Urban Stormwater Pollution Forecasting Using Recurrent Neural Networks

摘要

To reduce the impact of stormwater pollution on the environment, operators in charge of urban drainage systems and wastewater treatment plants need realtime predictions of pollutant concentrations during rainfall events. In this paper, this problem is addressed using nonlinear recurrent neural networks, which are used to simulate the rainfall-runoff transformation, as well as the production and transfer of solids in drainage catchments and sewer pipes. Prior knowledge provided by existing conceptual models has been used to design specific neural architectures with small numbers of parameters. Training of these networks can be performed using the epochwise or on-line backsweep algorithms recently formalised by Piche [IEEE Trans Neur Net 1994; 2: 198-211]. Experimental results demonstrate the efficiency of this approach for predicting suspended solid concentration at the outlet of urban drainage catchments.