From single-objective to multiple-objective multiple-rainfall events automatic calibration of urban storm water...

Area of Science:

• Environmental Engineering
• Hydrology
• Water Resource Management

Background:

• Storm water runoff model calibration is challenging, evolving from single-objective to multi-objective approaches.
• Recent advancements focus on utilizing extensive concurrent rainfall and flow measurement data.
• Existing methods struggle to fully exploit the wealth of available event-specific data.

Purpose of the Study:

• To present and evaluate a novel Multi-Objective Multiple-Event (MOME) calibration framework for storm water models.
• To compare the MOME approach against single-objective and traditional multi-objective calibration paradigms.
• To demonstrate the effectiveness of optimizing multiple performance criteria across multiple rainfall events simultaneously.

Main Methods:

• A case study involving a SWMM (Storm Water Management Model) calibration for a Singapore catchment was conducted.
• The study compared three calibration paradigms: single-objective, multi-objective, and the proposed MOME approach.
• The MOME framework was formulated as a multi-objective problem with m x r objective functions, optimizing m criteria across r events concurrently.

Main Results:

• The MOME framework demonstrated significantly superior performance in the tested case study.
• The new approach effectively integrates information from multiple rainfall events and performance criteria.
• Results indicate a substantial improvement over both single-objective and conventional multi-objective calibration methods.

Conclusions:

• The MOME framework offers a more effective approach to storm water model calibration by leveraging multiple events and objectives.
• This advanced methodology enhances the accuracy and reliability of storm water runoff models.
• The study highlights the potential of MOME for optimizing complex environmental models with extensive datasets.