Prediction of PM2.5 concentration based on a CNN-LSTM neural network algorithm
Ontology highlight
ABSTRACT:
SUBMITTER: Bai X
PROVIDER: S-EPMC11313410 | biostudies-literature | 2024 Jan
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:The National Institute of Environmental Research, under the Ministry of Environment of Korea, provides two-day forecasts, through AirKorea, of the concentration of particulate matter with diameters of ≤ 2.5 μm (PM2.5) in terms of four grades (low, moderate, high, and very high) over 19 districts nationwide. Particulate grades are subjectively designated by human forecasters based on forecast results from the Community Multiscale Air Quality (CMAQ) and artificial intelligence (AI) models in conjunction with weather patterns. This study evaluates forecasts from the long short-term memory (LSTM) algorithm relative to those from CMAQ-solely and AirKorea using observations from 2019. The skills of the one-day PM2.5 forecasts over the 19 districts were 39-70% for CMAQ, 72-79% for LSTM, and 73-80% for AirKorea; the AI forecasts showed comparable skills to the human forecasters at AirKorea. The one-day forecast skill levels of high and very high PM2.5 pollution grades are 31-98%, 31-74%, and 39-81% for the CMAQ-solely, the LSTM, and the AirKorea forecasts, respectively. Despite good skills for forecasting the high and very high events, CMAQ-solely forecasts also generate substantially higher false alarm rates (up to 86%) than the LSTM and AirKorea forecasts (up to 58%). Hence, applying only the LSTM model to the CMAQ forecasts can yield reasonable forecast skill levels comparable to the operational AirKorea forecasts that elaborately combine the CMAQ model, AI models, and human forecasters. The present results suggest that applications of appropriate AI models can greatly enhance PM2.5 forecast skills for Korea in a more objective way.Supplementary informationThe online version contains supplementary material available at 10.1007/s13143-022-00293-2.
| S-EPMC9483905 | biostudies-literature
Project description:The finished precision rolling bearings after processing are required to pass the life test before they can be put into the market. The life testing takes a lot of time and expense. Aiming to solve the problem of time and expense, the 1D-CNN and 1D-CNN-LSTM hybrid neural networks are used for deep learning based on the existing rolling bearing life big data results (a total of 791152 date). Taking the wear of bearing as the target, the life prediction of bearing is carried out by using Python. The results show that: (1) 1D-CNN-LSTM algorithm and "all parameters" are selected as the best prediction options. (2) "XYZ direction displacement" and "all parameters" have the best fitting effect on the predicted wear value, and the MAPE is 4.18877, 1.2102, 2.68903 and 1.19981, respectively. The 1D-CNN-LSTM algorithm is slightly better than the 1D-CNN algorithm. (3) Using 1D-CNN-LSTM algorithm and "all parameters" to predict the bearing wear life will obtain good results. Compared with the highest 1D-CNN and "Four Bearing Temperatures" parameters, it is reduced by 14.7 times. (4) The prediction process and results provide a wear prediction method for relevant bearing enterprises in the experimental running-in stage. It can also provide reliable research ideas for subsequent related enterprises and scholars.
| S-EPMC11399617 | biostudies-literature
Project description:In order to obtain the pattern of variation of PM2.5concentrations in the atmosphere in Nanchang City, we build a Support Vector Regression(SVR) with modified Whale Optimization Algorithm(WOA) hybrid model (namely mWOA-SVR model) that can predict the PM2.5concentration. Firstly, according to the Pearson correlation coefficient (PCC) method to examine the dynamic relationship between air pollutants and meteorological factors together with them, PM10, SO2and CO were selected as air pollutant concentration characteristics, while daily maximum and minimum temperatures, and wind power levels were selected as meteorological characteristics; then, using modified WOA algorithm for parameter selection of SVR model, four sets of better parameter combinations were found; finally, the mWOA-SVR model was built by the four sets parameters to predict PM2.5concentration. The results show that the prediction accuracy of mixed mWOA-SVR model with pollutant concentration plus weather factors as the feature was higher than single pollutant concentration.
