Forecasting Urban Air Quality via a Back-Propagation Neural Network and a Selection Sample Rule

Liu, Yonghong; Zhu, Qing; Yao, Dawen; Xu, Wei

doi:10.3390/atmos6070891

Cited by 13 publications

(7 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…MAPE is a measure of error, thus high values suggest a bad fit, and subtracting this statistical measure from 100 gives the percentage accuracy of the model. MAPE is commonly used to evaluate the performance of obtained regression models, and is applied to air quality analysis (Cheng et al, 2014;Liu et al, 2015).…”

Section: General Trends Of Pm Concentrationmentioning

confidence: 99%

Long-term trends in PM2.5 mass and particle number concentrations in urban air: The impacts of mitigation measures and extreme events due to changing climates

Jesus

Thompson

Knibbs

et al. 2020

Environmental Pollution

View full text Add to dashboard Cite

Urbanisation and industrialisation led to the increase of ambient particulate matter (PM) concentration. While subsequent regulations may have resulted in the decrease of some PM matrices, the simultaneous changes in climate affecting local meteorological conditions could also have played a role. To gain an insight into this complex matter, this study investigated the long-term trends of two important matrices, the particle mass (PM2.5) and particle number concentrations (PNC), and the factors that influenced the trends. Mann-Kendall test, Sen's slope estimator, the generalised additive model, seasonal decomposition of time series by LOESS (locally estimated scatterplot smoothing) and the Buishand range test were applied. Both PM2.5 and PNC showed significant negative monotonic trends (0.03-0.6 µg.m -3 .yr -1 and 0.40-3.8 x 10 3 particles.cm -3 .yr -1 , respectively) except Brisbane (+0.1 µg.m -3 .yr -1 and +53 particles.cm -3 .yr -1 , respectively). For the period covered in this study, temperature increased (0.03-0.07 °C.yr -1 ) in all cities except London; precipitation decreased (0.02-1.4 mm.yr -1 ) except in Helsinki; and wind speed was reduced in Brisbane and Rochester but increased in Helsinki, London and Augsburg. At the change-points, temperature increase in cold cities influenced PNC while shifts in precipitation and wind speed affected PM2.5. Based on the LOESS trend, extreme events such as dust storms and wildfires resulting from changing climates caused a positive step-change in concentrations, particularly for PM2.5. In contrast, among the mitigation measures, controlling sulphur in fuels caused a negative step-change, especially for PNC. Policies regarding traffic and fleet management (e.g. low emission zones) that were implemented only in certain areas or in a progressive uptake (e.g. Euro emission standards), resulted to gradual reductions in concentrations. Therefore, as this study has clearly shown that PM2.5 and PNC were influenced differently by the impacts of the changing climate and by the mitigation measures, both metrics must be considered in urban air quality management.

show abstract

Section: General Trends Of Pm Concentrationmentioning

confidence: 99%

Long-term trends in PM2.5 mass and particle number concentrations in urban air: The impacts of mitigation measures and extreme events due to changing climates

Jesus

Thompson

Knibbs

et al. 2020

Environmental Pollution

View full text Add to dashboard Cite

show abstract

“…Zolghadri [ 20 ] and Hoi [ 21 ] used the KF (Kalman filter) algorithm for predicting air quality parameters. Liu [ 22 ] used a Back-Propagation Neural Network and a Selection Sample Rule for forecasting Urban Air Quality. Li Xiang [ 23 ] also used GAB and fuzzy BP neural network for Air quality forecasting.…”

