A Three-Dimensional Modeling Method of Substation Surface Temperature Distribution Based on Decoupling of Thermal and Geometric Information

doi:10.3969/j.issn.1008-0198.2025.05.016

Abstract

Abstract: Aiming at the limitations that the traditional substation three-dimensional temperature re?construction technology not only relies on high-precision laser radar point cloud completion and three?-?dimensional scanning technology, but also frequently encounters issues such as insufficient point cloud density and distortion of appearance representation, given the inherent characteristics of infrared images of low textures and high noise, a dual-path three-dimensional modeling system based on infrared images is proposed. This system completes high-density temperature distribution reconstruction through the cross?-domain fusion of thermal structural features and geometric priors, combined with a three?-dimensional Gaussian splatting algorithm, ultimately generating three-dimensional models of substation scenes and equipment that achieve high-density temperature distribution reconstruction while maintaining extremely high morphological fidelity. Compared with existing methods, the model accuracy is improved by about 5%. The method proposed in this paper can achieve an upgrade in three-dimensional temperature reconstruction effects for substations, enhancing the intelligent operational inspection capabilities of substations.

Key words: substation, three-dimensional(3D) temperature reconstruction, three-dimensional(3D) Gaussian splatting

CLC Number:

TM63

KANG Wen, DING Yuzhu, DENG Wei, ZHONG Yuming, TANG Jiangqi. A Three-Dimensional Modeling Method of Substation Surface Temperature Distribution Based on Decoupling of Thermal and Geometric Information[J]. Hunan Electric Power, 2025, 45(5): 117-123.

References

[1] 范继锋,张彦斌,李智,等. 变电站多旋翼无人机巡检红外图像目标检测[J]. 激光与红外,2025,55(4):599-606.
[2] 郑叶龙,李长勇,夏宁宁,等. 基于“十字”标志物的红外图像与三维点云融合方法[J]. 天津大学学报(自然科学与工程技术版),2024,57(10):1090-1099.
[3] 靳丹,刘晓光,石刚,等. 基于红外图像特征融合的变电站机器人巡视轨迹三维点云配准方法[J]. 红外技术,2023,45(6):678-684.
[4] 宋丽梅,张继鹏,李云鹏,等. 基于多视角红外传感器的三维重建方法[J]. 液晶与显示,2023,38(6):759-769.
[5] 曾晗,林文萱,郭思雨,等. 基于SLIC超像素分割和ICP配准的牵引变压器套管红外三维立体重构方法[J]. 铁道学报,2022,44(12):37-45.
[6] 郭连朋,陈向宁,徐万朋,等. 基于kinect传感器的物体三维重建[J]. 四川兵工学报,2014,35(11):119-123.
[7] 罗智斌,张彦峰,曹宁. 一种基于融合地面LIDAR点云与红外视频信息的热红外三维建模方法[J]. 广东技术师范学院学报,2014,35(3):55-58.
[8] 刘畅文,李波,潘江涛,等. 基于ISS-3DSC的NDT三维点云配准算法研究[J]. 激光与红外,2023,53(5):777-783.
[9] 张志鹏,邵学君,庞庆. 三维激光倒置扫描关键技术研究[J]. 红外技术,2021,43(8):752-756.
[10] 王欢,汪同庆,李阳. 利用Kinect深度信息的三维点云配准方法研究[J]. 计算机工程与应用,2016,52(12):153-157.
[11] 涂兵,黄丹平,李明. 飞行器模型表面三维温度场重建方法[J/OL]. 空气动力学学报,2025:1-9.(2025-04-25)[2025-06-14]. https://kns.cnki.net/kcms/detail/51.1192.tk.20250424.1652.002.html.
[12] 聂根良,程晓红,周超,等. 基于热红外深度估计的路用骨料三维特征评价[J]. 红外技术,2024,46(9):1099-1105.
[13] 陈寂驰,魏国华,柴娟芳. 一种基于红外图像三维重建效果与精度的评价方法[J]. 空天防御,2023,6(4):42-50.
[14] 陈寂驰,魏国华,郭聪隆,等. 一种基于红外图像序列的深度学习三维重建仿真方法初探[J]. 空天防御,2020,3(4):21-29.
[15] 沈振一,孙韶媛,赵海涛. 基于PP-MRF模型的单目车载红外图像三维重建[J]. 东华大学学报(自然科学版),2015,41(3):341-347.
[16] VASWANI A,SHAZEER N,PARMAR N,et al.Attention is all you need[C]//31st Conference on Neural Information Processing Systems(NIPS 2017). Long Beach,CA,USA. 2017.
[17] YU A,LI R L,TANCIK M,et al.Plenoctrees for real-time rendering of neural radiance fields[C]//2021 IEEE/CVF International Conference on Computer Vision(ICCV). Montreal,QC,Canada. IEEE,2022:5732-5741.
[18] BARRON J T,MILDENHALL B,VERBIN D,et al.Mip-NeRF 360:unbounded anti-aliased neural radiance fields[C]//2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR). New Orleans,LA,USA. IEEE,2022:5460-5469.
[19] MILDENHALL B,SRINIVASAN P P,TANCIK M,et al.NeRF:Representing scenes as neural radiance fields for view synthesis[C]//Computer Vision-ECCV 2020. Cham:Springer,2020:405-421.
[20] KERBL B,KOPANAS G,LEIMKÜHLER T,et al. 3D Gaussian splatting for real-time radiance field rendering[J]. ACM Transactions on Graphics,2023,42(4):1-14.