The Research On Visual Positioning Technologies Of Condenser Cleaning Robot In Thermal Power Plant

With advantages of increasing labor efficiency, improving working condition and overcoming operating limit, the cleaning robots can replace the manual labor for cleaning operations. So tnat, robot researchers pay great attention to them in recent years. Condenser cleaning robot is one kind of industrial cleaning robots, taking an important part in cleaning equipments of large thermal power plant condenser, is a the typical complex intelligent system gathering multi-disciplinary. This paper takes the condenser cleaning robot as main study object and starts a series of studies about its visual positioning technologies:Firstly, the necessity of fouling cleaning is explained by investigating the cause of fouling accumulation and the influence on heat transfer performance of condenser. Also, the advantages and disadvantages of the ordinary fouling cleaning methods are analyzed, the developments of the worldwide cleaning robots and their key technologies are summarized at the same time.Secondly, the working principle of intelligent condenser cleaning equipment is introduced. As main part of it, the cleaning robot is focused on. Especially, the cleaning system and mechanical structure design are explained by detail, as well as the control system with host computer system and hypogynous system which includes high-pressure water cleaning subsystem and chemical cleaning subsystem.A novel strategy of underwater visual positioning system of condenser cleaning robot is taken into considering for solving the former tubes positioning problems caused by coordinates input. In consideration, key matters in the process of this strategy implementation are analyzed,and two underwater visual positioning system structures of PC-based and DSP-based are presented in detail.As the visual positioning mainly relies on the image information, the applications of image preprocessing technologies, edge detecting methods in nozzle positioning process are introduced in this paper. As well, the OTSU segmentation algorithm and morphology border extraction method are given. At last of this part, a nozzle positioning algorithm combining gravitational center method and improved Hough transform circle detecting method is presented.Finally, the motion simulation based on virtual reality technology in the nozzle positioning of the condenser cleaning robot is mainly studied. A 3D model of the robot using 3ds max software is created, its positioning and movement process are analyzed. And based on above, the simulation of robot arm's movement during the cleaning task with virtual modeling language is proposed in focus.