Adaptive Fuzzy Control Research On Weld In Hydraulic Turbine Repairing Robot

The repair and maintenance work of turbine blades is an international problem for all the hydro power plant. A lot of manpower and resources is cost every year. Because the rivers contain plentiful sand in our country and the erosion of turbine blades is more serious, repair work is more arduous in our country. We introduce and explore the special robot repairing hydraulic turbines which is suitable for our country, and use it in welding and grinding turbine blades. In order to improve the intellectualized level of the robot and the quality of the blades repaired, I made a research of adaptive fuzzy control on weld in hydraulic turbine repairing robot. The theme is an important part of province project. Study of the theme is for the welding process of the special repairing robot. Welding voltage, welding current and the feeding velocity of the roller are important parameters in weld process, so I design the data acquisition and control system for the weld course. The system is SCM (Single Chip Micyoco) system with the core AT89C52 chip and it can measure the parameters timely and transmit the data to main computer for further processing. It is in the pit where the robot repairs hydraulic turbines. The weld course is non-linear, time-variable and the parameters are intercoupling. So many factors influence the quality of weld bead and obvious randomicities accompanies with the process, so it's difficult to describe the process with precise mathematical models. Classical control theory and modern control theory are all based on the traditional mathematical models. We can't achieve ideal result by using the two methods during the complex weld process. So we come forward using fuzzy logic control to imitate the thinking manner and operating experience of advanced weld experts to realize the fuzzy control in weld course. I design a two-dimension fuzzy controller with its inputs, the error of welding voltage and the change of the error. Rules of simple fuzzy logic controller are fixed, the fixed rules seem to be more coarse or faulty and it can influence the control effect. In this paper I adopt adaptive fuzzy control theory to make the rules adjusted and bettered by the controller itself, and it overcomes the defect that the control rules which the controller depends on can not be adjusted on line automatically. The result of simulation indicates that adptive fuzzy controll method can improve the control performance obviously.