城市污水系统的规划受到系统重构和深度不确定性的双重挑战，要求系统规划理念从针对单一情景的“先预测后规划”向适应性规划转变。同时，在可持续发展要求的污水回用、节能降耗、韧性目标下，系统空间布局的重要性日益凸显。鉴于此，本研究旨在建立能够实现多目标下系统多设施协同规划、针对不同未来情景产出多阶段系统适应性路径的污水系统规划方法。本研究将适应性路径规划理论与城市污水系统相结合，以兼顾系统当前和未来的成本与效能为目的，构建了由问题概化、信息搜集与不确定性识别、初期方案集生成、适应性路径构建、路径评价与决策、方案执行与监测6个环节构成的城市污水系统适应性路径规划方法。其中，初期方案集生成和适应性路径构建是方法的核心环节。前者以初期的确定性城市空间边界、系统用户排放和需求等规划条件为输入，开发可持续性城市污水系统方案生成器，以建设和运行成本、污水回用量、运行能耗、系统结构韧性为目标，实现污水-再生水系统厂网布局和规模的协同优化。该生成器以有向无权图表征系统布局，以图的节点和边属性表征系统设施规模，构建系统多设施协同优化的数学模型，并开发基于蚁群的排水系统用户-管网-水厂全局路径搜索算法，实现模型的高效求解。适应性路径构建环节将每一时期的城市扩张、污水排放、再生水需求等因素影响下的系统适应性措施构建拆分为4个相对独立但互相约束的子问题，即污水子系统更新、再生水子系统更新、污水厂间调水管道建设和系统空间扩张，开发城市污水系统适应性路径模拟器对其逐一求解，从而模拟未来所有规划条件演变路径下系统建设方案的变化，最终产出系统适应性路径集。最后，方法在对各初期方案的当前成本效益、未来多情景下的适应成本综合评价比选后，选取确定性的初期方案进入执行阶段。以雄安新区起步区2025年至2035年间的污水系统规划为例，利用所建立的方法和工具进行适应性路径规划。以污水全处理、全回用为约束，以经济性和韧性为目标，生成6个帕累托最优方案作为2025年候选方案集。在不同城市扩张、不同污水和再生水量变化、不同决策偏好组成的810条未来规划条件演变路径下，构建了各候选方案的适应性路径。将规划结果和传统规划方案进行比较，表明了方法在案例地区的可行性和有效性。基于计算结果，进一步识别了污水厂数量和平均规模、管网总长度等系统结构指标与系统成本、性能间的关系。
The planning of urban wastewater systems （UWSs） is subject to the combined challenges of system reconfiguration and deep uncertainty. Adaptive planning, instead of ＂predict-then-act＂, is required as the guiding ideology in system planning. In the meantime, as sustainable development goals are added to the system‘s planning objectives, such as water reuse, energy conservation and resilience, the system‘s spatial layout plays an important role in system planning. To address this issue, this study aims to establish a method for UWS planning to achieve optimal design of multiple system facilities under multiple objectives, while generating multi-staged adaptation pathways for various future scenarios.This study combines the theory of Adaptation Pathway planning to UWSs. In order to consider the system‘s current and future cost and benefit, we establish the Urban Wastewater System Adaptation Pathway Planning Framework, which consist of six major steps, i.e. problem description, information collection and uncertainty identification, candidate solution generation for the initial planning stage, adaptation pathways development, pathways evaluation for decision making, execution and monitoring. The two core steps are candidate solution generation for the initial planning stage and adaptation pathways development. The Sustainable Urban Wastewater System Generator （SUWStor） is developed for candidate solution generation. Based on the certain planning condition for the first stage, including urban boundary, system user’s discharge and demands, and design constraints, it conducts a multi-objective optimization of both the layout and the scales of the sewage and reclaimed water system simultaneously. The SUWStor generalizes the system‘s layout as an unweighted directed graph, and uses the attributes of the nodes and edges to characterize the scale of the system facilities, in order to build a mathematical model for the optimization. The ant-colony based ＂user-pipe-plant global path search algorithm＂ is developed to solve the model efficiently. The step of adaptation pathways development characterizes the rehabilitation of each solution at each decision time step as 4 interconnected sub-problems, i.e. the rehabilitation of sewage subsystem, the rehabilitation of reclaimed water subsystem, system spatial expansion, and transfer pipeline construction between treatment plants. The Urban Wastewater System Adaptation Pathway Simulator （UWSAPS） is developed to successively solve these sub-problems for each future planning stage, in order to simulate the change in the UWS under all planning conditions in the future. The planning of the UWS in the starting area of Xiong‘an New District between 2025 and 2035 is taken as an example to test the established method and models. Considering 100% wastewater treatment rate and 100\% water reuse rate as constraints, with economic cost and structural resilience as objectives, 6 Pareto optimal solutions are generated as candidate solutions for the first planning stage, i.e. year 2025. The future adaptation pathways of each candidate solution in years 2030 and 2035 are developed under 810 transient scenarios, each containing different scenarios of urban expansion, sewage discharge, reclaimed water demand and decision preferences. Compared to the result of a traditional planning method, the result of the established method is proven to be more cost-effective in the first stage and more adaptive in the future, thus demonstrates the applicability and effectiveness of the established method and tools. Based on the calculation, the relationship between system performance and its structural characteristics, such as the number and average capacity of the treatment plants and the total length of the pipe network, is further identified.
