遺傳子 알고리듬을 利用한 2段 遊星齒車機構의 最適設計에 關한 硏究
(A) Study on the Design Optimization of Two Stage Simple Planetary Gear Train using Genetic Algorithm
유전자 알고리듬 유성치차기구 유성기어장치;
- 원문 URL
One of the most common components in the machine is gear, defined as toothed members transmitting rotary motion from one shaft to another. Among the various means of mechanical power transmission (including primarily gears, belts, and chains), gears have been used most widely in small-sized machine like watch or measuring instrument to large machine like automobile, construction equipment, airplane have brought an improvement to their precise reduction ratio, low noise, high efficiency and high torque capacity. Recently cost competition has been very competive in most of all the industries. Every company regarding power train is making every effort to achieve improving performance and cost reduction through reducing weight for power train components. In the excavator field of construction equipment, gears are mainly beeing used at the track and swing device. Meanwhile there are many design variables at the one pair of gears which conjugate each other to transmit power. Each design variable is connected with non-linear equations like involute function. Therefore it is very complicated to decide suitable specifications of gears to meet the working condition. Especially calculations become more and more difficult and complicated in the planetary gear train that makes it possible to design compact gearbox. There are several critical design limitations to support in the planetary gear train such as assembly condition, co-axis condition, boundary condition to ensure smooth operation of each gear and every gear should meet the basic geometrical condition to have precise operation. The current design process of trial and error method requires designer to have much time consuming and efforts to calculate and decide all the gear specifications. And the design result can not be optimal and it is depend on designer's experience. Thus it is necessary to introduce a method of design optimization which can meet the all design limitations and requirements of working condition and search the most cost-competitive design on the gearbox. There have been several attempts to design planetary gear train using the method of optimal design. But some of design variables were fixed and they attempted to search the optimum for others. In this study, GA(Genetic Algorithm) has been used for searching optimal design to minimize the weight of gears for 2 stage simple planetary gear train. Besides maximum load capacity, target gear ratio and mounting dimension, no design variables were fixed. My encouragement to research in this study, I have provided the whole chance for searching possible sets of design variables. As the result of this study, I have gain to learn that getting the improved design could cause reducing the weight of 9.2% product weight and I could confirm that GA could be able to work very effectively in the design optimization for planetary gear train.