Research Of The Detection Technology For Mineral Oil In Water Based On Fluorescence Mechanisrn

Oil in water pollution degree and adverse effects has become a worldwide disaster. Inorder to maintain the ecological balance of environment, protect people’s health andharmony, promote the sustainable development of society, it is particularly important torapidly detect and analyse the pollution of oil in water, timely detect the pollutionaccidents, accurately assess the degree of pollution. Using the fluorescence characteristicsof mineral oil, joining in the high performance DSP (Digital Signal Processor) chip, adesign scheme of a real-time measurement system for mineral oil concentration in water ispresented. At the same time, wavelet entropy threshold algorithm is introduced to thefluorescence signal denoising. Therefore, the detecting accuracy of the system is furtherimproved.With the fluorescence emission mechanism as a beginning, this paper studies thefluorescence detection principle of the oil. By linear fitting of the kerosene experimental,Lambert–Beer law is proved, and the feasibility of fluorescence method to measure oilconcentration in water is discussed in-depthly. According to the existing data in petroleumexcitation spectra and fluorescence spectra and the data from experiment, optimalexcitation wavelength and fluorescence emission wavelength is determined in the mineraloil measurement in this paper.Based on the optimal excitation wavelength and emission wavelength, light-emittingdiode and photoelectric diode are used as excitation light source and photoelectric detectorrespectively. Together with the optical fiber transmission module, fluorescence excitationlight path transmission system is built. What’s more, the excitation light sourcesynchronization driving circuit is designed by using PWM pulse width modulationwaveform. It makes up for the instrument defects of large volume and high energyconsumption.Combined with the superior performance of the DSP chip TMS320F2812, the higherintegration chip peripheral circuit, data acquisition module and serial communicationmodule hardware circuit is designed in the paper. Finally, the acquisition function, processing function, memory function and communication function for fluorescencesignal is realized by the main control chip. It conformes to the miniaturization,intelligentize, integration development trend in detection instrument.Finally, according to the wavelet transform and the information entropy theory, amore reasonable method with wavelet entropy threshold denoising to calculate thealgorithm is put forward. The optimal wavelet denoising parameters of fluorescence signalis determined by experiments. Through the simulation compare between traditionalwavelet denoising algorithm and wavelet entropy analysis denoising algorithm, aquantitative analysis and comparison of data is given. The results show that, waveletentropy threshold denoising analysis algorithm could be better to keep the signal edge anddetail information, and has a better effect on the separation of signal and noise influorescence signal. It is more conducive to improving the accuracy of the measurementsystem.