Artificial Intelligence based Performance Models to Support Hydrologic Rainfall Conditions using Ensembling Approach

The architectural design of the BC and the scale of its installation have a major impact on the effectiveness of the BC to lower the peak flow load of stormwater that enters an urban drainage system. In order to get the most of BC's benefits, designers must constantly make sure they are using the appropriate settings. In all, there are 18 design parameters, each of which may be customized in a variety of ways. Therefore, it could be difficult to find the characteristics that would provide the stormwater management model (SWMM) the most accurate results for the India models. The purpose of this study was to investigate the effects that BC design factors that are essential to hydrologic dynamics, such as depth, inflow and overflow, length, and time-to-peak, have on the rates of surface infiltration and outflow as well as storage as a function of a broad spectrum of a wide variety of different types and amounts of precipitation. In the first part of the methodology, the one-factor-at-a-time (OAT) method was used to choose the seven BC design criteria that were deemed to be the most important. This sampling was done in a completely arbitrary manner. The simulations were run by making use of the SWMM Python Wrapper, also known as Py SWMM. This wrapper ensured that the values of the other parameters stayed unchanged while randomly picking samples for one of the parameters. It is essential to examine these three aspects side by side.