Prediction of corrosion inhibition efficiency of imidazoline derivatives using fuzzy artificial neural network based on quantum chemical characteristics

doi:10.16085/j.issn.1000-6613.2018-1478

Abstract

Abstract:

In order to build the complicated nonlinear relationship between quantum chemical characteristics of imidazoline derivatives and corrosion inhibition efficiency, the fuzzy artificial neural network adopting quantum chemical characteristics, including the highest occupied molecular orbital energy, the lowest unoccupied orbital energy, molecular dipole moment, single point energy, hardness, softness, nucleophilic attack index, electrophilic attack index, electron transfer parameter and the sum of static charges of non-hydrogen atoms on the imidazole ring as inputs, corrosion inhibition efficiency as outputs, was established to determine their correlation based on multi-factor variance analysis. The results revealed that there was a very significant correlation between the mentioned quantum chemical characteristics and the corrosion inhibition efficiency. With the help of above research, the obtained prediction model of Takagi-Sugeno fuzzy artificial neural network with 10-30-1 structure using momentum optimization algorithm was trained repeatedly until its mean square error less than convergence tolerance 0.005 was reached. The model output values were approximately linear with actual desired values in the training and testing stage and demonstrated superior correlation due to determination coefficient 0.9999. The good reliability of prediction model was also displayed in the validating stage. Therefore, the fuzzy artificial neural network model based on quantum chemical characteristics accurately predicted the capacities of corrosion inhibition efficiency of various imidazoline derivatives.

Key words: imidazoline derivatives, computer simulation, quantum chemistry, corrosion inhibition efficiency, neural networks

摘要：

针对咪唑啉衍生物的量子化学特征参数与缓蚀效率存在复杂非线性关系，在利用多因素方差分析判断其相关性的基础上建立以最高占据轨道能量、最低未占据轨道能量、分子偶极矩、单点能、硬度、软度、亲核进攻指数、亲电进攻指数、电子转移参数以及咪唑环上非氢原子静电荷之和等量子化学特征参数为输入，以缓蚀效率为输出的模糊人工神经网络。结果表明，咪唑啉衍生物的量子化学特征参数及其缓蚀效率之间具有非常显著的相关性，据此所创建的Takagi-Sugeno型模糊人工神经网络预测模型采用10-30-1网络结构，通过Momentum优化算法对其进行反复训练直至其均方误差小于容许收敛误差限0.005，训练、测试阶段模型输出值与期望值近似呈线性关系，决定系数为0.9999，关联度较高，验证阶段该模型亦表现出良好的可靠性。因此利用量子化学特征的模糊人工神经网络预测模型能够准确预测不同咪唑啉系列衍生物的缓蚀效率。

关键词: 咪唑啉衍生物, 计算机模拟, 量子化学, 缓蚀效率, 神经网络

CLC Number:

TE983

Zheng FAN, Zhao LIU, Xiaoyan JING, Panpan JI, Hui ZHAO, Jian KANG. Prediction of corrosion inhibition efficiency of imidazoline derivatives using fuzzy artificial neural network based on quantum chemical characteristics[J]. Chemical Industry and Engineering Progress, 2019, 38(04): 1961-1969.

范峥, 刘钊, 井晓燕, 姬盼盼, 赵辉, 康建. 利用量子化学特征的模糊人工神经网络预测咪唑啉衍生物缓蚀效率[J]. 化工进展, 2019, 38(04): 1961-1969.

