Dissertation
Dissertation > Industrial Technology > Automation technology,computer technology > Automated basic theory > Artificial intelligence theory

Research of Bee Evolutionary Genetic Algorithm and Ant Colony System to Surface Mount Technology Optimization

Author SunYouTian
Tutor WangChaoXue; DongHui
School Xi'an University of Architecture and Technology
Course Signal and Information Processing
Keywords surface mount technology surface mount-placement machine beeevolutionary genetic algorithm ant colony system optimization
CLC TP18
Type Master's thesis
Year 2012
Downloads 22
Quotes 0
Download Dissertation

With the rapid development of modern electronic technology, electronicinformation industry has been one of the indispensable industries. In the modernnational defense, modern industrial and social life, electronic information industry isplaying an increasingly important role. Electronic manufacturing is the foundation ofelectronic information industry, surface mount technology (SMT) as an important partof the advanced manufacturing technology, the development of it gets the world’sattention. In the electronic assembly line, the most and important equipment isMulti-head Surface Mounting Machine. Meanwhile, the Multi-head Surface MountingMachine’s production efficiency has been the bottleneck of the entire electronicassembly line. The optimal of Multi-head Surface Mounting Machine is a reasonableway to increase productivity and to reduce production cycle.At first, the assembly planning of surface mount-placement machine was analyzed,and then an integrated optimization model was founded according. A bee evolutionarygenetic-ant colony algorithm (BGAA) was proposed by combining with the advantagesof bee evolutionary genetic algorithm and ant colony system, whose structure was serial.Then, because BGAA was slow, a hybrid intelligent algorithm using parallel structurebased on bee evolutionary genetic algorithm and ant colony system (BAHA) isproposed. The key of the hybrid intelligent algorithm lies in improving the convergencespeed by the combination between the two population;the improved OX crossoveroperator retained the sequence of the good genes;a local search operator wasintroduced,which has more elaborate search ability in the neighborhood of theiteration-best; pheromone resetting is used to jump from local optimal solution.The hybrid intelligent algorithm was tested using traveling salesman problems; by making the6suction nozzle vertical rotary placement machines as a prototype, thecomponent mounting sequence problem of5PCB cards was emulated by BAHA. Theresults show that BAHA has good global search ability and fast convergence rate, andcould slove the problem optimization of SMT effectively.

Related Dissertations
More Dissertations