多元线性分析在储粮真菌生长预测中应用研究

真菌生长是导致粮食储藏过程中粮食损失的重要因素之一,有效控制真菌生长对于保障储粮质量安全至关重要。储粮真菌生长受储藏环境等诸多因素影响,因此建立多因素条件下储粮真菌生长预测模型具有实际指导意义。本研究以稻谷为例,将不同水分梯度的稻谷样品,置于不同温度梯度恒温箱中模拟储藏180 d,定期取样检测真菌生长数量。选取静态模拟储藏检测的1 140组数据,利用MATLAB软件建立取对数后储粮真菌孢子数与稻谷含水量、储藏温度及储藏时间的多元线性回归模型,并对该回归模型进行F检验和t检验,通过残差分析消除异常数据优化模型,方程拟合优度R~2达到0.77。由回归模型可知,真菌生长数量与储藏温度、稻谷含水量和储藏时间呈指数关系,其中稻谷含水量影响最大,其次是储藏温度和储藏时间。结合华北地区粮库实仓检测数据对模型进行实仓初步验证,以储粮安全等级作为评价标准,储粮真菌危害程度预测的正确率达83.3%。通过多元线性回归方法得到的储粮真菌生长数量预测模型为实仓储粮安全状况预测提供了一个新的方法和途径。

Application of multiple linear analyses in prediction on growth of rice storage fungi

Fungus growth is one of the important factors leading to grain loss during grain storage. Effective control of fungus growth is very important to ensure the quality and safety of stored grain. The growth of grain storage fungi is affected by many factors, such as storage environment. Therefore, it is of practical significance to establish a prediction model. In this study, rice samples with different water content were placed in different temperature incubators for 180 days, and were regularly sampled to detect the fungi. Based on the static simulated storage test data, a multivariate linear regression model was established for the storage of fungal spores, ambient temperature, rice moisture and storage days using MATLAB software. The newly developed regression model was then subjected to the F test and the t-test. Residual analysis was also performed to improve the regression model by eliminating the abnormal data. The results showed that the goodness of fit of the equation reached 0.77. The established regression model indicated that the number of fungal spores is exponentially related to ambient temperature, rice moisture content and storage days, where the rice moisture content shows the highest influence, followed by the ambient temperature, and the storage days. The model was validated with the real warehouse detection data in North China, the accuracy of the prediction was 83.3%, which can be used to predict the growth of fungi in grain storage.