Research On Comprehensive Optimization Of Multi-head Surface Mounting Machine Scheduling

The placement machine is the core equipment of the surface mount production line,which mainly completes the process of assembling components on the printed circuit board(PCB).The process has the characteristics of high technical difficulty,high precision requirements,and high production cost,which is the bottleneck of the entire production line.With the continuous development of placement machine technology,the structure of placement machines is becoming more and more complicated,which makes the difficulty of comprehensive optimization of placement scheduling strategies of placement machines greatly increased.Important theoretical significance and practical value.In order to study the scheduling planning strategy in depth,we first summarized various placement machines with different characteristics.According to the different mounting characteristics and mechanical characteristics of the placement machine,it can be divided into three different placement machine types:arch type placement machine,turret placement machine and composite placement machine.The main components of the placement machine including feeder,placement head,control system and alignment system were analyzed and summarized.According to the placement process of the placement machine,by analyzing the factors affecting the placement efficiency,reasonable assumptions are given,and the optimization problem of placement scheduling is mathematically modeled using the 0-1 integer linear programming model.For high search efficiency,the initial solution generation algorithm is designed in detail.The problem is decomposed into three sub-problems of nozzle selection and allocation,feeder position allocation and placement sequence,and an effective approximation algorithm is designed respectively,and the complexity of the algorithm is calculated.Next,the design of heuristic optimization algorithm is completed.Using the principle of minimum criterion,a genetic algorithm with minimum criterion is proposed.The algorithm uses the minimum criterion of dynamic change to screen the offspring individuals,so that the optimization process maintains a high species diversity,and thus makes the algoritm not easy to fall into the local optimal.At the same time,a coding method and corresponding genetic operator are designed based on the characteristics of the placement scheduling optimization problem.The convergence analysis of the algorithm shows that the algorithm converges to the global optimal solution with probability 1.In order to verify the effectiveness of the algorithm,the mounting data of the 12 PCBs to be tested were first extracted,and the relevant parameters of the placement machine were determined;then through offline optimization experiments,the optimal parameters of the offline algorithm were obtained;Experiments verify the correctness and effectiveness of the algorithm.Comparative experiments show that the offline optimization algorithm proposed in this paper reduces the average placement time by 15.90%compared with the hybrid genetic algorithm,and the average placement time is reduced by 3.85%compared to the hybrid evolution algorithm;finally,the optimization process is completed using Python Preparation.