In view of the frequent waterlogging caused by rapid urbanization and the public’s dissatisfaction with the drainage system, the article based on the concept of the green ecological drainage system, constructed the urban green ecological drainage comprehensive simulation research system, and quickly evaluated pipe network operation and surface water of an industrial park under typical rainfall conditions. The results showed that the drainage capacity of the designed green ecological rainwater drainage system reached 100%, and there was no ponding phenomenon, which indicated that the green ecological rainwater drainage system could effectively solve the practical problems of urban drainage. The green ecological rainwater comprehensive simulation research system had good adaptability. The research results provided the scientific theoretical basis and reference significance for planning, designing, constructing, operating, and managing urban rainwater system scientifically and systematically.

In recent years, with the acceleration of urbanization, the impermeable area on the earth’s surface has increased sharply, and the gathering of rainwater has speeded up. Urban waterlogging has become another “high incidence” in cities after traffic congestion and haze [

At present, scholars worldwide have carried out researches on concepts of storm flood management and the waterlogging disaster simulation. In terms of management concepts, the US has successively proposed new concepts such as Best Management Practices and Application (BMP) [

This exploration focused on urban waterlogging in the process of urban development at this stage. The rainwater system and green ecological system were combined to form a green ecological concept of the urban rainwater system with Chinese characteristics. Besides, an ecological integrated simulation system of urban rainwater was built and applied to an industrial park. The operation of the regional pipe network and impounded surface water before and after the regulation and control were compared to explore the measures and methods of controlling excessive rainfall and reducing the load of the rainwater pipe network.

The green ecological rainwater system mainly emphasized the on-site recycling of urban rainwater resources, which integrated the drainage function of traditional rainwater facilities, the rainwater utilization functions of rainwater reuse facilities, the ecological restoration functions of rainwater infiltration facilities, and the disaster reduction functions of urban flood control facilities. Compared with the simple urban rainwater system for rainwater discharge, the green ecological rainwater system focused on the idea of recycling rainwater as a resource according to local conditions. It could comprehensively solve the problems of water resource shortage, water pollution, flood control and drainage, ecological restoration, and groundwater supply. Therefore, the urban green ecological rainwater system was a rainwater management system integrating the concepts of environment, ecology, and sustainable development.

Equations and mathematical expressions must be inserted into the main text. Two different types of styles can be used for equations and mathematical expressions. They are: in-line style, and display style.

Based on the theory of urban rainfall hydrology and hydraulics, with the support of computer simulation technology and numerical solution, a 1D drainage network model was established for an industrial park in semi-humid and semi-arid areas of China, and the real-time dynamic model of excess surface water was established by coupling the urban surface characteristics. At the same time, the green ecological rainwater facilities were generalized into the math module and added to this model under different rainfall conditions. Rapid qualitative and quantitative analysis of the rainwater drainage process improved the ability of the study area to deal with excess surface water induced by rainfall.

The 1D model of the rainwater pipe network was built by using MIKE URBAN software and could accurately describe the rainfall-runoff process and pipe flow process.

The rainfall-runoff model mainly adopted the equal flow time method (namely time-area curve method) [

where:

The pipe flow model could evaluate the unsteady flow dynamics in the whole pipe network in detail. The calculation was based on the Saint-Venant equations of 1D free surface flow, namely, the continuity equation (mass conservation) and momentum equation (momentum conservation-Newton’s second law), as shown in

The equation of mass conservation:

The equation of momentum conservation:

where: Q–cross-section discharge, m^{3}/s; A–cross-section area, m^{2}; y–the depth of water in the pipeline, m; X–a distance of pipeline along water direction, m; g–gravitational acceleration, m/s^{2}; t–time coordinates, s; α–momentum correction coefficient;

The 2D rapid assessment model of surface water was built by using the same software and described the interaction between surface free flow and rainwater pipe network, combining with the terrain and the distribution characteristics of buildings. The governing equation was the 2D shallow water equation along the water depth of Reynolds averaged Navier–Stokes equation, as shown in

The continuity equation of water flow:

The momentum equation of water flow along the

The momentum equation of water flow along the

where:

Based on the above principle, the basic principle of node generalization was used, and a green ecological rainwater model was designed. The model combined the relevant parameters of geometry size and hydrological parameters. According to the node water balance equation, the node storage and detention, infiltration, inflow, and outflow processes under rainfall conditions were simulated, as shown in

where: ^{2}; ^{2}.

