| Accurately measuring various key parameters of atomized droplets in the flow field plays an important guiding and optimizing role for improving combustion efficiency,reducing pollutant emissions and fine optimization control.Droplet with inclusions formed by dispersing tiny particles into immiscible liquids are widespread,but it has received less attention due to the difficulty of characterization.The transient evaporation of a droplet chain can mathematically model the interaction between droplets in the evaporation of droplet group,and there is also a lack of high-precision experimental research in this area.The difficulty of the above-mentioned researches lies in the lack of advanced testing methods for the droplet with inclusions and transient evaporating droplet train.As an advanced optical measurement technology,rainbow refractometry can simultaneously measure thermodynamic parameters(refractive index,temperature,composition,etc.)and geometric parameters(size),and has great potential to solve the above-mentioned difficulties.At the same time,the measurement requirements for complex multiphase flow have also promoted the development of measurement technology in the direction of higher dimensions.Increasing the spatial dimension of the field to be measured greatly facilitates the evolution measurement of the key parameters of the atomized droplets,which has prompted the research of atomized droplets from 1D"line"to 2D"plane"measurement.There is currently no algorithm that can process the inversion of standard and global rainbow signals at the same time and still a lack of evaluation of the accuracy and speed of rainbow signal inversion algorithms based on different iterative methods.Regarding the issues above,this thesis carried out the two-dimensional research of rainbow refractometry,the characterization of the droplet with inclusions,the transient evaporation measurement of the droplet chain,and the research of rainbow signal inversion algorithm through the combination of theoretical analysis,simulation and experimental verification.Based on theoretical analysis,the two-dimensional rainbow refractometry is proposed,including the design of a two-dimensional rainbow measurement system with a simple and reliable configuration,a two-dimensional scattering angle calibration method and a high-precision inversion algorithm for calibration coefficients,and building a droplet generation system and two-dimensional rainbow measurement system.The measurement system has been calibrated for two-dimensional scattering angle and the pneumatic spray of deionized water with a plane field of view of 130.5 mm×81.5 mm has been tested at a room temperature of 8°C.Performing image recognition and positioning on a typical experimental two-dimensional rainbow image to obtain the plane position information of the two droplets to be measured from the rainbow signal profile.Combined with the calibration of the two-dimensional scattering angle,the online measurement of the atomized droplets in two-dimensional plane has been performed by the proposed two-dimensional rainbow refractometry.The composite errors from the algorithm and image recognition lead to an estimated maximum measurement error of 7×10-4 in refractive index and a relative error of 1.4%in size.Based on the idea of fitting inversion and extinction of second scattering order signal from rainbow to characterize liquid phase parameters(host droplet refractive index and size)and solid phase parameters(volume concentration and size of inclusions),the signal intensity ratio of second to zeroth scattering order method is proposed,eliminating the influence of random intensity,and its calculation formula is theoretically derived.The Monte Carlo-based ray tracing method simulates and analyzes the influence of various factors on the light scattering signal around the geometric rainbow angle of droplet with inclusions.Setting up single&dual wavelength standard rainbow measurement system and droplet generation system,and a series of experiments with known and unknown inclusion size are carried out.Experiments verify the feasibility and effectiveness of the rainbow refractometry combined with the extinction method to characterize the droplet containing nanoparticles.Combing phase rainbow refractometry(PRR)and high-speed microscopic shadowgraphy,the transient evaporation of micron-sized moving ethanol droplet chain jetting into the air has been quantitatively studied.A temperature-controlled droplet chain generation and high-speed microscopic shadowgraphy imaging system are built to generate ethanol droplet chain with a controllable size,speed,spacing parameter and temperature.A simple and compact improved PRR measurement system is set up to record the PRR images of the droplet chain under different excitation frequencies,flow rates and initial heating temperatures.The resolution of droplet size reduction in the range of 100~180 nm through the measurement line is achieved and the evaporation rates of ethanol droplet chains are measured to be(0.7~4.4)×10-8(m2/s).The influence of the interaction between the droplets in the droplet chain is quantified by calculating the ratio of the evaporation rate of the droplet chain to the single droplet evaporation rate predicted by the Abramzon&Sirignano model.A large amount of experimental measurement data summarizes an improved empirical correlation.Aiming at the inversion process of standard/global rainbow signals,a general inversion algorithm based on local minimum is proposed.The algorithm is based on the CAM theory with correction coefficients,establishes a nonlinear optimization objective function with inequality constraints,and uses different iterative methods to solve.For the standard rainbow signal,the Active-set method performs best in accuracy(refractive index error<2×10-4,size relative error<1.3%)and speed(0.45 s averaged time);For the global rainbow signal,Active-set method is suggested as the iteration method of an inversion that requires high inversion accuracy and does not pay attention to speed.The refractive index inversion error is less than 1×10-4,the size relative error is less than 2.0%,the averaged time is 13.2 s;on the contrary,the Brent method is recommended to use.The maximum error of the refractive index is about 3.5×10-4,and the size relative error is mostly less than 10%,but the average time is less than 1 s. |