Figures/Tables 11

References 22

1	薛丹, 胡敏 . 陕北气田气井腐蚀速率影响因素及规律[J]. 表面技术, 2016, 45(2): 169-174.
	XUE Dan , HU Min . Factors and rules affecting gas well corrosion rate in Shanbei gasfield[J]. Surface Technology, 2016, 45(2): 169-174.
2	REN Y M , ZHANG J , DU M , et al . The synergistic inhibition effect between imidazoline-based dissymmetric bis-quaternary ammonium salts and thiourea on Q235 steel in CO₂ corrosion process[J]. Researchon Chemical Intermediates, 2016, 42(2): 641-657.
3	郭睿, 程敏, 杨江月, 等 . 月桂酸咪唑啉硫酸酯盐缓蚀剂在A3钢表面吸附成膜行为[J]. 化工进展, 2017, 36(1): 336-342.
	GUO Rui , CHENG Min , YANG Jiangyue , et al . Adsorption behavior of lauryl imidazoline sulfuric ester inhibitor on the surface of A3 steel[J]. Chemical Industry and Engineering Progress, 2017, 36(1): 336-342.
4	吴刚, 郝宁眉, 陈银娟, 等 . 新型油酸咪唑啉缓蚀剂的合成及其性能评价[J]. 化工学报, 2013, 64(4): 1485-1492.
	WU Gang , HAO Ningmei , CHEN Yinjuan , et al . Synthesis and performance evaluation of a new oleic imidazoline corrosion inhibitor[J]. CIESC Journal, 2013, 64(4): 1485-1492.
5	ZHANG J , NIU L W , ZHU F M , et al . Theoretical and experimental studies for corrosion inhibition performance of Q235 steel by imidazoline inhibitors against CO₂ corrosion[J]. Journal of Surfactants and Detergents, 2013, 16(6): 947-956.
6	AL-AZAWI K F , AL-BAGHDADI S B , MOHAMED AYAD Z , et al . Synthesis, inhibition effects and quantum chemical studies of a novel coumarin derivative on the corrosion of mild steel in a hydrochloric acid solution[J]. Chemistry Central Journal, 2016, 10(1): 1-9.
7	ZHANG W W , MA R, LI S , et al . Electrochemical and quantum chemical studies of azoles as corrosion inhibitors for mild steel in hydrochloric acid[J]. Chemical Research in Chinese Universities, 2016, 32(5): 827-837.
8	胡松青, 胡建春, 石鑫, 等 . 咪唑啉衍生物缓蚀剂的定量构效关系及分子设计[J]. 物理化学学报, 2009, 25(12): 2524-2530.
	HU Songqing , HU Jiangchun , SHI Xin , et al . QSAR and molecular design of imidazoline derivatives as corrosion inhibitors[J]. Acta Physico-Chimica Sinica, 2009, 25(12): 2524-2530.
9	梁占伟, 陈鸿伟, 杨新, 等 . 混燃煤气气氛下NO _x 排放特性与建模预测[J]. 化工进展, 2017, 36(11): 4265-4271.
	LIANG Zhanwei , CHEN Hongxin , YANG Xin , et al . Characteristic of NO _x emissions in co-firing gases and modeling prediction[J]. Chemical Industry and Engineering Progress, 2017, 36(11): 4265-4271.
10	GHIASI M M. , BAHADORI A , ZENDEHBOUDI S . Estimation of triethylene glycol (TEG) purity in natural gas dehydration units using fuzzy neural network[J]. Journal of Natural Gas Science and Engineering, 2014，17: 26-32.
11	付柯, 谢良才, 闫雨瑗, 等 . 改进BP神经网络预测Ni/Al₂O₃催化CH₄-CO₂重整反应[J]. 化工进展, 2017, 36(7): 2393-2399.
	FU Ke , XIE Liangcai , YAN Yuyuan , et al . Predicting model of CH₄-CO₂ reforming on Ni/Al₂O₃ catalyst by improved back propagation (BP)neural network[J]. Chemical Industry and Engineering Progress, 2017, 36(7): 2393-2399.
12	PENG X H XIE S Y , YU Y H , et al . Fuzzy neural network based prediction model applied in primary component analysis[J]. Cluster Computing, 2017, 20(1): 131-140.
13	ZOU K , GE X S . Neural-network-based fuzzy logic control of a 3D rigid pendulum[J]. International Journal of Control, Automation and Systems, 2017, 15(5): 2425-2435.
14	朱鹏飞, 夏陆岳, 周猛飞, 等 . 基于混合建模技术的聚氯乙烯粒径分布预测[J]. 高校化学工程学报, 2014, 28(2): 384-389.
	ZHU Pengfei , XIA Luyue , ZHOU Mengfei , et al . Prediction for the particle size distribution of polyvinyl chloride based on hybrid modeling technique[J]. Journal of Chemical Engineering of Chinese Universities, 2014, 28(2): 384-389.
15	TSAI C C , TAI F C , CHANG Y L , et al . Adaptive predictive PID control using fuzzy wavelet neural networks for nonlinear discrete-time time-delay systems[J]. International Journal of Fuzzy Systems, 2017, 19(6): 1718-1730.
16	QI X N , LIU Y Q , GUO Q J , et al . Performance prediction of a shower cooling tower using wavelet neural network[J]. Applied Thermal Engineering, 2016, 108(7): 475-485.
17	SAGHATOLESLAMI N , VATANKHAH G H , KARIMI H , et al . Prediction of the overall sieve tray efficiency for a group of hydrocarbons, an artificial neural network approach[J]. Journal of Natural Gas Science and Engineering, 2011, 3(1): 319-325.
18	YANG H J , HU X . Wavelet neural network with improved genetic algorithm for traffic flow time series prediction[J]. Optik International Journal for Light and Electron Optics, 2016, 127(19): 8103-8110.
19	马晓丹, 刘刚, 周薇, 等 . 基于量子遗传模糊神经网络的苹果果实识别[J]. 农业机械学报, 2013, 44(12): 227-232, 251.
	Xiaodan MA , LIU Gang , ZHOU Wei , et al . Apple fruit recognition based on quantum genetic fuzzy neural network[J]. Transactions of the Chinese Society for Agricultural Machinery, 2013, 44(12): 227-232, 251.
20	向敏, 何川, 田力, 等 . 基于模糊神经网络的配电网无线通信系统性能评估方法[J]. 电力科学与技术学报, 2016, 31(2): 64-71.
	XIANG Min , HE Chuan , TIAN Li , et al . Evaluation method for wireless communication performance of distribution network based on fuzzy neural network[J]. Journal of Electric Power Science and Technology, 2016, 31(2): 64-71.
21	袁清珂, 石亚平, 张明天, 等 . 基于变论域电阻点焊模糊神经网络控制方法[J]. 焊接学报, 2010, 31(1): 25-28, 32, 114.
	YUAN Qingke , SHI Yaping , ZHANG Mingtian , et al . Fuzzy neural network control method for resistance spot welding based on variable universe[J]. Transactions of the China Welding Institution , 2010, 31(1): 25-28, 32, 114.
22	杨智琴 . 山西省地质灾害气象预警模型探讨[J]. 中国地质灾害与防治学报, 2015, 26(1): 117-121.
	YANG Zhiqin . Meteorological early warning model research geologic disaster in Shanxi province[J]. The Chinese Journal of Geological Hazard and Control, 2015, 26(1): 117-121.

