More than 40 years of research work in the domain of photochemical engineering has shown that quantitative modelling and solving of the radiant energy conservation equation coupled to the momentum, mass and heat balances are very difficult and therefore, even taking into account the most recent developments, of little impact on photochemical reactor design and process optimization. The kinetics of photochemical processes depend on light absorption, and the modelling of photochemical reactors and processes must take into account the spatial distribution of radiation emitted by a given light source (radiation field) and of the radiation absorbed. This task has proven to be extremely difficult, even under the most favourable experimental conditions (e.g. sensitized reactions) and simplest reactor geometries. On the other hand, empirical methods of reactor design and up-scaling do not necessarily lead to optimal results, and modelling by means of artificial neural networks holds the promise of solving problems so far beyond the scope of methods using the transport phenomena approach. In this work, neural networks have been applied to the modelling of a homogeneous liquid phase photochemical system : the photolysis of an aqueous uranyl oxalate solution. The algorithm used to adjust the weights in neural network application was back-propagation. The comparison between the calculated and experimental data show good agreement, even when simulation was performed outside the range of the learning set (extrapolated result).