Air Quality Forecasting using LSTM RNN and Wireless Sensor Networks

Belavadi, Sagar V.; Rajagopal, Sreenidhi; Ranjani, R.; Mohan, Rajasekar

doi:10.1016/j.procs.2020.03.036

Cited by 58 publications

(19 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A variety of influencing factors are considered from the time dimension and space dimension to improve the predictive ability. Sagar V Belavadi [15] used a scalable architecture to monitor and gather real-time air pollutant concentration data from wireless sensor network in various places and to forecast future air pollutants concentrations. Xiaotong Sun [16] established a spatio-temporal GRU-based (Gated Recurrent Units) prediction framework which takes the spatial information into consideration to predict PM2.5 concentrations in the hour scale.…”

Section: Prediction On Spatio-temporal Factorsmentioning

confidence: 99%

“…The prediction models based on deep learning [9][10][11][12][13][14] can extract the features existing in the air quality data and can achieve higher prediction accuracy. Some methods [15][16][17][18][19][20][21][22][23][24][25] simulate the temporal and spatial dependence of air quality data at the same time. But widely-used machine learning methods often suffer from high variability in performance in different circumstances.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Air Quality Prediction Based on Integrated Dual LSTM Model

Chen

Guan

2021

IEEE Access

View full text Add to dashboard Cite

Air quality prediction is an important reference for meteorological forecast and air controlling, but over fitting often occurs in prediction algorithms based on a single model. Aiming at the complexity of air quality prediction, a prediction method based on integrated dual LSTM (Long Short-Term Memory) model was proposed in this paper. Firstly, the Seq2Seq (Sequence to Sequence) technology is used to establish a single-factor prediction model which can obtain the predicted value of each component in air quality data, independently. Each component of air quality is regarded as time series data in the forecasting process. Then, the LSTM model with attention mechanism is used as the multi-factor prediction model. The influencing factors of air quality, like the data of neighboring stations and weather data, are considered in the model. Finally, XGBoosting (eXtreme Gradient Boosting) tree is used to integrate two models. The final prediction results can be obtained by accumulating the predicted values of the optimal subtree nodes. Through evaluation and analysis using five evaluation methods, the proposed method has better performance in terms of error and model expression power. Compared with other various models, the precision of prediction data has been greatly improved in our model.

show abstract

Section: Prediction On Spatio-temporal Factorsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Air Quality Prediction Based on Integrated Dual LSTM Model

Chen

Guan

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…As for the hidden layers, they allow adjustment and minimize the differences between input and output. The challenge for neural network users is to find the optimal size (number of hidden layers and number of neurons) as well as the best activation function for each application [3,5]. The output layer represents the parameter that we want to model, calculated by the following formula:…”

Section: A Narx (Non-linear Autoregressive Neural Network With Multip...mentioning

confidence: 99%

The Use of a Recurrent Neural Network for Forecasting Ozone Concentrations in the City of Agadir (Morocco)

Adnane

Leghrib

Chaoufi

et al. 2020

Jour. Atom. Mol. Conden. Matt. Nano Physics

View full text Add to dashboard Cite

Air quality is a complex issue which depends mutually on source emission, land topography, meteorological parameters and used mathematical tools for forecasting its dispersion. One of major toxic gas is ozone which could be dangerous to human health. The present work has been developed to forecast ozone concentrations in the city of Agadir using a Recurrent Neural Network (RNN). Predicting ozone concentrations will provide useful information, especially, for decision makers in order to prevent and reduce the ozone human health impacts. The data was collected in the most polluted area in the city using a mobile monitoring station during a period of 60 days. We have tested different neural network architectures and we found that the 1-hour forecast model, whose input parameters are a combination of meteorological parameters as well as CO and SO 2 , give the most optimal results. Coefficient of Correlation (CC), Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) were used to evaluate the statistical agreement between observed and predicted values. The model successfully predicts the ozone concentration by a bias of 4 µg.m −3 over 24 hours and which a correlation coefficient is more than 80%. This work highlights the ability of the recurrent neural networks to forecast air pollutant concentrations in urban areas.

show abstract

“…RNN is made of high dimension hidden states with non-linearity units that are able to learn complex input-output relations and extract features automatically since it includes a input association that permits the past data to pass and hold on [6]. Some of the time-series applications fields where RNN is applied include speech recognition [7], sensor calibration [8,9], machine translation [10], buildings' energy demand forecasting [11,12], air quality prediction [13] and stock market [14,15]. In general, the development of a forecasting model using any neural network model (such as RNN) consists of several steps and can be illustrated in the conceptual Figure 1.…”

Section: Introductionmentioning

confidence: 99%

Time-Lag Selection for Time-Series Forecasting Using Neural Network and Heuristic Algorithm

Surakhi

Zaidan²,

Fung³

et al. 2021

Electronics

View full text Add to dashboard Cite

The time-series forecasting is a vital area that motivates continuous investigate areas of intrigued for different applications. A critical step for the time-series forecasting is the right determination of the number of past observations (lags). This paper investigates the forecasting accuracy based on the selection of an appropriate time-lag value by applying a comparative study between three methods. These methods include a statistical approach using auto correlation function, a well-known machine learning technique namely Long Short-Term Memory (LSTM) along with a heuristic algorithm to optimize the choosing of time-lag value, and a parallel implementation of LSTM that dynamically choose the best prediction based on the optimal time-lag value. The methods were applied to an experimental data set, which consists of five meteorological parameters and aerosol particle number concentration. The performance metrics were: Mean Absolute Error (MAE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and R-squared. The investigation demonstrated that the proposed LSTM model with heuristic algorithm is the superior method in identifying the best time-lag value.

show abstract

Air Quality Forecasting using LSTM RNN and Wireless Sensor Networks

Cited by 58 publications

References 10 publications

Air Quality Prediction Based on Integrated Dual LSTM Model

Air Quality Prediction Based on Integrated Dual LSTM Model

The Use of a Recurrent Neural Network for Forecasting Ozone Concentrations in the City of Agadir (Morocco)

Time-Lag Selection for Time-Series Forecasting Using Neural Network and Heuristic Algorithm

Contact Info

Product

Resources

About