Knowledge-driven optimization method for combined activation scheme of flood storage and detention areas and gate control and flood diversion strategy

Abstract

The present invention relates to the technical field of flood control and scheduling of water projects, and discloses a knowledge-driven flood storage and detention area combination activation scheme and gate control flood diversion strategy optimization method, comprising the following steps: collecting basic data and generalizing a joint flood control model of a flood storage and detention area engineering group; constructing a real-time optimization model for joint flood control of a flood storage and detention area engineering group; extracting knowledge on the impact of flood storage and detention area utilization and predicting the impact of combined flood storage and detention area utilization; and optimizing the flood storage and detention area combination activation scheme and gate control flood diversion strategy based on a multi-objective coordination mechanism. The present invention's knowledge-driven flood storage and detention area combination activation scheme and gate control flood diversion strategy optimization method quickly and quantitatively determines the impact of flood diversion on the flood discharge level of a river section, analyzes and formulates a coordination mechanism for overall flood control safety and flood diversion damage in the basin, designs a flood storage and detention area combination activation and gate control flood diversion strategy optimization method, automatically compares different flood storage and detention area combination activation schemes and different flood diversion strategies for the same flood storage and detention area, and achieves fast and convenient calculations.

Description

Technical Field

[0001] The present invention relates to the technical field of flood control and scheduling of water projects, and in particular to a knowledge-driven combined activation scheme for flood storage and detention areas and a gate control and flood diversion strategy optimization method. Background Art

[0002] Floods pose a serious threat to the safety of life and property, as well as the stable development of the social economy. To combat floods, we can vigorously develop water conservancy projects. Currently, major river basins have essentially completed a system of flood control and disaster reduction projects, including reservoirs, embankments, and flood storage and detention areas. In actual operation, through the rational operation of these project groups, they fully utilize the flood control functions of storage, discharge, and distribution, effectively enhancing the river basin's flood control capacity and significantly reducing flood damage. Among them, flood storage and detention areas are a crucial component of the river basin's flood control engineering system. Some of these areas also serve to prevent excessive floods, acting as the "trump card" for ensuring flood control safety in the river basin. The scientific use of flood storage and detention areas can enhance the river basin's defense capabilities and is an important manifestation of the further scientific development of bottom-line and limit-based thinking in flood control.

[0003] According to the Flood Control Law, flood storage and detention areas refer to low-lying areas and lakes outside the backwaters of river embankments, including flood diversion outlets, that temporarily store floodwater. When a major flood occurs in a river basin, and despite appropriate reservoir regulation and sufficient river flow, the incoming flood volume still exceeds the river's discharge capacity, the flood storage and detention areas must be activated to store the excess flood volume, preventing embankment breaches and potentially devastating disasters. In other years, restricted production and daily life activities can be carried out within the flood storage and detention areas. my country has established 98 national-level flood storage and detention areas, covering a total area of ​​approximately 34,000 square kilometers and with a storage capacity exceeding 100 billion cubic meters. Formulating a rational plan for the activation of flood storage and detention areas not only helps fully utilize limited flood diversion resources and maximize overall flood control safety in the river basin, but also effectively controls flood diversion losses and minimizes the socioeconomic impact of flood diversion. In particular, with the recent progress in flood storage embankments, flood diversion outlet sluices, and safety zone development, it has become possible to divert floodwaters from these areas in a timely and appropriate manner according to established plans.

