Liang Zhensheng, Xie Wenlang, Li Hao, Li Yu, Jiang Feng
School of Environmental Science & Engineering, Guangdong Provincial Key Lab of Environmental Pollution Control and Remediation Technology, Sun Yat-sen University, Guangzhou, 510275, China.
Guangdong Provincial International Joint Research Center on Urban Water Management and Treatment, Sun Yat-sen University, Guangzhou, 510275, China.
Water Res X. 2024 Jun 17;23:100230. doi: 10.1016/j.wroa.2024.100230. eCollection 2024 May 1.
The frequent occurrence of safety incidents in sewer systems due to the emergency toxicity of hydrogen sulfide (HS) necessitate timely and efficient prediction, early warning and real-time control. However, various factors influencing HS generation and emission leads to a substantial computational burden for the existing dynamic sewer process models and fails to timely control the HS exposure risk. The present study proposed a swift prediction model (SPM) that combined the validated dynamic sewer process model (the biofilm-initiated sewer process model, BISM) with a high-speed machine learning algorithm (MLA), achieving accurately and swiftly predict the dissolved sulfide (DS) concentration and HS concentration in a specific sewer network. Based on Gradient Boosting Decision Tree-based SPM, the simulated concentrations of DS and HS are 1.95 mg S/L and 214 ppm, respectively, which are closely to the field-measured values of 1.82 mg S/L and 219 ppm. Notably, SPM achieved a computation time of less than 0.3 s, and a significant improvement over BISM (> 5000 s) for the same task. Moreover, the real-time and dynamic dosing scheme facilitated by SPM outperformed the conventional constant dosing scheme provided by dynamic sewer process model, which significantly improved the HS control completion rate from 69 % to 100 %, and achieved a significant reduction in chemical dosage. In conclusion, the integration of dynamic sewer process models with MLA addresses the inadequacy of monitoring data for MLA training, and thus pursues swift prediction of HS generation and emission, and achieving real-time, effective, and economic control of HS in complex sewer networks.
由于硫化氢（HS）的急性毒性，下水道系统中安全事故频繁发生，因此需要及时、高效的预测、预警和实时控制。然而，影响HS生成和排放的各种因素给现有的动态下水道过程模型带来了巨大的计算负担，并且无法及时控制HS暴露风险。本研究提出了一种快速预测模型（SPM），该模型将经过验证的动态下水道过程模型（生物膜启动下水道过程模型，BISM）与高速机器学习算法（MLA）相结合，能够准确、快速地预测特定下水道网络中的溶解硫化物（DS）浓度和HS浓度。基于梯度提升决策树的SPM，模拟的DS和HS浓度分别为1.95 mg S/L和214 ppm，与现场测量值1.82 mg S/L和219 ppm非常接近。值得注意的是，SPM的计算时间不到0.3秒，与执行相同任务的BISM（>5000秒）相比有显著改进。此外，由SPM促成的实时动态加药方案优于动态下水道过程模型提供的传统恒定加药方案，这显著提高了HS控制完成率，从69%提高到100%，并实现了化学药剂用量的显著减少。总之，动态下水道过程模型与MLA的整合解决了MLA训练监测数据不足的问题，从而实现了对HS生成和排放的快速预测，并在复杂的下水道网络中实现了对HS的实时、有效和经济控制。
