Univariate Time Series Based Back Propagation Neural Network Modeling of Air Quality Inside a Public Transportation Bus Using Available Software

作者：Kadiyala Akhil; Kumar Ashok^*

来源：Environmental Progress & Sustainable Energy, 2015, 34(2): 319-323.

DOI：10.1002/ep.12119

• back propagation
• neural networks
• time series
• ARMA
• ARIMA
• indoor air quality
• in-bus air contaminants
• biodiesel
• public transportation buses
• operational performance measures

The development of reliable and accurate indoor air quality (IAQ) models is essential to predict occupant exposures within a considered microenvironment, in addition to the assessment of ventilation design characteristics (influencing air flow rates) to ensure indoor air contaminant levels are within the permissible IAQ guidelines. Time series and artificial neural networks (ANNs) are two distinct methodologies that present environmentalists with the resources in developing valid IAQ models. Over the years, the simple structure and robustness in prediction made the use of time series attractive to environmentalists in modeling their respective databases; while, the ease of modeling complex nonlinear multivariate environmental databases made ANNs gain prominence in the last 2 decades. The use of time series and ANNs together though has not been extensively examined in the field of environmental engineering and science. This software review article presents a methodology that combines the use of univariate time series and back propagation neural network (widely used ANN) methods in the development and evaluation of IAQ models for the monitored contaminants of carbon dioxide and carbon monoxide inside a public transportation bus using available software.

• 出版日期2015-4