Reliability Based Design, Analysis and Control of the Remote Handling Maintenance System for Fusion Reactor

Abstract

The in-vessel components of fusion reactor, such as the divertor and blanket that face the hot plasma during plasma operation, need regular maintenance due to the damage caused by the heavy heat load, the thermal stress and the complex electromagnetic force during the reactor operation. All the maintenance work of the key components in fusion power plant cannot be carried out by human directly without the remote handling (RH) technology due to the Deuterium-Tritium (D-T) reaction in the tokamak machine. The RH system has a significant effect to the fusion reactor’ components design, the maintenance efficiency and cost as well. The RH maintenance technology is one of the most interesting fields for the fusion technology research. It has been identified as one of the most critical issues on the road to the commercial fusion power plant.

The RH maintenance system includes many complex subsystems, such as the mechanical, electrical, hydraulic, and control systems. The failure of the RH system can cause great damage to the tokamak machine due to its complex structure and functions. Therefore, it is important to improve the reliability of the RH system of the future fusion reactor in the field of nuclear fusion technology. The purposes of this dissertation is to carry out the fusion reactor maintenance research work based on the functional requirements and the maintenance feasibility of the fusion reactor. The necessity and importance of the remote handling and maintenance of fusion reactor and its basic principles were expounded. The key technical problems and the future development of remote handling and maintenance were discussed. Different maintenance schemes were evaluated based on the availability, reliability, efficiency and cost of different systems. The integrated upper maintenance scheme with the comparison advantages disadvantages of these proposals was adopted. Different RH systems, aiming for the in-vessel components maintenance in the fusion reactor, were designed based on the optimum design methods of the reliability theory. The organization of this dissertation was briefly described below:

Firstly, the structure design and kinematics analysis of the divertor RH were performed according to the functional requirements of the maintenance system. And its simulation model was built to verify the kinematics model. The actuator system and transmission mechanism scheme were designed and determined.

Secondly, the reliability based optimization analysis of the 3-DOF lifting platform for the divertor RH maintenance was studied. The mathematical model of the 3-DOF lifting platform was built based on the reliability optimization and design. The optimized parameters of the 3-DOF platform were obtained by using the intelligent optimization algorithm to meet the requirements on reliability, structural stiffness and strength.

Then, the overall control strategy and the architecture design of the control system were studied. The hydraulic driving system of the robot was designed, and the simulation model of the hydraulic driving system was built. The experiment platform was set up to verify the hydraulic servo driving system’s dynamic simulation and optimization control with the differential evolution (DE) algorithm. Based on the reliability design and analysis of the RH maintenance 3-DOF platform, the water hydraulic servo control experiment was built to verify the influence on accuracy with different control strategies. The accurate position control of the hydraulic servo was realized by performing the comparative analysis between the DE optimization control and the fuzzy adaptive PID control. The experiment results showed that the DE optimization control algorithm has the advantages of small overshoot, fast response. Therefore, the DE based optimization control algorithm was applied to the water hydraulic servo control system.

Finally, further work was addressed at the end of thesis. In the design phase, several critical issues were addressed and required to be solved by both qualitative and quantitative approaches. Dynamic modelling of RH system and advanced robust control algorithms need to be developed in near future.