Exploring a deep LSTM neural network to forecast daily PM2.5 concentration using meteorological parameters in Kathmandu Valley, Nepal

Dhakal, Sandeep; Gautam, Yogesh; Bhattarai, Aayush

doi:10.1007/s11869-020-00915-6

Cited by 26 publications

(20 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similarly, the information not relevant are eliminated using the forget gate (Γf ) while the output gate (Γo) determines the output of the recurrent cell [15]. The current state is defined as…”

Section: Long Short Term Memorymentioning

confidence: 99%

Forecasting tropospheric wet delay using LSTM neural network

Ogunjo

Dada

Ajayi

2022

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

Tropospheric wet delay is a critical factor in radio communication. Accurate estimation of the wet delay is difficult due to the variability in water vapour. In this study, we aim to model and predict tropospheric wet delay over four tropical locations using Long Short-Term Memory (LSTM) neural network. Results obtained in this study showed RMSE and MAE within the range 18.96–21.16 and 14.08–16.38 respectively. LSTM model was able to capture the different regimes of wet delays in each of the locations under consideration. This approach can significantly improve link budget and planning within tropical regions.

show abstract

Section: Long Short Term Memorymentioning

confidence: 99%

Forecasting tropospheric wet delay using LSTM neural network

Ogunjo

Dada

Ajayi

2022

IOP Conf. Ser.: Earth Environ. Sci.

View full text Add to dashboard Cite

show abstract

“…As changes in air quality were mainly caused by human activities and natural factors, the contribution of meteorological factors to air pollutants was estimated based on the assumption that human activities remain unchanged during the period. The machine learning algorithm (LSTM neural network) was used in much research and proved its efficiency and accuracy [33,34]. In this paper, the LSTM neural network model was proposed to forecast daily pollutants concentration in Guangdong Province by employing historical pollutant (PM 2.5 , PM 10 , NO 2 , O 3 ) concentrations and meteorological parameters (temperature, relative humidity, wind speed, precipitation, and wind direction) as inputs.…”

Section: The Impact Of Covid-19 Control Measures On Air Qualitymentioning

confidence: 99%

The Impact of COVID-19 Control Measures on Air Quality in Guangdong Province

Mao

et al. 2022

Sustainability

View full text Add to dashboard Cite

COVID-19 control measures had a significant social and economic impact in Guangdong Province, and provided a unique opportunity to assess the impact of human activities on air quality. Based on the monitoring data of PM2.5, PM10, NO2, and O3 concentrations from 101 air quality monitoring stations in Guangdong Province from October 2019 to April 2020, the PSCF (potential source contribution factor) analysis and LSTM (long short-term memory) neural network were applied to explore the impact of epidemic control measures on air quality in Guangdong Province. Results showed that during the lockdown, the average concentration of PM2.5, PM10, NO2, and O3 decreased by 37.84%, 51.56%, 58.82%, and 24.00%, respectively. The ranges of potential sources of pollutants were reduced, indicating that air quality in Guangdong Province improved significantly. The Pearl River Delta, characterized by a high population density, recorded the highest NO2 concentration values throughout the whole study period. Due to the lockdown, the areas with the highest concentrations of O3, PM2.5, and PM10 changed from the Pearl River Delta to the eastern and western Guangdong. Moreover, LSTM simulation results showed that the average concentration of PM2.5, PM10, NO2, and O3 decreased by 46.34%, 54.56%, 70.63%, and 26.76%, respectively, which was caused by human-made impacts. These findings reveal the remarkable impact of human activities on air quality and provide effective theoretical support for the prevention and control of air pollution in Guangdong Province.

show abstract

“…Therefore, the time series forecasting model could be represented as a data-based model that aims to study the relationship between observations of data flows from past events or history to the current or future. Dhakal et al [3] used LSTM to forecast accurate particulate matter (PM2.5), while Mussumeci and Coelho [4] applied LSTM to predict dengue in 790 cities in Brazil and reported that the model is better than other machine learning approaches, such as LASSO and Random Forest. Further, Dubey et al [5] compared LSTM to ARIMA (autoregressive integrated moving average) and SARIMA (seasonal ARIMA) in the forecasting case to the daily energyconsumption time series data.…”

Section: Introductionmentioning

confidence: 99%

Robust LSTM With Tuned-PSO and Bifold-Attention Mechanism for Analyzing Multivariate Time-Series

et al. 2022

View full text Add to dashboard Cite

The need for accurate time-series results is badly demanding. LSTM has been applied for forecasting time series, which is generated when variables are observed at discrete and equal time intervals. Nevertheless, the problem of determining hyperparameters with a relatively high random rate will reduce the accuracy of the prediction results. This paper aims to promote LSTM with tuned-PSO and Bifold-Attention mechanism. PSO optimizes LSTM hyperparameters, and Bifold-attention mechanism selects the optimal input for LSTM. An accurate, adaptive, and robust time-series forecasting model is the main contribution, compared with ARIMA, MLP, LSTM, PSO-LSTM, A-LSTM, and PSO-A-LSTM. The model comparison is based on the accuracy of each model in forecasting Beijing PM2.5, Beijing Multi-Site, Air Quality, Appliances Energy, Wind Speed, and Traffic Flow. Proposed PSO Bifold-Attention LSTM has lower MAPE and RMSE than baselines. In other words, the model outperformed all LSTM base models in this study. The proposed model's accuracy is adaptable in daily, weekly, and monthly multivariate time-series datasets. This ground-breaking innovation is valuable for time-series analysis research, particularly the implementation of deep learning for time-series forecasting.

show abstract

Exploring a deep LSTM neural network to forecast daily PM2.5 concentration using meteorological parameters in Kathmandu Valley, Nepal

Cited by 26 publications

References 25 publications

Forecasting tropospheric wet delay using LSTM neural network

Forecasting tropospheric wet delay using LSTM neural network

The Impact of COVID-19 Control Measures on Air Quality in Guangdong Province

Robust LSTM With Tuned-PSO and Bifold-Attention Mechanism for Analyzing Multivariate Time-Series

Contact Info

Product

Resources

About