Abstract:Due to the uneven market of rattan pepper oil, based on near-infrared spectroscopy technology, rattan pepper oil is the research object, and the research on the adulteration detection of rattan pepper oil is carried out. First, pure rattan pepper oil was used as the base oil, and the adulterated soybean oil, corn oil, and sunflower oil were prepared in proportion to obtain oil samples. The near-infrared spectroscopy was used to collect the spectral data of the oil samples to obtain the adulterated near-infrared spectral data of rattan pepper oil. The spectral data are normalized and preprocessed by Standard Normal Variation (SNV) and MultivariativeScatter Correction (MSC). And then, the feature data is processed by Competitive Adaptive Reweighting Sampling (CARS) and SuccessiveProjection Algorithm (SPA). Extraction, combining different preprocessing algorithms and feature data extraction algorithms, and establishing a prediction model of vine pepper oil adulteration through Support Vector Machine regression (SVR). The results show that the coefficient of determination (R2) of the calibration set and prediction set of the MSC-CARS-SVR model is the highest, the calibration set R2 reaches 0.756 1, and the prediction set R2 reaches 0.705 2; the root mean square error (RMSE) is the smallest, and the calibration set RMSE reaches 0.743, The prediction set RMSE reaches 0.794. In order to improve the accuracy of the model, the Whale Optimization Algorithm (WOA) and the Improved Whale Optimization Algorithm (BAS-WOA) are used to optimize the SVR model. The left and right beards are moved forward, and the objective function after the advance is calculated. If the objective function is better than the current optimal whale value, the position of the whale is replaced by the position of the beetle after the move forward, thereby realizing the improvement of the beetle operator on the whale algorithm. When WOA optimizes the SVR model, compared with the MSC-CARS-WOA-SVR model with the highest accuracy, the R2 of the calibration set can reach 0.859 1, and the R2 of the prediction set can reach 0.821 6; the RMSE of the calibration set is reduced to 0.374, and the RMSE of the prediction set is reduced to 0.495. Compared with the traditional SVR model, the accuracy and performance of the SVR model are significantly improved. When BAS-WOA optimizes the SVR model, the MSC-CARS-BAS-WOA-SVR model has the highest accuracy. The calibration set R2 is as high as 0.955 1, and the prediction set R2 is as high as 0.943 9; the calibration set RMSE is reduced to 0.054, and the prediction set RMSE is reduced to 0.081. Compared with the WOA optimization algorithm, the model accuracy and performance of the BAS-WOA optimization have been further improved. The model prediction set R2 is increased from 0.821 6 to 0.943 9, and the prediction set RMSE is reduced from 0.495 to 0.081. Whale Optimization Algorithm easily falls into local extremum and convergence rate problems when optimizing the SVR model. The improved Whale Optimization Algorithm uses the left and right baleen search of the beetle algorithm to improve the lack of the Whale Optimization Algorithm, thereby improving the global optimization ability of the algorithm. The research shows that near-infrared spectroscopy technology combined with an intelligent optimization algorithms can effectively identify the adulteration of vine pepper oil, which provides a reference for the research on the adulteration of vine pepper oil.
许素安,王家祥,刘 勇. 近红外光谱结合改进鲸鱼算法优化模型BAS-WOA-SVR检测藤椒油掺伪[J]. 光谱学与光谱分析, 2023, 43(02): 569-576.
XU Su-an, WANG Jia-xiang, LIU Yong. Detection of Adulteration of Vine Pepper Oil by Near-Infrared
Spectroscopy Combined With Improved Whale Optimization
Algorithm Model BAS-WOA-SVR. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(02): 569-576.
[1] SHI Fang-fang(史芳芳). Southwest University of Science and Technology(西南科技大学), 2020,(8): 1.
[2] LI Min, WEN Guo-ji, YAN Guang-yuan(李 敏,温国基,闫光远). Experimental Technology and Management(实验技术与管理),2021,(12): 29.
[3] WANG Wu, WANG Jian-ming, LI Ying, et al(王 武,王建明,李 颖,等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2017, 37(4): 1064.
[4] ZHU Bao-qiang, WANG Shu-hong, ZHANG Ze, et al(朱宝强,王述红,张 泽,等). Journal of Zhejiang University·Engineering Science Edition(浙江大学学报·工学版),2021, 55(12): 2275.
[5] LIU Jing-sen, ZHENG Zhi-yuan, LI Yu(刘景森,郑智远,李 煜). Control and Decision(控制与决策), 2023, 38(1): 75. (doi: 10.13195/j.kzyjc.2021.0807).
[6] HU Zhi-gang, YANG Na, LIU Wei(胡志刚,杨 娜,刘 伟). Journal of Hunan University·Natural Science Edition(湖南大学学报·自然科学版), 2021, 48(10): 131.
[7] LI An-dong, LIU Sheng(李安东,刘 升). Computer Application Research(计算机应用研究), 2022, 39(5): 1415.
[8] GUO Qi-cheng, DU Xiao-yu, ZHANG Yan-yu, et al(郭启程,杜晓玉,张延宇, 等). Computer Science(计算机科学), 2021, 48(12): 304.
[9] WU Shuai, LI Yan-jun, CAO Yu-yuan, et al(吴 帅,李艳军,曹愈远,等). Journal of Nanjing University of Aeronautics and Astronautics(南京航空航天大学学报), 2021, 53(6): 965.
[10] XIE Hai-xiang, CHEN Fang-fang, LIU Yi, et al(解海翔,陈芳芳,刘 易,等). Automation Technology and Application(自动化技术与应用), 2021, 40(11): 131.