本文介紹了使用神經網路技術估算螺絲的物理性能。這項研究的目的在於研究熱處理和球化過程中,各種控制參數對合金鋼絲在於製造過程中之物理性能的影響。神經網路用於分析合金鋼絲熱處理過程中收集的數據,期望可以開發具有智慧分析能力的先進螺絲製造系統,該智慧系統可以即時提供最佳控制參數,用以生產具有理想物理性能的合金鋼螺絲。進而使公司可以在隨後的螺絲製造過程中生產出低缺陷率的高質量螺絲。從研究結果可以看出,在球化和淬火回火熱處理後,神經網路確實可以對合金鋼絲的物理性能做出相當準確的估計。這一結果表明,人工智慧對螺絲工藝最佳化機制的發展是非常有前途和可行的。
In this thesis, the estimation of screw’s physical properties by using neural network technique is studied and presented. The aim of this research is to study the effects of various control parameters of heat treatment and spheroidization on the physical properties of Alloy Steel Wire in its manufacturing process. The neural network was used to analyze the data collected during the heat treatment of Alloy Steel Wire. It is expected that an advanced screw manufacturing system with intelligent analysis ability can be developed. This smart system is able to provide the optimal control parameters in real time to produce Alloy Steel Wire with ideal physical properties. Therefore, the company can produce the high-quality screw with low defect rate in the later screw manufacturing process. From the study results shown, it can be found that the neural network indeed can reach a fairly accurate estimation of the physical properties of the Alloy Steel Wire after the spheroidization and the quenching and tempering heat treatments. This result shows that the development of an artificial intelligence screw process optimization mechanism is very promising and feasible.