[0004] Currently, flood management plans for various river basins provide general guidance for the use of flood storage and detention areas. These plans categorize groups of flood storage and detention areas within a river reach based on factors such as activation probability and importance, clarifying the order in which each type of area should be activated. When water levels at key control stations are judged to be at risk of exceeding the guaranteed level, each type of area is deployed in batches as needed. However, these plans lack a clear sequence for the sequential activation of flood storage and detention areas within the same category, as well as their flood diversion strategies. This makes it difficult to meet the practical needs of flood control decisions requiring the individual activation of flood storage and detention areas and precise flood diversion. Activating different combinations of flood storage and detention areas with varying distances to control stations and upstream and downstream relationships will result in different drops in water levels at control stations. Furthermore, activating flood storage and detention areas with varying overall socioeconomic development will result in different flooding losses. Activating the same flood storage and detention area using different strategies will also result in different drops in water levels at control stations and different flooding losses. Therefore, it is necessary to balance the flood diversion effects and flood losses of different combinations of flood storage and detention area activation plans and different flood diversion strategies within the same flood storage and detention area to determine the appropriate combination of flood storage and detention area activation plans. In actual flood control operations, several combined activation schemes for flood storage and detention areas are typically developed based on expert experience. Hydrodynamic models are then used to calculate the effects of each scheme and compare and select the appropriate options. While practical, this approach requires multiple trial calculations and analyses, lacks automation, results in a small number of options, and inadequate comparisons. The resulting scheme is not optimal, and relies on complex hydrodynamic models that require multiple global calculations across the entire study area, resulting in slow computation and inability to meet the real-time requirements of flood control operations.

[0005] Therefore, in order to give full play to the flood control and disaster reduction potential of flood storage and detention areas, to ensure the flood control safety of the basin to the greatest extent, to reduce flood diversion losses, and at the same time, to quickly and automatically generate scheduling plans, it is urgent to propose a technology for optimizing and determining the combined activation plans of flood storage and detention areas that is coupled with the knowledge of flood storage and detention area application. Summary of the Invention

[0006] The purpose of the present invention is to address the deficiencies of the above-mentioned technologies and provide a knowledge-driven flood storage and detention area combination activation plan and gate control flood diversion strategy optimization method, which can quickly and quantitatively determine the impact of flood diversion on the flood discharge level of the river section, analyze and formulate the overall flood control safety and flood diversion disaster coordination mechanism of the basin, design the flood storage and detention area combination activation and gate control flood diversion strategy optimization method, automatically compare different flood storage and detention area combination activation plans and different flood diversion strategies for the same flood storage and detention area, and calculate quickly and conveniently.

[0007] To achieve the above objectives, the present invention provides a knowledge-driven flood storage and detention area combined activation scheme and gate control flood diversion strategy optimization method, which includes the following steps:

[0008] 1) Collect basic data and generalize the joint flood control model of flood storage and detention area project groups;

[0009] 2) Constructing a real-time optimization model for joint flood control of flood storage and detention area project groups;

[0010] 3) Extracting knowledge about the impact of flood storage and detention area utilization and predicting the impact of combined flood storage and detention area utilization;

[0011] 4) Optimization of flood storage and detention area combination activation scheme and gate control and flood diversion strategy based on a multi-objective coordination mechanism.

[0012] Preferably, in said step 1), the basic data collected include basic data of flood storage and detention areas and flood control stations, and also include basic data of flood control scheduling in the study area. The basic data of flood storage and detention areas include the conditions for activation of the flood storage and detention areas, effective flood storage capacity, the curve of the relationship between flood storage capacity and water storage level in the area, the curve of the relationship between flood storage capacity and flooding loss, and socio-economic information. The basic data of flood control stations include the flood process predicted by the flood control stations, the relationship between water level and flow, the characteristic water level for flood control, and the safe discharge volume. The basic data of flood control scheduling in the study area include the flood scheduling plan, flood defense plan, flood control manual, and flood risk map of the study area.

[0013] Preferably, in said step 1), when generalizing the joint flood control model of the flood storage and detention area project groups, the flood control objectives and scheduling objects are determined, starting from the flood control needs of the river basin, with important flood control sections as nodes, the river sections in the area are divided into several sections, and flood control regulation is carried out separately for each river section. On this basis, the flood storage and detention areas are divided into different project groups, including important flood storage and detention area project groups, general flood storage and detention area project groups and reserved flood storage and detention area project groups, by utilizing the existing flood control control objectives of each river section and with each flood storage and detention area project group as the scheduling object.

