A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Abstract Globally, over 80% of wastewater is discharged into water bodies without adequate treatment (UNESCO 2017:10–15), making accurate water quality prediction essential for safeguarding aquatic ecosystems and public health. This study presents a novel QPSO-CNN-LSTM model that significantly advances water quality prediction by combining Quantum Particle Swarm Optimization (QPSO) with a CNN-LSTM architecture. Unlike traditional models, the QPSO-CNN-LSTM leverages CNN to capture complex spatial features from water quality data and LSTM to model long-term temporal dependencies. The QPSO algorithm optimizes key hyperparameters, mitigating the need for manual tuning and improving the model’s adaptability to dynamic environmental data. The model outperforms traditional methods with a 15–50% improvement in RMSE, MSE, MAE, and MAPE for dissolved oxygen and pH predictions. These enhancements demonstrate the model’s superior accuracy and robustness, making it an invaluable tool for real-time water quality monitoring, pollution prevention, and cost-effective water management strategies. The practical implications of this model offer a step forward in preserving aquatic ecosystems through data-driven environmental stewardship.
Abstract Globally, over 80% of wastewater is discharged into water bodies without adequate treatment (UNESCO 2017:10–15), making accurate water quality prediction essential for safeguarding aquatic ecosystems and public health. This study presents a novel QPSO-CNN-LSTM model that significantly advances water quality prediction by combining Quantum Particle Swarm Optimization (QPSO) with a CNN-LSTM architecture. Unlike traditional models, the QPSO-CNN-LSTM leverages CNN to capture complex spatial features from water quality data and LSTM to model long-term temporal dependencies. The QPSO algorithm optimizes key hyperparameters, mitigating the need for manual tuning and improving the model’s adaptability to dynamic environmental data. The model outperforms traditional methods with a 15–50% improvement in RMSE, MSE, MAE, and MAPE for dissolved oxygen and pH predictions. These enhancements demonstrate the model’s superior accuracy and robustness, making it an invaluable tool for real-time water quality monitoring, pollution prevention, and cost-effective water management strategies. The practical implications of this model offer a step forward in preserving aquatic ecosystems through data-driven environmental stewardship.
Application of a QPSO-optimized CNN-LSTM model in water quality prediction
Yue Zhu (author)
2024
Article (Journal)
Electronic Resource
Unknown
Metadata by DOAJ is licensed under CC BY-SA 1.0
Application of a QPSO-optimized CNN-LSTM model in water quality prediction
| Springer Verlag | 2024
A novel hybrid model for water quality prediction based on VMD and IGOA optimized for LSTM
| Springer Verlag | 2023