A mathematical model regarding water filtration with a nanofiltration (NF) titanium dioxide ceramic membrane is presented. The experiments aimed to use the excitation-emission matrix (EEM) spectroscopy method to highlight the existence of humic acid (HA) in water, before and after the NF process. Following the established operating conditions, experiments were performed for each quantity of AH separately, leaving the installation to work at the appropriate parameters for 15 minutes. for each quantity of AH. The analyzes for EEM fluorescence were performed using the FP-8300 spectrofluorimeter. The collected samples were analyzed with Spectra Manager II software on fluorescence intensity (au - arbitrary units), with an emission wavelength (nm) between 460 and 640 nm and with an excitation wavelength (nm) between 350 and 600 nm. Following the experiments carried out, mathematical correlations were established between the parameters that influence the filtering process and the studied parameters. It is worth mentioning that as a result of the experiments carried out, a number of 20,450 values were obtained, which were used for the elaboration of mathematical models. These models, for sets of values of the order of tens of thousands, verified both from the point of view of the real values and from the point of view of the regression coefficients (coefficients close to the value 1), demonstrate the quantity and the very good quality of the experimental data, respectively of the measured and calculated sizes. In order to validate the generated equations, they were subjected to checks, the difference being obtained between the value obtained by experimental means and the value obtained within the mathematical model. And the value of the resulting relative error, gives information on the accuracy (truth) of the mathematical model, so that it can be extended to other experiences. It turns out that this method cannot quantitatively determine the value of a parameter, but it can highlight the presence and differences between two samples.