[0014] Preferably, in step 2), when constructing a real-time optimization model for joint flood control of a flood storage and detention area project group, during the scheduling period t , based on the water level forecast process of each flood control station Z τ ,τ=t,t+1,…,t+h-1 As model input, h For the forecast period, the model boundary conditions are set according to the approved flood control dispatching rules, and the dispatching period is constructed as [ t , t + h ]'s combined flood control optimization scheduling model for flood storage and detention area project groups is proposed. The operation of flood storage and detention area projects should coordinate the two competing control objectives of overall flood control safety in the basin and flood diversion and inundation losses.

[0015] Preferably, in order to ensure the overall flood control safety of the basin, the water level at the flood control station is set to be lower than the guaranteed water level as a hard constraint. On this basis, the maximum water level of the flood control station is controlled as much as possible to ensure the safe operation of the embankment, and is represented by the minimum sum of the squares of the water levels at the flood control station in each period; the flood diversion and inundation losses cover both economic and social losses. The social losses are represented by the number of people transferred, and the economic losses are represented by the reduction in regional GDP.

[0016] Preferably, the objective function of the joint flood control optimization scheduling model of the flood storage and detention area project group includes: the minimum sum of the squares of the water levels at the flood control station in each period , the minimum number of transferred population , where n is the number of activated flood storage areas, P j Flood storage area j The number of people involved in the activation will be relocated, and the economic losses from flood storage and detention area flooding will be minimal. , where GDP j Flood storage area j The reduction in regional GDP after flooding.

[0017] Preferably, the constraints of the joint flood control optimization scheduling model for the flood storage and detention area project group include:

[0018] Water balance constraints: S j,τ+1 = S j,τ + W j,τ , where S j,τ 、 S j,τ+1 Periods τ Initial and final flood storage areas j The flood storage capacity, W j,τ for τ Periodic flood storage area j Flood diversion capacity;

[0019] Flood flow constraint for flood storage and detention areas: 0≤ Q j,τ ≤ Q j,max , where Q j,τ Flood storage area j Time τ The diversion flow, Q j,max Flood storage area j The maximum permissible flood diversion flow rate;

[0020] Flood storage capacity constraint of flood storage area: 0≤ S j,τ ≤ S j,max , where S j,max Flood storage area j effective flood storage capacity;

[0021] Conditions for activation of flood storage and detention areas: Zc,max,before ≥ Z c,safe , where Z c,max,before The highest water level of the associated flood control station in the forecast period when the flood storage area of ​​this group does not divert floodwaters. Z c,safe Maintaining water levels for associated flood control stations;

[0022] Constraints on the number of flood storage areas in this group: n ≤ b , where n is the number of activated flood storage areas, b is the total number of flood storage and detention areas in this group;

[0023] The activation order constraints of various flood storage and detention area groups: The activation order of various flood storage and detention areas is to activate important flood storage and detention areas first, then general flood storage and detention areas, and finally reserved flood storage and detention areas;

[0024] Coordination constraints for flood control safety of main and tributary rivers: If the water level of the tailwater control station exceeds its control level, the corresponding flood storage area of ​​the tailwater will be used first; otherwise, the corresponding flood storage area of ​​the main stream will be used first;

[0025] Basin flood control constraints: Z c,max,after ≥ Z c,safe , where Z c,max,after This is the highest water level during the foreseeable period at the flood control station after the flood storage and detention area is activated to divert flood water.

[0026] Preferably, in step 3), when extracting the knowledge of the impact of the use of the flood storage and detention area, the impact of the use of the flood storage and detention area is represented by the amplitude of the change in water level and flow at the flood control station before and after the use of the flood storage and detention area;

[0027] When the impact of a single flood storage and detention area on the hydrological conditions of a flood control station is insensitive to water regime changes, a knowledge base on the impact of flood storage and detention area application is constructed. Using a calibrated and verified one-dimensional hydrodynamic model, the changing process of the hydrological conditions of the flood control station corresponding to a typical engineering application strategy is simulated and generated. In real-time applications, the corresponding knowledge base can be retrieved based on a given flood storage and detention area application strategy to obtain the corresponding changing process of the hydrological conditions of the flood control station.

