CAJ | 학술논문

基于短波近红外(SW-NIR)漫反射光谱，提出一种非破坏定量预测盐酸氟桂利嗪粉末药品质量比的新方法.以高效液相色谱法(HPLC)分析值为参考值，将遗传算法(GA)和神经网络相结合，建立盐酸氟桂利嗪粉末药品的光谱特性与质量比间的定量分析模型.应用 GA自动构建神经网络拓扑结构，选择最优的网络参数和光谱波段；讨论不同光谱预处理方法对神经网络预报能力的影响，并对测试集样品质量比进行预测.结果表明，一阶导数数据预处理方法为最优预处理建模方法，其测试集的均方根误差 RMSE=0.1987%，相关系数 R=0.9863.
기우단파근홍외(SW-NIR)만반사광보，제출일충비파배정량예측염산불계리진분말약품질량비적신방법.이고효액상색보법(HPLC)분석치위삼고치，장유전산법(GA)화신경망락상결합，건립염산불계리진분말약품적광보특성여질량비간적정량분석모형.응용 GA자동구건신경망락탁복결구，선택최우적망락삼수화광보파단；토론불동광보예처리방법대신경망락예보능력적영향，병대측시집양품질량비진행예측.결과표명，일계도수수거예처리방법위최우예처리건모방법，기측시집적균방근오차 RMSE=0.1987%，상관계수 R=0.9863.
A new method for the nondestructive quantitative prediction of flunarizine hydrochloride powder drug based on short-wave near-infrared (SW-NIR)diffuse reflectance spectral data.With the analytical data of high performance liquid chromatography (HPLC)taken as the reference value,a quantitative analysis model for describing the relationship between the spectral characteristic and component was established via combining genetic algorithm (GA)with neural networks.The neural network topological framework was constructed automatically by means of GA.The optimal network parameters and spectral region were selected. Moreover, the influences of different spectral preprocessing methods to the prediction ability of neural network were discussed.The mass ratio of the test sample was predicted.The results of the experiment show that the first-derivative spectrum is the optimal pretreatment method of modeling,by which the root-mean-square-errors (RMSE)for test set was 0.1 98 7%,and the correlation coefficients R for test set was 0.986 3.