
人工神经网络优化油莎豆油亚临界萃取工艺
Optimization of subcritical butane extraction for tiger nut oil based on artificial neural network coupled with PSO
为优化亚临界丁烷萃取脱皮油莎豆油工艺,采用单因素试验确定因素水平,中心复合表面设计(CCF)安排寻优试验,在此基础上分别构建了响应面(RSM)和反向传播人工神经网络(BP-ANN)模型,运用粒子群算法(PSO)对BP-ANN模型进行优化,并对RSM和PSO-BP-ANN模型的寻优结果进行了比较。结果表明,RSM模型优化的萃取条件为:料液比(脱皮油莎豆∶丁烷)1∶10.36 g/mL、萃取时间45 min、萃取温度30℃、坯料厚度0.5 mm;PSO-BP-ANN模型优化的萃取条件为:料液比1∶10.67 g/mL、萃取时间40.10 min、萃取温度34℃、轧坯厚度0.5 mm。在最佳条件下,RSM模型预测提取率为91.63%,验证值为94.27%,相对误差2.56%;PSO-BP-ANN模型预测值为95.58%,验证值为95.14%,相对误差0.46%。采用人工神经网络耦合粒子群算法(PSO-BP-ANN)优化油莎豆油亚临界萃取工艺,具有提取率高、相对误差小等优势。本研究可为亚临界萃取技术在油莎豆油高效制取中应用提供参考。
In order to optimize the subcritical butane extraction process for dehulled tiger nut oil, single factor experiment was taken to determine the levels of the factor, central composite face-centered design (CCF) was used to optimize the subcritical extraction conditions, based on which response surface methodology (RSM) and back propagation artificial neural network (BP-ANN) models were constructed, respectively. The BP-ANN was optimized by particle swarm optimization (PSO), and the optimization results of the RSM model and PSO-BP-ANN model were compared. The optimal extraction conditions optimized by RSM and PSO-BP-ANN models were as follows: solid-liquid ratio (dehulled tiger nut: butane) was 1:10.36 g/mL, incubation time for 45 min, extraction temperature was 30℃, the rolling thickness was 0.5 mm; the solid-liquid ratio was 1:10.67 g/mL, the extraction time was 40.10 min, the extraction temperature was 34℃, and the thickness of the rolled billet was 0.5 mm. Under the optimal conditions, the predicted extraction rate of the RSM model was 91.63%, the experimental result was 94.27%, and the relative error was 2.56%. The prediction value of the PSO-BP-ANN model was 95.58%, the validation value was 95.14%, and the relative error was 0.46%. The artificial neural network coupled particle swam optimization (PSO-BP-ANN) was used to optimize the subcritical extraction process of tiger nut oil, which had advantages of high extraction rate and small error. This study can provide a reference for the application of subcritical extraction technology in the efficient production of tiger nut oil.
反向传播人工神经网络 / 粒子群优化算法 / 亚临界丁烷萃取 / 脱皮油莎豆 / 工艺优化 {{custom_keyword}} /
back propagation artificial neural network / particle swarm optimization / subcritical butane extraction / dehulled tiger nut / process optimization {{custom_keyword}} /
表1 CCF因素与水平Table 1 CCF factors and levels |
编码 Code | 因子 Factor | 水平 Level | ||
---|---|---|---|---|
-1 | 0 | 1 | ||
A | 料液比 Solid-liquid ratio /(g/mL) | 1∶8 | 1∶10 | 1∶12 |
B | 萃取时间 Extraction time/min | 25 | 35 | 45 |
C | 萃取温度 Extraction temperature/℃ | 30 | 40 | 50 |
D | 轧坯厚度 Thickness of rolled billet/mm | 0.5 | 1 | 1.5 |
表2 CCF试验结果Table 2 CCF arrangement and results |
