Abstract: A double-mirror focusing inside-beam powder-feeding laser-cladding device includes: first and second reflection beam splitters, a focusing lens, a vertical piping powder feeding device, and a collimated lens group. The first reflection beam splitter is disposed at a side of the collimated lens group; the second reflection beam splitter is disposed outside the first reflection beam splitter; the focusing lens is disposed at a side of the second reflection beam splitter facing away the collimated lens group; and the vertical piping powder feeding device is disposed at a side of the first reflection beam splitter facing away the collimated lens group. When the device is in use, a collimated beam is formed through the collimated lens group, and then reflected by the first and second reflection beam splitters to form an annular laser beam that can be focused by the focusing lens to form a light spot.

Abstract: Since defect areas in collected hydro turbine defect images account for a small proportion and there is a significant color difference between surrounding background and defects, which creates too much redundant information and greatly affects detection speed, a method for detecting defects in a hydro turbine top cover based on an improved YOLOv8 model involves first cropping collected defect images to obtain cropped images, then expanding the cropped images by image enhancement techniques such as image flipping and image blurring to obtain a training dataset. A defect detection network based on YOLOv8-CBAM is constructed and trained to generate a defect detection model for detecting the defects in the hydro turbine top cover. The method achieves high precision in defect detection for the hydro turbine top cover.

Abstract: A device for repairing a top cover of a hydro turbine by adding or removing materials includes a rotating support base, a connecting arm, a movable frame, and a machining assembly. The rotating support base includes a base and a rotating seat disposed on the base, a driving assembly is disposed on the base, and an output shaft of the driving assembly is drivably connected to the rotating seat. An end of the connecting arm is connected to the rotating seat, and the other end of the connecting arm is connected to the movable frame. A lifting plate is disposed on the movable frame, a sliding table assembly and a measuring system are disposed on an upper side of the lifting plate. A swivel seat is disposed on an upper side of the sliding table assembly, and the machining assembly is detachably mounted on the swivel seat.

Abstract: An in-situ robot material-reducing processing method and system for a hydraulic turbine top cover are provided. The method includes: S1, obtaining frequency response data of an end of a cutter controlled by a robot by performing an impact hammer test; S2, obtaining a modal parameter of the end of the cutter controlled by the robot by using a modal analysis software; S3, obtaining a damping matrix [C] and a stiffness matrix [K] by using a free vibration equation of a damped system; S4, establishing a dynamic model of a three-degree-of-freedom robot processing system; S5, obtaining a milling force coefficient of the cutter by performing a calibration experiment; S6, solving a dynamic equation; S7, drawing a lobe diagram of flutter stability of a milling process performed by the robot; and S8, obtaining stable milling process parameters according to the lobe diagram of flutter stability.

Abstract: An in-situ robot programming method for a hydraulic turbine top cover based on binocular structured light vision is provided, which relates to the field of robot vision-guided intelligent processing technologies. The method performs hand-eye calibration on a binocular structured light camera and a robot to construct a coordinate conversion relationship thereof, collects an annular to-be-processed area of the hydraulic turbine top cover by the binocular structured light camera for joining, and obtains a trajectory and a trajectory plan through point cloud processing algorithms, to thereby achieve a rapid programming for the hydraulic turbine top cover based on guidance of binocular structured light vision. The method adapts to changes in sizes of the hydraulic turbine top cover and take into account real-time performance and robustness, so that impacts of uncertainty factors on the system are greatly weakened, and the method has characteristics of intelligence and high degree of systematization.