【作者】 齐海铭；

【导师】 张晓华；

【作者基本信息】 哈尔滨工业大学 ， 电力电子与电力传动， 2010， 博士

【摘要】 管内移动机器人示踪定位技术是一种在管道外部确定机器人于管内所处位置的技术,现已成为管道无损检测和管内作业等技术领域发展与工程应用的必要条件。由于金属管道及其埋藏环境的屏蔽作用,使得常规的电磁技术在管内移动机器人示踪定位技术中无法应用。因此,如何实现管内移动机器人的无线、可靠、快速甚至全程示踪定位已经成为提高管内移动机器人工作性能和实用价值的重要课题之一。本文结合国家“863”计划项目“海底管道内爬行器及其检测技术”与“基于超长波的管道机器人示踪定位技术”,针对管内移动机器人的示踪定位理论与系统实现问题进行了深入的研究。本文提出了管内移动机器人示踪定位系统的总体设计方案,同时分析了示踪定位系统实际工程应用中的工艺流程和要求,详细描述了管内发射系统与管外传感器接收系统的具体构成和性能指标。针对金属管壁及其埋藏环境对常规电磁信号的屏蔽问题,本文提出了应用极低频电磁场作为管道内外信息传达的媒介,解决了示踪定位技术的瓶颈问题。从Maxwell电磁理论出发,建立了金属管道环境下极低频电磁场分布的数学模型,通过数值仿真分析了管道及埋藏环境的几何尺寸和电磁参数对管外空间极低频电磁场分布特性的影响,并确定了最佳的发射频率。为了便于示踪定位算法的理论分析与工程应用,本文提出了应用等效磁偶极子模型来描述管道环境下的磁场分布,通过对发射器线圈结构的优化设计,有效降低了模型误差的影响。通过实物实验对极低频电磁场在金属管道环境下的适用性进行了验证。示踪是指在管道外部粗略地获取管内移动机器人的方位,精度要求较低。本文提出了两种解决方案:(1)单轴激磁与单轴检测方式,(2)三轴激磁与三轴检测方式。前者通过归纳各类管道环境下极低频电磁场的分布规律,定性分析管内移动机器人的位置,估算精度较低,单次示踪过程耗时较长;后者是一种定量计算的示踪方法,通过对三轴发射向量和三轴传感器输出向量分别进行正交旋转变换,实现对发射器与传感器相对位置关系的定量求解;该示踪算法采用传感器输出信号“能量”作为参考输入,较好地解决了管内移动机器人动态示踪和相对位置较远时传感器输出信噪比较低的问题。将上述两种示踪方法分别进行实验验证,结果表明:前者在可视管线、管道分布已知或分布未知但机器人运行速度较低情况下具有较好的示踪效果;而后者适用于各类管内移动机器人的示踪过程,位置估算误差较小,具有较好的抗扰性能,满足一般管道工程需求。由于极低频电磁场在金属管道环境下的传播距离有限,为了提高管内移动机器人示踪定位技术的实用价值,本文提出了全程示踪定位的概念。针对管内移动机器人检测到管壁缺陷或机器人出现故障的静态定位问题,提出了基于传感器阵的定位方法,一方面提高了极低频电磁场的探测范围,另一方面通过融合多个传感器的信息有效提高了定位精度。根据传感器轴向和数目不同,给出了基于矩阵变换的线性定位算法和基于最优原理的非线性定位算法。理论分析表明:基于最优原理的非线性定位算法需要传感器数目较少、算法简单且便于工程实现。分析了传感器运动感应噪声产生的原因,并提出了相应的方法来减弱噪声对定位精度的影响。在管道平台上进行管内机器人的静态定位实验,实验结果表明所设计的定位算法和定位系统均满足管道示踪定位工程的需求。更多还原

【Abstract】 The tracking and locating of in-pipe mobile robot is a technology to detect the position of robot by measuring outside the pipeline. This technology decides the further developments and practical applications of pipeline nondestructive tests and in-pipe operations, etc. As the normal electromagnetic methods can not be applied in pipeline engineering, because of the intense shielding effects of metallic pipe wall and buried pipeline environment. So, the wireless, reliable, fast, even the global tracking and locating technology is one of the key technologies to improve the system performance and practical application value of in-pipe mobile robot. Based on the projects“Seabed in-pipe mobile robot and inspection technology”and“Tracking and locating technology of pipeline robot based on ultra-long electromagnetic wave”, this thesis studies the tracking and locating theory and system applications of in-pipe mobile robot.The thesis presents the completed design methods of the tracking and locating system of the in-pipe mobile robot, and analyzes the working process and requirements of the system. The specific structure and performance of the in-pipe emitter system and the outside sensor system are described in detail.In order to solve the shielding problem by metallic pipe wall and buried pipeline environment, which is a great bottleneck for the application of tracking and locating technology, the extreme low frequency (ELF) magnetic field is applied to realize the communication between inside and outside of pipeline. Based on the Maxwell electromagnetic theory, the analytical propagation model of the ELF magnetic field in the buried pipeline environment is presented. The thesis studies the affects of the geometric sizes and electromagnetic parameters of the pipeline environment on the propagation character of the ELF magnetic field and the optimum emitting frequency. In order to simply the analysis of the tracking and locating algorithm, the thesis presents an equivalent magnetic dipole model to approximately describe the propagation of the ELF magnetic field. The imitation error is reduced to minimum by the structural optimum design of the emitting coil. The actual propagation performance of the ELF magnetic field is tested by various experiments, and the results show that the ELF magnetic field is practicable in the tracking and locating technology of the in-pipe mobile robot.Tracking is roughly estimating the position of in-pipe mobile robot with low accuracy. There are two tracking methods are presented in the thesis, i.e. (1) single axis emitting and point measure, (2) three axes emitting and point measure. The first method can only estimate the rough position of the in-pipe robot by concluding the propagation character of ELF magnetic field in various pipeline environments. The estimation precision of position is low and this method wastes time a lot. The second method is based on the orthogonal rotations of the emitting vectors and the receiving vectors. The relative position and orientation between emitter and sensor are calculated respectively. This tracking algorithm adopts the energy of the sensor’s output, which is helpful for the dynamic tracking of robot. It also solves the problem of low sensor to noise ratio (SNR) when the relative distance between emitter and sensor is too long. The two methods are verified by field tests. It is concluded that the first tracking method is suitable for the visible pipeline, the specific distribution pipeline, or unknown distribution but the moving speed of robot is low enough. With little estimation error, the second method is verified practicable for tracking of in-pipe mobile robot, and satisfies the requirements of pipeline engineering.As the propagation distance of ELF magnetic field is limited, the thesis presents a global tracking and locating concept in order to improve the practical application value of the tracking and locating technology. The static locating algorithm, which is applied when defects on pipe wall are inspected or there is something wrong with the robot, is presented based on the sensor array. The sensor array is help to improve the detection area of ELF magnetic field and the accuracy of the locating algorithm. According to the number of axis and sensor, linear and nonlinear algorithms are presented, respectively. The first algorithm is based on the matrix transformation method and the second one is based on the optimum theory. With theoretical analysis, it is shown that the second method is simple and more suitable for practical application, and requires fewer sensors. The principle of motion inductive noise of sensor is analyzed and some methods have been presented to reduce the affects. The static locating algorithms have been verified by experiments. It is concluded that the locating algorithms and the locating system satisfy the requirements of pipeline engineering.更多还原