电测与仪表2026，Vol.63Issue(8)：95-101,7.DOI:10.19753/j.issn1001-1390.2026.08.010
注意力机制驱动的DeepLabv3+电力系统路径规划方法研究
Research on attention mechanism-driven DeepLabv3+path planning method for power system
摘要
Abstract
In response to the accuracy and efficiency issues of power system path planning,this paper proposes a planning method for novel power system path based on the DeepLabv3+network,which views the power system path planning problem as an image segmentation problem.On the basis of the DeepLabv3+method,the spatial at-tention mechanism is initially introduced to focus on more important information of the image,which is conducive to the restoration of image detail information.A feature reconstruction module is designed by introducing asymmetric convolution and depthwise separable convolution to reconstruct the feature information and obtain the output of the network.Furthermore,a new loss function is devised to optimize the features of different scales of the network.The test is conducted based on the Vaihingen dataset,and compared with the methods of SegNet,UNet,DANet,and DeepLabv3+,which shows that the proposed method has higher accuracy.The effectiveness of each module is veri-fied through ablation experiments.关键词
电力系统路径规划/注意力机制/非对称卷积/特征重建Key words
power system path planning/attention mechanism/asymmetric convolution/feature reconstruction分类
信息技术与安全科学引用本文复制引用
董添,李博强,楚云飞,王勇,王雨薇,彭泽朴,关珊珊..注意力机制驱动的DeepLabv3+电力系统路径规划方法研究[J].电测与仪表,2026,63(8):95-101,7.基金项目
国网吉林省电力有限公司高质量发展战略研究课题(SGJLJY00ZLJS2400059) （SGJLJY00ZLJS2400059）
