Dynamic Parameter Identification And Feedforward Control Of Hydraulic Heavy-duty Manipulator

Because the heavy-duty industrial manipulator has a large load capacity,it has been widely used in major offshore equipment maintenance,dangerous nuclear power plant replacement,and heavy-duty mining workpiece installation in the field of international engineering applications.Developing in the direction of speed,high precision,and high intelligence,it has become an important equipment for reducing labor costs,realizing highly dexterous operations,and improving the working intensity of heavy loads.In mining engineering,the ball mill is an important large-scale equipment for ore crushing in modern mining enterprises.Due to the impact of grinding bodies and materials for a long time,the internal liner of the ball mill is extremely easy to wear and damage,so it needs to be replaced regularly.This paper takes the sevendegree-of-freedom hydraulic heavy-duty manipulator used for ball mill liner replacement as the research object,and aims to realize the stable clamping and highprecision control of the manipulator.The dynamic parameter identification and feedforward control of the manipulator are studied.Conduct relevant research.The main research content of this paper is as follows:Firstly,the three-dimensional structure of the hydraulic heavy-duty manipulator was quantitatively analyzed,and its parameters were designed and kinematic model was constructed by D-H parameter method.The transformation matrix and terminal position vector of the adjacent linkage mechanism were derived.The working space cloud map of the end grabber of the hydraulic heavy-duty manipulator was drawn by Monte Carlo method to test the working characteristics of the manipulator in space motion.The dynamic model of the manipulator was established by using Newton-Euler method,and the influence of internal friction factors at the connecting rod was fully considered.Then the complete mathematical expression of joint torque of the manipulator was deduced,and the equation was decomposed and linearized.Finally,the minimum inertial parameter set for the identification of the dynamic parameters of the manipulator was obtained.Secondly,in order to fully stimulate the dynamic characteristics of the manipulator,the design and optimization of the excitation trajectory are studied,and the form and advantages of the improved Fourier series excitation trajectory are discussed.Taking the minimum condition number of the regression matrix as the objective function,taking the joint motion limit and avoiding the start-stop jitter as the optimization constraints,the trajectory parameter optimization problem is converted into a nonlinear optimization mathematical problem,and the joint space optimization is obtained through the genetic algorithm with penalty function Incentive trajectory.Then,let the manipulator carry out multiple work experiments according to the trajectory model excited by the improved finite term Fourier series,filter the motion data collected in real time in the experiment,and bring them into the linearized dynamic equation.The dynamic parameters of the manipulator are identified by the least squares method,and another trajectory is selected for verification to prove the correctness of the parameter identification results.Finally,aiming at the problems of low control precision and poor dynamic performance of traditional PID closed-loop control used for heavy-duty manipulators,this paper researches and implements a torque feedforward control method based on dynamic model.The expected output torque value of each joint of the manipulator is calculated by the dynamic model constructed with the identification parameters,and superimposed with the control output in the current loop to realize real-time compensation of the joint torque of the manipulator,thereby effectively improving the trajectory tracking accuracy of the manipulator,to improve dynamic performance.