多目标优化下的景观水文规划设计研究：优化框架与方法应用

Multi-Objective Optimization in Landscape Hydrology Planning and Design: Optimization Framework and Method Implementation

关键词:  风景园林  景观水文  雨洪管理  多目标优化  海绵城市  优化算法

Key words:  landscape architecture  landscape hydrology  stormwater management  multi-objective optimization  sponge city  optimization algorithm

摘要:

多目标优化理论和算法广泛的应用领域和前景为景观水文规划设计研究提供了交叉、整合和创新的切入口。全球气候变化和城市化背景下，规划设计人员在平衡水文效益、生态功能、空间质量和社会福祉等方面面临诸多挑战。在处理建成环境中景观涉水要素问题时，多目标优化寻求协同景观水文规划设计方法，建立起识别关键变量与设计目标的优化框架，提供应对多目标和多尺度层级的解决方案。探讨了多目标优化与景观水文规划设计相互整合的契机，并从优化变量、优化目标、约束条件和优化算法入手，剖析了设计导向下的优化框架建立流程。再者，对优化尺度层级进行分类，归纳寻优方法的应用，并辅以从后期决策视角的参考。旨在为城市雨洪管理、海绵城市建设中存在的典型多重效益协同与优化问题提供改善途径与解决思路。

Abstract:

The theory and algorithms of multi-objective optimization, with their wide range of applications and prospects, offer an interdisciplinary, integrative and innovative entry point for research in landscape hydrology planning and design. As urbanization intensifies and the effects of climate change become increasingly complex, planners and designers face growing challenges in balancing hydrological performance, ecological function, spatial quality, and social well-being. When dealing with landscape issues related to water elements in the built environment, multi-objective optimization can be integrated with landscape hydrology planning and design methods to establish an optimization framework that identifies key variables and design objectives, thereby providing solutions that address multiple objectives across different spatial scales. This study explores the opportunities for integrating multi-objective optimization with landscape hydrology planning and design, and analyzes the process of establishing a design-oriented optimization framework from the perspectives of optimization variables, objectives, constraints, and algorithms. Within this framework, optimization variables represent the controllable parameters of design, such as land use type, elevation modification, spatial layout, or rainwater facility attributes, that define the search space of possible solutions. In landscape hydrology, these variables are often multi-dimensional, involving not only single parameters but also combinations of spatial and functional attributes. Optimization objectives exhibit mutual exclusivity that improvement in one objective may compromise another, and reflect the desired performance metrics derived from four major groups: hydrological management, landscape benefits, environmental benefits, and economic costs. Constraints serve as the boundary conditions that ensure the feasibility and realism of optimization. They are classified into three categories: variable constraints, which limit the parameter ranges; objective constraints, which restrict performance indicators to acceptable thresholds; and computational constraints, which regulatealgorithmic efficiency and convergence. In the urban built environment, additional contextual constraints - such as land-use regulations, climatic conditions, and planning policies - must also be incorporated to ensure that optimization results align with practical design and regulatory frameworks. Furthermore, this study classifies optimization into hierarchical spatial scales that correspond to both hydrological and landscape design perspectives. These include four main levels: 1) the urban catchment scale, emphasizing blue-green spatial structure optimization to improve ecological protection and flood resilience; 2) the sub-catchment scale, focusing on coordinated planning among green, blue, and gray infrastructure systems; 3) the district or watershed unit scale, addressing layout optimization of impervious and pervious surfaces, stormwater facilities, and terrain adjustments; and 4) the site-scale hydrological unit, focusing on detailed design parameters such as vegetation, soil composition, and infiltration structure. Each scale requires specific variable definitions, objectives, and algorithmic strategies, forming a nested and scalable optimization hierarchy. Finally, this study systematically summarizes the application of various optimization methods, supplemented by insights from a post-decision-making perspective to enhance the interpretability and practical relevance of the optimization results. By integrating these analytical and decision-support dimensions, the research aims to provide comprehensive improvement pathways and solution strategies for addressing typical multi-benefit synergy and optimization challenges encountered in urban stormwater management and sponge city construction. The proposed integration framework not only advances the theoretical understanding of how optimization can inform landscape planning and hydrological design but also offers practical tools for achieving co-benefits across environmental, social, and spatial dimensions.