Optimization of methanol distillation process based on chemical mechanism and industrial digital twinning modeling

doi:10.16085/j.issn.1000-6613.2023-1286

Abstract

Abstract:

In the production of the chemical industry, the data from the distributed control system (DCS) is crucial for reflecting the production status of the process. However, due to measurement errors, it often fails to meet the requirements for accurate process modeling and optimization. Conventional process modeling and optimization studies do not fully consider the deviations caused by industrial production and design data. In this work, we proposed a twin modeling framework based on industrial production data of methanol distillation and chemical mechanisms, combined with industrial experience, to guide more accurate optimization of industrial processes. We established material and energy conservation constraints and assigned weights to measurement variables based on the measurement range of instruments. Using nonlinear programming algorithms and chemical mechanism constraints, we calibrated and solved for the calibrated values of the measurement variables. We also proposed a confidence score model for process variables based on the calibrated values and industrial experience to evaluate the confidence of the measurement variables. By constructing a methanol distillation process model which was closer to industrial reality based on the calibrated measurement variables, we achieved more accurate process optimization. The twin modeling approach combining chemical mechanisms and industrial data proposed in this work has significant scientific and practical value for the construction of digital twin systems and intelligent chemical plants.

Key words: data-driven, confidence analysis, data reconciliation, process optimization, methanol distillation, digital twin

摘要：

在化学工业生产中，分布式控制系统（distributed control system，DCS）数据作为反映过程生产状况的关键信息，通常由于测量误差影响而不能满足过程精准建模与优化的要求。常规过程建模与优化研究一般无法充分考虑工业实际生产与设计数据产生的偏差。为此，本工作在结合化工机理、DCS工业数据及工业经验的基础上，从底层逻辑出发，提出一种基于甲醇精馏工业生产数据与化工机理孪生建模框架来指导工业过程更精准的优化。将仪表的测量范围作为权重赋予测量变量，使用非线性规划算法基于化工机理约束对测量变量进行校正并求解校正值。结合校正值和工业经验提出过程变量的置信分数模型，实现对测量变量的置信评价。基于校正后的测量变量构建更贴近工业实际的甲醇双效精馏过程模型并实现对其更精准的过程工艺优化。该工作提出的化工机理与工业数据孪生建模思想，对构建数字化工孪生系统和智能化工厂的数据甄别、工艺优化等过程具有重要的科学意义和实际应用价值。

关键词: 数据驱动, 置信分析, 数据协同, 过程优化, 甲醇精馏, 数字孪生

CLC Number:

TQ021.8

WANG Xiong, YANG Zhenning, LI Yue, SHEN Weifeng. Optimization of methanol distillation process based on chemical mechanism and industrial digital twinning modeling[J]. Chemical Industry and Engineering Progress, 2024, 43(1): 310-319.

王雄, 杨振宁, 李越, 申威峰. 基于化工机理与工业数据孪生建模的甲醇精馏过程优化[J]. 化工进展, 2024, 43(1): 310-319.

Figures/Tables 11

组成	质量分数/%
CH₃OH	90.54
H₂O	9.334
C₂H₅OH	0.081
C₄H₁₀O	0.045

参数	加压塔	常压塔	回收塔
进料流量/kg·h^-1	12640.99	7458.99	1647.99
进料压力/kPa	900	561	500
进料温度/℃	82.87	120.12	90.11
塔顶压力/kPa	560	57	120
塔顶温度/℃	115.42	41	68.84
塔底压力/kPa	581	132	150
塔底温度/℃	121.35	90.497	111.4
理论板数	72	70	78
进料位置	65	55	47
塔顶回流量/kg·h^-1	13473.3	10846	3536

组分	f₄质量分率		组分	f₅质量分率
组分	设计数据	模拟数据	组分	设计数据	模拟数据
CH₃OH	0.99999	0.999966	CH₃OH	0.99998	0.99992
H₂O	0.000003	0.000002	H₂O	0	0.00001
C₂H₅OH	0.000007	0.000032	C₂H₅OH	0.00002	0.00007

