University of Twente Student Theses

Login

Analysis, optimization and evaluation of a pipe inspection robot

Boer, Harm (2008) Analysis, optimization and evaluation of a pipe inspection robot.

[img]
Preview
PDF
3MB
Abstract:Gas main inspection in urban areas is very time consuming and expensive. This is why a first robot prototype has been built which can inspect autonomously the pipe network from inside the pipe. However the currently available prototype does not satisfy all the requirements that have to be fulfilled for the first prototype. This master's thesis focuses on a small subset of those requirements e.g. the manoeuvring requirements. The currently constructed prototype is analyzed, optimised and evaluated in relation to the manoeuvring requirements. First the shortcomings of the existing prototype in relation to the required manoeuvres, were determined. Moving up a slope of 30° and through a T-joint is problematic. Also the average speed requirement of 80 mm/s cannot be met. The shortcomings of the existing prototype in relation to these manoeuvres, are analysed for the different prototype modules. The mechanical construction of the bending module has been optimized to reduce friction, which improved the clamping torque. The maximum necessary clamping torque could not be theoretically determined, because the friction coefficient between the wheel and pipe wall is too high to measure. This high friction coefficient is caused by the deformation of the wheels. However, it can be concluded that the necessary clamp torque that the bending modules can deliver, is sufficient for the different manoeuvres. The overall efficiency of the bending module is low, which is caused by the low clamping speed of approximately (14 mm/s). Using the motor current as clamp torque feedback seems to be a better solution than using the torsion spring elongation. An advantage of using the motor current for control, is that excessive stress on the motors is limited. Also the zero reference is much more accurate and the deviation of the torque as function of the motor current is low. A disadvantage is that only one motor at a time can be used. From the driving module analysis it can be concluded that the driving motors cannot provide the maximum necessary torque at the nominal motor current. This maximum necessary torque is approximately 131 mNm, while the driving torque at the nominal motor current is approximately 120 mNm. The efficiencies of the driving module are low. These low efficiencies are caused by friction between the gears. The driving speed is approximately 70 mm/s and less above driving torques of 120 mNm. Because the bending speed is also low, the speed requirement of 80 mm/s cannot be met. For driving autonomously, it is recommended to use position/speed control in combination with the motor current control. This prevents excessive stress on the motor. From the rotation module analysis it can be concluded that the rotation motor can deliver the necessary theoretical torque. During the analysis of the limitations of the current prototype, it has also been determined that the rotation module can practically deliver the necessary rotation torque in the worst case situation. Therefore, the rotation module does not have to be optimized. From the performance evaluation of the total pipe inspection robot with respect to all the required manoeuvres, it can be concluded that the bending module optimizations increased the manoeuvrability of the prototype. Almost all manoeuvre requirements can be met, except for driving sideways up a slope of 30°. A reason for not meeting this manoeuvre is that the wheels deform slightly when driving sideways, giving a reduction of the possible torque of 40 mNm. Also the speed requirement cannot be met by the optimized prototype. To increase the driving torque the current gearbox can be replaced by the Faulhaber 15A, which has a higher reduction. However the driving speed will then decrease because of these extra reductions. Instead of increasing the driving torque, it would be better to reduce the friction caused by the clamping torque. Experimenting with different kinds of rubber and shape might create a more suitable trade-off between the traction and driving torque.
Item Type:Essay (Master)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:53 electrotechnology
Programme:Electrical Engineering MSc (60353)
Link to this item:http://purl.utwente.nl/essays/58144
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page