近红外结合Si-ELM检测食醋品质指标

Measurement of quality index in vinegar using near infrared(NIR) combined with Si-ELM

为了提高近红外光谱技术检测食醋中可溶性无盐固形物含量(SSFSC)的精度和稳定性,提出采用联合区间偏最小二乘(Si-PLS)筛选光谱特征区间,再利用极限学习机(ELM)算法建立非线性回归模型,并对该方法的优越性进行系统比较;试验通过交互验证优化模型相关参数,以预测时的相关系数(Rp)和预测均方根误差(RMSEP)作为模型的评价指标。结果表明,Si-PLS结合ELM算法(Si-ELM)所建模型最佳,预测结果:Rp=0.973 9,RMSEP=1.232g/100mL。说明利用近红外光谱技术可以快速准确检测食醋中的SSF-SC,Si-ELM的应用可以适当提高该预测模型的精度。

To address the performance of NIR predicted model in measurement of soluble salt-free solid content（SSFSC） in vinegar,synergy interval partial least square（Si-PLS） was employed to select efficient spectral regions,and then extreme learning machine（ELM） algorithm was employed to develop the non-linear regression model.The relevant parameters of the model were optimized by cross-validation.The performance of the model was evaluated according to the correlation coefficient（Rp）and root mean square error of prediction（RMSEP） in prediction set.Experimental results showed that the model based on Si-PLS and ELM（i.e.Si-ELM model） was superior to others,and the optimum results were achieved as follows：Rp=0.973 9,RMSEP = 1.232 g/100 mL.The work demonstrated that NIR spectroscopy can be applied in rapid measurement of SSFSC in vinegar,and Si-PLS and ELM algorithms has the potentials in increasing the performance of NIR prediction model.