Research On Some Problems In The Development Of Raman Intelligent Identification Module

Raman detection,as an excellent detection method,has been widely used in the field of substance identification.At present,the rapid development of Raman detector provides the conditions for the popularization of Raman technology.In this paper,Raman technology is studied in depth on the basis of the previous research.Due to the different performance of different Raman detectors,the collected Raman spectra are quite different from the standard spectra.At the same time,many Raman detectors are not very good at identifying substances.Aiming at the above problems,this chapter designs and develops the general Raman intelligent identification module.It realizes the functions of adaptive acquisition,multi-dimensional calibration and multi-core acceleration of Raman spectra,thus improving the accuracy and speed of spectral recognition.The main work content of this paper includes the following points:(1)For samples with uncertain signal strength,a Raman automatic integral control algorithm with strong robustness based on fuzzy PID control is designed.Parameters such as Raman integral time and laser power are automatically controlled by real-time evaluation of the signal-to-noise ratio or peak strength of the Raman signal,so that for different samples,Raman signal acquisition modules with different performance can automatically obtain high quality Raman spectral data.(2)To increase the Raman spectra of versatility and accuracy of intelligent identification algorithm,designed the Raman spectra of multi-dimensional calibration algorithm,in the Raman spectra for filtering de-noising,on the basis of design based on the material of the calibration of the Raman shift calibration method and the relative intensity calibration method,improved the performance of different Raman signal acquisition module to obtain the Raman spectra of the characteristic information of difference,Thus,the accuracy of spectrum recognition is improved.(3)Based on ARM embedded system,the parallel acceleration of intelligent recognition algorithm is realized.By adopting multi-core and multi-thread parallel processing and hash table database retrieval method,the calculation speed of intelligent recognition algorithm of Raman spectra is improved,and the performance of intelligent recognition module is greatly improved.At the same time,this paper also develops a communication bridge based on the serial port communication,and realizes the communication between the front and back end programs based on the HTTP communication under the serial port.In this paper,a micro Raman intelligent identification module is developed and designed,the algorithm and control program are written,and the experimental analysis and algorithm verification are carried out.The results show that the Raman intelligent identification module can adapt to different Raman spectrum detection modules,and can provide offline and online fast Raman spectrum recognition services.