A Comparative Study of Different Artificial Intelligence Models and Response Surface Methodology for Heptachlor Removal Using Fe/Cu Nanoparticles

Abstract

Water is a basic and essential resource in the human body. Every structure in the body including cells, tissues and organs needs water to work properly. Hence, without food, humans can last up to several weeks, but just a few days without water. Meanwhile, before water is consumed by a human’s body, harmful impurities such as heptachlor which is a highly toxic organochlorine compound in the water must be removed. To remove heptachlor from the wastewater, the adsorption using the bimetallic iron/cupper (Fe/Cu) nanoparticles can be a solution. However, the effectiveness of the elimination of heptachlor using the Fe/Cu nanoparticles could be affected by environmental factors including pH, adsorbent dosage, contact time, initial adsorbate concentration, and stirring rate. Response surface methodology (RSM) is widely used to correlate these factors with the heptachlor removal efficiency to achieve performance optimisation. However, the artificial intelligence models may perform better than RSM to optimise the heptachlor removal process. Therefore, this study aims to compare the performance of different artificial intelligence models with RSM for heptachlor removal using Fe/Cu nanoparticles. These different artificial intelligence models include principal component regression (PCR), artificial neural network (ANN), locally weighted kernel partial least square regression (LW-KPLSR), partial least square regression (PLSR), and least-square support vector regression (LSSVR). Based on the obtained results, the LW-KPLSR model performed better than other artificial intelligence models and RSM. Its root means square error, and mean absolute error are around 159% to 3,297% lower than other models and RSM. Moreover, its coefficient of determination which is so-called R2 is the highest among others. These results denote that LW-KPLSR is more convincing as compared to RSM to predict optimum performance of heptachlor removal.