TY - GEN
T1 - Precise cutting force estimation by hybrid estimation of DC/AC components
AU - Sato, Taiki
AU - Yamato, Shuntaro
AU - Imabeppu, Yasuhiro
AU - Irino, Naruhiro
AU - Kakinuma, Yasuhiro
N1 - Funding Information:
This work was supported by JSPS 18H01353.
Publisher Copyright:
Copyright © 2020 ASME
PY - 2020
Y1 - 2020
N2 - External sensor-less cutting force estimation using a load-side disturbance observer (LDOB) has potential to estimate the cutting force with high accuracy in both feed and cross-feed directions. However, the accuracy of its low frequency components in feed direction decrease due to effect of the friction and heat of a ball-screw-driven stage. In this study, DC and AC components of the cutting force is estimated by different methods; friction-compensated motor thrust force and LDOB, and the cutting force was estimated in real time by hybridizing them. In particular, regarding the friction model, the dynamic and static characteristics of the friction force in each axis (X, Y, Z) were identified from the idling test results. In addition to the model that depends on the velocity, the characteristics of the friction that depend on the position was also identified and considered when compensating for the motor thrust force. Then, a simple moving average filter with an appropriate window length is applied to the cutting force by LDOB and motor thrust force, and the DC component error of LDOB is corrected by that of motor thrust force. The validity of the proposed method was evaluated through end-milling tests. The experimental results showed that estimation accuracy of cutting force using the proposed method can be greatly improved in feed directions. On the other hand, in cross-feed direction, the cutting estimation was performed using the conventional LDOB.
AB - External sensor-less cutting force estimation using a load-side disturbance observer (LDOB) has potential to estimate the cutting force with high accuracy in both feed and cross-feed directions. However, the accuracy of its low frequency components in feed direction decrease due to effect of the friction and heat of a ball-screw-driven stage. In this study, DC and AC components of the cutting force is estimated by different methods; friction-compensated motor thrust force and LDOB, and the cutting force was estimated in real time by hybridizing them. In particular, regarding the friction model, the dynamic and static characteristics of the friction force in each axis (X, Y, Z) were identified from the idling test results. In addition to the model that depends on the velocity, the characteristics of the friction that depend on the position was also identified and considered when compensating for the motor thrust force. Then, a simple moving average filter with an appropriate window length is applied to the cutting force by LDOB and motor thrust force, and the DC component error of LDOB is corrected by that of motor thrust force. The validity of the proposed method was evaluated through end-milling tests. The experimental results showed that estimation accuracy of cutting force using the proposed method can be greatly improved in feed directions. On the other hand, in cross-feed direction, the cutting estimation was performed using the conventional LDOB.
KW - Cutting force observer
KW - Friction compensation
KW - Process monitoring
UR - http://www.scopus.com/inward/record.url?scp=85092661811&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85092661811&partnerID=8YFLogxK
U2 - 10.1115/ISFA2020-9629
DO - 10.1115/ISFA2020-9629
M3 - Conference contribution
AN - SCOPUS:85092661811
T3 - 2020 International Symposium on Flexible Automation, ISFA 2020
BT - 2020 International Symposium on Flexible Automation, ISFA 2020
PB - American Society of Mechanical Engineers (ASME)
T2 - 2020 International Symposium on Flexible Automation, ISFA 2020
Y2 - 8 July 2020 through 9 July 2020
ER -