变量	DCS仪表	描述	汽化潜热/kJ·kg^-1
H₁	FI_4007.PV	至E0405低压蒸汽流量	2752.88
H₂	FI_4012.PV	至E0409低压蒸汽流量	2752.88

变量	DCS仪表	描述	质量等压热容/kJ·kg^-1·K^-1
$c p 3$	FI_4005.PV	T0402甲醇液流量	3.111
$c p 4$	FI_4018.PV	E0407甲醇液流量	2.918
$c p 5$	FI_4011.PV	T0403精甲醇产品流量	2.474
$c p 6$	FI_4023.PV	E0411甲醇液流量	2.627
$c p 7$	FI_4014.PV	P0407A/B含醇水流量	4.147
$c p 8$	FI_4025.PV	T0404采出杂醇油流量	5.658
$c p 9$	FI_4006.PV	T0402回流量	3.018
$c p 10$	FI_4008.PV	T0403回流量	2.396
$c p 11$	FI_4013.PV	T0404回流量	3.012

变量	DCS仪表	描述	质量等压热容/kJ·kg^-1·K^-1
$c p 3$	FI_4005.PV	T0402甲醇液流量	3.111
$c p 4$	FI_4018.PV	E0407甲醇液流量	2.918
$c p 5$	FI_4011.PV	T0403精甲醇产品流量	2.474
$c p 6$	FI_4023.PV	E0411甲醇液流量	2.627
$c p 7$	FI_4014.PV	P0407A/B含醇水流量	4.147
$c p 8$	FI_4025.PV	T0404采出杂醇油流量	5.658
$c p 9$	FI_4006.PV	T0402回流量	3.018
$c p 10$	FI_4008.PV	T0403回流量	2.396
$c p 11$	FI_4013.PV	T0404回流量	3.012

变量	描述	DCS测量值/kg·h^-1	校正值/kg·h^-1	相对误差/%
f₁	至E0405低压蒸汽流量	10038.199	9862.606	1.75
f₂	至E0409低压蒸汽流量	2134.483	1705.557	20.10
f₃	至T0402甲醇液流量	12939.601	11486.436	11.23
f₄	自E0407甲醇液流量	4961.233	5049.447	-1.78
f₅	自T0403精甲醇产品流量	5712.498	5275.642	7.65
f₆	自E0411甲醇液流量	66.816	64.845	2.95
f₇	自P0407A/B含醇水流量	465.631	833.960	-79.10^①
f₈	T0404侧线采出杂醇油流	48.562	48.655	0.19
f₉	T0402回流量	13334.351	13193.590	1.06
f₁₀	T0403回流量	13087.428	13062.239	0.19
f₁₁	T0404回流量	3889.730	4285.911	-10.19
f₁₂	至V0407A/B甲醇液流量	10014.640	9917.230	0.97

变量	描述	DCS测量值/℃	校正值/℃	相对误差/%
t₁	自V0402甲醇液温度	104.311	106.718	-2.31
t₂	T0403回流液温度	33.986	34.039	-0.16
t₃	T0404回流液温度	66.158	66.023	0.20
t₄	T0404塔釜温度	109.540	110.570	-0.94
t₅	自V0408杂醇油温度	39.284	39.196	0.22
t₆	至V0402甲醇液温度	97.365	110.971	-13.97^①
t₇	T040塔顶气温度	64.564	64.741	0.27
t₈	自E0408甲醇液温度	38.533	38.531	0.01
t₉	T0404塔顶气温度	67.805	66.557	1.84
t₁₀	自E0410温度	63.558	63.800	-0.38

