| The path planning problem in intelligent cylinder changing is the key to realize the efficient cylinder changing in unmanned knitting workshops.At present,various intelligent technologies in the knitting workshop are quite mature,such as automatic cylinder changing robots,automatic empty cylinder identification and automatic cylinder coordinate positioning in each scene,and other problems and methods have been proposed by scholars one after another.However,there are few studies on the path planning of cylinder change for truss robots in knitting shops.Due to the complex distribution of yarn cylinders in the knitting shop and the large number of yarn cylinders,the actual production also needs to consider the remaining yarn quantity of yarn cylinders to avoid empty cylinder breakage to ens ure the normal operation of the yarn supply machine,and the path planning of the truss robot is needed to complete the timely and efficient change from empty to full cylinder.The main research is about:(1)A set of bobbin-changing robot control system and overall planning scheme of bobbin-changing path planning method are designed for the characteristics of knitting workshop and the motion characteristics of truss bobbin-changing robot.A PLC lower control system with 1769 L33 ERM as the main control mod ule is designed,mainly focusing on hardware,electrical and control process,etc.It is verified that the designed bobbin-changing robot control system scheme meets the production requirements.(2)To address the problem of a single group of intelligent algorithms each with defects,the advantages of common algorithms such as simulated annealing algorithm,ant colony algorithm,and forbidden search algorithm on the TSP problem similar to the problem are analyzed,and the genetic algorithm is selected for optimization analysis.For the problem of low quality of initialized population of genetic algorithm,a dual initialization strategy of nearest neighbor + particle swarm optimization is proposed;for the problem of weak global optimization ability of conventional crossover operator and easy premature maturity,a splicing insertion crossover strategy is proposed;the 2-opt local optimization operator,adaptive crossover and variation method and restart mechanism are introduced to improve the local optimizatio n ability and avoid premature maturity of the algorithm.(3)The analysis and modeling of the cylinder change path planning is carried out for the problems of cylinder distribution,yarn residual calibration,and positional transformation consumption caused by the structure of the yarn frame and robot in the unmanned knitting workshop,and a solution with the minimum cylinder change time as the optimization objective is determined as the key to solve the intelligent cylinder change path planning of the truss cylinder change robot based on the above conditions.Based on the improved algorithm to solve the path planning of the truss cylinder changing robot,after the validation test,it has greater advantages in initial solution quality,optimal solution quality and average solution quality compared with other algorithms.In the factory test,depending on the size of the empty cylinder,the test set reduces the time consumption by up to 55.5% compared with the manual sorting method,and the improved genetic algorithm improves the convergence speed by 48.3%compared with the traditional cylinder changing algorithm.The test results meet the production requirements and prove the effectiveness of the improved genetic algorithm for truss barrel changing problem.(4)Design the upper computer software for the control system and path planning method,design and verify the practical effect of the software function module and path planning method in the upper computer,and the practical effect in the factory is as expected. |
PDF Full Text Request