Optimization of caustic current efficiency in a zero-gap advanced chlor-alkali cell with application of genetic algorithm assisted by artificial neural networks

The effects of various process parameters on caustic current efficiency (CCE) in a zero-gap oxygen-depolarized chlor-alkali cell employing a state-of-the-art silver plated nickel screen electrode (ESNS®) were studied. For doing a thorough research, we selected the process parameters from both cathodic and anodic compartments. Seven process parameters were studied including anolyte pH, temperature, flow rate and brine concentration from the anode side, oxygen temperature and flow rate from the cathode side and the applied current density. The effect of these parameters on CCE was determined quantitatively. A feed forward neural network model with the Levenberg–Marquardt (LM) back propagation training methodwas developed to predict CCE. Then genetic algorithm (GA)was implemented to neural network model. The highest CCE (98.53%) was found after 20 times running GA at the following conditions: brine concentration (287 g/L), anolyte temperature (80 ◦C), anolyte pH (2.7), anolyte flow rate (408 cm3/min), oxygen flow rate (841 cm3/min), oxygen temperature (79 ◦C), and current density (0.33 A/cm^2).