Effective Monitoring of an air quality network

Air pollution in urban areas could be considered as one of the most dangerous types of pollution that can cause impact health and the ecosystem. Hence, monitoring air quality networks has captivated the interest of various research studies. In this context, this paper deals with Fault Detection of an Air Quality Monitoring Network. The proposed approach is based on nonlinear principal component analysis to cope with modeling of nonlinear data. In addition, the fault detection would be improved by combining exponentially weighted moving average with hypothesis testing technique: generalized likelihood ratio test. The evaluation was carried out on an Air Quality Monitoring Network (AQMN). The results revealed a good results compared to the classical PCA.