Genetic Algorithm Optimization of a Filament Winding Process Modeled in WITNESS
نویسندگان
چکیده
With the advent of smaller, less expensive, and generally more effective computers, simulation models and simulation environments have become increasingly popular tools for solving engineering problems. This paper describes an effort to link a genetic algorithm with WITNESS, a modeling program developed by AT&T and Istel, in order to optimize a model of a manufacturing process called filament winding. 1 FILAMENT WINDING MODEL The filament winding process is an assembly process used by the advanced composites industry to make composite material filaments that are used in such products as rocket-motor cases, helicopter blades, piping, tubing, and drive shafts. For this research, a genetic algorithm is connected to a filament winding model (designed in WITNESS) in order to optimize it for a minimum setup cost per filament produced with the constraint that a minimum number of filaments need to be produced within a certain time limit. Figure 1 shows a schematic of the filament winding model, along with how the genetic algorithm interacts with the filament winding model. 2 RESULTS For the purposes of this research, only two methods of determining the lowest cost per part are used. The first is the genetic algorithm, and the second is a random search. Both searches are allowed the same number of search points in their respective runs. Figure 2 shows how the genetic algorithm performs against the random search. Figure 1: Filament Winding Model Schematic Figure 2: Minimum Cost, Both Optimization Methods WITNESS itself has recently released its own optimization program for use. However, a comparison between the genetic algorithm and the WITNESS optimization package was not researched, mainly because the WITNESS optimization package is very expensive. Given its effetiveness, this genetic algorithm serves as a viable low-cost alternative for those WITNESS programmers looking to do optimization of their models. Mandrel Preparation Output Recycled Mandrels Filament Winding
منابع مشابه
Genetic Algorithm Optimization of a Filament Winding Process
This paper describes a research effort to improve the efficiency of a filament winding process by combining the simulation capabilities of the WITNESS modeling program with the search capabilities of a genetic algorithm. Results show that the genetic algorithm is able to reduce the cost of producing filament wound mandrels used in the defense industry.
متن کاملAn empirical study on statistical analysis and optimization of EDM process parameters for inconel 718 super alloy using D-optimal approach and genetic algorithm
Among the several non-conventional processes, electrical discharge machining (EDM) is the most widely and successfully applied for the machining of conductive parts. In this technique, the tool has no mechanical contact with the work piece and also the hardness of work piece has no effect on the machining pace. Hence, this technique could be employed to machine hard materials such as super allo...
متن کاملImprovement of Surface Finish when EDM AISI 2312 Hot Worked Steel using Taguchi Approach and Genetic Algorithm
Nowadays, Electrical Discharge Machining (EDM) has become one of the most extensively used non-traditional material removal process. Its unique feature of using thermal energy to machine hard to machine electrically conductive materials is its distinctive advantage in the manufacturing of moulds, dies and aerospace components. Howevere, EDM is a costly process and hence proper selection of its ...
متن کاملOptimization of Cement Spacer Rheology Model Using Genetic Algorithm (RESEARCH NOTE)
The primary cement job is a critical step in successful well completion. To achieve effective cementing job, complete mud removal from the annular is recommended. Spacer and flushers are used widely to achieve this goal. This study is about weighted cement spacer systems containing a surfactant package, weighting agent and rheological modifiers. Weighted spacer systems are utilized when a high ...
متن کاملSimulation and Design of Adaptive Control for Filament Winding Multivariable Process System
This paper mainly presents the simulation and design of adaptive control scheme for filament winding automatic control system. According to the model reference adaptive control principle, an algorithm of fuzzy immune PID is adopted, which is suitable for the velocity and position servo system control. The simulation results show the controller design is practical, and not only has achieved fine...
متن کامل