Fig. 4: Experimental results.
From: Suppressing unknown disturbances to dynamical systems using machine learning
a, b Unknown and reconstructed disturbance functions [gx(t), gy(t)] (black curve) and [ux(t), uy(t)] (red curve) along with the training forcing functions [fx(t), fy(t)] (thick blue curves and symbols). For each case the reservoir was trained with (a) a 5 hz square wave out of phase by π/2 and (b) \([{f}_{x}(t),{f}_{y}(t)]=[(\cos (10\pi t)),(\sin (10\pi t))]\) and disturbed with combinations of sinusoidal functions. c, d Time series for the unknown (solid black) and recovered (dashed red) disturbance functions in the y component, gy(t) and uy(t).