参数	优化前	优化后
加压塔塔顶CH₃OH质量分率	0.999996	0.999978
常压塔塔顶CH₃OH质量分率	0.99995	0.999975
低压蒸汽流量f₁/kg·h^-1	10038.20	9862.61
低压蒸汽的潜热H₁/kJ·kg^-1	2752.88	2752.88
加压塔精甲醇流量f₄/kg·h^-1	4868.35	4935.5
常压塔精甲醇流量f₅/kg·h^-1	5042.71	5027.87
精甲醇价格 $C C H 3 O H$ /CNY·kg^-1	2.7
水电价格C_水电/CNY·kg^-1	0.27
运行时间/d	300
经济收益/10⁴CNY·a^-1	19267.1	19368.8
加压塔塔釜的热负荷/kW	7676.10	7541.83
换热器E0406热负荷/kW	4742.99	4453.62
公用工程费用/10⁴CNY·a^-1	1951.43	1917.29

参数	优化前	优化后
加压塔塔顶CH₃OH质量分率	0.999996	0.999978
常压塔塔顶CH₃OH质量分率	0.99995	0.999975
低压蒸汽流量f₁/kg·h^-1	10038.20	9862.61
低压蒸汽的潜热H₁/kJ·kg^-1	2752.88	2752.88
加压塔精甲醇流量f₄/kg·h^-1	4868.35	4935.5
常压塔精甲醇流量f₅/kg·h^-1	5042.71	5027.87
精甲醇价格 $C C H 3 O H$ /CNY·kg^-1	2.7
水电价格C_水电/CNY·kg^-1	0.27
运行时间/d	300
经济收益/10⁴CNY·a^-1	19267.1	19368.8
加压塔塔釜的热负荷/kW	7676.10	7541.83
换热器E0406热负荷/kW	4742.99	4453.62
公用工程费用/10⁴CNY·a^-1	1951.43	1917.29

