Optimizing the prediction models of the air quality state in cities

Skrzypski, J.; Jach-Szakiel, E.; Kami, W.; ski,

doi:10.2495/air070091

Cited by 4 publications

(2 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The previously conducted research of neural models was used in aspect of the optimisation of structures and functioning of models [17]. The optimisation concerned the possibility of simplification of the model architecture by: a. the qualitative and quantitative selection of input vector structure including the assessment of transformation effects using the Method Principal Component Analysis (PCA).…”

Section: Neural Model Structurementioning

confidence: 99%

Neural Models for Prediction of Maximum Daily Particulate Matter PM10 Concentration in the Air in Big Cities as Ecological Safety Management Tools

Skrzypski

Jach-Szakiel

Kamiński

2008

2008 19th International Conference on Systems Engineering

Self Cite

View full text Add to dashboard Cite

The aim of the study was to examine the possibilities of the development of a prognostic instrument for the air quality management in cities. The study was focused on the development of the neural network models for prediction of the classes of the air quality state in relation to maximum daily dust PM 10 concentration. The air quality class was predicted for the next day in relation to maximal daily concentrations. The models MLP and RBF were tested.The tests were carried out in the city of Lodz in central Poland. The results of the modelling were satisfactory. In the optimally constructed models false prognosis (in testing series) were only 7.4% in the case of predicting maximal daily concentration (test series) and 2.7% (training series). A low level of error prediction confirmed the fact, that the neural network models is an effective instrument of the air quality management in cities.

show abstract

Section: Neural Model Structurementioning

confidence: 99%

Neural Models for Prediction of Maximum Daily Particulate Matter PM10 Concentration in the Air in Big Cities as Ecological Safety Management Tools

Skrzypski

Jach-Szakiel

Kamiński

2008

2008 19th International Conference on Systems Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

“…These models are simpler, lighter, and easier to implement, and when enough historical data are provided, statistical models show better forecast results for specific locations. Popular statistical models for predicting the air quality include artificial neural networks [16][17][18][19][20][21][22][23][24], multiple linear regression [25], autoregressive integrated moving average (ARIMA) [26], support vector machine (SVM) [27], nonlinear regression [28] and random forest [29]. Neural networks (NN) are a popular choice for air pollution forecasting, since they do not require any prior assumptions about the data distribution, and are capable of modeling complex nonlinear processes.…”

Section: Introductionmentioning

confidence: 99%

A Hybrid Spatiotemporal Deep Model Based on CNN and LSTM for Air Pollution Prediction

2022

View full text Add to dashboard Cite

Nowadays, air pollution is an important problem with negative impacts on human health and on the environment. The air pollution forecast can provide important information to all affected sides, and allows appropriate measures to be taken. In order to address the problems of filling in the missing values in the time series used for air pollution forecasts, the automation of the allocation of optimal subset of input variables, the dependency of the air quality at a particular location on the conditions of the surrounding environment, as well as automation of the model’s optimization, this paper proposes a deep spatiotemporal model based on a 2D convolutional neural network and a long short-term memory network for predicting air pollution. The model utilizes the automatic selection of input variables and the optimization of hyperparameters by a genetic algorithm. A hybrid strategy for missing value imputation is used based on a combination of linear interpolation and a strategy of using the average between the previous value and the average value for the same time in other years. In order to determine the best architecture of the spatiotemporal model, the architecture hyperparameters are optimized by a genetic algorithm with a modified crossover operator for solutions with variable lengths. Additionally, the trained models are included in various ensembles in order to further improve the prediction performance—these include ensembles of models with the same architecture comprising the best architecture obtained by the evolutionary optimization, and ensembles of diverse models comprising the k best models of the evolutionary optimization. The experimental results for the Beijing Multi-Site Air-Quality Data Set show that the proposed spatiotemporal model for air pollution forecasting provides good and consistent prediction results. The comparison of the suggested model with other deep NN models shows satisfactory results, with the best performance according to MAE, based on the experimental results for the station at Wanliu (16.753 ± 0.384). Most of the model architectures obtained by the optimization of the model hyperparameters using the genetic algorithm have one convolutional layer with a small number of kernels and a small kernel size; the convolutional layers are followed by a max-pooling layer, and one or two LSTM layers are utilized with dropout regularization applied to the LSTM layer using small values of p (0.1, 0.2 and 0.3). The utilization of ensembles from the k best trained models further improves the prediction results and surpasses other deep learning models, according to MAE and RMSE metrics. The used hybrid strategy for missing value imputation enhances the results, especially for data with clear seasonality, and produces better MAE compared to the strategy using average values for the same hour of the same day and month in other years. The experimental results also reveal that random searching is a simple and effective strategy for selecting the input variables. Furthermore, the inclusion of spatial information in the model’s input data, based on the local neighborhood data, significantly improves the predictive results obtained with the model. The results obtained demonstrate the benefits of including spatial information from as many surrounding stations as possible, as well as using as much historical information as possible.

show abstract

Multivariate air pollution classification in urban areas using mobile sensors and self-organizing maps

Oliveira

Carneiro

Filho

et al. 2018

Int. J. Environ. Sci. Technol.

View full text Add to dashboard Cite

Optimizing the prediction models of the air quality state in cities

Cited by 4 publications

References 7 publications

Neural Models for Prediction of Maximum Daily Particulate Matter PM10 Concentration in the Air in Big Cities as Ecological Safety Management Tools

Neural Models for Prediction of Maximum Daily Particulate Matter PM10 Concentration in the Air in Big Cities as Ecological Safety Management Tools

A Hybrid Spatiotemporal Deep Model Based on CNN and LSTM for Air Pollution Prediction

Multivariate air pollution classification in urban areas using mobile sensors and self-organizing maps

Contact Info

Product

Resources

About