[0028] When the impact of the use of a single flood storage and detention area on the hydrological conditions of the flood control station is sensitive to changes in water conditions, a hydrodynamic proxy model is constructed. Based on the one-dimensional hydrodynamic model, the impact samples of the use of a single flood storage and detention area under different hydrological boundary conditions and engineering application strategies are simulated and generated. Based on the samples, the artificial neural network data-driven model is used to construct the data relationship between the hydrological boundary conditions, the use strategy of the flood storage and detention area and the application impact, forming the knowledge of the impact of the use of a single flood storage and detention area. In actual applications, according to the real-time hydrological boundary conditions and the use strategy of the flood storage and detention area, the corresponding hydrological condition change process of the flood control station is obtained from this data relationship. When there is sufficient historical flood data, different historical flood processes can be directly used as hydrological boundary conditions. If the historical flood data is insufficient, the probability distribution characteristics of historical floods are analyzed, and the multi-station runoff random simulation method is used to generate artificial flood processes.

[0029] When predicting the impact of the combined use of flood storage and detention areas, the knowledge of the impact of the use of a single flood storage and detention area is iteratively applied to obtain the impact of the combined use of flood storage and detention areas. The knowledge of the impact of the use of a single flood storage and detention area is applied one by one to predict the changes in the hydrological conditions of the flood control stations during the forecast period after each flood storage and detention area is used according to the set activation plan and flood diversion strategy. The hydrological condition process of the flood control station is updated, and the initial boundary of the next flood storage and detention area use prediction is set accordingly until the impact of the use of all flood storage and detention areas has been calculated.

[0030] Preferably, in step 4), when optimizing the combined activation plan of the flood storage and detention areas and the gate control strategy based on the multi-objective coordination mechanism, the water level of the flood control station is set to be lower than the guaranteed water level as a hard constraint to ensure the overall flood control safety of the basin, and the primary goal of disaster loss control is to minimize the number of people to be relocated. On this basis, the economic losses caused by the diversion of flood storage projects are further controlled.

[0031] Preferably, when optimizing the combined activation plan of flood storage and detention areas, in various groups of flood storage and detention area projects, the combined activation plan of flood storage and detention areas is determined by the activation objects and activation order, and the population of the transfer area is sorted from small to large. In response to the incoming water situation, the flood storage and detention areas are activated one by one in sequence until the solution that the water level at the flood control station is lower than the guaranteed water level is found. Then, the flood storage projects that need to be activated and the activation order can be determined;

[0032] When optimizing the flood control and diversion strategy of the flood storage and detention area, the flood control and diversion strategy of the flood storage and detention area is determined by three decision variables: the activation time, the diversion flow and the number of diversion days. The knowledge of the impact of the use of the flood storage and detention area is used to simulate the water level of the flood control point after the flood storage and detention area is activated according to a certain flood diversion strategy. If the flood control constraints of the flood storage and detention area are satisfied, that is, the water level of the flood control station does not exceed the protection level and the levee is within the safe operation range, which can ensure the flood control safety of the basin, the flood diversion strategy is optimized with the goal of minimizing the economic loss of the flood diversion operation of the flood storage and detention area. Otherwise, the flood diversion strategy is optimized with the goal of minimizing the sum of the squares of the water levels of the flood control stations in each period to maximize the protection of the flood control safety of the basin.

[0033] Compared with the prior art, the present invention has the following advantages:

[0034] 1. A comprehensive knowledge base and data-driven model are used to mine knowledge on the impact of flood storage and detention area scheduling, replacing the hydrodynamic mechanism model to quickly and quantitatively determine the impact of flood diversion on the flood flow level of the river section;

[0035] 2. By analyzing and formulating a coordination mechanism for overall flood control safety and flood diversion losses in the basin, we can design a combination activation plan for flood storage and detention areas and an optimization strategy for gate control and flood diversion strategies. This allows for automatic comparison of different combination activation plans for flood storage and detention areas, as well as different flood diversion strategies for the same flood storage and detention area, with fast and convenient calculations.

