Rare labeled data are difficult to recognize by using conventional methods in the process of radar emitter recogni-tion.To solve this problem,an optimized cooperative semi-supervised learning radar emitter recognition...Rare labeled data are difficult to recognize by using conventional methods in the process of radar emitter recogni-tion.To solve this problem,an optimized cooperative semi-supervised learning radar emitter recognition method based on a small amount of labeled data is developed.First,a small amount of labeled data are randomly sampled by using the bootstrap method,loss functions for three common deep learning net-works are improved,the uniform distribution and cross-entropy function are combined to reduce the overconfidence of softmax classification.Subsequently,the dataset obtained after sam-pling is adopted to train three improved networks so as to build the initial model.In addition,the unlabeled data are preliminarily screened through dynamic time warping(DTW)and then input into the initial model trained previously for judgment.If the judg-ment results of two or more networks are consistent,the unla-beled data are labeled and put into the labeled data set.Lastly,the three network models are input into the labeled dataset for training,and the final model is built.As revealed by the simula-tion results,the semi-supervised learning method adopted in this paper is capable of exploiting a small amount of labeled data and basically achieving the accuracy of labeled data recognition.展开更多
The broken efficiency of cell wall and the release amount of Pichia pastoris intracellular protein under different cell breaking conditions were investigated in this paper. The results showed that broken efficiency us...The broken efficiency of cell wall and the release amount of Pichia pastoris intracellular protein under different cell breaking conditions were investigated in this paper. The results showed that broken efficiency using hot alkali combined with high-pressure homogenizing method was higher than that of enzyme hydrolysis, hot alkali treatment and high-pressure homogenation, respectively. Suspended medium had little effect on the broken efficiency of yeast cell, but had significant effect on the protein release yield. The results indicated that optimal condition for intracellular proteins extraction was 30% (wet weight, w/v) of yeast cells suspend in 50 mM phosphate buffer (pH 10.0), water bathed at 60?C for 2 hours, homogenized twice at 100 MPa pressure. The broken efficiency of Pichia pastoris cell could reach 87.6% and the protein yield was 35.48 g per 100 g cells.展开更多
Motivated by the study on the spontaneous potential well-logging, this paper deals with the homogenization of boundary conditions for a class of elliptic problems with jump interface conditions.
A two-scale analysis (TSA) method for predicting the heat transfer performance of composite materials with the random distribution of same-scale grains is presented. First the representation of the materials with the ...A two-scale analysis (TSA) method for predicting the heat transfer performance of composite materials with the random distribution of same-scale grains is presented. First the representation of the materials with the random distribution is briefly described. Then the two-scale analysis formulation of heat transfer behavior of the materials with random grain distribution of small periodicity is formally derived by means of construction way for each cell. Finally the numerical result on the heat transfer parameters of composite materials is shown. The numerical result shows that TSA is effective to predict the heat transfer performance of composite materials with random grain distribution.展开更多
文摘Rare labeled data are difficult to recognize by using conventional methods in the process of radar emitter recogni-tion.To solve this problem,an optimized cooperative semi-supervised learning radar emitter recognition method based on a small amount of labeled data is developed.First,a small amount of labeled data are randomly sampled by using the bootstrap method,loss functions for three common deep learning net-works are improved,the uniform distribution and cross-entropy function are combined to reduce the overconfidence of softmax classification.Subsequently,the dataset obtained after sam-pling is adopted to train three improved networks so as to build the initial model.In addition,the unlabeled data are preliminarily screened through dynamic time warping(DTW)and then input into the initial model trained previously for judgment.If the judg-ment results of two or more networks are consistent,the unla-beled data are labeled and put into the labeled data set.Lastly,the three network models are input into the labeled dataset for training,and the final model is built.As revealed by the simula-tion results,the semi-supervised learning method adopted in this paper is capable of exploiting a small amount of labeled data and basically achieving the accuracy of labeled data recognition.
文摘The broken efficiency of cell wall and the release amount of Pichia pastoris intracellular protein under different cell breaking conditions were investigated in this paper. The results showed that broken efficiency using hot alkali combined with high-pressure homogenizing method was higher than that of enzyme hydrolysis, hot alkali treatment and high-pressure homogenation, respectively. Suspended medium had little effect on the broken efficiency of yeast cell, but had significant effect on the protein release yield. The results indicated that optimal condition for intracellular proteins extraction was 30% (wet weight, w/v) of yeast cells suspend in 50 mM phosphate buffer (pH 10.0), water bathed at 60?C for 2 hours, homogenized twice at 100 MPa pressure. The broken efficiency of Pichia pastoris cell could reach 87.6% and the protein yield was 35.48 g per 100 g cells.
文摘Motivated by the study on the spontaneous potential well-logging, this paper deals with the homogenization of boundary conditions for a class of elliptic problems with jump interface conditions.
基金This work was supported by the Special Funds for Major State Basic Research Projectthe National Natural Science Foundation of China(Grant No.19932030).
文摘A two-scale analysis (TSA) method for predicting the heat transfer performance of composite materials with the random distribution of same-scale grains is presented. First the representation of the materials with the random distribution is briefly described. Then the two-scale analysis formulation of heat transfer behavior of the materials with random grain distribution of small periodicity is formally derived by means of construction way for each cell. Finally the numerical result on the heat transfer parameters of composite materials is shown. The numerical result shows that TSA is effective to predict the heat transfer performance of composite materials with random grain distribution.
