Prediction of voltage required for nonthermal plasma based diesel exhaust treatment for removal of nitrogen oxides.

Abstract

A prior prediction of voltage required for the nonthermal plasma (NTP)-based diesel exhaust treatment toward removal of nitrogen oxides, i.e., NOX (sum of NO and NO2) would help the researchers to plan for the resources required for the treatment. In this present study, experimental data set of 4032 number of input-output pairs have been collected by taking flow rate (Fr), temperature (T), discharge gap (Dg), initial NOX concentration (NOXi), and NOX removal efficiency (ηr) as the 5 inputs and the voltage required for the treatment (Vrt) as the output. Then, an attempt has been made to predict Vrt for achieving particular ηr, while other input parameters have fixed values, using multivariable linear and quadratic polynomial regression techniques. The performances of these two predictive models have been assessed during training and testing. The root mean square error (RMSE) between experimental and predicted values of Vrt with both the models has been compared, where quadratic polynomial regression-based model has been found to be performing well with an RMSE of 0.85kV for a test data of 1210 sets. This study shows that the voltage required for the treatment, to achieve a particular amount of NOX removal from the exhaust, can be predicted with very good accuracy so that the necessary arrangements for the treatment can be made by the researchers.