口感预测型食品物性仿生检测系统

针对目前在使用质构仪进行食品物性检测时，存在的检测数据准确性较差，操作系统功能单一，测试结果与口感 评价间无关联性分析等缺点，提出了基于六自由度并联机器人的口感预测型仿生检测系统。设计了仿生咀嚼检测平台； 利用数理统计原理和最小二乘法建立了物性检测数据与口感评价之间的数学模型，实现了仿生检测系统的口感预测功 能。实验结果表明，利用数学模型获得的口感预测值与人口感评价结果之间的误差小于9%，满足实际应用的需要，验 证了口感预测数学模型构建的正确性及口感预测功能的可行性。

Food Texture Bionic Detection System of Achieving the Function of Humans Taste Prediction

When using an existing texture analyzer to detect physical properties of food.there are some drawbacks.such as, poor accuracy of detection data, single function of operating system. lack of correlation analysis between test results and texture evaluation. For these shortcomings, this paper presented a bionic detection system with taste prediction capabilities. and the bionic chew detection platform was designed. A mathematical model was established between food texture test data and taste evaluation using mathematical statistics principle and the least squares method to achieve taste prediction function of bionic detection system. Experiments show the deviation between the result of taste prediction and the value of person''s sensory evaluation is less than 9% . it is acceptable in prcatice. The results demonstrate the rationality of method to build the equation and the feasibility to achieve the function of predicting person''s taste to food.