Li Bowen, Liu Li, Xu Zikang, Li Kexun
College of Environmental Science and Engineering, Nankai University, Tianjin, 300350, China; MOE Key Laboratory of Pollution Processes and Environmental Criteria, Tianjin Key Laboratory of Environmental Remediation and Pollution Control, Tianjin Key Laboratory of Environmental Technology for Complex Trans-Media Pollution, Nankai University, Tianjin, 300350, China.
Tianjin Medical University General Hospital, Tianjin, 300052, China.
Environ Res. 2025 Feb 15;267:120653. doi: 10.1016/j.envres.2024.120653. Epub 2024 Dec 17.
Appropriate carbon source addition can save operational costs and reduce surplus sludge yield in the wastewater treatment plant (WWTP). However, the link between carbon source and surplus sludge yield remains neglected although machine learning (ML) has become a powerful tool for WWTP, and is a challenge due to more complex multidimensional pattern recognition. Herein, weighted average ensemble strategy was conducted to assemble multiple diverse basic models to obtain better prediction capability to optimize carbon source addition (Model-1) and further control surplus sludge yield (Model-2). The ensemble models significantly outperformed all single models with MAE of 5.82 g/m, MSE of 60.59 and R value of 0.98 in Model-1 and MAE of 15.09 g/m, MSE of 449.01 and R value of 0.93 in Model-2. The optimal input feature subset was explored to reduce model complexity, indicating that the final ensemble models can predict with high precision using relatively few features with MAE of 6.41 g/m, MSE of 78.49 and R value of 0.97 in Model-1 and MAE of 12.82 g/m, MSE of 232.71 and R value of 0.95 in Model-2. Furthermore, the final models were deployed into an offline web application to facilitate their utility in real-world settings, demonstrating 47.25 % savings in carbon source addition and 15.89 % reductions in surplus sludge yield for an extra month of running. This work offers an efficient approach for the WWTP to optimize carbon source addition and provides new insights into controlling surplus sludge yield.
在污水处理厂（WWTP）中，适当添加碳源可以节省运营成本并减少剩余污泥产量。然而，尽管机器学习（ML）已成为污水处理厂的强大工具，但碳源与剩余污泥产量之间的联系仍被忽视，并且由于更复杂的多维模式识别，这是一个挑战。在此，采用加权平均集成策略来组合多个不同的基本模型，以获得更好的预测能力，从而优化碳源添加（模型-1）并进一步控制剩余污泥产量（模型-2）。在模型-1中，集成模型的表现明显优于所有单一模型，平均绝对误差（MAE）为5.82 g/m，均方误差（MSE）为60.59，R值为0.98；在模型-2中，MAE为15.09 g/m，MSE为449.01，R值为0.93。探索了最优输入特征子集以降低模型复杂度，这表明最终的集成模型可以使用相对较少的特征进行高精度预测，在模型-1中，MAE为6.41 g/m，MSE为78.49，R值为0.97；在模型-2中，MAE为12.82 g/m，MSE为232.71，R值为0.95。此外，最终模型被部署到一个离线网络应用程序中，以促进其在实际环境中的应用，结果表明，在多运行一个月的情况下，碳源添加量节省了47.25%，剩余污泥产量减少了15.89%。这项工作为污水处理厂优化碳源添加提供了一种有效的方法，并为控制剩余污泥产量提供了新的见解。