序号	名称	I /μA·cm^-2	R /%
1	1-(2-氨基乙基)-2-苄基咪唑啉	57.80	87.87
2	1-(2-氨基乙基)-2-硬脂酸咪唑啉	29.26	93.86
3	1-(2-氨基乙基)-2-油酸咪唑啉	81.39	82.92
4	1-(2-氨基乙基)-2-丙基咪唑啉	108.26	77.28
5	1-(2-乙基)-2-辛基咪唑	177.59	62.73
6	1-(2-羟乙基)-2-丙基咪唑啉	185.69	61.03
7	2-甲基砜-4, 5-二氢-1H-咪唑	267.94	43.77
8	2-乙基砜-4, 5-二氢-1H-咪唑	313.92	34.12
9	2-苯基-2-咪唑啉	18.96	96.02
10	2-甲基-2-咪唑啉	35.98	92.45
11	咪唑	236.06	50.46
12	2-丙基-2-咪唑啉	170.54	64.21
13	2-甲基-2-咪唑啉	212.42	55.42
14	4-羟甲基-5-甲基咪唑	67.76	85.78
15	苯并咪唑	34.12	92.84
16	萘胺唑啉	24.25	94.91
17	苯甲唑啉	216.81	54.50
18	4, 5-二氢-2-十一烷基-1-乙醇-1H-咪唑	138.04	71.03
19	2-甲基苯并咪唑	58.37	87.75
20	2-氨基苯并咪唑	55.23	88.41

序号	名称	I /μA·cm^-2	R /%
1	1-(2-氨基乙基)-2-苄基咪唑啉	57.80	87.87
2	1-(2-氨基乙基)-2-硬脂酸咪唑啉	29.26	93.86
3	1-(2-氨基乙基)-2-油酸咪唑啉	81.39	82.92
4	1-(2-氨基乙基)-2-丙基咪唑啉	108.26	77.28
5	1-(2-乙基)-2-辛基咪唑	177.59	62.73
6	1-(2-羟乙基)-2-丙基咪唑啉	185.69	61.03
7	2-甲基砜-4, 5-二氢-1H-咪唑	267.94	43.77
8	2-乙基砜-4, 5-二氢-1H-咪唑	313.92	34.12
9	2-苯基-2-咪唑啉	18.96	96.02
10	2-甲基-2-咪唑啉	35.98	92.45
11	咪唑	236.06	50.46
12	2-丙基-2-咪唑啉	170.54	64.21
13	2-甲基-2-咪唑啉	212.42	55.42
14	4-羟甲基-5-甲基咪唑	67.76	85.78
15	苯并咪唑	34.12	92.84
16	萘胺唑啉	24.25	94.91
17	苯甲唑啉	216.81	54.50
18	4, 5-二氢-2-十一烷基-1-乙醇-1H-咪唑	138.04	71.03
19	2-甲基苯并咪唑	58.37	87.75
20	2-氨基苯并咪唑	55.23	88.41

