Research On Parallel Mechanism Calibration Based On Swarm Intelligence Optimization Algorithm

The swarm intelligent optimization algorithm is currently a highly sought-after optimization method,which has developed rapidly in recent years and is applied to many fields.Different swarm intelligent optimization algorithms have commonality.They use the swarm behavior characteristics of simulated swarm animals,and use the information exchange and guidance between individuals to achieve swarm optimization.The algorithm has advantages that easy to implement,high efficiency and few control parameters.The optimization problem needs to be solved quickly and accurately,and the balance between the exploitation and the exploration of the optimization algorithm still needs to be studied.The differential evolution algorithm is an optimization algorithm for simulating biological evolution.It uses the methods of crossover,mutation and selection to update the swarm.The algorithm is robust and has good individual diversity,which can achieve global convergence and strong exploration ability.The particle swarm optimization algorithm is an optimization algorithm for simulating the behavior of bird swarm.The optimal particle information is used to guide individual motion.The algorithm has strong local optimization,fast convergence and strong development ability.This paper proposes a hybrid differential evolution and particle swarm optimization algorithm.The individuals of swarm in the algorithm are updated by means of cross-variation which in order to keep the individual diversity of the swarm,so that the swarm reaches the global optimal region in the search space.Then the individual moves to optimize the local optimal region according to the overall optimization experience of the swarm.The new improved algorithm guarantees the robustness and individual diversity of the algorithm.On the basis of achieving global optimization,the convergence speed of the algorithm is increased.The individual movement mode of the swarm is adjusted by the newly introduced judgment factors.According to different optimization models and reasonable selection of judgment factors,the exploration and exploitation of the algorithm can be balanced.The artificial bee colony algorithm is a swarm intelligent optimization algorithm for simulating the movement of the swarm of bees.This paper proposes an adaptive artificial bee colony algorithm based on guard.The algorithm introduces an adaptive employed bee to improve the development ability of the algorithm,and the guard bees to improve the exploration ability of the algorithm.The employed bees behave different strategies according to the search situation.First,a wide area of honey source search is carried out to improve the convergence speed and quickly find the optimal area.Secondly,a small area of honey source search is performed to accurately search for the optimal position of the honey source.The onlooker bees introduce the guidance of the optimal honey source,and more effectively move to the honey source.The guard bee ensures that the adaptive employed bee can distinguish the honey source area and guard the effectiveness of its honey source search.The results show that the new algorithm proposed in this thesis has good search ability for benchmark function optimization,and can efficiently converge to the global optimal value.In the development of engineering technology,mechanical engineering has many complicated optimization problems,such as the forward and inverse solution of the series-parallel mechanism and the identification of the mechanism error calibration parameters.Traditional optimization methods usually use analytical methods and numerical methods.When solving such optimization problems,problems such as low solution precision,long solution time and inability to solve are always encountered.Therefore,the optimization methods based on swarm intelligence optimization algorithm are continuously proposed.This thesis studies the improvement of swarm intelligence optimization algorithm and effectively solves the optimization problem of nonlinear model in manipulators.In this paper,the 6-PSS parallel robot is analyzed with the size parameters and degrees of freedom of the mechanism.The Jacobian matrix of the mechanism is established by the screw theory.The kinematics analysis is performed on the mechanism.The forward kinematic model of the parallel mechanism is established by numerical method.The nonlinearity of the forward solution is obtained.The equations are transformed into a single-objective optimization problem,and a numerical forward solution equation is established.The swarm intelligent optimization algorithm can effectively solve the problem of forward optimization of parallel mechanism.The traditional swarm intelligent optimization algorithm and the new algorithm proposed in this paper are used to optimize the forward kinematc model,and the results are compared to obtain the best algorithm for the mechanism.The main error of 6-PSS parallel robot is analyzed,and the traditional vector chain error model is established.The vector chain error model of nonlinear matrix is transformed into the optimization problem,and the cost function of the error model is established.Different swarm intelligent optimization algorithms are used to identify the vector chain error parameters and compare the results.In this paper,a linearized parallel kinematic chain error model based on D-H is proposed.The six branches of the parallel mechanism are regarded as six independent series robots.The kinematic equation of the series motion branch is established by D-H method.The error model is linearized by the adjoint matrix and Taylor's theorem.A new kinematic chain error model is proposed.The two error models are simulated and compared.The simulation results show that the linearization error model provides a linear relationship between the terminal cumulative error and the motion chain joint motion error.The model can effectively identify the error and improve the accuracy of the end effector.