Search

Accurately predicting the spatial-temporal distribution of PM2.5 is challenging due to missing data and selecting an appropriate modeling method. Effective imputation of missing data must consider the relationships between variables while preserving their inherent variability and uncertainty. In this study, we employed machine learning techniques to impute missing data by analyzing the relationships between meteorological variables and other pollutants. Subsequently, we introduced an innovative spatiotemporal hybrid model, AC_GRU, which integrates a one-dimensional convolutional neural network (CNN), GRU, and an attention-based network to predict PM2.5 concentrations in urban areas. The AC_GRU model utilizes meteorological variables, PM2.5 concentrations from nearby air quality monitoring stations, and concentrations of other pollutants as inputs. This approach allows the model to learn spatiotemporal correlations within the time-series data, enhancing the accuracy of PM2.5 predictions. Additionally, the attention mechanism improves prediction accuracy by automatically weighting the past input variables based on their importance for future PM2.5 predictions. The experimental results demonstrate that our AC_GRU model outperforms state-of-the-art methods, making it a valuable tool for urban air quality management and public health protection. [ABSTRACT FROM AUTHOR]

By accurate predicting of pipe bursts, it is possible to schedule pipe maintenance, rehabilitation and improve level of services in Water Distribution Networks (WDNs). In recent years, artificial intelligence techniques have been used vastly in pipe burst prediction. These techniques need big input dataset for deep learning but big dataset might not be available in old WDNs. This paper aimed to compare the performance of multivariate adaptive regression splines (MARS) and regressive convolution neural network with least square support vector regression (RCNN-LSSVR) for predicting of Pipe Burst Rate (PBR) with limited dataset. Pipes age, diameter, depth of installation, length, average and maximum hydraulic pressure are considered as effective parameters of PBR. Data collected from real-case study in Iran include 158 cases for polyethylene and 124 cases for asbestos cement pipes during 2012–2019. The results indicate that RCNN-LSSVR model has a great performance of PBR prediction on limited dataset. [ABSTRACT FROM AUTHOR]

Building recognition is a core task for urban image classification (mapping), especially in optical high-resolution imagery. Convolutional Neural Networks (CNNs) have recently achieved unprecedented performance in the automatic recognition of objects (e.g. buildings, roads, or trees) in high-resolution imagery. Although these results are promising, questions remain about generalizability. This is a great challenge, as there is a wide variability in the visual characteristics of the building image scene across different geographic locations. CNNs are overfitted with limited and low diversity samples and are tested on the same or nearby geographic locations. In this work, we propose two scenarios with regard to transfer learning CNN features for building scene classification. We also investigate the generalizability of CNNs for building recognition across different geographic locations. The results of the two scenarios show that the final model, generalizable in different geographic locations, unseen areas. [ABSTRACT FROM AUTHOR]

Background: Evidence concerning the impact of long-term exposure to fine Particulate Matter ≤2.5 μm (PM2.5) on Cardio-Vascular Diseases (CVDs) for those people subject to ambient air pollution in developing countries remains relatively scant. This study assessed the relationship of 15-year PM2.5 exposure with cardiovascular incidence and mortality rate in Isfahan province, Iran.Methods: The cohort comprised 3081 participants over 35 years old who were free of CVDs. They were selected through multi-stage cluster sampling in Isfahan, Iran. PM2.5 exposure was determined separately for each individual via satellite-based spatiotemporal estimates according to their residential addresses. In this context, CVD is defined as either fatal and non-fatal Acute Myocardial Infarctions (AMI) or stroke and sudden cardiac death. The incidence risk for CVD and the ensuing mortality was calculated based on the average PM2.5 exposure within a study period of 15 years using the Cox proportional hazards frailty model upon adjusting individual risk factors. The mean annual rate of PM2.5 and the follow-up data of each residential area were combined.Results: Mean three-year PM2·5 exposure for the cohort was measured at 45.28 μg/m3, ranging from 20.01 to 69.80 μg/m3. The median time period for conducting necessary follow-ups was 12.3 years for the whole population. Notably, 105 cardiovascular and 241 all-cause deaths occurred among 393,786 person-months (27 and 61 per 100,000 person-months, respectively). In well-adjusted models, 10 μg/m3 increase in PM2.5 corresponded to a 3% increase in the incidence rate of CVDs [0.95 CI = 1.016, 1.036] (in case of p = 0.000001 per 10 μg/m3 increase in PM2.5, the Hazard Ratio (HR) for AMI and Ischemic Heart Disease (IHD) was 1.031 [0.95 CI = 1.005, 1.057] and 1.028 [0.95 CI = 1.017, 1.039]), respectively. No consistent association was observed between PM2.5 concentration and fatal CVD (fatal AMI, fatal stroke, SCD (Sudden Cardiac Death)) and all-cause mortality.Conclusions: Results from analyses suggest that the effect of PM2.5 on cardiovascular disease occurrence was stronger in the case of older people, smokers, and those with high blood pressure and diabetes. The final results revealed that long-term exposure to ambient PM2.5 with high concentrations positively correlated with IHD incidence and its major subtypes, except for mortality. The outcome accentuates the need for better air quality in many countries. [ABSTRACT FROM AUTHOR]

Standard penetration test (SPT) is one of the most widely used tools to predict the soil properties. In recent years, the dynamic probing light (DPL) test has been performed more frequently for geotechnical applications because it is more cost-effective and fast. Since the majority of empirical equations of soil properties are related to SPT N-value, it is beneficial to find the best correlation between SPT and DPL N-values. In this study, the adaptive neuro-fuzzy inference system (ANFIS) and multiple regression (MR) are used to predict the correlation between SPT and DPL N-values. To achieve this goal, the soil properties of 64 sample specimens of silty clay were calculated at various depths. Three parameters including depth, total density, and DPL N-value were chosen as the input of the models. Results show that both methods estimate the correlation between SPT and DPL N-values precisely. However, ANFIS predicts more accurately than multiple regression. [ABSTRACT FROM AUTHOR]

By accurate predicting of pipe bursts, it is possible to schedule pipe maintenance, rehabilitation and improve level of services in water distribution networks (WDNs). In this study we aimed to implement five artificial intelligence and machine learning regression models such as multivariate adaptive regression splines (MARS), M5' regression tree (M5'), Least square support vector regression (LS-SVR), fuzzy regression based on c-means clustering (FCMR) and regressive convolution neural network with support vector regression (RCNN-SVR) for predicting pipe burst rate and evaluating the performance of these models. The most effective parameters for regression models are pipes age, diameter, depth of installation, length, average and maximum hydraulic pressure. In the present study, collected data include 158 cases for polyethylene (PE) and 124 cases for asbestos cement (AC) pipes during 2012-2019. The results indicate that RCNN-SVR model has a great performance of pipe burst rate (PBR) prediction. [ABSTRACT FROM AUTHOR]