The mechanical condition prediction can realize the early warning of mechanical fault, which is of great sig-nificance to the safe operation of high-voltage circuit breaker (HVCB) and the enhancement of the reliability for the power system. However, at present, few researches for the mechanical state prediction of HVCBs are re-ported. This paper proposes a novel prediction method for the mechanical condition of HVCB based on long short-term memory (LSTM) neural network and support vector machine (SVM). Firstly, based on LSTM, break signal, contact travel, and coil current of HVCB are predicted. Then, the key mechanical characteristic param-eters are calculated by using the predicted results. Finally, using the predicted mechanical characteristic pa-rameters, the mechanical state of the circuit breaker is diagnosed based on the SVM model, to realize the mechanical state prediction. Research indicates that the proposed method can accurately predict HVCB me-chanical conditions and lay a foundation for realizing predictive maintenance of HVCB mechanical state.
针对胡瓜钝绥螨体积小、与粉螨相似、难以分类计数问题,提出一种基于YOLOv4胡瓜钝绥螨自动检测计数方案.在多时间段、分批次拍摄螨虫照片制作数据集,通过构建YOLOv4深度学习模型对螨虫图像进行特征提取和高精度分类检测.结果表明,基于YOLOv4的螨虫识别模型能在图像存在杂质且螨虫相连情况下准确检测胡瓜钝绥螨,平均检测单幅图片耗时约0.137 s.在测试集中对胡瓜钝绥螨检测精度达到99.45％,粉螨检测精度达到93.94％.对胡瓜钝绥螨和粉螨检测精度均值(mAP)达到96.69％,分别比YOLOv4-Tiny、YOLOv3、Faster-RCNN、EfficientDet模型高14.85、2.12、35.77、27.18百分点.设计螨虫GUI(图形用户界面)品质监控界面,方便检测人员操作.本研究提出的检测方案具有很高的应用价值,能够精准、快速识别胡瓜钝绥螨,可为胡瓜钝绥螨养殖质量检测提供技术支撑.
本文提出一种基于图像处理的复合绝缘子憎水性智能识别方法.针对不同憎水性等级的复合绝缘子图像,首先对图像进行直方图均衡化和滤波处理,再利用Otsu阈值分割方法将水珠/水迹与背景分割开,从而提取清晰且完整的水珠/水迹轮廓.采用水珠/水迹覆盖率、最大水珠/水迹面积比、水珠/水迹平均尺寸等特征量对水珠/水迹进行量化,利用支持向量机建立特征分类模型,从而实现对复合绝缘子憎水性的智能识别.结果表明:基于图像处理的复合绝缘子憎水性智能识别方法能够有效识别出7种憎水性等级,平均识别准确率保持在80％以上.
准确确定杆塔接地体冲击接地电阻对于提高输电线路耐雷水平具有重要的意义.基于有损长线原理的电路理论法,在仿真软件ATP-EMTP中建立接地体冲击散流模型,研究考虑火花效应时接地体冲击接地电阻的有效确定方法.以一接地体在波形为2.6/50μs、幅值为20 kA的雷电流作用下为例进行冲击特性仿真分析,结果表明,采用迭代修正法修正火花放电区域接地体半径,得到的冲击接地电阻更准确.本研究所得结论可为杆塔接地工程改造及优化设计提供有效参考依据.
Path planning is an important technique for mobile robots, which can effectively map out a safe path, whether in an open field or in a complex and changing environment. To address the problems of many inflection points and irregular routes in traditional path planning algorithms for mobile robot, this paper proposes a path planning approach that incorporates the improved A* algorithm and the dynamic window approach. First, the algorithm is based on the jump point search algorithm, which extends path finding by finding jump points, thus reducing unnecessary nodes and ultimately achieving the effect of shortening path finding time. Then, the generated paths are smoothed using the gradient descent method to make them smoother and more natural; finally, the improved A* algorithm and the dynamic window approach are fused. This makes the planned path smoother and the obstacle avoidance more accurate, which not only reflects the good path planning ability of the mobile robot, but also meets the practical needs of the mobile robot.
