收期/a
数值	54.6	86.7	22.3	32.9	14.3	156.2	3

参考文献 34

[1]	Zheng W, Yang W L, Chen J F, et al. H₂S induced in situ formation of recyclable metal sulfide-based sorbent for elemental mercury sequestration in natural gas[J]. Chemical Engineering Journal, 2024, 497: 154699.
[2]	Silva A M A, Britto de Faria A C, de Lima Ribeiro C, et al. Review of zeolite membranes for natural gas processing and treatment: state of the art and future perspectives[J]. Gas Science and Engineering, 2023, 117: 205056.
[3]	Zhao X, Ma X W, Chen B Y, et al. Challenges toward carbon neutrality in China: strategies and countermeasures[J]. Resources, Conservation and Recycling, 2022, 176: 105959.
[4]	Abdallah L, El-Shennawy T. Reducing carbon dioxide emissions from electricity sector using smart electric grid applications[J]. Journal of Engineering, 2013, 2013(1): 845051.
[5]	Bayoumy S H, El-Marsafy S M, Ahmed T S. Optimization of a saturated gas plant: meticulous simulation-based optimization—A case study[J]. Journal of Advanced Research, 2020, 22: 21-33.
[6]	Adib H, Kazerooni N, Falsafi A, et al. Prediction of sulfur content in propane and butane after gas purification on a treatment unit[J]. Oil & Gas Science and Technology, 2018, 73: 70.
[7]	Chew Y E, Putra Z A, Foo D C Y. Process simulation and optimisation for acid gas removal system in natural gas processing[J]. Journal of Natural Gas Science and Engineering, 2022, 107: 104764.
[8]	Cui X P, Lv W D, Ye H T, et al. Development of a selective sequential process for H₂S enrichment and CO₂ capture in aqueous MDEA solutions[J]. Journal of Cleaner Production, 2024, 478: 143858.
[9]	Moioli S, Giuffrida A, Romano M C, et al. Assessment of MDEA absorption process for sequential H₂S removal and CO₂ capture in air-blown IGCC plants[J]. Applied Energy, 2016, 183: 1452-1470.
[10]	Moioli S, Pellegrini L A, Romano M C, et al. Pre-combustion CO₂ removal in IGCC plant by MDEA scrubbing: modifications to the process flowsheet for energy saving[J]. Energy Procedia, 2017, 114: 2136-2145.
[11]	Dara S, AlHammadi A A, Berrouk A S, et al. Carbon footprint reduction of acid gas enrichment units in hot climates: a techno-economic simulation study[J]. Journal of Cleaner Production, 2018, 201: 974-987.
[12]	Dai Y Y, Peng Y W, Qiu Y, et al. Techno-economic analysis of a novel two-stage flashing process for acid gas removal from natural gas[J]. Energies, 2019, 12(21): 4213.
[13]	Al-Amri A, Zahid U. Design modification of acid gas cleaning units for an enhanced performance in natural gas processing[J]. Energy & Fuels, 2020, 34(2): 2545-2552.
[14]	Liu G H, Zhu L, Cao W H, et al. New technique integrating hydrate-based gas separation and chemical absorption for the sweetening of natural gas with high H₂S and CO₂ contents[J]. ACS Omega, 2021, 6(40): 26180-26190.
[15]	Tang W T, Chien C K, Ward J D. A review of energy intensification strategies for distillation processes: cyclic operation, stacking, heat pumps, side-streams, dividing walls and beyond[J]. Separation and Purification Technology, 2025, 357: 130030.
[16]	Felbab N, Patel B, El-Halwagi M M, et al. Vapor recompression for efficient distillation(1): A new synthesis perspective on standard configurations[J]. AIChE Journal, 2013, 59(8): 2977-2992.
[17]	Kishimoto A, Kansha Y, Fushimi C, et al. Exergy recuperative CO₂ gas separation in post-combustion capture[J]. Industrial & Engineering Chemistry Research, 2011, 50(17): 10128-10135.
[18]	Zhang K F, Liu Z L, Wang Y Y, et al. Flash evaporation and thermal vapor compression aided energy saving CO₂ capture systems in coal-fired power plant[J]. Energy, 2014, 66: 556-568.
[19]	Waheed M A, Oni A O, Adejuyigbe S B, et al. Performance enhancement of vapor recompression heat pump[J]. Applied Energy, 2014, 114: 69-79.
[20]	Shakerian F, Kim K H, Szulejko J E, et al. A comparative review between amines and ammonia as sorptive media for post-combustion CO₂ capture[J]. Applied Energy, 2015, 148: 10-22.
[21]	Oh S Y, Binns M, Cho H, et al. Energy minimization of MEA-based CO₂ capture process[J]. Applied Energy, 2016, 169: 353-362.
[22]	Song C F, Liu Q L, Ji N, et al. Reducing the energy consumption of membrane-cryogenic hybrid CO₂ capture by process optimization[J]. Energy, 2017, 124: 29-39.
[23]	Jang G G, Jung G S, Aye Meyer P, et al. Effective direct steam regeneration of bis-iminoguanidine solid sorbent used for carbon dioxide capture[J]. Chemical Engineering Journal, 2024, 495: 153469.
[24]	Mantingh J, Kiss A A. Enhanced process for energy efficient extraction of 1,3-butadiene from a crude C₄ cut[J]. Separation and Purification Technology, 2021, 267: 118656.
[25]	Christopher C C E, Dutta A, Farooq S, et al. Process synthesis and optimization of propylene/propane separation using vapor recompression and self-heat recuperation[J]. Industrial & Engineering Chemistry Research, 2017, 56(49): 14557-14564.
[26]	Sajid M, Nazal M K, Ihsanullah, et al. Removal of heavy metals and organic pollutants from water using dendritic polymers based adsorbents: a critical review[J]. Separation and Purification Technology, 2018, 191: 400-423.
[27]	Witek-Krowiak A, Chojnacka K, Podstawczyk D, et al. Application of response surface methodology and artificial neural network methods in modelling and optimization of biosorption process[J]. Bioresource Technology, 2014, 160: 150-160.
[28]	Bowden G D, Pichler B J, Maurer A. A design of experiments (DoE) approach accelerates the optimization of copper-mediated 18F-fluorination reactions of arylstannanes[J]. Scientific Reports, 2019, 9: 11370.
[29]	Lu H N, Dong Q, Yan S, et al. Development of flexible grouting material for cement-stabilized macadam base using response surface and genetic algorithm optimization methodologies[J]. Construction and Building Materials, 2023, 409: 133823.
[30]	Ferreira S L C, Bruns R E, Ferreira H S, et al. Box-Behnken design: an alternative for the optimization of analytical methods[J]. Analytica Chimica Acta, 2007, 597(2): 179-186.
[31]	Emmerich M, Giotis A, Özdemir M, et al. Metamodel: assisted evolution strategies[M]//Parallel Problem Solving from Nature— PPSN Ⅶ. Berlin, Heidelberg: Springer Berlin Heidelberg, 2002: 361-370.
[32]	Hao Y Y, Zhao C L, Zhang Y Q, et al. Constrained multi-objective optimization problems: methodologies, algorithms and applications[J]. Knowledge-Based Systems, 2024, 299: 111998.
[33]	Wang T, Feng J C. Multi-objective joint optimization for concurrent execution of design-construction tasks in design-build mode[J]. Automation in Construction, 2023, 156: 105078.
[34]	Zhai C H, Wang J J, Tu Y L, et al. Robust optimization of 3D printing process parameters considering process stability and production efficiency[J]. Additive Manufacturing, 2023, 71: 103588.