Section: Introductionmentioning

confidence: 99%

Urban air quality forecasting based on multi-dimensional collaborative Support Vector Regression (SVR): A case study of Beijing-Tianjin-Shijiazhuang

et al. 2017

View full text Add to dashboard Cite

Today, China is facing a very serious issue of Air Pollution due to its dreadful impact on the human health as well as the environment. The urban cities in China are the most affected due to their rapid industrial and economic growth. Therefore, it is of extreme importance to come up with new, better and more reliable forecasting models to accurately predict the air quality. This paper selected Beijing, Tianjin and Shijiazhuang as three cities from the Jingjinji Region for the study to come up with a new model of collaborative forecasting using Support Vector Regression (SVR) for Urban Air Quality Index (AQI) prediction in China. The present study is aimed to improve the forecasting results by minimizing the prediction error of present machine learning algorithms by taking into account multiple city multi-dimensional air quality information and weather conditions as input. The results show that there is a decrease in MAPE in case of multiple city multi-dimensional regression when there is a strong interaction and correlation of the air quality characteristic attributes with AQI. Also, the geographical location is found to play a significant role in Beijing, Tianjin and Shijiazhuang AQI prediction.

show abstract

“…An MLP is a type of neural network that is widely used for forecasting applications. It comes under the category of feedforward algorithms, as inputs are combined with the initial weights in a weighted sum and subjected to the activation function [34]. In an MLP-NN, each linear combination of data is propagated to the next layer through a perceptron and multiple layers of interconnected neurons process the input data to produce the output data.…”

Section: Multilayer Perceptron Neural Network (Mlp-nn)mentioning

confidence: 99%

“…Lu et al [33] adopted an ant colony optimization (ACO) model to train the perceptron and to predict the pollutant levels. The approach proved to be feasible and effective in solving air-quality problems, particularly when compared to the simple backpropagation (BP) approach [34]. A modified ACO in conjugation with a simulated annealing technique was also studied [35].…”

Section: Introductionmentioning

confidence: 99%

Ant Colony Based Artificial Neural Network for Predicting Spatial and Temporal Variation in Groundwater Quality

Bhavya

Sivaraj

Elango

2023

Water

View full text Add to dashboard Cite

The quality of groundwater is of utmost importance, as it directly impacts human health and the environment. In major parts of the world, groundwater is the main source of drinking water, hence it is essential to periodically monitor its quality. Conventional water-quality monitoring techniques involve the periodical collection of water samples and subsequent analysis in the laboratory. This process is expensive, time-consuming and involves a lot of manual labor, whereas data-driven models based on artificial intelligence can offer an alternative and more efficient way to predict groundwater quality. In spite of the advantages of such models based on artificial neural network (ANN) and ant colony optimization (ACO), no studies have been carried out on the applications of these in the field of groundwater contamination. The aim of our study is to build an ant colony optimized neural network for predicting groundwater quality parameters. We have proposed ANN comprising of six hidden layers. The approach was validated using our groundwater quality dataset of a hard rock region located in the northern part of Karnataka, India. Groundwater samples were collected by us once every 4 months from March 2014 to October 2020 from 50 wells in this region. These samples were analyzed for the pH, electrical conductivity, Na+, Ca+, K+, Mg2+, HCO3−, F−, Cl− and U+. This temporal dataset was split for training, testing and validation of our model. Metrics such as R2 (Coefficient of Determination), RMSE (Root Mean Squared Error), NSE (Nash–Sutcliffe efficiencies) and MAE (Mean Absolute Error) were used to evaluate the prediction error and model performance. These performance evaluation metrics indicated the efficiency of our model in predicting the temporal variation in groundwater quality parameters. The method proposed can be used for prediction and it will aid in modifying or reducing the temporal frequency of sample collection to save time and cost. The study confirms that the combination of ANN with ACO is a promising tool to optimize weights while training the network, and for prediction of groundwater quality.

show abstract

Forecasting Urban Air Quality via a Back-Propagation Neural Network and a Selection Sample Rule

Cited by 13 publications

References 31 publications

Long-term trends in PM2.5 mass and particle number concentrations in urban air: The impacts of mitigation measures and extreme events due to changing climates

Long-term trends in PM2.5 mass and particle number concentrations in urban air: The impacts of mitigation measures and extreme events due to changing climates

Urban air quality forecasting based on multi-dimensional collaborative Support Vector Regression (SVR): A case study of Beijing-Tianjin-Shijiazhuang

Ant Colony Based Artificial Neural Network for Predicting Spatial and Temporal Variation in Groundwater Quality

Contact Info

Product

Resources

About