| S-EPMC11458793 | biostudies-literature
Project description:BackgroundIn this work, we leverage state-of-the-art deep learning-based algorithms for blood glucose (BG) forecasting in people with type 1 diabetes.MethodsWe propose stacks of convolutional neural network and long short-term memory units to predict BG level for 30-, 60-, and 90-minute prediction horizon (PH), given historical glucose measurements, meal information, and insulin intakes. The evaluation was performed on two data sets, Replace-BG and DIAdvisor, representative of free-living conditions and in-hospital setting, respectively.ResultsFor 90-minute PH, our model obtained mean absolute error of 17.30 ± 2.07 and 18.23 ± 2.97 mg/dL, root mean square error of 23.45 ± 3.18 and 25.12 ± 4.65 mg/dL, coefficient of determination of 84.13 ± 4.22% and 82.34 ± 4.54%, and in terms of the continuous glucose-error grid analysis 94.71 ± 3.89% and 91.71 ± 4.32% accurate predictions, 1.81 ± 1.06% and 2.51 ± 0.86% benign errors, and 3.47 ± 1.12% and 5.78 ± 1.72% erroneous predictions, for Replace-BG and DIAdvisor data sets, respectively.ConclusionOur investigation demonstrated that our method achieved superior glucose forecasting compared with existing approaches in the literature, and thanks to its generalizability showed potential for real-life applications.
| S-EPMC10658677 | biostudies-literature
Project description:BackgroundOutdoor fine particulate air pollution, <2.5 µm (PM2.5) mass concentrations can be constructed through many different combinations of chemical components that have varying levels of toxicity. This poses a challenge for studies interested in estimating the health effects of total outdoor PM2.5 (i.e., how much PM2.5 mass is present in the air regardless of composition) because we must consider possible confounders of the version of treatment-outcome relationships.MethodsWe evaluated the extent of possible bias in mortality hazard ratios for total outdoor PM2.5 by examining models with and without adjustment for sulfate and nitrate in PM2.5 as examples of potential confounders of version of treatment-outcome relationships. Our study included approximately 3 million Canadians and Cox proportional hazard models were used to estimate hazard ratios for total outdoor PM2.5 adjusting for sulfate and/or nitrate and other relevant covariates.ResultsHazard ratios for total outdoor PM2.5 and nonaccidental, cardiovascular, and respiratory mortality were overestimated due to the confounding version of treatment-outcome relationships, and associations for lung cancer mortality were underestimated. Sulfate was most strongly associated with nonaccidental, cardiovascular, and respiratory mortality suggesting that regulations targeting this specific component of outdoor PM2.5 may have greater health benefits than interventions targeting total PM2.5.ConclusionsStudies interested in estimating the health impacts of total outdoor PM2.5 (i.e., how much PM2.5 mass is present in the air) need to consider potential confounders of the version of treatment-outcome relationships. Otherwise, health risk estimates for total PM2.5 will reflect some unknown combination of how much PM2.5 mass is present in the air and the kind of PM2.5 mass that is present.
| S-EPMC11254114 | biostudies-literature
Project description:Today, with the rapid growth of Internet technology, the changing trend of real estate finance has brought great an impact on the progress of the social economy. In order to explore the visual identification (VI) effect of Convolutional Neural Network and Long Short-Term Memory (CNN-LSTM) algorithm based on neural network optimization on China's real estate index and stock trend, in this study, artificial neural network (ANN) algorithm is introduced to predict its trend. Firstly, LSTM algorithm can effectively solve the problem of vanishing gradient, which is suitable for dealing with the problems related to time series. Secondly, CNN, with its unique fine-grained convolution operation, has significant advantages in classification problems. Finally, combining the LSTM algorithm with the CNN algorithm, and using the Bayesian Network (BN) layer as the transition layer for further optimization, the CNN-LSTM algorithm based on neural network optimization has been constructed for the VI and prediction model of real estate index and stock trend. Through the performance verification of the model, the results reveal that the CNN-LSTM optimization algorithm has a more accurate prediction effect, the prediction accuracy is 90.55%, and the prediction time is only 52.05s. At the same time, the significance advantage of CNN-LSTM algorithm is verified by statistical method, which can provide experimental reference for intelligent VI and prediction of trend of China real estate index and property company stocks.
