Search

In order to optimize the rehydration process of whole milk powder and explore the effect of low temperature rehydration on the flavor of reconstituted milk, headspace solid phase microextraction (HS-SPME) combined with comprehensive two-dimensional gas chromatography-mass spectrometry (GC-GC-MS) was used in this study. The percentage of main volatile compounds in the reconstituted milk was investigated under the process conditions of single-stage rehydration (55 ℃ for 30 min) and two-stage rehydration (4 ℃ for 4, 8 and 12 h after rehydration 55 ℃ for 30 min). The ROAV method and principal component analysis were used to analyze the flavor contribution of volatile substances to the reconstituted milk. The results showed that a total of 68 kinds of volatile substances divided into 9 types, such as alcohol, ether and furan were identified in the 4 groups of reconstituted milk samples. Among these substances, dodecanol, n-hexanal, n-octaldehyde, nonanal, trans-2-octenal, decanal, and trans-2-nonanal were the common characteristic flavor substances in the four groups of reconstituted milk, the contribute of aldehydes to the reconstituted milk was more than that of other volatile compounds. The single-stage hydration reconstituted milk was well differentiated from the two-stage hydration reconstituted milk by principal component analysis. The rehydration process at 4 ℃ was beneficial to reduce the smell caused by 1-octene-3-alcohol, dodecanol and trans-2-decenoaldehyde. The rehydration process at 4 ℃ for 8 h was the most beneficial to improve the beneficial flavor of the reconstituted milk, which was the best rehydration process. The results could provide a reference for the industrial production of reconstituted milk.

Dependency parsing aims to analyze the syntactic structure of sentences from the perspective of linguistics.Existing studies suggest that combining such graph-like data with graph convolutional network(GCN) can help model better understand the text semantics.However,when dealing with dependency syntactic information as adjacency matrix,these methods ignore the types of syntactic dependency tags and the word semantics related to the tags,which makes the model unable to capture the deep emotional features.To solve the preceding problem,this paper proposes a Chinese short text sentiment analysis model CDSI(context and dependency syntactic information).This model can use BiLSTM(bidirectional long short-term memory) network to extract the context semantics of the text.Moreover,a dependency-aware embedding representation method is introduced to mine the contribution weights of different dependent paths to the sentiment classification task based on the syntactic structure.Then the GCN is used to model the context and dependent syntactic information at the same time,so as to strengthen the emotional features in the text representation.Based on SWB,NLPCC2014 and SMP2020-EWEC datasets,experimental results show that CDSI can effectively integrate the semantic and structural information in sentences,which achieves good results in both the Chinese short text sentiment binary classification and multi-classification tasks.

Aiming at the inconsistency evaluation standards caused by using the gear designed by theory for simulation and using the gear produced by actual production for roll inspection certificate test，by using the NURBS surface fitting method，the real tooth surface is reconstructed and the construct of the 3D model in CATIA software is carried out. The Abaqus software is used to analyze the error of the hypoid gear transmission of the real tooth surface，and the transmission error of the real tooth surface meshing and the pure theoretical tooth surface meshing is compared，the existence of real tooth surface error and its negative influence on transmission error are verified. Changing the installation error of the gear found that the influence of the shaft angle error，the small wheel installation distance error，the offset distance error and the large wheel installation distance error on the transmission error decreased in turn. The rationality of the real tooth profile simulation model is verified by the rolling machine experiment.