A kinetic determination of L-ascorbic acid based on the Belousov-Zhabotinskii oscillating system was proposed. Optimal conditions are: [BrO₃^-] = 0.0625 mol l^-1, [malonic acid] = 0.1675 mol l^-1, [Ce(IV)] = 0.002 mol l^-1, [H₂SO₄] = 0.8 mol l^-1, temperature 30 ?C. Absorbance was measured at 320 nm. For model samples, the calibration curve method gives the values with relative error lower than 10 %. The application of artificial neural networks (ANN) for the calculation of results was demonstrated. The best results were obtained for the optimal ANN architecture using the oscillation amplitude as the value for layer of input neurons and L-ascorbic acid concentration as the layer of output neuron (relative error < 5 %). The proposed method is suitable for the analysis of pharmaceutical preparations (Celaskon ™) but not for analysis of citrus fruits due to the presence of other oxidizable compounds.