- 學位論文
基於能量釋放來預測刀具壽命的可靠度研究方法
An Energy-Based Reliability Method for Prediction of Tool Life
摘要
在過去的十年中，由於機械、汽車、航空和醫療設備等科技不斷發展，設計者及使用者對於機械元件的尺寸精度與表面品質的要求不斷提高，因此元件加工的優劣決定了產品的品質。在加工過程中，隨著切削量的累積，導致刀具逐漸產生磨耗，進而影響工件表面粗糙度。傳統的加工方式憑著過去的經驗或技術，選擇較保守的切削方式(如較慢的切削速度、較少量的切削深度)，不僅掌握表面加工品質也減少刀具產生的磨耗，但無形中也浪費許多時間與金錢(即降低加工效率)；現代的工具機廠商實際加工時常採用高轉速與大進刀量，以減短切削時間，但增加對刀具的磨耗。如何正確掌握刀具的壽命除了能讓操作者在容許磨耗程度內完成加工任務，並在適當的時機更換刀具以工件的加工精度與表面品質。 本研究擬記錄加工過程中釋放的能量(如摩擦能、震動能)以評估刀具的可靠度並預測其壽命，故稱此研究方法為能量可靠度評估法(Energy-Based Reliability Method, ERM)。由於加工過程中能量的絕對釋放量難以量測，能量可靠度評估法記錄各種能量的累積相對釋放量，以預測刀具磨耗程度。本研究使用車床與銑床兩種傳統工具機驗證本演算法的正確性，實驗結果發現車床與銑床加工時的累積釋放能量與刀具的累積加工體積皆成線性關係。此關係可以取代加工面隨累積加工而漸漸增加的表面粗糙度量測值，比傳統粗糙度量測方式具更小變異性，並可即時量測。 本研究設置的車床與銑床加工實驗皆以固定的加工參數(如進刀速度、進刀量、機台轉速等)為基礎，重複實驗驗證能量釋放的可靠度預測具有重複性，將此重複性失效狀態定義為第一失效模式。此外，重複的實驗過程中發現較少數加工過程具有異常失效情況，本研究並未專注於異常失效發生的原因，但異常失效狀態歸類為第二失效模式。經過實驗數據擬合成的指數可靠度模型將可適用於不同類型的工具機加工過程中，用來評估刀具磨耗程度、預測失效模式並預測刀具壽命。
並列摘要
In the past decade, the developments of technologies of machinery, automotive, aerospace and medical equipment had been tremendous while higher requirements of dimensional accuracy and tooling quality were needed for designs of mechanical components. These two factors had been the standard to determine the quality of the product. It is well known for manufacturers that the accumulated amount of cutting results in gradual tool wear generation, which affects the surface roughness of the machined product. The traditional processing methods and experience would often rely on more conservative cutting means (such as a slower cutting speed, a smaller amount of cutting depth). It does not only affect the quality of surface finish which also reduces tool wear, but also waste a lot of time and money (i.e., reduce processing efficiency). Modern machine tool manufacturer’s actual processing often uses high cutting speed and large feed to shorten the cutting time, inevitably leading to greater growth of tool wear. This research aims to estimate the reliability of tool and predict the tool life for quality tooling processes under acceptable wear conditions. Also, the developed method is exected to inform the manufacturer when is the best timing for tool replacement and maintain the acceptable tooling quality. The developed research method collects the dissipated energy during the tooling process, including vibrational and frictional energy dissipations, for estimation of the tool reliability and prediction of the tool life. This method is called the Energy-based Reliability Method (ERM). Because the exact measurement of energy dissipation is difficult to obtain, the ERM utilizes the relative measurements of energy dissipation to build the its relationship with the accumulated cutting volume. The experiemts of the designed cutting tests in a lathe and a milling machine have been comducted to validate the proposed ERM. The experimental results showed the accumulated energy dissipation has a linear growth with the accumulation of the cutting volume in both machine tools. This energy model can be utilized to estimate the tooling quality in real time instead of taking roughness measurements at the workpiece surface after each cutting process. Furthermore, the ERM has fewer variance than the tranditional measurement of surface roughness of workpiece. The experiments in the lathe and milling machine were both performed under the fixed cutting parameters, such cutting speed, feed, spindle speed, etc. The repeated experimental results showed the proposed ERM is repeatable for most failure conditions. The usual failure conditions were specified as the first kind of failure mode. The other unusual and unexpected failure conditions were considered as the second kind of failure mode. Both the first and second kinds of failure modes were modeled using an exponential mathematical equation and fitted with the experiemental data. The developed ERM is suitable for other kinds of machine tools and can be used for estimation of tool reliability, determination of failure mode, and prediction of tool life.