序号	E _HOMO/eV	E _LUMO/eV	μ /debyes	φ/eV	η/eV	σ/eV^-1	F_i ⁺	F_i ^-	ΔN	Q _ring	R/%
1	-5.22867	0.03157	3.2261	-17186.0	2.6301	0.3802	-0.1401	-0.1536	0.8367	-0.8372	87.87
2	-5.59766	-0.46341	3.3270	-33141.4	2.5671	0.3895	-0.0741	-0.0084	0.7731	-0.6099	93.86
3	-5.38378	-0.48926	2.9395	-33108.0	2.4473	0.4086	-0.2015	-0.1984	0.8302	-0.5810	82.92
4	-5.17071	1.15349	3.4285	-13039.2	3.1621	0.3162	-0.1877	-0.1649	0.7893	-0.8337	77.28
5	-5.22132	1.02723	3.1723	-18386.9	3.1243	0.3201	0.0064	-0.1108	0.7847	-0.8281	62.73
6	-5.36663	1.02723	3.1517	-13579.6	3.1969	0.3128	-0.1466	-0.1577	0.7555	-0.8354	61.03
7	-5.71439	0.78777	1.9873	-18090.2	3.2511	0.3076	-0.0778	-0.1241	0.6977	-1.1807	43.77
8	-5.68256	0.76083	1.8880	-19159.8	3.2217	0.3104	-0.1081	-0.0890	0.7045	-1.1884	34.12
9	-5.59657	-0.75103	2.7308	-12472.3	2.4228	0.4128	-0.0809	-0.0204	0.7896	-1.0130	96.02
10	-5.43384	1.17336	3.2384	-7255.9	3.3036	0.3027	-0.1784	-0.1468	0.7370	-0.9167	92.45
11	-6.36148	0.65335	3.9745	-6153.6	3.5074	0.2851	-0.0554	-0.1107	0.5910	-0.7971	50.46
12	-5.44718	1.10669	3.1385	-9394.9	3.2769	0.3052	-0.1621	-0.1486	0.7369	-0.9283	64.21
13	-6.04120	0.73144	4.2307	-7223.3	3.3863	0.2953	-0.0714	-0.1255	0.6416	-0.6249	55.42
14	-6.10787	0.31946	3.8462	-10338.7	3.2137	0.3112	-0.0640	-0.0977	0.6388	-0.4985	85.78
15	-6.16692	-0.45144	3.5905	-10333.7	2.8577	0.3499	-0.1787	-0.1502	0.6458	-0.5350	92.84
16	-5.37125	0.27374	3.0434	-18223.2	2.8225	0.3543	-0.1178	-0.1125	0.7885	-0.7353	94.91
17	-5.57262	-0.14149	2.9965	-13541.7	2.7156	0.3682	0.0103	-0.1308	0.7628	-0.9170	54.50
18	-5.37153	1.02179	3.3133	-22135.7	3.1967	0.3128	-0.0946	-0.0692	0.7547	-0.8356	71.03
19	-6.00528	-0.30939	3.6124	-11403.6	2.8479	0.3511	-0.1677	-0.1544	0.6746	-0.3740	87.75
20	-5.26214	0.22477	4.6576	-11839.8	2.7435	0.3645	-0.1345	-0.1308	0.8167	-0.1756	88.41