The planned industrial park covers 13.867 hm^{2}, which is divided into a working area of 12.746 hm^{2} and a living area of 1.121 hm^{2}. The specific distribution is shown in

Land-use type | Area (hm^{2}) |
Percentage (%) |
---|---|---|

Buildings | 5.010 | 36.13 |

Road | 2.802 | 20.21 |

Greenbelt | 1.404 | 10.12 |

Square | 2.341 | 16.89 |

Parking lots | 0.332 | 2.39 |

Other areas | 1.978 | 14.26 |

Total | 13.867 | 100 |

Due to the lack of measured rainfall data and the small study area, the Chicago storm pattern method was adopted to synthesize rainfall considering that the short duration of heavy rainfall has a greater impact on the region. The rainstorm equation was derived from the latest annual maximum sampling method in the

where:

A large number of studies at home and abroad showed that the short duration of rainfall was mainly in the form of a single peak. The relative position of the rain peak was basically between 0.30 and 0.5. This time, 0.35 was taken, and the recurrence periods were 1a, 1.4a, 2a, 3a, and 5a. The rainfall process was shown in

1) Generalization of the Pipe Network

The model was used for the digital processing of pipe network data. 102 sections of circular pipe with a total length of 2916.4 m, 101 inspection wells, 1 drainage outlet, and 101 catchment areas were generalized. The pipe network topology was shown in

The processed data was loaded into the model in the form of Geodatabase.

2) Terrain Generalization

According to the measured 1:1000 topographic map, a digital elevation map (DEM) was obtained after processing in the model. The grid accuracy was 5 m × 5 m, the number of data nodes was 2,144,597, and the number of grids was 250,104, as shown in

Land-use type | Area (ha) | Percentage (%) | Coefficient of imperviousness (%) |
---|---|---|---|

Buildings | 5.01 | 36.13 | 90 |

Road | 2.802 | 20.21 | 90 |

Greenbelt | 1.404 | 10.12 | 20 |

Square | 3.973 | 28.65 | 65 |

Parking lots | 0.332 | 2.39 | 100 |

Other areas | 0.346 | 2.50 | 60 |

Total | 13.867 | 100 | / |

3) Boundary Condition

The upstream boundary of the pipe network model was the loaded rainfall time series file, and the downstream boundary was the water level value of the pipe network outlet, which would not be processed this time. The surface waterlogging model needed to establish the land closed boundary, which was to set a particular land value outside the study area.

4) Simulation Time

The time step was consistent with the time interval of rainfall data, taking 5 min. The groundwater collection time in the catchment area of the pipe network model was set at 5–10 min. The time step of the simulation calculation was the self-adaptive adjustment step size, and the variable range of the step size was 10–60 s. The simulation step length of the surface water accumulation model was 1 s.

Referring to

5) Model Parameter Setting

The rapid assessment model was based on the lack of detailed rainfall data. The parameters of the 1D pipe network model and excess surface water model are set referring to the 2016 edition of

Models | Parameters | Parameter values |
---|---|---|

Pipe network model | Initial rainfall loss (mm) | 0.6 |

Hydrological attenuation coefficient | 0.9 | |

Surface confluence velocity (m/s) | 0.3 | |

Pipe roughness (M) | 80 | |

Excess surface water model | Dry water depth (m) | 0.002 |

Water depth (m) | 0.003 | |

The surface eddy viscosity coefficient | 0.02 | |

Ground roughness (M) | 32 | |

Coupled model | maximum flow (m^{3}/s) |
0.1 |

inlet area (m^{2}) |
0.16 |

The hydrological parameters of green ecological rainwater facilities were determined according to the regional demand and referring to previous studies. The parameters were facility type, infiltration form, infiltration rate, porosity, and soil hydraulic conductivity. The values were shown in

Facility type | Infiltration form | Infiltration rate | Porosity | Saturated hydraulic |
---|---|---|---|---|

Green roof | Unsteady infiltration | / | 0.35–0.5 | 10^{−6}−10^{−3} |

Permeable pavement | No infiltration | / | 0.12–0.21 | / |

Sunken greenbelt | Constant velocity infiltration | 0.0006–0.001 | 0.5–0.75 | / |

6) Model Calibration and Verification

Since this study was for planning an industrial park and lacks measured data, the model was calibrated and verified based on the runoff coefficient. The basic principle was that the runoff coefficient was used as the objective calibration function of the model parameters, and the empirical parameters of the model were calibrated by comparing the actual value of the urban comprehensive runoff coefficient with the simulated value of the model. After calibration and verification, the values of parameters in the model were shown in

The drainage capacity of the pipe network and the water accumulation in the study area were taken as the main research evaluation indexes. The comparison between the designed model and the traditional model was mainly analyzed. The annual runoff control rate was set at 80%, and the control comparison between the green ecological rainwater system model and the traditional model was analyzed and compared. ArcGIS10.2 software (Esri, Redlands, USA) was mainly used for model simulation to judge the effect of the green ecological rainwater system model.

