Stick Slip Friction Models Control Design Approach For Friction Compensation In Machine Tools Drive System

• Main Author: Rafan, Nur Aidawaty
• Format: Thesis
• Language: English; English
• Published: 2017
• Subjects: T Technology (General); TJ Mechanical engineering and machinery
• Online Access: http://eprints.utem.edu.my/id/eprint/21898/1/Stick%20Slip%20Friction%20Models%20Control%20Design%20Approach%20For%20Friction%20Compensation%20In%20Machine%20Tools%20Drive%20System%20-%20Nur%20Aidawaty%20Rafan%20-%2024%20pages.pdf; http://eprints.utem.edu.my/id/eprint/21898/2/Stick%20Slip%20Friction%20Models%20Control%20Design%20Approach%20For%20Friction%20Compensation%20In%20Machine%20Tools%20Drive%20System.pdf; http://eprints.utem.edu.my/id/eprint/21898/; https://plh.utem.edu.my/cgi-bin/koha/opac-detail.pl?biblionumber=104918

In machining process, positioning accuracy of the drives system is always the key element in producing good products with great precision and minimal or zero defects. Positioning accuracy of an electrical drive system is measured by two types of errors; tracking and contour errors. Reducing tracking error will reduce positioning error and thus increase motion accuracy. Meanwhile, reducing contour error will improve quality of machined surface that leads to improvement in overall precision. Position accuracy and precision are subjected to input disturbance acting on the drive system. A special form of error produced as a result of friction is quadrant glitches. Quadrant glitches are spikes occurred at each quadrant angle in a circular motion due to the effect of highly non-linear friction force acting on the feed drive mechanism influenced by pre-sliding friction characteristics at low velocity. At pre-sliding, friction is pre-dominantly a function of displacement that behaves as hysteresis function with non-local memory. This thesis aims at enhancing knowledge and contributes towards compensating quadrant glitches in circular motion for a ball screw driven XY milling positioning table by means of control design approach using enhanced friction force models. The objectives are to model friction behaviour, design and validate the friction compensation performance at low tracking velocity. Two models of friction forces were introduced; the Sigmoid-Like-Curve-Slip (SLCS) model and the Pseudo-Like-Curve-Slip (PLCS) model. Compensation via friction model based method was implemented in this thesis with different position controllers; namely, Proportional Integral Derivative Controller (PID), Cascade Proportional/Proportional-Integral (P/PI) Controller and Sliding Mode Controller (SMC). The effectiveness of the two proposed friction models were validated against the Generalized Maxwell Slip (GMS) friction model – a model known for effective friction compensation in pre-sliding regime. The numerical analyses and experimental validation performed showed improved performance with reduced contour errors. The SLCS model managed to produce a 99% reduction in the magnitude of the quadrant glitches when combined with cascade P/PI position controller at tracking velocity of 2 mm/s. For similar position controller, the PLCS model was able to produce a maximum quadrant glitches reduction of 70%. In comparison, the GMS model was only able to produce a maximum reduction of 40%. Also, both SLCS and PLCS models demonstrate better friction compensation performance when applied with cascade P/PI position controller compared to SMC. Whereas, PID controller has limited ability to sufficiently compensate quadrant glitches even with feedforward of friction models. In conclusion, this thesis has successfully presented significant improvement in accuracy of drives system made with implementation of the two new improved friction models combined with a cascade P/PI position controller. The new models are able to accurately describe friction behaviour in pre-sliding regime by providing smooth transition between pre-sliding and sliding regimes. However, further researches are desired in enhancing the capability of the friction compensation performance in terms of adaptive ability and robustness. Also, further analyses are necessary in the design of SMC robust controller for friction compensation.