• Home
  • Search
  • Rooftop wind field reconstruction using sparse sensors: From deterministic to generative learning methods
  • https://doi.org/10.1016/j.buildenv.2026.114480Copy DOI Icon

Rooftop wind field reconstruction using sparse sensors: From deterministic to generative learning methods

Show More
  • Abstract
  • Literature Map
  • References
  • Similar Papers
Abstract

Real-time rooftop wind-speed distribution is essential for safe operation of drones or urban air mobilities, wind control systems, and rooftop utilization. However, rooftop flows exhibit strong nonlinearity, separation, and cross-direction variability, which challenge flow field reconstruction from sparse sensors. This study develops a learning-from-observation framework using wind-tunnel experiment data obtained by Particle Image Velocimetry (PIV) and benchmarks Kriging interpolation against three deep learning models—UNet, Vision Transformer Autoencoder (ViTAE), and Conditional Wasserstein GAN (CWGAN). We evaluate two training strategies—single wind-direction training (SDT) and mixed wind-direction training (MDT)—across sensor densities of 5–30, test robustness under 1-grid sensor position perturbations, and optimize sensor placement via Proper Orthogonal Decomposition with QR decomposition. Results indicate the potential of deep learning methods to reconstruct rooftop wind fields from sparse sensor data. Compared with traditional Kriging interpolation, the deep learning models achieved improvements of up to 32.7% in SSIM, 24.2% in FAC2, and 27.8% in NMSE. In addition, training with mixed wind directions was found to be essential for achieving better model performance, resulting in further improvements of up to 173.7% in SSIM, 16.7% in FAC2, and 98.3% in MG compared with single-direction training. Furthermore, the comparison among methods indicates that sensor configuration, optimization, and training strategy should be jointly considered to ensure reliable deployment. For example, QR-based optimization enhances robustness by up to 27.8% under sensor perturbations, although with metric-dependent performance trade-offs. By training on experimental rather than simulated data, the proposed framework also offers practical guidance for method selection and sensor placement across different scenarios. The source code for this project is publicly available as open source on Github ( https://github.com/Yng314/windreconstruction ). • Comparing deterministic and generative learning methods against Kriging interpolation. • Evaluating methods and strategies under real-world constraints using experimental data. • Investigating a sensor position optimization method and its effectiveness. • Providing method selection advice under various application scenarios.

Similar Papers
  • PDF
  • Research Article
  • Citations2

Sensor Placement in an Irregular 3D Surface for Improving Localization Accuracy Using a Multi-Objective Memetic Algorithm.

  • Jul 11, 2023
  • Sensors
  • Paula A Graça +2
  • Research Article
  • Citations7

A novel tensor-based modal decomposition method for reduced order modeling and optimal sparse sensor placement

  • Aug 30, 2024
  • Aerospace Science and Technology
  • Zhirong Zhong +3
  • Research Article
  • Citations3

A Deep Reinforcement Learning Approach to Sensor Placement under Uncertainty

  • Jan 01, 2022
  • IFAC PapersOnLine
  • Amin Jabini +1
  • Research Article
  • Citations170

Computational methodologies for optimal sensor placement in structural health monitoring: A review

  • Oct 03, 2019
  • Structural Health Monitoring
  • Yi Tan +1
  • Research Article
  • Citations12

A physics-driven sensor placement optimization methodology for temperature field reconstruction

  • Sep 27, 2024
  • Applied Thermal Engineering
  • Xu Liu +5
  • Conference Article
  • Citations16

Optimal reporter placement in sparsely measured genetic networks using the Koopman operator

  • Dec 01, 2019
  • Aqib Hasnain +2
  • Conference Article
  • Citations8

Observability analysis and optimal sensor placement in stereo radar odometry

  • May 01, 2016
  • Andreu Corominas-Murtra +4
  • Research Article
  • Citations7

Performance-based optimal sensor placement method for single-layer reticulated shells considering modal observability and damage identifiability

  • May 13, 2023
  • Thin-Walled Structures
  • Xiaonong Guo +4
  • Research Article
  • Citations15

Optimal Sensor Placement and Velocity Configuration for TDOA–FDOA Localization and Tracking of a Moving Source

  • Dec 01, 2024
  • IEEE Transactions on Aerospace and Electronic Systems
  • Yang Yang +5
  • PDF
  • Research Article
  • Citations16

Multiple underwater target positioning with optimally placed acoustic surface sensor networks

  • May 01, 2018
  • International Journal of Distributed Sensor Networks
  • David Moreno-Salinas +2
  • Dissertation

Human-centric 3D representation learning

  • Jan 01, 2025
  • Fangzhou Hong
  • Conference Article
  • Citations6

Determination of an Effective Sensor Configuration for Suppression of the von Karman Vortex Street

  • Jan 05, 2004
  • 42nd AIAA Aerospace Sciences Meeting and Exhibit
  • Jeffrey Cameron +4
  • PDF
  • Research Article
  • Citations13

Optimal sensor placement for modal testing on wind turbines

  • Sep 01, 2016
  • Journal of Physics: Conference Series
  • Andreas Schulze +5
  • Research Article
  • Citations26

Optimal sensor placement methods and metrics – comparison and implementation on a timber frame structure

  • Feb 16, 2018
  • Structure and Infrastructure Engineering
  • Claude Leyder +4
  • Conference Article
  • Citations1

Multi-Objective optimization of Sensor Placement in a 3D Body for Underwater Localization

  • Oct 17, 2022
  • Paula A Graca +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.