Summary Converting craft beer to a powder form allows the concentration of phenolic and volatile compounds. In this study, craft beer was brewed in laboratory brewing equipment and converted to a powder using freeze‐drying and spray‐drying processes. When the freeze‐dried powder was dissolved in water, its physicochemical properties were identical to those of craft beer. Freeze‐drying effectively encapsulated the bioactive compounds in craft beer, with an encapsulation rate of 99.6%, which was 27.6% higher than that of spray‐drying. Headspace solid‐phase microextraction coupled to gas chromatography–mass spectrometry analysis revealed that the esters were effectively retained after freeze‐drying and had the highest retention of 90% for ethyl hexanoate, whereas spray‐drying resulted in the loss of most of the esters. Spray‐dried particles were all spherical, whereas lyophilized particles were irregularly shaped. The particle size ranged from 6.28 to 12.15 μm for spray‐dried powders and from 14.19 to 50.97 μm for lyophilized powders. The findings indicate that both encapsulation methods are effective for the preparation of craft beer powder. However, freeze‐drying can encapsulate the bioactive and volatile compounds in craft beer more effectively than spray‐drying and aid in the better retention of craft beer characteristics.

In order to prepare a more stable coconut oil emulsion, sodium caseinate (SC) and xanthan gum (XG) were combined as emulsifiers, and coconut oil was used as the oil phase, and the coconut oil emulsion was prepared by ultrasonic method. The average droplet sizes, Zeta-potentials, centrifugal stability and turbidity of emulsions were characterized to choose reasonable parameters concerning the ultrasonic treatment time, ultrasonic treatment power, mass fraction of oil phase, and pH values of aqueous phase, by the single factor test. The Box-Behnken response surface methodology (RSM) was used to optimize the parameter (including ultrasonic treatment power, ultrasonic treatment time, and pH values of aqueous phase) to prepare the stable emulsions. The optimal conditions were obtained as follows: Ultrasonic power 480 W, ultrasonic time 18 min, and pH7 of aqueous phase. Under this condition, the smallest droplets size of emulsions was obtained 304.5±13.2 nm. All coconut oil-based emulsions showed an extraordinary stability with heat treatment temperature of 40~90 ℃, pH6~8 and ion concentration of 0~0.5 mol/L, and the emulsion remained stable after three freeze-thaw cycles. The findings would provide a facile strategy to prepare stable coconut oil-based emulsions in the food processing.

【目的】改良传统农业生产预测方法，实现油茶产量的快速高效预测，为油茶实际生产提供参考。【方法】收集油茶栽种较为集中的湖南、江西、浙江、广西的油茶籽年度总产量、实有油茶林面积、气象等数据，选择17个气象指标作为油茶产量的影响因素，使用MATLAB软件，通过主成分分析提取出主成分，再将主成分作为BP神经网络的输入集，在传统神经网络模型基础上构建主成分分析与BP神经网络组合模型，对4个地区油茶籽单位面积年产量进行预测。BP神经网络的训练集和测试集选用1990—2018年的数据，采用2019年的数据对模型预测效果进行验证，最后应用模型对2025年油茶籽单位面积产量进行预测。【结果】对主成分有重要贡献的气象因子有日照时长、6—11月气温、3—5月降水量、平均最低气温、露点温度、平均风速、最大持续风速以及海平面气压。改进后模型迭代耗时更少，拟合度较高，对4个地区油茶产量的预测结果的平均相对误差均低于3%。应用模型预测得到2025年湖南、江西、浙江、广西的单位面积油茶干籽产量分别为0.831、0.583、0.449、0.512 t/hm~2。【结论】与传统预测模型相比，改进后的主成分分析与BP神经网络组合模型的预测效率和预测精度均有提高，在今后一段时期内4个地区的油茶产量有较好的发展趋势。