| In recent years,with the rapid development of Traditional Chinese Medicine(TCM)and its manufacturing techniques,especially since the novel Coronavirus outbreak in 2019,TCM has played an important role in the fight against COVID-19.At present,TCM has begun to go abroad and populated in the world.The production process of finished products of TCM is diversified and complex,however the consistency of TCM manufacturing process is difficult to guarantee.Turbidity is one of the key indexes in the extraction,concentration and drying of TCM pills and essential oils,as well as one of the significant indexes during the quality inspection of manufacturing process.Compared with the traditional visual turbidimetry and colorimeter,the turbidity detection methods based on transmission turbidimetry,scattering turbidimetry and ratio turbidimetry have higher measurement accuracy and reliability.Because of more spectrum signal information than direct measurement of the transmission/scattering light intensity with square signal excitation,a combination method is implemented for the research of low-cost online TCM turbidimetric sensor based on near-infrared pulse LED.The main research contents are as follows:(1)Based on Beer-Lambert law,Mie scattering theory and Rayleigh scattering theory,the turbidity detection principle of transmission method and scattering method are analyzed,and the turbidity calculation formula of ratio method is derived.Spectrum analysis algorithm is proposed to pave a way to theoretical foundation for sensor design.In addition,a near-infrared LED with a wavelength of 860 nm is used as the light source of the sensor,which cuts down the cost and power consumption and eliminates the influence of ambient light.(2)Design a low-cost and low-power sensor hardware circuit: 1.The constant current driving circuit with negative current feedback network is designed to stabilize the light source and improve the stability of the sensor.2.Photo-electric signal is converted by a low-noise FDS100 photodiode,and is preprocessed by a transimpedance amplifier and a low-pass filter to improve the Signal-to-noise ratio.3.A MCU of STM32F405 integrated with DSP kernel is adopted for fast Fourier transform(FFT)calculation and turbidity inversion,which improves the response time of the sensor.(3)By combining FFT algorithm and wavelet filtering algorithm based on DB4,turbidity inversion algorithm is realized by using spectral components.A homemade turbidity signal acquisition and analysis system is built to verify the feasibility of the algorithm.The original transmission/scattering signals of Formazin with the range of 2-290 NTU are collected and processed by wavelet transform filtering.The fundamental frequency,the third harmonic and the fifth harmonic components of the filtered signal are extracted by FFT algorithm.The three harmonic components are linearly fitted with turbidity values by transmission method,scattering method and ratio method respectively.Experimental results show that the linearity of ratio method is optimum with the minimum error of 0.4% and the maximum error of 6% in the range of 4-280 NTU.(4)A low-cost and low-power turbidity sensor is developed for the turbidity measurement of TCM essential oil by using ratio method and spectrum analysis algorithm.The measurement range of the sensor is expended from 0-300 NTU to0-1000 NTU via ratio method optimization and embedded transplantation of the algorithm.The calibration experiment of the optimized sensor shows the correlation coefficient is up to 0.9946.The validation experiment ranging from 0to 1000 NTU shows that the optimized sensor measurement accuracy is significantly improved,the maximum error is 3.768%,the minimum error is0.471%.Finally,the optimized portable sensor is used to detect the turbidity of angelica essential oil.The application indicates that the linearity is up to 0.99176,which can satisfy the online measurement requirements of essential oil extraction. |
PDF Full Text Request