Day-Ahead PM2.5 Concentration Forecasting Using WT-VMD Based Decomposition Method and Back Propagation Neural Network Improved by Differential Evolution

Wang, Deyun; Liu, Yanling; Luo, Hongyuan; Yue, Chenqiang; Cheng, Sheng

doi:10.3390/ijerph14070764

Cited by 30 publications

(12 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…ANN includes a family of intelligence models that mimic biological neural networks and present in various layers. The BP are networks that can self‐distribute to fit the specific correction algorithm by using one or more hidden layers in the ANN to optimize the weights of each layer (Wang, Liu, Luo, Yue, & Cheng, ). The BP‐ANN does not require an explicit understanding of the mechanisms underlying the processes it examines but instead uses data sets (input, hidden and output data) pertaining throughout the mapping process.…”

Section: Introductionmentioning

confidence: 99%

A prediction model of sleep disturbances among female nurses by using the BP‐ANN

et al. 2019

View full text Add to dashboard Cite

Aim To describe sleep disturbances and fatigue among female registered nurses in Beijing and to develop a prediction model for sleep disturbances. Background Chinese nurses are required to work rotating shifts on a weekly basis, which could negatively impact their sleep and well‐being. Method A total of 647 registered nurses participated in this study. Self‐reported sleep‐related data and selected physiological data were collected. Back propagation artificial neural networks was used to develop the prediction model by using the risk management and population health framework. Results Majority of them reported clinically significant poor sleep (69.4%) and fatigue (75.4%). A total of eight predictors were identified for sleep disturbances, and the top four normalized importance predictors are morning fatigue (100%), body mass index (30.5%), gastrointestinal symptoms (17.6%) and drinking caffeinated beverages at work (17.3%). The cross‐entropy error was 206.58, and the model accounted for 77.6% of the variance in sleep disturbances. Conclusions and implications for nursing management Female registered nurses in China experience clinically significant sleep disturbances. Morning fatigue severity along with seven significant influencing factors may be used to identify shift nurses who face a higher risk of sleep disturbances. The back propagation artificial neural networks model could be used as the foundation for health promotion interventions for registered nurses.

show abstract

Section: Introductionmentioning

confidence: 99%

A prediction model of sleep disturbances among female nurses by using the BP‐ANN

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Then, Experiment II is designed to compare with better previous models made in the prediction of PM 2. 5 [22,54], VCEEMDAN-GRNN (Zhou Q. et al 2014) [39], BPNN, GRNN, ARIMA [55]). It is found that after the feature selection, only a small number of input features can be selected to obtain higher prediction accuracy, and it is also found that WOA used in our model is better than some other meta-heuristic optimization algorithms such as CS in PM 2.5 concentration prediction.…”

Section: Methodsmentioning

confidence: 99%

A Combined Model Based on Feature Selection and WOA for PM2.5 Concentration Forecasting

Zhao

2019

Atmosphere

View full text Add to dashboard Cite

As people pay more attention to the environment and health, P M 2.5 receives more and more consideration. Establishing a high-precision P M 2.5 concentration prediction model is of great significance for air pollutants monitoring and controlling. This paper proposed a hybrid model based on feature selection and whale optimization algorithm (WOA) for the prediction of P M 2.5 concentration. The proposed model included five modules: data preprocessing module, feature selection module, optimization module, forecasting module and evaluation module. Firstly, signal processing technology CEEMDAN-VMD (Complete Ensemble Empirical Mode Decomposition with Adaptive Noise and Variational Mode Decomposition) is used to decompose, reconstruct, identify and select the main features of P M 2.5 concentration series in data preprocessing module. Then, AutoCorrelation Function (ACF) is used to extract the variables which have relatively large correlation with predictor, so as to select input variables according to the order of correlation coefficients. Finally, Least Squares Support Vector Machine (LSSVM) is applied to predict the hourly P M 2.5 concentration, and the parameters of LSSVM are optimized by WOA. Two experiment studies reveal that the performance of the proposed model is better than benchmark models, such as single LSSVM model with default parameters optimization, single BP neural networks (BPNN), general regression neural network (GRNN) and some other combined models recently reported.

show abstract

“…Backpropagation (BP) network is a multilayered feedforward neural network trained by error backpropagation. It has good self-organizing learning ability and can implement any nonlinear mapping from input to output [ 20 , 21 ]. The network prediction model mainly realizes the training process through the forward propagation of the input signal and the backpropagation of the error signal.…”

Section: Basic Theorymentioning

confidence: 99%

A Hybrid Model for Forecasting Sunspots Time Series Based on Variational Mode Decomposition and Backpropagation Neural Network Improved by Firefly Algorithm

Yang

2018

Computational Intelligence and Neuroscience

View full text Add to dashboard Cite

The change of the number of sunspots has a great impact on the Earth's climate, agriculture, communications, natural disasters, and other aspects, so it is very important to predict the number of sunspots. Aiming at the chaotic characteristics of monthly mean of sunspots, a novel hybrid model for forecasting sunspots time-series based on variational mode decomposition (VMD) and backpropagation (BP) neural network improved by firefly algorithm (FA) is proposed. Firstly, a set of intrinsic mode functions (IMFs) are obtained by VMD decomposition of the monthly mean time series of the sunspots. Secondly, the firefly algorithm is introduced to initialize the weights and thresholds of the BP neural network, and a prediction model is established for each IMF. Finally, the predicted values of these components are calculated to obtain the final predict results. Comparing BP model, FA-BP model, EMD-BP model, and VMD-BP model, the simulation results show that the proposed algorithm has higher prediction accuracy and can be used to forecast the time series of sunspots.

show abstract

Day-Ahead PM2.5 Concentration Forecasting Using WT-VMD Based Decomposition Method and Back Propagation Neural Network Improved by Differential Evolution

Cited by 30 publications

References 41 publications

A prediction model of sleep disturbances among female nurses by using the BP‐ANN

A prediction model of sleep disturbances among female nurses by using the BP‐ANN

A Combined Model Based on Feature Selection and WOA for PM2.5 Concentration Forecasting

A Hybrid Model for Forecasting Sunspots Time Series Based on Variational Mode Decomposition and Backpropagation Neural Network Improved by Firefly Algorithm

Contact Info

Product

Resources

About