The challenge of this study is to investigate on applicability of a data-driven Gaussian Process Emulator (GPE) technique to develop a surrogate model for a computationally expensive and detailed urban drainage simulator. The novelty is the consideration of (short) time series for the simulation inputs and outputs. Such simulation setup is interesting in applications such as Model Predictive Control (MPC) in which numerous, fast and frequent simulation results are required. Here, an emulator is developed to predict a storage tank’s volume in a small case study in Luxembourg. Three main inputs are considered as the GPE’s parameters: Initial volume in the tank, the level in which the outlet pump of the tank must start to work, and the time series of expected rainfall in the upcoming 2 h. The output of interest is the total volume of the storage tank for the next 24 h. A dataset of 2000 input-output scenarios were produced using different possible combinations of the inputs and running the detailed simulator (InfoWorks® ICM). 80% of the dataset were applied to train the emulator and 20% to validate the results. Distributions of Nash-Sutcliffe efficiency and Volumetric Efficiency are presented as indicators for quantification of the emulation error. Based on the preliminary results, it can be concluded that the introduced technique is able to reduce the simulations runtime significantly while imposing some inevitable accuracy cost. More investigation is required to validate the more generic applicability of this technique for multiple outputs and interactions between different urban drainage components.

In this study, applicability of a data-driven Gaussian Process Emulator (GPE) technique to develop a dynamic surrogate model for a computationally expensive urban drainage simulator is investigated. Considering rainfall time series as the main driving force is a challenge in this regard due to the high dimensionality problem. However, this problem can be less relevant when the focus is only on short-term simulations. The novelty of this research is the consideration of short-term rainfall time series as training parameters for the GPE. Rainfall intensity at each time step is counted as a separate parameter. A method to generate synthetic rainfall events for GPE training purposes is introduced as well. Here, an emulator is developed to predict the upcoming daily time series of the total wastewater volume in a storage tank and the corresponding Combined Sewer Overflow (CSO) volume. Nash-Sutcliffe Efficiency (NSE) and Volumetric Efficiency (VE) are calculated as emulation error indicators. For the case study herein, the emulator is able to speed up the simulations up to 380 times with a low accuracy cost for prediction of the total storage tank volume (medians of NSE = 0.96 and VE = 0.87). CSO events occurrence is detected in 82% of the cases, although with some considerable accuracy cost (medians of NSE = 0.76 and VE = 0.5). Applicability of the emulator for consecutive short-term simulations, based on real observed rainfall time series is also validated with a high accuracy (NSE = 0.97, VE = 0.89).