Wyniki wyszukiwania - Biblioteka Nauki

1

A Deep Q-Learning Network for ship stowage planning problem

100%

Shen Y. , Zhao N. , Xia M. , Du X.

|

2017

|

tom S 3

102--109

EN

Ship stowage plan is the management connection of quae crane scheduling and yard crane scheduling. The quality of ship stowage plan affects the productivity greatly. Previous studies mainly focuses on solving stowage planning problem with online searching algorithm, efficiency of which is significantly affected by case size. In this study, a Deep Q-Learning Network (DQN) is proposed to solve ship stowage planning problem. With DQN, massive calculation and training is done in pre-training stage, while in application stage stowage plan can be made in seconds. To formulate network input, decision factors are analyzed to compose feature vector of stowage plan. States subject to constraints, available action and reward function of Q-value are designed. With these information and design, an 8-layer DQN is formulated with an evaluation function of mean square error is composed to learn stowage planning. At the end of this study, several production cases are solved with proposed DQN to validate the effectiveness and generalization ability. Result shows a good availability of DQN to solve ship stowage planning problem.

2

Analysis of uniform illumination system with imperfect Lambertian LEDs

100%

Tan J. , Yang K. , Xia M. , Yang Y.

|

2011

|

tom Vol. 41, nr 3

507--517

EN

This paper offers a novel algorithm to better design LEDs arrays for illumination systems. First, the illuminance distribution of LED arrays is studied through theoretical analysis. Second, we present the algorithm criterion and steps. And finally, we use computer to simulate and verify this method. The results show that the illumination uniformity is significantly affected by the spacing of each light in the LED array. The analysis method presented here in this thesis can be usefully applied in LED in the illumination engineering.

3

Storage allocation in automated container terminals: the upper level

100%

Xia M. , Zhao N. , Mi W.

|

2016

|

tom S 1

160--174

EN

Nowadays automation is a trend of container terminals all over the world. Although not applied in current automated container terminals, storage allocation is indispensable in conventional container terminals, and promising for automated container terminals in future. This paper seeks into the storage allocation problem in automated container terminals and proposed a two level structure for the problem. A mixed integer programming model is built for the upper level, and a modified Particle Swarm Optimization (PSO) algorithm is applied to solve the model. The applicable conditions of the model is investigated by numerical experiments, so as the performance of the algorithm in different problem scales. It is left to future research the lower level of the problem and the potential benefit of storage allocation to automated container terminals.

4

Microbiota and chemical compounds in fermented Pinelliae Rhizoma (Banxiaqu) from different areas in the Sichuan Province, China

75%

Shu B. , Ying J. , Wang T. , Xia M. , Zhao W. , You L.

|

nr 1

XX

This study focused on the microbiota and chemical compounds of the fermented Pinelliae Rhizoma produced in Longchang (LC), Zizhong (ZZ) and Xindu (XD), in Sichuan Province (China). High-throughput sequencing was used to analyze the microbiota. GC-MS and LC-MS were used to detect the compounds produced during the three different Pinelliae Rhizoma fermentation processes. The bacteria and fungi of the three fermented Pinelliae Rhizoma differed substantially, with the bacterial content mainly composed of the Bacillus genus, while the common fungi were only included in four OTUs, which belong to three species of Eurotiomycetes and Aspergillus cibarius. 51 volatile compounds were detected; they varied between LC, XD, and ZZ fermented Pinelliae Rhizoma. C10 and C15 terpenes were most frequently detected, and only curcumene and β-bisabolene were detected in the three fermented Pinelliae Rhizoma. 65 non-volatile compounds were detected by LC-MS, most were of C16, C18, C20, C21 and C22 structures. Cluster analysis showed more similarity between LC and XD fermented Pinelliae Rhizoma with regards to volatile compound content, but more similarity between the XD and ZZ fermented Pinelliae Rhizoma for non-volatiles. Moreover, no correlation between geographical distance and microflora or compounds of fermented Pinelliae Rhizoma was observed. These results showed that hundreds of compounds are produced by the natural mixed fermentation of Pinelliae Rhizoma, and may mostly relate to the microorganisms of five species.