| 摘要 |
PURPOSE:To automatically find and set an optimum speed feedforward coefficient. CONSTITUTION:The quantity (u) of variation between a torque command TCM which is one sampling cycle before and that two cycles before is found (S1). The estimated speed y' of a motor is found from the variation quantity (u), motor speeds y(i-1) and y(i-2) which are one cycle and two cycles before, and a torque constant which is found in the last cycle and the estimated rate I/beta(i-1) of inertia (S2). A speed loop process is performed to find a torque command TCMD(i), which is outputted to the motor (S4). The estimated speed error e(i) which is the,difference between the actual speed and estimated speed of the motor is found (S5). Then the estimated rate 1/beta(i) in this cycle is found from 1/beta(i)=1/beta(i-1)+b.u.e(i)/(1+bu<2>), where (b) is a parameter value determining the speed of convergence. The differential value of 2nd order of a position command at the time of speed feedforward control is multiplied by the reciprocal beta of the value when the estimated rate 1/beta(i) is converged to obtain a coefficient for finding a feedforward quantity. |
