Research On Intelligent Posture Adjustment Method And Teaching Reproduction Trajectory Optimization Technology Of Parallel Drilling Machine

The parallel drilling machine has the advantages of compact structure,high precision and good flexibility.It can realize the drilling operation along the normal direction of the complex surface.Compared with the traditional drilling machine,the control complexity of the parallel drilling machine is greatly increased.How to reduce the control complexity of the parallel drilling machine,improving the processing quality and processing efficiency of parallel drilling machines are key technologies in the industrial application of parallel drilling machines,and have important theoretical significance and practical value for the research of these key technologies.This dissertation takes the 1PT+3TPS parallel drilling machine as the research object,and conducts in-depth and systematic research on the intelligent posture control method and teaching reproduction trajectory optimization technology of the parallel drilling machine.First this thesis introduces the manual posture adjustment method and teaching reproduction control technology currently used in parallel drilling machines,analyzes the shortcomings and deficiencies in it,and points out that manual control joystick for posture adjustment of parallel drilling machines have low working efficiency and poor drilling accuracy.The shortcomings such as direct use of teaching track points for motion reproduction have the disadvantages of poor motion stability and low processing efficiency.An intelligent position and attitude control method for parallel drilling machines is proposed.The intelligent position and attitude control method has the advantages of fast attitude adjustment speed and high alignment accuracy.An optimization technique for teaching reproduction trajectories of parallel drilling machines is proposed.The reproducing trajectory of the drilling machine has the shortest movement time and the highest processing efficiency.BP neural network hardware based on FPGA is designed and implemented.The FPGA-based BP neural network is divided into modules,the basic function modules are designed,the overall operation process of the neural network is completed,and the hardware implementation of each module and the overall neural network is carried out using Verilog hardware description language.The parallel operation and pipeline technology of the neural network hardware have improved the operation speed of the BP neural network hardware.Digital simulation and error analysis have been performed on each module of the neural network.The approximation experiment of the sine function(y=sinx)has been carried out using the neural network hardware.The FPGA-based BP neural network hardware is used to realize the intelligent posture adjustment of the parallel drilling machine.The intelligent attitude adjustment method uses three laser ranging sensors to detect the distance between the moving platform and the surface of the part,and outputs the distance parameters to the trained neural network hardware unit configured on the FPGA.After the mapping output,the length of each driving leg and the displacement of the XY table along the X axis and Y axis of the parallel drilling machine are obtained,by controlling the motion parameters,the parallel drilling machine can be automatically aligned.Intelligent posture adjustment in the normal direction of the hypoid surface.In this paper,the parameters of the intelligent posture model based on the neural network hardware are determined,and the designed intelligent posture model is implemented using Verilog.The collected sensor data and the calculated control parameters are used to train and predict the neural network.After the convergence,the weight threshold is input into the model,and the digital simulation and error analysis of the intelligent pose adjustment model are carried out.Finally,the model is mounted on the FPGA,and the on-board experiment is verified.A trajectory optimization technique for teaching and reproducing parallel drill press is proposed.Simulated Annealing algorithm is added to the Beetle Antennae Search algorithm to improve it,so that the Beetle Antennae Search algorithm can avoid falling into local extremum,and is more suitable for global optimization in multidimensional space;the improved Beetle Antennae Search algorithm is used for the processed parallel drilling machine to optimize the trajectory of teaching,the designed fitness function takes the shortest moving time as the optimization goal,and the maximum speed is limited by the penalty function,so that the parallel drilling machine can not only meet the speed limit of the drive during the teaching and reproduction process,but also reach the motion The shortest time reduces the difficulty of teaching.The research work in this thesis reduces the operation complexity of the parallel drilling machine,improves the automation/intelligence level of the parallel drilling machine,and improves the processing efficiency of the parallel drilling machine.