Advanced Engineering Informatics, Journal Year: 2025, Volume and Issue: 65, P. 103343 - 103343
Published: April 18, 2025
Language: Английский
Citations
0A novel approach for multivariate time series interval prediction of water quality at wastewater treatment plants
Water Science & Technology, Journal Year: 2024, Volume and Issue: 90(10), P. 2813 - 2841
Published: Nov. 12, 2024
ABSTRACT This study proposes a novel approach for predicting variations in water quality at wastewater treatment plants (WWTPs), which is crucial optimizing process management and pollution control. The model combines convolutional bi-directional gated recursive units (CBGRUs) with adaptive bandwidth kernel function density estimation (ABKDE) to address the challenge of multivariate time series interval prediction WWTP quality. Initially, wavelet transform (WT) was employed smooth data, reducing noise fluctuations. Linear correlation coefficient (CC) non-linear mutual information (MI) techniques were then utilized select input variables. CBGRU applied capture temporal correlations series, integrating Multiple Heads Attention (MHA) mechanism enhance model's ability comprehend complex relationships within data. ABKDE employed, supplemented by bootstrap establish upper lower bounds intervals. Ablation experiments comparative analyses benchmark models confirmed superior performance point prediction, analysis forecast period, fluctuation detection Also, this verifies broad applicability robustness anomalous contributes significantly improved effluent efficiency control WWTPs.
Language: Английский
Citations
2Flood prediction with optimized gated recurrent unit-temporal convolutional network and improved KDE error estimation
AIMS Mathematics, Journal Year: 2024, Volume and Issue: 9(6), P. 14681 - 14696
Published: Jan. 1, 2024
<abstract> <p>Flood time series forecasting stands a critical challenge in precise predictive models and reliable error estimation methods. A novel approach utilizing hybrid deep learning model for both point interval flood prediction is presented, enhanced by improved kernel density (KDE) comparison simulation. Firstly, an optimized gated recurrent unit-time convolutional network (GRU-TCN) constructed tuning the internal structure of TCN, activation function, L2 regularization, optimizer. Then, Pearson Correlation used feature selection, hyperparameters GRU-TCN are subtraction-average-based optimizer (SABO). To further assess uncertainty, predictions provided via Non-parametric KDE, with bandwidth setting accurate distribution Experimental comparisons made on 5-year hydro-meteorological daily data from two stations along Yangtze River. The proposed surpasses long short-term memory (LSTM), GRU, TCN-LSTM, GRU-TCN, reduction more than 13% root mean square (RMSE) approximately 15% absolute (MAE), resulting better control. curves errors closer to value confidence intervals, reflecting trend distribution. This research enhances accuracy reliability improves capacity humans cope climate environmental changes.</p> </abstract>
Language: Английский
Citations
1Forecasting the Mitigation Potential of Greenhouse Gas Emissions in Shenzhen through Municipal Solid Waste Treatment: A Combined Weight Forecasting Model
Atmosphere, Journal Year: 2024, Volume and Issue: 15(4), P. 507 - 507
Published: April 20, 2024
As a significant source of anthropogenic greenhouse gas emissions, the municipal solid waste sector’s emission mode remains unknown, hampering effective decision-making on possible reductions. Rapid urbanization and economic growth have resulted in massive volumes trash. result, identifying reduction routes for treatment is critical. In this research, we investigate potential methods lowering (GHG) emissions Shenzhen, typical Chinese major city. The results showed that combined 58% incineration, 2% landfill, 40% anaerobic digestion (AD) had lowest about 5.91 million tons under all scenarios. implementation sorting organic after separate collection can reduce by simply increasing incineration ratio.
Language: Английский
Citations
1A multistage exergy evaluation-cooperated liquid level optimization approach for multi-equipment evaporation process
Chemical Engineering Science, Journal Year: 2024, Volume and Issue: 298, P. 120403 - 120403
Published: June 24, 2024
Language: Английский
Citations
0ConvFormer-KDE: A Long-Term Point–Interval Prediction Framework for PM2.5 Based on Multi-Source Spatial and Temporal Data
Toxics, Journal Year: 2024, Volume and Issue: 12(8), P. 554 - 554
Published: July 30, 2024
Accurate long-term PM
Language: Английский
Citations
0Research on regional carbon emission scenario simulation based on GA-BP-KDE under uncertain conditions
Atmospheric Pollution Research, Journal Year: 2024, Volume and Issue: unknown, P. 102375 - 102375
Published: Dec. 1, 2024
Language: Английский
Citations
0An Improved Kernel Density Estimation Method for Characterizing Buoy Offset
IEEE Access, Journal Year: 2024, Volume and Issue: 12, P. 114495 - 114511
Published: Jan. 1, 2024
The AtoN department is responsible for mastering the drift characteristics of buoys, conducting targeted buoy inspections and resets, providing accurate position information ship navigation. To analyze pattern, a k-nearest neighbor(KNN) improved Kernel Density Estimation(KED) method(KNN-KDE) proposed to optimize single bandwidth in more complex distribution dataset can only depict approximate trend data, while details data changes not be accurately estimated. KNN-KDE utilized estimate coordinates buoy's gyration center its radius gyration, which establishes mathematical model drift. Based on this approach, telemetry from typical buoys main channel Xiamen Harbor collected pattern buoys. This analysis provides useful reference safety
Language: Английский
Citations
0An Interval Prediction Method for Quantifying the Uncertainties of Gate Lifting Force Under Sediment Deposition
Published: Jan. 1, 2024
Language: Английский
Citations
0Short-Term Wind Power Prediction Based on GS-PCA-RF
Published: Dec. 13, 2024
Language: Английский
Citations
0