Risk forecasting of pollution accidents based on an integrated Bayesian Network and water quality model for the South to North Water Transfer Project

For a long distance water transfer system, guaranteeing the safety of water transferred in a long open canal is a challenge. Potential pollution sources and locations are identified first. Bridges above the canal are considered as the typical location where pollution accidents could happen. Truck responses, road conditions, and human responses are the main factors resulting in sudden pollution accidents. A Bayesian Network model is developed to calculate the risk of water pollution and evaluate the effects of pollutants in water. The key causes (human judgment, truck condition) were determined to calculate the unfavorable accidental risk. Six types of hydraulic structures were included in this canal reach. MIKE 11, a one-dimensional hydrodynamic and water quality model, was used to simulate the fluid field and pollutants transportation process. The water quality situations of cyanide at four pollution loads (0.5 t, 5 t, 10 t, 20 t) were simulated. Emergent countermeasures and measures, including the management of diversion and check gate, have been proposed according to the calculate result.