为了探索电子鼻对猪肉新鲜度检测的可能性，以STM32和CC2430为核心设计了新型无线电子鼻，并对5种不同新鲜度的猪肉样品进行了分析。对数据进行平滑处理后提取稳态响应值，并分别利用主成分分析和概率神经网络建立新鲜度识别模型。结果表明，主成分分析的前2个主元累计贡献率达92.79%，分类效果明显；概率神经网络模型识别率达到100%。

The detecting possibility of a novel wineless electronic nose were explored for 5 different freshness pork samples, which was designed with the STM32 and CC2430. Data processing included extracting the steady-state response after smoothing, and using principal component analysis and probabilistic neural network to establish model for freshness recognition. The results showed that contribution rate of the first two principal components total reached 92.79%. Classification effect is obvious and the identification rate of probabilistic neural network model achieved 100%.

国家自然科学基金(61203056)