Genetic Algorithm And Its Application To Traveling-wave Tube To Optimize The Design Of The Study

Traveling wave tube (TWT) is one of the most important types of the microwave/millimeter wave vacuum tubes. Due to its high power, high gain, high efficiency and broadband, it is widely used in modern military electronic equipments, such as communications radars, electron wareface, etc. However, the manufacture process of TWT is so complicated. With developments and applications of computers, the technologies on TWT have been greatly improved so that the design abilities are enhanced, production period is reduced, hardware experimentes are decreased, the performance is improved, and the intrinsic erperiences are solidified. Now CAD technology has become a main researcher tool in TWT's research and manufacture. However, special CAD softwares on microwave tube generally lack the function of optimization, even the experienced designers of microwave tube pipe use these CAD softwares, it is difficult to make full use of them. So, optimization algorithm will be introducted to these special microwave tube CAD softwares. In some key parameters, we allow the software to optimize them automatically, which is a new direction of the microwave tube CAD softwares'development.Genetic algorithm is a kind of method searching optimal solution in high-dimensional space using natural selection and evolution thought. It provides a common framework to solve optimization problems of complex system, which is not dependent on the specific areas of problems, and it has a strong robustness to the different types of questions. So this algorithm has a good adaptability. And, it is a kind of intelligent search algorithm using heuristic knowledge, so it often has better results than the other previous algorithms on highly complex problem. Several important and valuable results are achieved and listed as follows:1,An improved genetic algorithm has been developed.An improved genetic algorithm (IGA) was introduced. In this improved algorithm, real-valued coding and some improved genetic mechanisms are adopted, also the elitist strategy is introduced. Compared with simple genetic algorithm (SGA), IGA is a high performance genetic algorithm.2,A genetic algorithm for multi-objective optimizationWe often met the problems about the design of multi-criteria or multi-objective. If these goals are contrary, we need to find the best to satisfy these goals. Genetic algorithm in solving this problem has been proved to be an effective method. The algorithm in this paper about multi-objective problem combines genetic search with local search, which improves not only the global optimization performance, but also the local search performance.3,Optimal design of Travelling Wave TubeGenetic algorithm which is a global optimization algorithm for functions of many continuous variables is intruoduced. In order to optimize the output power of Helix Travelling-Wave Tubes, we implement and improve genetic algorithm into 1D CHRINSTINE, which is a large-signal helix TWT code, and computational results are shown, too.