Optimization of CED dry film thickness using artificial neural network for automobile industries

To optimize the cathodic electro deposition paint film on automobile body for corrosion protection, AI driven solutions are generated. Cathodic electro deposition technique has been adopted for coating process by controlling process variables such as surface area, non-volatile matter, processing time, bath temperature, and electrode voltage. An artificial neural network model has been developed using the experimental dataset to train and enable it to predict dry film thickness. The proposed ANN model has been evaluated with an error rate of 6% which is well within the accepted limit of 10%. The study also shows that mild steel’s dry film thickness was more responsive to electrode voltage and bath temperature than other metals. The study suggests that to achieve the most economical output, it is recommended to maintain the bath chemistry while adjusting the physical characteristics such as voltage and temperature, as indicated by the findings.