| 刊名 | Agricultural Biotechnology |
| 作者 | Zhiyu ZHANG |
| 作者单位 | College of Information and Electrical Engineering, Heilongjiang Bayi Agricultural University, Daqing 163319, China |
| DOI | DOI:10.19759/j.cnki.2164-4993.2026.02.009 |
| 年份 | 2026 |
| 刊期 | 2 |
| 页码 | 46-48 |
| 关键词 | Near-infrared spectroscopy; Alcohol content; Genetic algorithm; Non-destructive testing |
| 摘要 | [Objectives] This study was conducted to establish a new method for rapid determination of alcohol content in beer using near-infrared spectroscopy (NIRS). [Methods] Genetic algorithm (GA) was used to select characteristic wavelengths from the original spectra, and 158 key wavelength variables were identified. The Kennard-Stone (KS) algorithm was adoped to divide 216 beer samples into a calibration set and a prediction set at a ratio of 7:3. Partial least squares regression (PLSR) and support vector regression (SVR) models were subsequently established. [Results] The GA-SVR model constructed after GA screening exhibited the best performance, with a coefficient of determination () of 0.963, a root mean square error of prediction (RMSEP) of 0.318% vol, and a residual predictive deviation (RPD) of 5.633 for the prediction set. [Conclusions] This method enables rapid, non-destructive, and high-precision prediction of alcohol content in beer, providing an effective technical approach for quality control in beer production. |
