Application Of Spectral Droplet Analysis In Liquid Safety

With the current changes in the international environment , citizens in public places of personal safety issues become increasingly prominent. At present, the security equipment in the traffic field is mainly based on the X-ray imaging principle to obtain the information of the passengers carrying the goods, and most of the security channel using open cover test and odor detection method to detect liquid carriers, so the detection of liquid dangerous substances is the focus of the field of security technology. In this paper, the methods of fiber droplet analysis, capacitance droplet analysis and near infrared spectroscopy were used to study the method of identifying dangerous liquid. The feasibility of each scheme was analyzed by experimental analysis.In this paper, the spectral droplet analysis system is constructed, the optical fiber signal detection circuit and the capacitance signal detection circuit are designed.Through the combination of the hardware circuit and the acquisition software, the experimental platform of the droplet detection experiment is completed. At the same time, the absorbance information of liquid samples was collected by the spectrometer.Using the experimental platform to do preliminary experiments, some typical liquid samples for testing and analysis to verify the availability of the system. The three-dimensional fingerprints of the samples were obtained by using the simulation technology to fuse the optical fiber, capacitance and spectral signals of the samples,and the characteristic parameters were extracted from them and the recognition ability of these parameters was tested.Based on the collected spectral data, the method of rapid identification of flammable liquids by near infrared spectroscopy was studied. The near infrared absorption spectra of typical flammable and nonflammable liquid samples were collected and the data were smoothed and baseline corrected. When the discriminant prediction model is established, the principal component analysis is carried out on the spectral data of the liquid sample, and the characteristic wavelength point is selected to extract the characteristic parameters of the sample. The discriminant prediction model is established by distance discrimination,Bayes discriminant and Fisher discriminant. The results show that the Fisher discriminant analysis method is accurate and fast.At the same time,after denoising and normalizing to the information data- of fiber, capacitor droplet fingerprint, the characteristic parameters of the liquid samples are extracted with waveform analysis method to extract the characteristic parameters,and then the limit learning machine algorithm is used to build classification and identification model. The feasibility of identifying the dangerous liquid of the optical fiber capacitor droplet system is verified.