Layout Optimization for Stormwater Harvesting Facilities in Coal Ports Considering Stochasticity of Underlying Surface Types

Abstract

Dust removal in ports exacerbates water shortages and coastal pollution, particularly in coal ports with significant dust production. Constructing green ports for water conservation and emissions reduction is the future direction for port development. Stormwater harvesting, especially through low-impact development (LID), emerges as an attractive solution in line with green ports. concept. However, determining optimal LID layouts is complex due to multiple objectives. Fluctuations in runoff coefficients, stemming from changing underlying surface types in coal ports, are often overlooked, resulting in costly and ineffective LID layouts that fail to adequately control runoff across varying scenarios. This study innovatively addresses the impact of underlying surface stochasticity on optimizing LID layouts in coal ports. First, the storm water management model (SWMM) is employed to simulate runoff changes in coal ports under various representative underlying surface scenarios, generated through the K-Medoids method. The analysis reveals a significant 70.6% variation in the stockyard’s total runoff during a 1-year 2-h design rainfall, ranging from 14,100 m3 to 48,000 m3. Subsequently, a multiobjective stochastic programming model for LID layout optimization is proposed, coupled with a surrogate model for SWMM. Two objectives for LID layout optimization are considered: investments and total runoff harvesting. Finally, the Nash bargaining solution is applied to balance the trade-off between the two objectives and obtain the optimal LID layout considering underlying surface stochasticity. Results indicate that the optimal LID layout has a cost of 1.636 billion Chinese yuan (CNY) and achieves a 50.98% runoff harvesting rate. Compared to previous studies ignoring underlying surface stochasticity, it demonstrates a 2% improvement in harvesting rate, a cost reduction of 200 million CNY, and shows higher robustness with 96 compliance instances out of 100 simulations. This study offers methodological support for developing economically efficient planning and construction schemes for stormwater harvesting facilities in coal ports.