A novel spatial optimization approach for the cost-effectiveness improvement of LID practices based on SWMM-FTC

Due to the increasingly frequent occurrence of urban waterlogging, the spatial optimization of low impact development (LID) practices has been commonly used to detain and reduce storm water runoff in the most costeffective way. In this study, the flow transmission chain (FTC) was proposed to replace the routing portion of the Storm Water Management Model (SWMM) and was combined with the runoff component of the SWMM to simulate LID practices (SWMM-FTC). In the SWMM-FTC, the third Evolution Step of Generalized Differential Evolution (GDE3) was employed to optimize the LID layout design. The results showed that the relative error between the modified SWMM-FTC and the calibrated SWMM was less than 0.25% under various LID scenarios, and the computational efficiency of the SWMM-FTC was improved by 19.3 times. Moreover, the GDE3 outperformed the commonly used non-dominated sorting genetic algorithm (NSGA-II), the strength Pareto evolutionary algorithm (SPEA2), and the multi-objective shuffled frog leaping algorithm (MOSFLA) due to its ability to find the most cost-effective solution. The LID layout obtained from the SWMM-FTC with the GDE3 saved $210-1067 to achieve a 1% reduction in storm water runoff. This result demonstrates that the SWMM-FTC with the GDE3 can achieve higher environmental benefits than comparable models, providing better guidance for managers and stakeholders.

Automated summary

This study proposes a method that combines flow transmission chain and generalized differential evolution to simulate low impact development practices. The method achieves low relative error and improved computational efficiency under different LID scenarios. Compared to other algorithms, the generalized differential evolution algorithm can find the most cost-effective solution. The LID layout obtained through this method saves costs while reducing storm water runoff, providing better environmental benefits.