| S-EPMC9955670 | biostudies-literature
Project description:In recent years, the software industry has invested substantial effort to improve software quality in organizations. Applying proactive software defect prediction will help developers and white box testers to find the defects earlier, and this will reduce the time and effort. Traditional software defect prediction models concentrate on traditional features of source code including code complexity, lines of code, etc. However, these features fail to extract the semantics of source code. In this research, we propose a hybrid model that is called CBIL. CBIL can predict the defective areas of source code. It extracts Abstract Syntax Tree (AST) tokens as vectors from source code. Mapping and word embedding turn integer vectors into dense vectors. Then, Convolutional Neural Network (CNN) extracts the semantics of AST tokens. After that, Bidirectional Long Short-Term Memory (Bi-LSTM) keeps key features and ignores other features in order to enhance the accuracy of software defect prediction. The proposed model CBIL is evaluated on a sample of seven open-source Java projects of the PROMISE dataset. CBIL is evaluated by applying the following evaluation metrics: F-measure and area under the curve (AUC). The results display that CBIL model improves the average of F-measure by 25% compared to CNN, as CNN accomplishes the top performance among the selected baseline models. In average of AUC, CBIL model improves AUC by 18% compared to Recurrent Neural Network (RNN), as RNN accomplishes the top performance among the selected baseline models used in the experiments.
| S-EPMC8627227 | biostudies-literature
Project description:In China, severe haze is a major public health concern affecting residents' health and well-being. This study used hourly air quality monitoring data from 285 cities in China to analyze the effect of green coverage (GC) and other economic variables on the incremental PM2.5 concentration (ΔPM2.5) during peak hours. To detect possible non-linear and interaction effect between predictive variables, a kernel-based regularized least squares (KRLS) model was used for empirical analysis. The results show that there was considerable heterogeneity between cities regarding marginal effect of GC on ΔPM2.5, which could potentially be explained by different seasons, latitude, urban maintenance expenditure (UE), real GDP per capita (PG), and population density (PD). Also described in this study, in cities with high UE, the growth of GC, PG, and PD always remain a positive impact on mitigation of haze pollution. This shows that government expenditure on urban maintenance can reduce or mitigate the environmental pollution from economic development. In addition, the influence of other urban elements on air quality had also been analyzed so that different combinations of mitigation policies are proposed for different regions in this study to meet the mitigation targets.
| S-EPMC7701305 | biostudies-literature
Project description:Retrieving historical fine particulate matter (PM2.5) data is key for evaluating the long-term impacts of PM2.5 on the environment, human health and climate change. Satellite-based aerosol optical depth has been used to estimate PM2.5, but estimations have largely been undermined by massive missing values, low sampling frequency and weak predictive capability. Here, using a novel feature engineering approach to incorporate spatial effects from meteorological data, we developed a robust LightGBM model that predicts PM2.5 at an unprecedented predictive capacity on hourly (R2 = 0.75), daily (R2 = 0.84), monthly (R2 = 0.88) and annual (R2 = 0.87) timescales. By taking advantage of spatial features, our model can also construct hourly gridded networks of PM2.5. This capability would be further enhanced if meteorological observations from regional stations were incorporated. Our results show that this model has great potential in reconstructing historical PM2.5 datasets and real-time gridded networks at high spatial-temporal resolutions. The resulting datasets can be assimilated into models to produce long-term re-analysis that incorporates interactions between aerosols and physical processes.
| S-EPMC8566180 | biostudies-literature
Project description:Long-term air quality observations are seldom analyzed from a dynamic view. This study analyzed fine particulate matter (PM2.5) pollution processes using long-term PM2.5 observations in three Chinese cities. Pollution processes were defined as linearly growing PM2.5 concentrations following the criteria of coefficient of determination R2 > 0.8 and duration time T ≥ 18 hrs. The linear slopes quantitatively measured pollution levels by PM2.5 concentrations rising rates (PMRR, μg/(m3·hr)). The 741, 210 and 193 pollution processes were filtered out, respectively, in Beijing (BJ), Shanghai (SH), and Guangzhou (GZ). Then the relationships between PMRR and wind speed, wind direction, 24-hr backward points, gaseous pollutants (CO, NO2 and SO2) concentrations, and regional PM2.5 levels were studied. Inverse relationships existed between PMRR and wind speed. The wind directions and 24-hr backward points converged in specific directions indicating long-range transport. Gaseous pollutants concentrations increased at variable rates in the three cities with growing PMRR values. PM2.5 levels at the upwind regions of BJ and SH increased at high PMRRs. Regional transport dominated the PM2.5 pollution processes of SH. In BJ, both local contributions and regional transport increased during high-PMRR pollution processes. In GZ, PM2.5 pollution processes were mainly caused by local emissions.
| S-EPMC5404509 | biostudies-literature