The design and control on intelligent underwater cleaning robot for power plant condenser

Abstract

A novel method for online steam condenser fouling cleaning is proposed in this paper. In the approach, high pressure cleaning is chosen as the main method, chemical cleaning is applied to complement and this cleaning method is achieved by a new autonomous cleaning robot. In this paper, firstly, the robot system structure is designed which consists of mechanical and electrical structure. Secondly, the distributed control system is used to improve cleaning efficiency. Based on these, the cleaning control strategy based on fuzzy-gaussian neural network is presented and the BP algorithm is adopted to train the network weights. Then, the simulation results show that the proposed control strategy has better robustness and dynamic performance than traditional fuzzy control, which can not only reduce the phenomenon of chattering in effect, but also has good robustness and dynamic performance. Hence, the control strategy is very fit for underwater robotic arm control.