선택한 단어 수는 입니다.
최소 단어 이상 선택하여야 합니다.
최소 단어 이상 선택하여야 합니다.
선택한 단어 수는 30입니다.
최대 10 단어까지만 선택 가능합니다.
최대 10 단어까지만 선택 가능합니다.
제안 방법
- Finally this system could adjust the welding conditions so as to prevent the /Id defects such as undercut, overlap and excess convexity. To evaluate the performance of the proposed method, experiments were executed for three typical weld defects occurring in horizontal fillet "welding.
- The fuzzy 1 system was to check the welding conditions for the weld defects. Although this method could determine if there would be weld defects, it was difficult to adjust the welding conditions to prevent the weld defects by keeping the weld bead geometry obtained from the neural network. Therefore, another fuzzy rule-based method(fuzzy 2) was used to achieve satisfactory results.
- A considerable amount of experiments and experiences for the arc welding process are required to achieve the appropriate welding conditions. In this paper, a mathematical model based on various experimental results(non-linear regression method) and the welding data from the literature were used to determine the delicate relationship between the welding conditions and weld bead geometry7,8,9,12).
- In this paper, the neural network method for selecting the appropriate welding conditions and the fuzzy rulebased method for examining the welding conditions for weld defects were investigated. Because of the nonlinear characteristics and complexity of the arc phenomena, the fuzzy rule-based method utilized was based on the mathematical modeling derived from the experimental results
- To achieve a satisfactory fillet welded joint geometry it is necessary to study the relationship between the weld bead shape and welding conditions, which can be represented by a mathematical model based on the experimental results. In this study, the WPV included the welding current, arc voltage, welding speed, gas flow rate and offeet distance for horizontal fillet welding. All other parameters except these parameters under consideration were kept constant.
- The experiments were carried out for various ranges of the base metal thickness, welding current, arc voltage and welding speed. Fig.
- 5. The experiments were performed twelve times for estimating he performance of the neural network. The error percentages covered the range from +20% to -20%, which indicates that the neural network could successfiilly express the non-linear welding process.
- Generally fillet welded joint shapes are classified into four parts such as the leg length, penetration, tiiroat thickness and reinforcement height. To achieve a satisfactory fillet welded joint geometry it is necessary to study the relationship between the weld bead shape and welding conditions, which can be represented by a mathematical model based on the experimental results. In this study, the WPV included the welding current, arc voltage, welding speed, gas flow rate and offeet distance for horizontal fillet welding.
이론/모형
- In this study, the experiment was conducted based on the European Standard..
- To overcome the previous di#culties, the neural network method and the fuzzy rule-based method on the basis of the 2n-1 fractional factorial design method were used in this study. The neural network method based on the back propagation algorithm was used to predict the appropriate welding conditions for the desired weld bead geometry.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.