CityTransformer; A Transformer-based model for contaminant dispersion prediction in a realistic urban area

Asahi, Yuichi; Onodera, Naoyuki; Hasegawa, Yuta; Shimokawabe, Takashi*; Shiba, Hayato*; Idomura, Yasuhiro

Boundary-Layer Meteorology, 34 Pages, 2023/01

https://doi.org/10.1007/s10546-022-00777-8

We develop a Transformer-based deep learning model to predict the plume concentrations in the urban area under uniform flow conditions. Our model has two distinct input layers: Transformer layers for sequential data and convolutional layers in convolutional neural networks (CNNs) for image-like data. Our model can predict the plume concentration from realistically available data such as the time series monitoring data at a few observation stations and the building shapes and the source location. It is shown that the model can give reasonably accurate prediction with orders of magnitude faster than CFD simulations. It is also shown that the exactly same model can be applied to predict the source location, which also gives reasonable prediction accuracy.

Multi-resolution steady flow prediction with convolutional neural networks

Asahi, Yuichi; Hatayama, Sora*; Shimokawabe, Takashi*; Onodera, Naoyuki; Hasegawa, Yuta; Idomura, Yasuhiro

Keisan Kogaku Koenkai Rombunshu (CD-ROM), 26, 4 Pages, 2021/05

We develop a convolutional neural network model to predict the multi-resolution steady flow. Based on the state-of-the-art image-to-image translation model Pix2PixHD, our model can predict the high resolution flow field from the signed distance function. By patching the high resolution data, the memory requirements in our model is suppressed compared to Pix2PixHD.

Development of a GUI-based operation system for building a 3D point cloud classifier

Tanifuji, Yuta; Kawabata, Kuniaki; Hanari, Toshihide

Proceedings of 2019 IEEE Region Ten Conference (TENCON 2019) (Internet), p.36 - 40, 2019/10

Integrating deep learning-based object detection and optical character recognition for automatic extraction of link information from piping and instrumentation diagrams

Dong, F.*; Chen, S.*; Demachi, Kazuyuki*; Hashidate, Ryuta; Takaya, Shigeru

no journal, , 

Piping and Instrumentation Diagrams contain information about the piping and process equipment together with the instrumentation and control devices, which is essential to the design and management of Nuclear Power Plants. There are abundant complex objects on P&IDs, with imbalanced distribution of these objects and their linked information across different diagrams. Therefore, the content of P&IDs is generally extracted and analyzed manually, which is time consuming and error prone. To efficiently address these issues, we integrate state-of-the-art deep learning-based object detection and Optical Character Recognition models to automatically extract link information from P&IDs. Besides, we propose a novel image pre-processing approach using sliding windows to detect low resolution small objects. The performance of the proposed approach was experimentally evaluated, and the experimental results demonstrate it capable to extract link information from P&IDs of NPPs.

Development of a deep learning model for predicting plume concentrations in the urban area

Asahi, Yuichi; Onodera, Naoyuki; Hasegawa, Yuta; Idomura, Yasuhiro

no journal, , 

We have developed a convolutional neural network (CNN) model to predict the plume concentrations in the urban area under uniform flow condition. By combining the Transformer or Multilayer Perceptron (MLP) layers with CNN model, our model can predict the plume concentrations from the building shapes, release points of plume and time series data at observation stations. It is also shown that the exactly same model can be applied to predict the source location, which also gives reasonable prediction accuracy.

Developing data driven analysis methods for extreme scale numerical simulations

Asahi, Yuichi; Maeyama, Shinya*; Bigot, J.*; Garbet, X.*; Grandgirard, V.*; Fujii, Keisuke*; Shimokawabe, Takashi*; Watanabe, Tomohiko*; Idomura, Yasuhiro; Onodera, Naoyuki; et al.

no journal, , 

We have established an in-situ data analysis method for large scale fluid simulation data and developed deep learning based surrogate models to predict fluid simulation results. Firstly, we have developed an in-situ data processing approach, which loosely couples the MPI application and python scripts. It has been shown that this approach is simple and efficient which offers the speedup of 2.7 compared to post hoc data processing. Secondly, we have developed a deep learning model for predicting multiresolution steady flow fields. The deep learning model can give reasonably accurate predictions of simulation results with orders of magnitude faster compared to simulations.

Method for creating deep learning datasets to improve accuracy of depth images

Murayama, Masahiro*; Harazono, Yuki*; Ishii, Hirotake*; Shimoda, Hiroshi*; Taruta, Yasuyoshi; Koda, Yuya

no journal, , 

no abstracts in English