试验号 Test number | A | B | C | D | 提取率 Extraction rate/% |
---|---|---|---|---|---|
1 | 8 | 25 | 30 | 0.5 | 88.99 |
2 | 12 | 25 | 30 | 0.5 | 91.76 |
3 | 8 | 45 | 30 | 0.5 | 89.19 |
4 | 12 | 45 | 30 | 0.5 | 91.21 |
5 | 8 | 25 | 50 | 0.5 | 86.18 |
6 | 12 | 25 | 50 | 0.5 | 89.98 |
7 | 8 | 45 | 50 | 0.5 | 87.55 |
8 | 12 | 45 | 50 | 0.5 | 89.18 |
9 | 8 | 25 | 30 | 1.5 | 82.79 |
10 | 12 | 25 | 30 | 1.5 | 81.36 |
11 | 8 | 45 | 30 | 1.5 | 84.09 |
12 | 12 | 45 | 30 | 1.5 | 85.99 |
13 | 8 | 25 | 50 | 1.5 | 78.15 |
14 | 12 | 25 | 50 | 1.5 | 83.76 |
15 | 8 | 45 | 50 | 1.5 | 81.60 |
16 | 12 | 45 | 50 | 1.5 | 87.56 |
17 | 10 | 35 | 40 | 1 | 87.80 |
18 | 10 | 35 | 40 | 1 | 88.82 |
19 | 10 | 35 | 40 | 1 | 86.50 |
20 | 10 | 35 | 40 | 1 | 87.84 |
21 | 8 | 35 | 40 | 1 | 85.76 |
22 | 12 | 35 | 40 | 1 | 87.46 |
23 | 10 | 25 | 40 | 1 | 85.78 |
24 | 10 | 45 | 40 | 1 | 89.14 |
25 | 10 | 35 | 30 | 1 | 89.22 |
26 | 10 | 35 | 50 | 1 | 86.96 |
27 | 10 | 35 | 40 | 0.5 | 89.56 |
28 | 10 | 35 | 40 | 1.5 | 84.31 |
29 | 10 | 35 | 40 | 1 | 87.70 |
30 | 10 | 35 | 40 | 1 | 90.54 |
表3 回归模型的方差分析Table 3 ANOVA of the regression model |
来源 Source | 自由度 df | Adj SS | Adj MS | F值 F value | P值 P value |
---|---|---|---|---|---|
模型 Model | 15 | 272.711 | 18.181 | 11.13 | 0 |
A | 1 | 33.392 | 33.392 | 20.45 | 0** |
B | 1 | 17.032 | 17.032 | 10.43 | 0.006** |
C | 1 | 8.326 | 8.326 | 5.1 | 0.04** |
D | 1 | 171.183 | 171.183 | 104.84 | 0** |
A2 | 1 | 8.732 | 8.732 | 5.35 | 0.036** |
B2 | 1 | 0.129 | 0.129 | 0.08 | 0.783 |
C2 | 1 | 0.238 | 0.238 | 0.15 | 0.708 |
D2 | 1 | 0.336 | 0.336 | 0.21 | 0.657 |
AB | 1 | 0.037 | 0.037 | 0.02 | 0.883 |
AC | 1 | 8.623 | 8.623 | 5.28 | 0.037** |
AD | 1 | 0.208 | 0.208 | 0.13 | 0.726 |
BC | 1 | 0.314 | 0.314 | 0.19 | 0.668 |
BD | 1 | 10.454 | 10.454 | 6.4 | 0.024** |
CD | 1 | 1.634 | 1.634 | 1 | 0.334 |
误差 Error | 14 | 22.86 | 1.633 | ||
失拟 Lack of fit | 10 | 15.778 | 1.578 | 0.89 | 0.601 |
纯误差 Pure error | 4 | 7.081 | 1.77 | ||
合计 Total | 29 | 295.57 |
图4 PSO优化BP神经网络模型均方误差效果图Fig. 4 Mean square error effect graph of BP neural network model optimized by PSO |
图5 PSO优化BP-ANN模拟仿真效果Fig. 5 Simulation effect of BP neural network model optimized by PSO |
表4 响应面、PSO-BP-ANN和BP-ANN优化亚临界丁烷萃取脱皮油莎豆油工艺验证试验结果(n=3)Table 4 Validation test results of RSM, PSO-BP-ANN and BP-ANN optimized subcritical butane extraction of dehulled tiger nut oil process (n=3) |
因素 Factor | RSM | PSO-BP-ANN | BP-ANN |
---|---|---|---|
料液比 Ratio of solvent to material/(g/mL) | 1∶10.36 | 1∶10.67 | 1∶12 |
萃取时间 Extraction time/min | 45 | 40.10 | 45 |
萃取温度 Extraction temperature/℃ | 30 | 34 | 31 |
轧坯厚度 Thickness of rolled billet/mm | 0.5 | 0.5 | 0.5 |
提取效率 Extraction rate/% | 94.52±0.83 | 95.14±0.71 | 93.18±0.53 |
模型预测值 Model predicted value/% | 91.86 | 95.58 | 92.19 |
1 |
郭婷婷, 万楚筠, 黄凤洪, 等. 油莎豆主要营养成分及生理功能研究进展[J]. 中国油料作物学报, 2021, 43(6): 1174-1180. DOI: 10.19802/j.issn.1007-9084.2020145 .
