Research Article Open Access

COMPARATIVE STUDY BETWEEN GENETIC ALGORITHM AND NEURAL NETWORK COUPLING AND MONITORING TOOLS ON THE DISPERSION OF POLLUTANTS FROM THE YASSA-DIBAMBA THERMAL POWER PLANT

Abstract

Prediction plays an important role in air quality monitoring. However, nowadays, the main difficulty is to accurately predict the impact of pollution sources on the environment due to atmospheric dispersion models that are not accurate enough. To solve these problems, this paper proposes a coupling between an optimisation model and a prediction model. In this work, a Genetic Algorithm Coupled With Neural Networks (GA-ANN) was used to predict the concentration of pollutants in the Yassa region. Thus, the genetic algorithm was used as an objective function optimisation tool and the neural networks were used as a data learning tool to predict the concentration values. The model takes into account the meteorological parameters of the study area and the source over 5 years, from January 2017 to December 2021. To evaluate the model, two indices are used to indicate the performance of the prediction model; the squared correlation coefficient R2 whose value in the test case is about 0.88 and the Root Mean Square Error (RMSE) whose best value is about 0.0044. We also evaluated the optimisation methods and found that compared to the particle optimisation swarm, the genetic algorithm gives a better “Fitness Function Curve” as a function of the number of iterations. The results show that the GA-ANN coupling is more accurate and efficient in estimating pollutant concentration values than CFD and the Gaussian model

A. C. GOUNE1, J. C. SEUTCHE 2*, R. Y. EKANI 3, B. E. ESSOMBO 4, J. L. NSOUANDELE 5, G. H. BEN-BOLIE 6

To read the full article Download Full Article | Visit Full Article