Prediction of Indoor PM2.5 Index Using Genetic Neural Network Model
Since people spend more than 80% of the daytime in indoor environment every day, the effect on people’s health of the indoor PM2.5 is much greater than outdoor PM2.5. This paper proposes a method based on genetic neural network to predict the indoor PM2.5. We use seven features including indoor ventilation rate, air temperature, relative humidity and others to train the model. The experiment results showed that the relative error is 5.60%, which is 7.55% lower than the traditional artificial neural network, 5.98% lower than the support vector regression method, 8.36% lower than the Random Forest.
KeywordsIndoor PM2.5 Genetic algorithm Ventilation rate
This paper is supported by the National Natural Science Foundation of China (61772357, 61502329, 61672371), Jiangsu 333 talent project and top six talent peak project (DZXX-010), Suzhou Foresight Research Project (SYG201704, SNG201610) and Postgraduate Research & Practice Innovation Program of Jiangsu Province (SJCX17_0680).
- 2.Cincinelli, A., Martellini, T.: Indoor air quality and health. Int. J. Environ. Res. Public Health 14(11), 4535–4564 (2017)Google Scholar
- 5.Huang, Y., Yuan, X.: Present situation and development of indoor PM2.5 pollution control. Shanxi Archit. 11(2), 85–90 (2017)Google Scholar
- 6.Kuang, C.L.: Influence of relative humidity on real-time measurement of indoor PM2.5 concentration. Environ. Sci. Technol. 40(1), 107–111 (2017)Google Scholar
- 7.Anders, U., Korn, O., Schmitt, C.: Improving the pricing of options: a neural network approach. J. Forecast. 17(5), 369–388 (2016)Google Scholar