{{custom_citation.content}}
{{custom_citation.annotation}}
|
2 |
李志军, 王绛, 刘和, 等. 油莎豆产业发展现状及建议[J]. 现代农业科技, 2022(8): 225-231. DOI: 10.3969/j.issn.1007-5739.2022.08.072 .
{{custom_citation.content}}
{{custom_citation.annotation}}
|
3 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
4 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
5 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
6 |
王志成, 李双寿, 梁雄, 等. 中国油莎豆产业发展现状与前景展望[J]. 科技和产业, 2022, 22(1): 62-67. DOI: 10.3969/j.issn.1671-1807.2022.01.009 .
{{custom_citation.content}}
{{custom_citation.annotation}}
|
7 |
郭婷婷, 万楚筠, 黄凤洪, 等. 亚临界流体萃取油脂及微量成分研究进展[J]. 中国油料作物学报, 2020, 42(1): 154-160. DOI: 10.19802/j.issn.1007-9084.2019176 .
{{custom_citation.content}}
{{custom_citation.annotation}}
|
8 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
9 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
10 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
11 |
马建春, 马灶亮, 张昊亮, 等. 补阳还五汤提取工艺的响应面法和人工神经网络模型优化[J]. 时珍国医国药, 2019, 30(2): 337-340.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
12 |
邹立飞, 郑鹏. 人工神经网络和响应面法优化薏苡仁酒发酵条件[J]. 中国酿造, 2021, 40(1): 142-147. DOI: 10.11882/j.issn.0254-5071.2021.01.027 .
{{custom_citation.content}}
{{custom_citation.annotation}}
|
13 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
14 |
张浩科, 高静, 赵智龙, 等. 人工神经网络结合粒子群优化算法优化不同产地广藿香挥发油提取工艺[J]. 中国中医药信息杂志, 2021, 28(5): 83-87. DOI: 10.19879/j.cnki.1005-5304.202007035 .
{{custom_citation.content}}
{{custom_citation.annotation}}
|
15 |
彭文珍. 基于CCOS小磨头抛光石英玻璃表面质量的评价与预测模型研究[D]. 天津: 天津大学, 2014.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
16 |
曹曼,刘艳,顾雪平.基于粒子群优化算法的神经网络在暂态稳定评估中的应用[C]// 中国高等学校电力系统及其自动化专业第二十二届学术会论文摘要集. 2006: 2039-2042.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
17 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
18 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
19 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
20 |
张志红, 何桢, 郭伟. 在响应曲面方法中三类中心复合设计的比较研究[J]. 沈阳航空工业学院学报, 2007, 24(1): 87-91. DOI: 10.3969/j.issn.2095-1248.2007.01.026 .
{{custom_citation.content}}
{{custom_citation.annotation}}
|
21 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
22 |
万楚筠, 黄凤洪, 张明, 等. 亚临界流体萃取胡麻籽低温压榨饼中油脂[J]. 农业工程学报, 2014, 30(21): 324-331. DOI: 10.3969/j.issn.1002-6819.2014.21.039 .
{{custom_citation.content}}
{{custom_citation.annotation}}
|
23 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
24 |
万楚筠, 张明, 黄凤洪, 等. 响应面优化亚临界萃取低温压榨菜籽饼中油脂的工艺及其品质评价[J]. 中国油料作物学报, 2018, 40(1): 146-153. DOI: 10.7505/j.issn.1007-9084.2018.01.019 .
{{custom_citation.content}}
{{custom_citation.annotation}}
|
25 |
丁志平, 乔延江. 不同粒径黄连粉体的吸湿性实验研究[J]. 中国实验方剂学杂志, 2004, 10(3): 5-7. DOI: 10.13422/j.cnki.syfjx.2004.03.003 .
{{custom_citation.content}}
{{custom_citation.annotation}}
|
26 |
弓志青, 陈相艳, 程安玮, 等. 喷雾干燥杨梅粉玻璃化转变温度和结块特性研究[J]. 食品科学技术学报, 2013, 31(2): 43-46, 66.
{{custom_citation.content}}
{{custom_citation.annotation}}
|
27 |
连四超, 刘玉兰, 陈璐, 等. 油莎豆油亚临界丁烷萃取条件优化及产品质量研究[J]. 中国油脂, 2022, 47(4): 9-14. DOI: 10.19902/j.cnki.zgyz.1003-7969.210700 .
{{custom_citation.content}}
{{custom_citation.annotation}}
|
28 |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
{{custom_ref.label}} |
{{custom_citation.content}}
{{custom_citation.annotation}}
|
/
〈 |
|
〉 |