Modeling and diagnosis of water quality parameters in wastewater treatment process based on improved particle swarm optimization and self-organizing neural network

Journal of environmental chemical engineering, Год журнала: 2024, Номер 12(4), С. 113142 - 113142

Опубликована: Май 28, 2024

Язык: Английский

---

Efficient greenhouse gas prediction using IoT data streams and a CNN-BiLSTM-KAN model

Alexandria Engineering Journal, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Язык: Английский

---

Interpretability and performance assessment of advanced machine learning models for α-factor prediction in wastewater treatment plants

Journal of Water Process Engineering, Год журнала: 2025, Номер 72, С. 107637 - 107637

Опубликована: Апрель 1, 2025

Язык: Английский

---

High-Precision Prediction of Total Nitrogen Based on Distance Correlation and Machine Learning Models—A Case Study of Dongjiang River, China

Water, Год журнала: 2025, Номер 17(8), С. 1131 - 1131

Опубликована: Апрель 10, 2025

Excessive total nitrogen (TN) in water bodies leads to eutrophication, algal blooms, and hypoxia, which pose significant risks aquatic ecosystems human health. Accurate real-time TN prediction is crucial for effective quality management. This study presents an innovative approach that combines the distance correlation coefficient (DCC) feature selection with a coupled Attention-Convolutional Neural Network-Bidirectional Long Short-Term Memory (At-CBiLSTM) model predict concentrations Dongjiang River China. A dataset of 28,922 time-series data points was collected from seven sampling sites along River, spanning November 2020 February 2023. The DCC method identified conductivity, Permanganate Index (CODMn), phosphorus as most predictors levels. At-CBiLSTM model, optimized time step three, outperformed other models, including standalone (LSTM), Bi-directional LSTM (Bi-LSTM), Convolutional Network (CNN-LSTM), Attention-LSTM variants, achieving excellent performance following metrics: mean absolute error (MAE) = 0.032, squared (MSE) 0.005, percentage (MAPE) 0.218, root (RMSE) 0.045. Importantly, increasing number input features beyond three variables led decline accuracy, underscoring importance DCC-driven selection. results highlight combining deep learning particularly At-CBiLSTM, effectively captures nonlinear temporal dependencies improves accuracy. provides solid foundation monitoring can inform targeted pollution control strategies river ecosystems.

Язык: Английский

---

Quantifying the mitigation potential of energy and chemical consumption for a full-scale wastewater treatment plant with deep learning methods

Applied Energy, Год журнала: 2025, Номер 394, С. 126123 - 126123

Опубликована: Май 26, 2025

Язык: Английский

---

An Interpretable Deep Learning Model for Predicting the Influent of Wastewater Treatment Plants

Journal of environmental chemical engineering, Год журнала: 2025, Номер unknown, С. 117437 - 117437

Опубликована: Июнь 1, 2025

Язык: Английский

---

Atmospheric Ozone Prediction by Linear Dendritic Neuron Model Combining Seasonal-Trend Decomposition Based on LOESS

Опубликована: Янв. 1, 2024

Язык: Английский

---

Atmospheric Ozone Prediction by Linear dendritic neuron model combining Seasonal-trend decomposition based on LOESS

Research Square (Research Square), Год журнала: 2024, Номер unknown

Опубликована: Авг. 14, 2024

Ozone in the atmosphere is not only closely related to people's lives and production, but can also endanger health at high concentrations. However, existing methods for predicting ozone levels are accurate enough. Therefore, a prediction model composed of Seasonal-trend decomposition based on LOESS(STL) improved dendritic neuron model(L-DNM) proposed predict atmospheric content by dividing time series into three parts: seasonal, trend residual. Then, part sequence smoothed using least squares method, most irregular residual used as input L-DNM. During training process, Back-propagation (BP) algorithm output predicted The seasonal with regularity other two processed parts added together obtain result. Through comparison, STL-L-DNM state-of-the-art method five datasets considering four indicators including Mean Squared Error (MSE), Absolute Percentage (MAPE), (MAE), Nash-Sutcliffe Efficiency (NSE). This includes popular Transformer its variants recent years. At same time, Taylor diagrams regression show that value stable fluctuation least. experimental results demonstrate accuracy stability algorithm, which play an important role level.

Язык: Английский

---

Application of Dual-Stage Attention Temporal Convolutional Networks in Gas Well Production Prediction

Mathematics, Год журнала: 2024, Номер 12(24), С. 3896 - 3896

Опубликована: Дек. 10, 2024

Effective production prediction is vital for optimizing energy resource management, designing efficient extraction strategies, minimizing operational risks, and informing strategic investment decisions within the sector. This paper introduces a Dual-Stage Attention Temporal Convolutional Network (DA-TCN) model to enhance accuracy efficiency of gas forecasting, particularly wells in tight sandstone reservoirs. The DA-TCN architecture integrates feature temporal attention mechanisms (TCN) framework, improving model’s ability capture complex dependencies emphasize significant features, resulting robust forecasting performance across multiple time horizons. Application data from two Block T Sulige field China demonstrated 19% improvement RMSE 21% MAPE compared traditional TCN methods long-term forecasts. These findings confirm that dual-stage not only increases predictive but also enhances forecast stability over short-, medium-, By enabling more reliable reduces uncertainties, optimizes allocation, supports cost-effective management unconventional resources. Leveraging existing knowledge, this scalable data-efficient approach represents advancement delivering tangible economic benefits industry.

Язык: Английский

---

BOD prediction model for wastewater treatment process based on IBKA-GRNN

Опубликована: Авг. 9, 2024

Язык: Английский

---