Research On 3D Printing Technology Based On Mechanicalarm

3D printing technology has developed rapidly in recent years,breaking through the bottleneck of traditional manufacturing methods in the production of complex parts,enabling personalized customization,and has been widely used in industrial production and daily life.However,at present,the main body structure of the traditional 3D printer is mainly parallel arms and three-coordinate there is no way to achieve the processing of inclined surfaces or curved surfaces,because of the limitation of the working space,it is impossible to print large-size items,and the complex posture cannot be performed due to the fixed print head posture.Printing and other issues limit the scope of application of 3D printers.At the same time,3D printing layering algorithms and filling methods affect the printing quality and efficiency of the model.Therefore,many scholars have focused on the research of 3D printing layering algorithms and filling methods.Based on the above problems,this paper uses a six-axis manipulator as the main body of the 3D printer to study the layering algorithm,the filling method and the inverse kinematics solution of the manipulator.The specific research contents are as follows:(1)By analyzing the existing 3D printing technology and methods,this paper proposes a 3D printing technology based on a robotic arm to achieve repair printing of parts.By establishing the kinematics model of the six-axis manipulator,its forward and inverse kinematics and working space are deduced and simulated according to the D-H rule.The simulation results show that the six-axis manipulator can realize multi-pose 3D printing,and the working space is larger and the space utilization rate is higher.(2)In order to reduce the step effect caused by layering in 3D printing,this paper designs an improved adaptive layering algorithm based on step slope height and volume error.The algorithm first considers the existence of tiny triangle patches in the STL model file,and obtains the initial layer thickness based on the minimum height difference of the triangle patches and the user-defined layer thickness value range,and then determines the maximum value generated when using this layer thickness for layering Whether the volume error meets the limit value.When the limit value of the maximum volume error is not satisfied,determine whether to use a smaller layer thickness to reduce the step effect of the printed part according to the step slope height.The simulation results prove that the use of an improved adaptive layering algorithm based on stepped slope height and volume error can alleviate the gradient effect and improve the accuracy of the print.(3)A gradient outline offset filling method based on parallel lines is proposed.Using the principle that the contour offset distance changes from small to large,parallel line filling is performed according to the offset contour,so that the contour and the filling path are effectively adhered without affecting the surface quality.When the contour is offset,the position close to the contour is densely filled,and the filling rate is reduced at a position away from the contour.When parallel lines are filled,odd-numbered layers are filled vertically and even-numbered layers are filled horizontally.This not only reduces the thermal shrinkage problem,but also improves printing efficiency.(4)Establish an improved BP(MPFGA-BP)neural network model based on multi-group fuzzy genetic algorithm to realize the inverse kinematics solution of the six-axis robot arm to control the robot arm for 3D printing.The training samples are obtained within the scope of the kinematics positive solution.The position and posture of the robot arm are used as input,and the joint angles are used as output.The inverse solution network model of the six-axis robot arm used in this paper is established.Among them,the multi-group fuzzy genetic algorithm is used to initialize the weights and thresholds of the network,the trained network model is tested,and the accuracy of the established MPFGA-BP neural network model is verified by controlling any curve in the end tracking space of the robotic arm.The research results prove that the MPFGA-BP neural network model of the robotic arm established in this paper has a high approximation accuracy.(5)Based on the inverse solution model of MPFGA-BP neural network,the improved adaptive layering algorithm based on stepped slope height and volume error is used to carry out 3D repair printing simulation.In order to verify the correctness of the simulation results,the communication between the PC and the six-axis manipulator arm was completed,and the established MPFGA-BP neural network model was used to obtain the angle of the manipulator arm to control the motion of the manipulator arm,and to achieve any plane with the required attitude Simulated tracking printing of complex contours.Simulations and experiments show that the improved adaptive layering algorithm designed in this paper and the established MPFGA-BP neural network model can be used to realize 3D printing based on robotic arms.The simulation and experimental results show that the 3D printing technology based on the six-axis manipulator can realize high-efficiency printing with multi-pose in a large space.