序号	E _HOMO/eV	E _LUMO/eV	μ /debyes	φ/eV	η/eV	σ/eV^-1	F_i ⁺	F_i ^-	ΔN	Q _ring	R/%
1	-5.22867	0.03157	3.2261	-17186.0	2.6301	0.3802	-0.1401	-0.1536	0.8367	-0.8372	87.87
2	-5.59766	-0.46341	3.3270	-33141.4	2.5671	0.3895	-0.0741	-0.0084	0.7731	-0.6099	93.86
3	-5.38378	-0.48926	2.9395	-33108.0	2.4473	0.4086	-0.2015	-0.1984	0.8302	-0.5810	82.92
4	-5.17071	1.15349	3.4285	-13039.2	3.1621	0.3162	-0.1877	-0.1649	0.7893	-0.8337	77.28
5	-5.22132	1.02723	3.1723	-18386.9	3.1243	0.3201	0.0064	-0.1108	0.7847	-0.8281	62.73
6	-5.36663	1.02723	3.1517	-13579.6	3.1969	0.3128	-0.1466	-0.1577	0.7555	-0.8354	61.03
7	-5.71439	0.78777	1.9873	-18090.2	3.2511	0.3076	-0.0778	-0.1241	0.6977	-1.1807	43.77
8	-5.68256	0.76083	1.8880	-19159.8	3.2217	0.3104	-0.1081	-0.0890	0.7045	-1.1884	34.12
9	-5.59657	-0.75103	2.7308	-12472.3	2.4228	0.4128	-0.0809	-0.0204	0.7896	-1.0130	96.02
10	-5.43384	1.17336	3.2384	-7255.9	3.3036	0.3027	-0.1784	-0.1468	0.7370	-0.9167	92.45
11	-6.36148	0.65335	3.9745	-6153.6	3.5074	0.2851	-0.0554	-0.1107	0.5910	-0.7971	50.46
12	-5.44718	1.10669	3.1385	-9394.9	3.2769	0.3052	-0.1621	-0.1486	0.7369	-0.9283	64.21
13	-6.04120	0.73144	4.2307	-7223.3	3.3863	0.2953	-0.0714	-0.1255	0.6416	-0.6249	55.42
14	-6.10787	0.31946	3.8462	-10338.7	3.2137	0.3112	-0.0640	-0.0977	0.6388	-0.4985	85.78
15	-6.16692	-0.45144	3.5905	-10333.7	2.8577	0.3499	-0.1787	-0.1502	0.6458	-0.5350	92.84
16	-5.37125	0.27374	3.0434	-18223.2	2.8225	0.3543	-0.1178	-0.1125	0.7885	-0.7353	94.91
17	-5.57262	-0.14149	2.9965	-13541.7	2.7156	0.3682	0.0103	-0.1308	0.7628	-0.9170	54.50
18	-5.37153	1.02179	3.3133	-22135.7	3.1967	0.3128	-0.0946	-0.0692	0.7547	-0.8356	71.03
19	-6.00528	-0.30939	3.6124	-11403.6	2.8479	0.3511	-0.1677	-0.1544	0.6746	-0.3740	87.75
20	-5.26214	0.22477	4.6576	-11839.8	2.7435	0.3645	-0.1345	-0.1308	0.8167	-0.1756	88.41

名称	平方和		自由度	均方	F值	P值	显著性
E _HOMO	组间	6.3160×10⁴	1	6.3160×10⁴	342.09	1.3456×10^-20	显著
	组内	7.0160×10³	38	1.8463×10²
E _LUMO	组间	5.3998×10⁴	1	5.3998×10⁴	292.26	1.9587×10^-19	显著
	组内	7.0210×10³	38	1.8476×10²
μ	组间	4.9834×10⁴	1	4.9834×10⁴	269.68	7.5562×10^-19	显著
	组内	7.0220×10³	38	1.8479×10²
φ	组间	2.3637×10⁹	1	2.3637×10⁹	83.90	4.1972×10^-11	显著
	组内	1.0705×10⁹	38	2.8171×10⁷
η	组间	5.0230×10⁴	1	5.0230×10⁴	272.06	6.5237×10^-19	显著
	组内	7.0160×10³	38	1.8463×10²
σ	组间	5.4067×10⁴	1	5.4067×10⁴	292.92	1.8855×10^-19	显著
	组内	7.0140×10³	38	1.8458×10²
F_i ⁺	组间	5.4729×10⁴	1	5.4729×10⁴	296.50	1.5359×10^-19	显著
	组内	7.0140×10³	38	1.8458×10²
F_i ^-	组间	5.4743×10⁴	1	5.4743×10⁴	296.58	1.5289×10^-19	显著
	组内	7.0140×10³	38	1.8458×10²
ΔN	组间	5.3481×10⁴	1	5.3481×10⁴	289.74	2.2666×10^-19	显著
	组内	7.0140×10³	38	1.8458×10²
Q _ring	组间	5.5697×10⁴	1	5.5697×10⁴	301.72	1.1434×10^-19	显著
	组内	7.0150×10³	38	1.8460×10²