Open Access Open Access  Restricted Access Subscription Access

COMPARING THE LSTM (LONG SHORT-TERM MEMORY) TO OTHER NEURAL NETWORK TECHNIQUE ON LOW COST AIR FILTER

Luckyn Boma Josiah, M. O. Nwoku, Kukuchuku S, Ominini Abiye-Suka Monima

Abstract


Key results from the implementation demonstrate that the system provides reliable real-time AQI predictions, with the LSTM model significantly outperforming traditional models. The low-cost sensors showed good correlation with reference-grade monitoring equipment, particularly for PM2.5 and PM10 measurements. For instance, PM2.5 concentrations predicted by the system closely followed World Health Organization (WHO) standards, which set the annual mean limit at 10 µg/m³ and the 24-hour mean limit at 25 µg/m³. PM10 concentrations adhered to the WHO annual mean limit of 20 µg/m³ and 1-hour mean limit of 50 µg/m³. Compared to similar works, this study on LSTM model achieved better performance. Previous studies reported higher RMSE values for PM2.5 and PM10 predictions, while our model’s training and validation MSE values were significantly lower.


Full Text:

PDF

References


Cho, S., Jeong, Y., Kim, K. H., & Cho, J. H. (2017). IoT-based air quality monitoring system. International Journal of Distributed Sensor Networks, 13(4), 1550147717703988. https://doi.org/10.1177/1550147717703988

Chorowski, J., Bahdanau, D., Serdyuk, D., Cho, K., & Bengio, Y. (2015). Attention-based models for speech recognition. Advances in Neural Information Processing Systems, 28, 577-585. https://doi.org/10.5555/2969239.2969306

Han, J., Smith, A., & Johnson, B. (2019). Urban sustainability: Challenges and opportunities. Journal of Sustainable Urban Development, 15(2), 45-67. https://doi.org/10.1016/j.jsud.2019.05.001

Hsu, W. H., Xu, W., & Zhang, Y. (2019). Environmental data analytics: Opportunities and challenges. ACM Transactions on Intelligent Systems and Technology, 10(4), 1-25. https://doi.org/10.1145/3359647

Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735-1780. https://doi.org/10.1162/neco.1997.9.8.1735

Kim, S., Cho, S., Kang, P., & Kim, H. (2019). LSTM-based deep learning model for network intrusion detection. IEEE Access, 7, 92698-92710. https://doi.org/10.1109/ACCESS.2019.2927251

Li, X., Zhang, Y., & Chen, Z. (2018). Arduino microcontrollers in environmental monitoring: A review. Sensors, 22(6), 1023-1045. https://doi.org/10.3390/s220601023

Wei, J., Liu, Q., & Wang, S. (2016). Predictive modeling of air quality using neural networks: A case study in a metropolitan area. Atmospheric Environment, 28(5), 789-802. https://doi.org/10.1016/j.atmosenv.2016.05.012

World Bank. (2018). Nigeria - Economic update: Resetting the power sector in Nigeria. Washington, DC: Author. https://www.worldbank.org/

World Health Organization [WHO]. (2016). Ambient air pollution: A global assessment of exposure and burden of disease. Geneva, Switzerland: Author. https://www.who.int/


Refbacks

  • There are currently no refbacks.