References 27

1	王子宗, 王基铭, 高立兵. 石化工业软件分类及自主工业软件成熟度分析[J]. 化工进展, 2021, 40(4): 1827-1836.
	WANG Zizong, WANG Jiming, GAO Libing. Classification and maturity analysis on petrochemical industrial software[J]. Chemical Industry and Engineering Progress, 2021, 40(4): 1827-1836.
2	GRIEVES Michael W. Product lifecycle management: The new paradigm for enterprises[J]. International Journal of Product Development, 2005, 2(1/2): 71.
3	苗田, 张旭, 熊辉, 等. 数字孪生技术在产品生命周期中的应用与展望[J]. 计算机集成制造系统, 2019, 25(6): 1546-1558.
	MIAO Tian, ZHANG Xu, XIONG Hui, et al. Applications and expectation of digital twin in product lifecycle[J]. Computer Integrated Manufacturing Systems, 2019, 25(6): 1546-1558.
4	林智成. 数字孪生技术框架及其在制造业中的应用[J]. 工业控制计算机, 2020, 33(6): 129-131, 133.
	LIN Zhicheng. Digital twins technology framework and its applications in manufacturing industry[J]. Industrial Control Computer, 2020, 33(6): 129-131, 133.
5	SHANKAR Narasimhan, CORNELIUS Jordache. Data reconciliation and gross error detection[M]. Gulf Professional Publishing, 1999:142-173.
6	XIE Sen, YANG Chunhua, WANG Xiaoli, et al. A data-driven adaptive multivariate steady state detection strategy for the evaporation process of the sodium aluminate solution[J]. Journal of Process Control, 2018, 68: 145-159.
7	WANG Zan, WANG Xianpeng. Multiobjective multifactorial operation optimization for continuous annealing production process[J]. Industrial & Engineering Chemistry Research, 2019, 58(41): 19166-19178.
8	WANG Yalin, PAN Zhuofu, YUAN Xiaofeng, et al. A novel deep learning based fault diagnosis approach for chemical process with extended deep belief network[J]. ISA Transactions, 2020, 96: 457-467.
9	YUAN Xiaofeng, Chen OU, WANG Yalin, et al. Deep quality-related feature extraction for soft sensing modeling: A deep learning approach with hybrid VW-SAE[J]. Neurocomputing, 2020, 396: 375-382.
10	KUEHN D R, DAVIDSON H. Computer control Ⅱ. Mathematics of control[J]. Chemical Engineering Progress, 1961, 57(6): 44-47.
11	ZHOU Lingke, SU Hongye, CHU Jian. A new method to solve robust data reconciliation in nonlinear process[J]. Chinese Journal of Chemical Engineering, 2006, 14(3): 357-363.
12	CROWE Cameron M. Recursive identification of gross errors in linear data reconciliation[J]. AIChE Journal, 1988, 34(4): 541-550.
13	Mabel SÁNCHEZ, ROMAGNOLI José. Use of orthogonal transformations in data classification-reconciliation[J]. Computers & Chemical Engineering, 1996, 20(5): 483-493.
14	ZHANG Zhengjiang, CHUANG Yingyu, CHEN Junghui. Methodology of data reconciliation and parameter estimation for process systems with multi-operating conditions[J]. Chemometrics and Intelligent Laboratory Systems, 2014, 137: 110-119.
15	MARTINEZ PRATA Diego, SCHWAAB Marcio, LUIS LIMA Enrique, et al. Simultaneous robust data reconciliation and gross error detection through particle swarm optimization for an industrial polypropylene reactor[J]. Chemical Engineering Science, 2010, 65(17): 4943-4954.
16	CHEBEIR J, WEBB Z T, ROMAGNOLI J A. An environment for topology analysis and data reconciliation of the pre-heat train in an industrial refinery[J]. Applied Thermal Engineering, 2019, 147: 623-635.
17	VASEBI Amir, Éric POULIN, HODOUIN Daniel. Dynamic data reconciliation in mineral and metallurgical plants[J]. Annual Reviews in Control, 2012, 36(2): 235-243.
18	KHODADADI Hossein, Hooshang JAZAYERI-RAD. Applying a dual extended Kalman filter for the nonlinear state and parameter estimations of a continuous stirred tank reactor[J]. Computers & Chemical Engineering, 2011, 35(11): 2426-2436.
19	PRAKASH J, HUANG Biao, SHAH Sirish L. Recursive constrained state estimation using modified extended Kalman filter[J]. Computers & Chemical Engineering, 2014, 65: 9-17.
20	LEIBMAN M J, EDGAR T F, LASDON L S. Efficient data reconciliation and estimation for dynamic processes using nonlinear programming techniques[J]. Computers & Chemical Engineering, 1992, 16(10/11): 963-986.
21	GANDHI Mital A, MILI Lamine. Robust Kalman filter based on a generalized maximum-likelihood-type estimator[J]. IEEE Transactions on Signal Processing, 2010, 58(5): 2509-2520.
22	FUENTE M J, GUTIERREZ G, GOMEZ E, et al. Gross error management in data reconciliation[J]. IFAC-PapersOnLine, 2015, 48(8): 623-628.
23	LLANOS Claudia E, SANCHÉZ Mabel C, MARONNA Ricardo A. Robust estimators for data reconciliation[J]. Industrial & Engineering Chemistry Research, 2015, 54(18): 5096-5105.
24	XIE Sen, YANG Chunhua, YUAN Xiaofeng, et al. A novel robust data reconciliation method for industrial processes[J]. Control Engineering Practice, 2019, 83: 203-212.
25	VALLERIO Mattia, TELEN Dries, CABIANCA Lorenzo, et al. Robust multi-objective dynamic optimization of chemical processes using the Sigma Point method[J]. Chemical Engineering Science, 2016, 140: 201-216.
26	LI Yue, YANG Zhenning, DENG Xianghui, et al. A mechanism-data hybrid-driven framework for identifying dynamic characteristic of actual chemical processes[J]. Chemical Engineering Research and Design, 2023, 199: 115-129.
27	LI Yue, HU Lijuan, LI Ning, et al. A light attention-mixed-base deep learning architecture toward process multivariable modeling and knowledge discovery[J]. Computers & Chemical Engineering, 2023, 174: 108259.

