DeepRadiation:An intelligent augmented reality platform for predicting urban energy performance just through 360 panoramic streetscape images utilizing various deep learning models

Urban energy simulation is critical for understanding and managing energy performance in cities.In this research,we design a novel framework called DeepRadiation,to enable automatic urban environmental performance prediction.By incorporating deep learning strategies,DeepRadiation predicts solar radiation on an urban scale using just panoramic streetscape images without any 3D modeling and simulation.New York City was chosen as the case study for this research.DeepRadiation is comprised of three different deep learning models organized into two stages.The first stage,named DeepRadiation modeling,serves as the framework's brain.At this stage,solar radiation analysis was performed using a Pix2Pix model,a type of conditional generative adversarial networks(GANs).After extracting GIS data and performing energy simulation analysis to prepare the dataset,the Pix2Pix model was trained on 10000 paired panoramic depth images of streetscapes with only building blocks and related panoramic images of streetscapes with only solar radiation analysis.Two GAN generator evaluation measures named qualitative evaluation and quantitative evaluation were used to validate the trained Pix2Pix model.Both demonstrated high levels of accuracy(qualitative evaluation:93%,quantitative evaluation:89%).DeepRadiation application as the DeepRadiation's sescond stage is the framework's eyes.At this stage,two convolutional neural network(CNN)models(DeepLabv3 and MiDaS)were used to perform computer vision tasks on panoramic streetscape images,such as semantic segmentation and depth estimation.The DeepRadiation application stage allows urban designers,architects,and urban policymakers to use the DeepRadiation framework and experience the final output via augmented reality.

Artificial intelligence aided design of film cooling scheme on turbine guide vane

In recent years,artificial intelligence(AI)technologies have been widely applied in many different fields including in the design,maintenance,and control of aero-engines.The air-cooled turbine vane is one of the most complex components in aero-engine design.Therefore,it is interesting to adopt the existing AI technologies in the design of the cooling passages.Given that the application of AI relies on a large amount of data,the primary task of this paper is to realize massive simulation automation in order to generate data for machine learning.It includes the parameterized three-dimensional(3-D)geometrical modeling,automatic meshing and computational fluid dynamics(CFD)batch automatic simulation of different film cooling structures through customized developments of UG,ICEM and Fluent.It is demonstrated that the trained artificial neural network(ANN)can predict the cooling effectiveness on the external surface of the turbine vane.The results also show that the design of the ANN architecture and the hyper-parameters have an impact on the prediction precision of the trained model.Using this established method,a multi-output model is constructed based on which a simple tool can be developed.It is able to make an instantaneous prediction of the temperature distribution along the vane surface once the arrangements of the film holes are adjusted.