Multi-Horizon Air Pollution Forecasting with Deep Neural Networks

Arsov, Mirche; Zdravevski, Eftim; Lameski, Petre; Corizzo, Roberto; Koteli, Nikola; Gramatikov, Sasho; Mitreski, Kosta; Trajkovik, Vladimir

doi:10.3390/s21041235

Cited by 27 publications

(13 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The forecast horizons are selected as 1, 6, 12, and 24 h ahead. This horizon configuration is based on the daily periodicity of the time series and is commonly adopted by recent works (Arsov et al, 2021;He et al, 2022). In adopted for training, with the batch size equal to 128 and a max epoch of 300.…”

Section: Dataset and Evaluation Settingmentioning

confidence: 99%

A Transformer-Based Multi-Entity Load Forecasting Method for Integrated Energy Systems

Wang

Zhu²,

Xiao³

et al. 2022

Front. Energy Res.

View full text Add to dashboard Cite

Energy load forecasting is a critical component of energy system scheduling and optimization. This method, which is classified as a time-series forecasting method, uses prior features as inputs to forecast future energy loads. Unlike a traditional single-target scenario, an integrated energy system has a hierarchy of many correlated energy consumption entities as prediction targets. Existing data-driven approaches typically interpret entity indexes as suggestive features, which fail to adequately represent interrelationships among entities. This paper, therefore, proposes a neural network model named Cross-entity Temporal Fusion Transformer (CETFT) that leverages a cross-entity attention mechanism to model inter-entity correlations. The enhanced attention module is capable of mapping the relationships among entities within a time window and informing the decoder about which entity in the encoder to concentrate on. In order to reduce the computational complexity, shared variable selection networks are adapted to extract features from different entities. A data set obtained from 13 buildings on a university campus is used as a case study to verify the performance of the proposed approach. Compared to the comparative methods, the proposed model achieves the smallest error on most horizons and buildings. Furthermore, variable importance, temporal correlations, building relationships, and time-series patterns in data are analyzed with the attention mechanism and variable selection networks, therefore the rich interpretability of the model is verified.

show abstract

Section: Dataset and Evaluation Settingmentioning

confidence: 99%

A Transformer-Based Multi-Entity Load Forecasting Method for Integrated Energy Systems

Wang

Zhu²,

Xiao³

et al. 2022

Front. Energy Res.

View full text Add to dashboard Cite

show abstract

“…It was found that exogenous variables like weather parameters have shown considerable improvement in performance. LSTMs have also been used in traffic forecasting (Awan et al, 2020) and pollution classification (Arsov et al, 2021). Our research compares different machine learning models ranging from linear regression to multiple kernel based SVR techniques with both traditional mathematical models like ARIMA and the popular LSTM based deep learning models.…”

Section: Related Workmentioning

confidence: 99%

Data-Driven Framework for Understanding and Predicting Air Quality in Urban Areas

et al. 2022

View full text Add to dashboard Cite

Monitoring, predicting, and controlling the air quality in urban areas is one of the effective solutions for tackling the climate change problem. Leveraging the availability of big data in different domains like pollutant concentration, urban traffic, aerial imagery of terrains and vegetation, and weather conditions can aid in understanding the interactions between these factors and building a reliable air quality prediction model. This research proposes a novel cost-effective and efficient air quality modeling framework including all these factors employing state-of-the-art artificial intelligence techniques. The framework also includes a novel deep learning-based vegetation detection system using aerial images. The pilot study conducted in the UK city of Cambridge using the proposed framework investigates various predictive models ranging from statistical to machine learning and deep recurrent neural network models. This framework opens up possibilities of broadening air quality modeling and prediction to other domains like vegetation or green space planning or green traffic routing for sustainable urban cities. The research is mainly focused on extracting strong pieces of evidence which could be useful in proposing better policies around climate change.

show abstract

“…In the experiments, the authors predicted PM 2.5 for the next 1-8 h. The results showed the superiority of the proposed approach against SVR, GBT, and a standard LSTM model. Another RNN based model was proposed in [24] to forecast the concentration level of PM 10 at different future time steps (6, 12, and 24 h). On the other hand, Guo et al [25] proposed a feature engineering pipeline as well as a deep ensemble network algorithm which combines RNN, LSTM, and GRU networks to predict the PM 2.5 concentration of the next hour.…”

Section: Introductionmentioning

confidence: 99%

Attention-Based Distributed Deep Learning Model for Air Quality Forecasting

et al. 2022

View full text Add to dashboard Cite

Air quality forecasting has become an essential factor in facilitating sustainable development worldwide. Several countries have implemented monitoring stations to collect air pollution particle data and meteorological information using parameters such as hourly timespans. This research focuses on unravelling a new framework for air quality prediction worldwide and features Busan, South Korea as its model city. The paper proposes the application of an attention-based convolutional BiLSTM autoencoder model. The proposed deep learning model has been trained on a distributed framework, referred to data parallelism, to forecast the intensity of particle pollution ( and ). The algorithm automatically learns the intrinsic correlation among the particle pollution in different locations. Each location’s meteorological and traffic data is extensively exploited to improve the model’s performance. The model has been trained using air quality particle data and car traffic information. The traffic information is obtained by a device which counts cars passing a specific area through the YOLO algorithm, and then sends the data to a stacked deep autoencoder to be encoded alongside the meteorological data before the final prediction. In addition, multiple one-dimensional CNN layers are used to obtain the local spatial features jointly with a stacked attention-based BiLSTM layer to figure out how air quality particles are correlated in space and time. The evaluation of the new attention-based convolutional BiLSTM autoencoder model was derived from data collected and retrieved from comprehensive experiments conducted in South Korea. The results not only show that the framework outperforms the previous models both on short- and long-term predictions but also indicate that traffic information can improve the accuracy of air quality forecasting. For instance, during prediction, the proposed attention-based model obtained the lowest MAE (5.02 and 22.59, respectively, for short-term and long-term prediction), RMSE (7.48 and 28.02) and SMAPE (17.98 and 39.81) among all the models, which indicates strong accuracy between observed and predicted values. It was also found that the newly proposed model had the lowest average training time compared to the baseline algorithms. Furthermore, the proposed framework was successfully deployed in a cloud server in order to provide future air quality information in real time and when needed.

show abstract

Multi-Horizon Air Pollution Forecasting with Deep Neural Networks

Cited by 27 publications

References 49 publications

A Transformer-Based Multi-Entity Load Forecasting Method for Integrated Energy Systems

A Transformer-Based Multi-Entity Load Forecasting Method for Integrated Energy Systems

Data-Driven Framework for Understanding and Predicting Air Quality in Urban Areas

Attention-Based Distributed Deep Learning Model for Air Quality Forecasting

Contact Info

Product

Resources

About