College of Environmental and Resource Sciences, Zhejiang University, Hangzhou, 310058, China.
Zhejiang Supcon Information Technology Co., Ltd, Hangzhou, 310052, China.
J Environ Manage. 2024 May;359:120887. doi: 10.1016/j.jenvman.2024.120887. Epub 2024 Apr 27.
The accurate effluent prediction plays a crucial role in providing early warning for abnormal effluent and achieving the adjustment of feedforward control parameters during wastewater treatment. This study applied a dual-staged attention mechanism based on long short-term memory network (DA-LSTM) to improve the accuracy of effluent quality prediction. The results showed that input attention (IA) and temporal attention (TA) significantly enhanced the prediction performance of LSTM. Specially, IA could adaptively adjust feature weights to enhance the robustness against input noise, with R increased by 13.18%. To promote its long-term memory ability, TA was used to increase the memory span from 96 h to 168 h. Compared to a single LSTM model, the DA-LSTM model showed an improvement in prediction accuracy by 5.10%, 2.11%, 14.47% for COD, TP, and TN. Additionally, DA-LSTM demonstrated excellent generalization performance in new scenarios, with the R values for COD, TP, and TN increasing by 22.67%, 20.06%, and 17.14% respectively, while the MAPE values decreased by 56.46%, 63.08%, and 42.79%. In conclusion, the DA-LSTM model demonstrated excellent prediction performance and generalization ability due to its advantages of feature-adaptive weighting and long-term memory focusing. This has forward-looking significance for achieving efficient early warning of abnormal operating conditions and timely management of control parameters.
准确的出水预测在提供异常出水的预警和实现废水处理前馈控制参数的调整方面起着至关重要的作用。本研究应用了一种基于长短期记忆网络（LSTM）的双阶段注意机制（DA-LSTM）来提高出水质量预测的准确性。结果表明，输入注意（IA）和时间注意（TA）显著提高了 LSTM 的预测性能。特别是，IA 可以自适应地调整特征权重，增强对输入噪声的鲁棒性，使 R 值提高了 13.18%。为了提高其长期记忆能力，使用 TA 将记忆跨度从 96 小时增加到 168 小时。与单一的 LSTM 模型相比，DA-LSTM 模型在 COD、TP 和 TN 的预测精度上分别提高了 5.10%、2.11%和 14.47%。此外，DA-LSTM 在新场景中表现出了出色的泛化性能，COD、TP 和 TN 的 R 值分别提高了 22.67%、20.06%和 17.14%，而 MAPE 值分别降低了 56.46%、63.08%和 42.79%。总之，由于 DA-LSTM 模型具有特征自适应加权和长期记忆聚焦的优势，因此具有出色的预测性能和泛化能力。这对于实现异常运行条件的高效预警和及时管理控制参数具有前瞻性意义。