1) Operation Analysis of Pipe Network

The pipe network load capacity was analyzed based on whether the pipe network reached the maximum pipe filling. The calculation was shown in

where:

Drainage capacity | Less than once a year | Once 1–1.4 years | Once 1.4–2 years | Once 2–3 years | Once 3–5 years | Once more than 5 years | Total |
---|---|---|---|---|---|---|---|

Length (m) | 1528.462 | 596.611 | 471.577 | 210.712 | 56.728 | 52.31 | 2916.4 |

Percentage (%) | 52.41 | 20.46 | 16.17 | 7.23 | 1.95 | 1.78 | 100 |

According to the simulation results, under the rainfall condition of once 1.4 years, most of the pipe filling has reached the critical value. The longer the recurrence period, the more filled pipe sections. The overall analysis shows that the drainage capacity of the current regional pipe network was inferior, which was mainly due to the low design standard of the pipe network. Concurrently, the traditional planning model fails to integrate the pipe network system with green ecological facilities to realize the on-site rainwater absorption and utilization.

2) Analysis of Excess Surface Water

The surface flow was mainly considered from the water accumulation after the rainwater overflows from the generalized node, ignoring the ponding phenomenon before the surface runoff enters the pipe network. The water accumulation range and depth at max depth time under the set rainfall condition were calculated as the analysis indexes to directly reflect the dynamic changes of range and depth of excess surface water. The distribution of accumulated water was shown in

Recurrence P | Max depth (m) | Max area (m^{2}) |
Percentage (%) |
---|---|---|---|

1a | 0.120 | 3400 | 2.45 |

1.4a | 0.240 | 6300 | 4.54 |

2a | 0.674 | 9600 | 6.92 |

3a | 0.676 | 11700 | 8.44 |

5a | 0.677 | 13500 | 9.74 |

The simulation results suggested that there was a positive correlation between the surface ponding area, ponding depth, and the design rainstorm recurrence period. The distribution of ponding in the study area was similar under different rainfall scenarios, and the water depth was mainly 0–0.677 m. With the increase of rainfall magnitude, the ponding area expands continuously, and the maximum ponding depth was increasing. The investigations tracing to the source of the pipe network

The green ecological rainwater system model was used to plan and design the rainwater system under the rainfall condition of 1.4a 2 h. The current simulation results were compared and analyzed to evaluate the improvement effect of green ecological rainwater facilities on urban water accumulation. The simulation and comparison results of the pipe network drainage capacity were shown in

Types | Total pipe length (m) | Pipe filling < 1 (m) | Percentage (%) |
---|---|---|---|

Current rainwater system | 2916.4 | 791.327 | 27.13 |

Green ecological rainwater system | 2916.4 | 2916.4 | 100 |

The results in ^{2}; there was no ponding when the green ecological rainwater system was used and the simulation effect of the model was good. The results above showed that after the improvement of green ecological rainwater facilities in the study area, the bearing capacity of the rainwater pipe network and the excess surface water were significantly improved. There was almost no runoff discharge in the study area, which has achieved the research goal. Therefore, the green ecological rainwater system model could effectively reduce the risk of ponding.

Types | Max water depth (m) | Max water area (hm^{2}) |
Percentage (%) |
---|---|---|---|

Current rainwater system | 0.240 | 0.63 | 4.54 |

Green ecological rainwater system | 0 | 0 | 0 |

With an industrial park as an example, a comprehensive simulation and research system of the green ecological drainage system was constructed to quickly evaluate the capacity of regional drainage facilities under different rainfall recurrence periods. At the same time, based on this model, rainwater regulation and control were carried out in the study area. Besides, the operation of the regional pipe network and drainage load, before and after regulation and control, were compared and evaluated. The conclusions are as follows:

The concept of an urban green ecological rainwater system has been defined, which is composed of rainwater engineering facilities including rainwater infiltration, storage, and reuse, storage and peak shaving, infiltration and drainage, rainwater purification, and new rainwater pipeline drainage. After upgrading the green ecological rainwater system model in the study area, the drainage pressure of the pipe network and surface water accumulation is effectively reduced. It proves the applicability of the model in the evaluation of green ecological measures. The simulation results show that the drainage capacity of the model is 100%, and there is no ponding phenomenon. The designed rainwater pipeline analysis model has the advantages of simple modeling and fast calculation. Therefore, the research content has a very important practical application value and reference value, which is very suitable for the needs of ecological city construction at this stage. However, due to limited data, only a rapid evaluation model has been built to simulate and analyze a typical area. Next, the simulation of the model will be expanded and compared with the real-time monitoring of the drainage and ponding process to enhance the reliability of the model and improve the simulation accuracy.