Measurement of Particle Velocity, Particle Size Distribution and Concentration in Particulate Suspension by Transmission Fluctuation Correlation Spectrometry

In coal-fired power generation industry, parameters such as particle size affect combustion efficiency. Especially in the application of two-phase flow clean energy, the parameters such as particle velocity, particle size distribution and concentration are very important, because the coal particle velocity, concentration or size range have an impact on the whole combustion process. This paper introduces an optical measurement setup based on the transmission fluctuation correlation spectrum measurement technique, which realizes the simultaneous measurement of particle velocity, particle size distribution and concentration. Compared with image method, ultrasonic spectrum method and other methods, the experimental device is simple and low-cost.

Study on the chaotic behavior of mining rock seepage system

One dimensional non-steady, non-Darcy flow of water in a rock stratum was reduced into a system described by six ordinary differential equations involving five controlling parameters. Through response computations and time series analysis, chaotic behavior in the reduced system was discussed in details. Firstly, the dynamical response of the reduced system under a set of parameters was calculated, and the power spectrum of the attractor was obtained through fast Lagrangian transformation; then the phase space was reconstructed by fixing embedding dimension to be 6 and delay time to range from 1 to 20, and the correlation dimension of the attractor was calculated based on the curves under the coordinates of logarithm of correlation integral vs. logarithm of covering radius; and lastly, the Lyapunov indices of the attractor were calculated by using Gram-Schmit's orthogonalization method. The results show that the power spectrum of the attractor is continuous; the correlation dimension of the attractor is equal to 2.36; among the Lyapunov indices, LE1, LE2, LE3 are positive, LE5, LE6 are negative, and LE4 fluctuates near zero. All the analysis indicates that there may exist chaos in the system of non-steady, non-Darcy flow.

Non-destructive determination and visualization of gel springiness of preserved eggs during pickling through hyperspectral imaging

The gel springiness of preserved egg during pickling is an important indicator of the maturity and gel quality of preserved eggs(PEs),for which there are no validated rapid means of detection.This work,hyperspectral im-aging was used to non-destructively detect the gel springiness of preserved eggs(PEs)at different maturity levels during the curing period.Firstly,principal component analysis was used to explain the characteristics of the changes in gel springiness during the curing process of PEs of different maturity levels.Secondly,based on two-dimensional correlation spectroscopy(2DCOS),the 427-843 nm band range was explored as the optimal study region for gel springiness when gel springiness was used as a perturbation factor.In turn,spectral factors representative of the change in gel springiness are stripped in this band region.Comparing three different variable selection methods(SPA、CARS and UVE),it was found that the UVE-PLSR model had the highest detection accuracy:The predicted coefficient of determination R2p and the predicted standard deviation SEp of the predicted gel springiness are 0.846 and 1.526,with a residual prediction deviation RPD of 2.35.Finally,the above optimal prediction model is applied to the pixel spectrum to calculate the springiness value of each pixel point on the hyperspectral image and is supplemented by a pseudo-color technique to visualize the prediction of the spatial distribution of the gel springiness value.The results show that using hyperspectral imaging can detect the gel springiness value of the PE during the curing process,which will provide a theoretical basis for future determination of the ripeness of the PE as well as high-throughput online sorting.

On Using Physico-Chemical Properties of Amino Acids in String Kernels for Protein Classification via Support Vector Machines

String kernels are popular tools for analyzing protein sequence data and they have been successfully applied to many computational biology problems. The traditional string kernels assume that different substrings are independent. However, substrings can be highly correlated due to their substructure relationship or common physico-chemical properties. This paper proposes two kinds of weighted spectrum kernels： The correlation spectrum kernel and the AA spectrum kernel. We evMuate their performances by predicting glycan-binding proteins of 12 glycans. The results show that the correlation spectrum kernel and the AA spectrum kernel perform significantly better than the spectrum kernel for nearly all the 12 glycans. By comparing the predictive power of AA spectrum kernels constructed by different physico-chemical properties, the authors can also identify the physico- chemical properties which contributes the most to the glycan-protein binding. The results indicate that physico-chemical properties of amino acids in proteins play an important role in the mechanism of glycamprotein binding.