Development of a hybrid artificial intelligence optimization method and its use in the composite material production process

In recent years, the high cost of raw material exploration and processing, as well as the decrease in raw material resources, have led industry and science to look for alternative solutions. Additionally, interest in composite production has increased due to the need to improving quality. At this point, the use of artificial intelligence technology in composite production processes is becoming increasingly widespread. However, studies on determining raw material mixture ratios are not common yet. In this thesis, firstly, experimental productions were made using mixture components and a data set was created with the results obtained. With this data set, the proportional optimization of the mixture components of hybrid thermoplastic composite materials was studied using the Particle Swarm Optimization Algorithm, and as a result of the proportional optimization, 95.3% more successful results were obtained compared to commercial products. Subsequently, a hybrid method was developed using Dragonfly, Gray Wolf and Differential Development Algorithms and the proportional optimization of the mixture components was carried out again. Composite samples were produced with the ratios obtained as a result of these processes, and it was observed that the final products produced were 98.1% more successful than the reference commercial product. This study has proven that the use of artificial intelligence technology, especially in the pre-composite production stages, can reduce costs and increase quality.