无人值守变电站三维实景目标协同识别模型构建

Construction of Collaborative Identification Model for 3D Real Object of Unattended Substation

为解决无人值守变电站三维实景目标识别存在的信息饱和及识别效果差的问题,构建了无人值守变电站三维实景目标协同识别模型,该模型使用三维激光扫描仪采集变电站空间数据,利用奇异值分解算法转换空间数据坐标,同时结合迭代最近点法完成点云数据配准,获取变电站三维实景模型,并使用改进原型模式重构的协同识别算法,针对变电站三维实景模型,重新设计反馈时机与反馈样本选择准则,实现无人值守变电站三维实景目标协同识别。试验验证结果表明,该模型在各种环境因素影响下均能准确识别无人值守变电站中设备情况,且避免原型模式重构信息饱和问题,识别率得到有效提高,具有一定应用前景。

In order to solve the problem that the information was saturated and the identification effect was poor in the three-dimensional(3 D)real scene object identification of unattended substation,a 3 Dcollaborative identification model was proposed.The model uses a 3 Dlaser scanner to collect the spatial data of the substation,transforms the spatial data coordinates using the singular value decomposition algorithm,and completes the point cloud data registration by combining the iterative nearest point method to obtain the 3 Dreal scene of the substation.By using the improved collaborative recognition algorithm of prototype pattern reconstruction,the feedback timing and feedback sample selection criteria are redesigned to realize the collaborative recognition of 3 Dreal scene targets.The experimental results show that the model can accurately identify the equipment in unattended substation under the influence of various environmental factors,and avoid the problem of information saturation of prototype pattern reconstruction,and the recognition rate is effectively improved,which has a certain application prospect.