| 摘要 |
<p><P>PROBLEM TO BE SOLVED: To quickly and exactly detect or predict tool breakage even if tool conditions have been changed by the use of the tool for a long period of time. <P>SOLUTION: When the tool breakage occursed, a cutting time T and the maximum absolute value G of the inclination of a descending cutting load are reduced. An accumulated cutting load value S is increased by tool wear and is reduced by the tool breakage. Accordingly, the time T, the maximum absolute value G of the inclination, and the accumulated cutting load value S are obtained as values of loading states from the detected cutting load (A2, A3). A movement variation threshold is renewed and obtained based on the values of loading states obtained during a machining cycle for a preceding workpiece (A5). The abnormality of a tool is determined by comparing the value of the loading state with the movement variation threshold in a present machining cycle (A7, A8). Because the movement variation threshold for determining the abnormality of the tool is renewed by the value of the loading state during the preceding machining cycle, the detection and the prediction of the tool breakage can be performed even if cutting conditions are changed and cutting states are changed by the aging of the tool. Because the number of data for determination is small, high-speed processing is possible. <P>COPYRIGHT: (C)2004,JPO</p> |