[1]	ZHANG Ruijie, LIU Zhilin, WANG Junwen, ZHANG Wei, HAN Deqiu, LI Ting, ZOU Xiong. On-line dynamic simulation and optimization of water-cooled cascade refrigeration system [J]. Chemical Industry and Engineering Progress, 2023, 42(S1): 124-132.
[2]	CHEN Sen, YIN Pengyuan, YANG Zhenglu, MO Yiming, CUI Xili, SUO Xian, XING Huabin. Advances in the intelligent synthesis of functional solid materials [J]. Chemical Industry and Engineering Progress, 2023, 42(7): 3340-3348.
[3]	SUO Hansheng, JIA Mengda, SONG Guang, LIU Dongqing. Digital twin-driving force for petrochemical smart factory [J]. Chemical Industry and Engineering Progress, 2023, 42(7): 3365-3373.
[4]	WANG Zizong, SUO Hansheng, ZHAO Xueliang, YAN Yakun. Design and construction of industrial internet platform for digital twin intelligent ethylene plant [J]. Chemical Industry and Engineering Progress, 2023, 42(10): 5029-5036.
[5]	WANG Zizong, GAO Libing, SUO Hansheng. Designing petrochemical smart plant of the future: state of the art, comparison and prospects [J]. Chemical Industry and Engineering Progress, 2022, 41(7): 3387-3401.
[6]	DONG Lin, CHEN Qingbai, WANG Jianyou, LI Pengfei, WANG Jin. Research progress in brackish water electrodialysis desalination technology [J]. Chemical Industry and Engineering Progress, 2022, 41(4): 2102-2114.
[7]	GAO Libing, LYU Zhongyuan, SUO Hansheng, LIU Xiaoyu. Market analysis and development trend of petrochemical process simulation software [J]. Chemical Industry and Engineering Progress, 2021, 40(S2): 1-14.
[8]	LI Qiao, TIAN Siqi, FENG Zeming, DONG Lichun. Simulation and optimization for separation processes of methanol and trimethoxysilane azeotrope [J]. Chemical Industry and Engineering Progress, 2021, 40(5): 2431-2439.
[9]	YAO Yuman, LUO Wenjia, DAI Yiyang. Research progress of data-driven methods in fault diagnosis of chemical process [J]. Chemical Industry and Engineering Progress, 2021, 40(4): 1755-1764.
[10]	Dongliang WANG, Wenliang MENG, Yong YANG, Peng DONG, Chunqiang LI. A novel methanol distillation process combining heat pump and multi-effect process [J]. Chemical Industry and Engineering Progress, 2020, 39(9): 3550-3555.
[11]	YU Bin, ZHU Weijia. Application exploration of digital twin technology in petrochemical industry [J]. Chemical Industry and Engineering Progress, 2019, 38(s1): 278-281.
[12]	Hong JIANG, Shijian ZHANG, Jiaqiang JING, Cong ZHU. Comparison of conventional and novel high-pressure NGL recovery processes [J]. Chemical Industry and Engineering Progress, 2019, 38(06): 2581-2589.
[13]	ZHU Henan, WANG Zhipu, XING Wenlong, SHU Xinqian. Process optimization and adsorption performance of biochars prepared by co-pyrolysis of sludge and biomasses [J]. Chemical Industry and Engineering Progress, 2018, 37(S1): 199-204.
[14]	ZOU Jiangwen, YANG Xinyun, ZHANG Yongming, XU Shuying. Application of uniform design in the preparation of nano-alumina by microemulsion method [J]. Chemical Industry and Engineering Progress, 2018, 37(07): 2748-2752.
[15]	QIAN Yu, YANG Siyu, MA Donghui, CUI Peizhe. Research advances in treatment of coal gasification wastewater with high phenol and ammonia [J]. Chemical Industry and Engineering Progree, 2016, 35(06): 1884-1893.