This is a decommissioned version of ERA which is running to enable completion of migration processes. All new collections and items and all edits to existing items should go to our new ERA instance at https://ualberta.scholaris.ca - Please contact us at erahelp@ualberta.ca for assistance!
- 6 views
- 2 downloads
A Semi-Autonomous Robotic-Based Repair System for Revolving Mechanical Components via Laser Metal Deposition Process
-
- Author / Creator
- Al-Musaibeli, Hamdan M. S.
-
Natural resources are finite, and the demand for such resources is increasing. Thus, an urgent need for systems to support the reusability and lifespan extendibility of products, such as those with a high demand and manufacturing cost, is essential. Furthermore, cylindrical components are widely used across many industries, such as mining, oil and gas, and agriculture, due to their mechanical and geometrical unique characterizations. Practically, the repair, restoration, and remanufacturing processes of such mechanical components are currently being achieved manually, which require professional manpower and long processing time. Therefore, this research aims to develop a semi-autonomous repair and restoration robotic-based system for revolving mechanical components using laser metal deposition technology to eliminate the aforementioned limitations, thus supporting a sustainable society.
In this project, a framework is developed for a semi-autonomous repair process of end-of-life mechanical components using additive manufacturing technology. The designed system consists of three main stages: an intelligent autonomous recognition method for detecting defective areas on a component, an automatic localization and spatial measurement process of the recognized faulty areas, and an automatic generation of the robot paths for the laser metal deposition repair process.
In the first stage, a vision-based pipeline was adopted to recognize the damaged sections on the component’s surface. A preliminary investigation was conducted to study the behavior and performance of the state-of-the-art model on the object detection task. This study leverages the transfer learning concept to build upon the knowledge gained on previous detection tasks to boost the learning process time and model convergence to the new detection task. Furthermore, the state-of-the-art was also trained on public metal defects datasets for further study of the model on different metal defects datasets. After the deep study, one dataset is selected to be used as a base for detecting defects on metal surfaces.
The second part of this framework is the localization and topography data acquisition step. A novel localization method is developed based on image-spatial mapping of cylindrical components. The method uses a vision sensor to locate bounding points of the damage in 3D space for the robot to acquire the point cloud scan using a proximity sensor. This method covers process planning, image mapping, and spatial mapping.
In the last stage, robot paths for the repair process of the damaged surface are then generated automatically to the nominal dimension of the nearby areas. Further, the build target is adjustable in the developed method to support the manufacturability of new features. Moreover, this method is capable of repairing the internal and external surfaces of a component. With this, a virtual repair process using the generated tool-paths is then produced to verify the repair process using an online/offline platform. Lastly, an actual repair process is carried out using the laser metal deposition process. -
- Subjects / Keywords
-
- Graduation date
- Spring 2023
-
- Type of Item
- Thesis
-
- Degree
- Master of Science
-
- License
- This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.