Peri-Net-Pro: the neural processes with quantified uncertainty for crack patterns

This paper develops a deep learning tool based on neural processes(NPs)called the Peri-Net-Pro,to predict the crack patterns in a moving disk and classifies them according to the classification modes with quantified uncertainties.In particular,image classification and regression studies are conducted by means of convolutional neural networks(CNNs)and NPs.First,the amount and quality of the data are enhanced by using peridynamics to theoretically compensate for the problems of the finite element method(FEM)in generating crack pattern images.Second,case studies are conducted with the prototype microelastic brittle(PMB),linear peridynamic solid(LPS),and viscoelastic solid(VES)models obtained by using the peridynamic theory.The case studies are performed to classify the images by using CNNs and determine the suitability of the PMB,LBS,and VES models.Finally,a regression analysis is performed on the crack pattern images with NPs to predict the crack patterns.The regression analysis results confirm that the variance decreases when the number of epochs increases by using the NPs.The training results gradually improve,and the variance ranges decrease to less than 0.035.The main finding of this study is that the NPs enable accurate predictions,even with missing or insufficient training data.The results demonstrate that if the context points are set to the 10th,100th,300th,and 784th,the training information is deliberately omitted for the context points of the 10th,100th,and 300th,and the predictions are different when the context points are significantly lower.However,the comparison of the results of the 100th and 784th context points shows that the predicted results are similar because of the Gaussian processes in the NPs.Therefore,if the NPs are employed for training,the missing information of the training data can be supplemented to predict the results.

Cultivated Land Changes and Their Driving Forces——A Satellite Remote Sensing Analysis in the Yellow River Delta,China

Taking Kenli County in the Yellow River Delta, China, as the study area and using digital satellite remote sensing techniques, cultivated land use changes and their corresponding driving forces were explored in this study. An interactive interpretation and a manual modification procedure were carried out to acquire cultivated land information. An overlay method based on classification results and a visual change detection method which was supported by land use maps were employed to detect the cultivated land changes. Based on the changes that were revealed and a spatial analysis between cultivated land use and related natural and socio-economic factors, the driving forces for cultivated land use changes in the study area were determined.The results showed a decrease in cultivated land in Kenli County of 5321.8 ha from 1987 to 1998, i.e.,an average annual decrement of 483.8 ha, which occurred mainly in the central paddy field region and the northeast dry land region. Adverse human activities, soil salinization and water deficiencies were the driving forces that caused these cultivated land use changes.

Numerical investigation on heat transfer in an advanced new leading edge impingement cooling configuration

It is known that the leading edge has the most critical heat transfer area of a gas turbine blade.The highest heat transfer rates on the airfoil can always be found on the stagnation region of the leading edge.In order to further improve the gas turbine thermal efficiency the development of more advanced internal cooling configurations at leading edge is very necessary.As the state of the art leading edge cooling configuration a concave channel with multi inline jets has been widely used in most of the blades.However,this kind of configuration also generates strong spent flow,which shifts the impingement off the stagnation point and weakens the impingement heat transfer.In order to solve this problem a new internal cooling configuration using double swirl chambers in gas turbine leading edge has been developed and introduced in this paper.The double swirl chambers cooling(DSC)technology is introduced by the authors and contributes a significant enhancement of heat transfer due to the generation of two anti-rotated swirls.In DSC-cooling,the reattachment of the swirl flows always occurs in the middle of the chamber,which results in a linear impingement effect.Compared with the reference standard impingement cooling configuration this new cooling system provides a much more uniform heat transfer distribution in the chamber axial direction and also provides a much higher heat transfer rate.In this study,the influences of different geometrical parameters e.g.merging ratio of two cylinder channels,the jet inlet hole configurations and radius of blunt protuberances in DSC have been investigated numerically.The results show that in the DSC cooling system the jet inlet hole configurations have large influences on the thermal performance.The rectangular inlet holes,especially those with higher aspect ratios,show much better heat transfer enhancement than the round inlet holes.However,as the price for it the total pressure drop is increased.Using blunt protuberances instead of sharp edges in the DSC cooling can improve the heat transfer enhancement and reduce the total pressure drop.