Research on key technologies of snake arm maintainers in extreme environments

Abstract

The China Fusion Engineering Test Reactor (CFETR) is a major international scientific project to solve the problem of providing the ultimate energy source for mankind. The safe and stable operation of the test reactor relies on regular monitoring and maintenance via the remote handling (RH) system. Traditional industrial robots are unable to meet the requirements of the narrow, obstacle-ridden and highly radioactive vacuum chamber. Additionally, maintenance efficiency is low, which seriously affects the process of promoting efficient automated maintenance equipment for the future commercial operation of nuclear fusion power plants. The hyper-redundant snake arm maintainers (SAM) with their excellent environmental adaptability and obstacle avoidance capabilities have attracted a lot of attention from researchers. Therefore, SAM is a highly significant and practical application in the monitoring and maintenance of large and complex equipment and confined spaces. SAM is a multi-jointed, hyper-redundant robot whose design is inspired by vertebrate and invertebrate animals in nature, such as snakes, elephant trunks, octopus tentacles and animal tails. A distinctive feature of the SAM is that the drive unit (motor, etc.) is placed outside the robot’s working space, which simplifies the overall structure of the robot and allows for a more complete streamlined posture. This externally driven “bionic tendon” makes the entire length of the arm free of excessive electronics and ideal for working in high-risk, radioactive, confined areas such as nuclear power plants.

The objective of this project was to develop SAM as one of the key subsystems of the CFETR remote handling maintenance system for tasks such as visual inspection and dust removal in the complex pipeline areas of the upper window and the bottom of the divertor in the vacuum chamber. By analyzing the skeleton characteristics of the snake and its geometric form of sinuous movement, an under-actuated SAM design method is proposed. Mounting it on the quick-change interface at the end of the CFETR multipurpose overload robot (CMOR) enables many types of maintenance operations in complex and confined spaces inside the vacuum chamber. The main research content of this dissertation includes (1) A SAM structure design method. In this work, a layered drive principle was adopted to design a rigid under-actuated SAM with a highly integrated composite capstan drive system to meet the design objectives for miniaturised and lightweight components in narrow vacuum chambers. (2) A SAM dynamics decoupling algorithm. Based on the SAM structure and drive characteristics, a non-linear decoupling algorithm for the cable traction force of each joint was investigated. The cable traction force was decoupled using Lagrange’s equation, force balance, torque balance and equivalent transformation. The strong coupling between the cable traction force and the joints, the end-effector and the servo motor and the cable forces in different positions are analysed. A closed-loop control strategy based on the SAM decoupled dynamics model was designed. The simulation results show that the stability and position accuracy of the motion was significantly improved. (3) A SAM adaptive trajectory control algorithm. Based on the tractrix principle, an improved trajectory tracking algorithm and minimum joint displacement principle, an adaptive trajectory control algorithm was designed and integrated for the whole working process of the SAM. This mainly includes trajectory tracking for the process of entering the narrow space at the initial position, trajectory planning for the process of completing the task and trajectory tracking for the process of exiting the narrow space at any position. (4) A SAM rigid-flexible coupling deformation position error compensation algorithm. By using a SAM rigid-flexible coupling dynamics model, the position error of the SAM end-effector under variable loads was obtained. Based on the Levenberg-Marquardt (LM) non-linear damped least squares algorithm, various parametric errors of the SAM are identified. It was also combined with gridded workspace and linearised variable load principles to achieve variable parameter compensation for different loads and positions. Finally, SAM prototypes were developed and validated with cable traction tests, trajectory control tests, load capacity and accuracy tests. The results of the research will be used to meet the practical maintenance requirements of the CFETR vacuum chamber in the divertor and upper window complex pipeline area.

The SAM designed in the project has a compact structure while offering high spatial curvature, positional accuracy and load capacity. In addition, the drive system, control system and kinematic model have been simplified to achieve bionic motion control of the robot. It can be mounted at the end of the CMOR for monitoring, maintenance, flaw detection and dust removal in the complex environment of the CFETR vacuum chamber. This is important for the stable operation, rapid inspection and automated maintenance of nuclear fusion reactors, which can accelerate the process of commercial operation of fusion reactors and realise the dream of an ultimate energy source for mankind. Expanding the scope of applications of the SAM designed in this project is of great significance in the aerospace, manufacturing, rescue and medical sectors.