霉变小麦的电子鼻区分及其传感器阵列选择优化

Mildew Wheat Discrimination by Electronic Nose and Its Sensor Array Optimization

采用电子鼻区分不同霉变程度的扬麦23号样品,连续检测不同霉变程度小麦样品,并记录检测数据。将检测数据耦合到双稳态随机共振系统,调解系统参数诱发产生共振,依据系统输出信噪比特征值建立小麦霉变程度预测模型。为了提高电子鼻对霉变小麦样品区分效果,进行了电子鼻传感器负荷加载分析,对电子鼻传感器阵列进行了优化研究,结果表明传感器阵列优化可有效提高电子鼻检测小麦霉变程度的准确度。采用华麦6号样品构建验证实验,结果证明所建立的方法具有较好的应用意义,并具有普遍意义上的适用性。

Electronic nose technique was utilized to discriminate Yangmai23 wheat samples in different mildew status. The experiments were continuously measured,and measurement data was recorded. The data was coupled into bistable stochastic resonance. Resonance status was achieved by adjusting system parameters. Output signal-to-noise ratio eigen value was selected to build wheat mildew status predicting model. In order to improve discriminating accuracy,electronic nose sensor array loadings analysis was conducted. Electronic nose gas sensor array optimization effectively improved the detecting accuracy of mildew wheat status prediction. Huamai6 samples were used as validation experiments,and results demonstrated that the developed method presented good application value. This method also has universal applicability.