[0036] 3. It fills the technical gap of real-time dispatching and application of flood storage and detention area engineering groups in my country, and is a shortcoming-filling work for real-time flood control dispatching of flood storage and detention areas in the post-engineering period, and has engineering application and promotion value. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 The technical roadmap for the knowledge-driven combined activation scheme of flood storage and detention areas and the optimization method of gate control and flood diversion strategy of the present invention;

[0038] Figure 2 A flow chart showing the impact prediction of combined utilization of flood storage and detention areas in the present invention;

[0039] Figure 3 This is the knowledge base of the impact of the 3000m³ / s flood diversion flow in the F3 flood storage area on the water level of the S4 station in this embodiment;

[0040] Figure 4 The water level change process of S4 before and after flood discharge in the flood storage and detention area in this embodiment;

[0041] Figure 5 Optimize the flood diversion strategy for the flood storage and detention area project group in this embodiment. DETAILED DESCRIPTION

[0042] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0043] like Figure 1 As shown in FIG, a knowledge-driven flood storage and detention area combination activation scheme and gate control flood diversion strategy optimization method includes the following steps:

[0044] 1) Collect basic data and generalize the joint flood control model of flood storage and detention area project groups;

[0045] 2) Constructing a real-time optimization model for joint flood control of flood storage and detention area project groups;

[0046] 3) Extracting knowledge about the impact of flood storage and detention area utilization and predicting the impact of combined flood storage and detention area utilization;

[0047] 4) Optimization of flood storage and detention area combination activation scheme and gate control and flood diversion strategy based on a multi-objective coordination mechanism.

[0048] Specifically, in step 1), the basic data collected include the basic data of flood storage and detention areas and flood control stations, as well as the basic data of flood control scheduling in the study area. The basic data of flood storage and detention areas include the conditions for activation of flood storage and detention areas, effective flood storage capacity, the curve of the relationship between flood storage capacity and water storage level in the area, the curve of the relationship between flood storage capacity and flooding loss, and socio-economic information. The basic data of flood control stations include the flood process predicted by the flood control station, the relationship between water level and flow, the characteristic water level for flood control, and the safe discharge volume. The basic data of flood control scheduling in the study area include the flood scheduling plan, flood defense plan, flood control manual, and flood risk map of the study area.

[0049] In step 1), when generalizing the joint flood control model of flood storage and detention area project groups, the flood control objectives and scheduling objects are determined. Starting from the flood control needs of the basin, the river sections in the area are divided into several sections with important flood control sections as nodes, and flood control and regulation are carried out separately for each river section. On this basis, the flood storage and detention areas are divided into different project groups, including important flood storage and detention area project groups, general flood storage and detention area project groups, and reserved flood storage and detention area project groups, based on the flood control control objectives of each river section and with each flood storage and detention area project group as the scheduling object. The joint flood control model of the flood storage and detention area project groups is generalized separately.

[0050] In step 2), when constructing the real-time optimization model for joint flood control of flood storage and detention area project groups, during the scheduling period t , based on the water level forecast process of each flood control station Z τ ,τ=t,t+1,…,t+h-1As model input, h For the forecast period, the model boundary conditions are set according to the approved flood control dispatching rules, and the dispatching period is constructed as [ t , t + h ]'s combined flood control optimization scheduling model for flood storage and detention area project groups is proposed. The operation of flood storage and detention area projects should coordinate the two competing control objectives of overall flood control safety in the basin and flood diversion and inundation losses.

[0051] Among them, in order to ensure the overall flood control safety of the basin, the water level of the flood control station is lower than the guaranteed water level as a hard constraint. On this basis, the maximum water level of the flood control station is controlled as much as possible to ensure the safety of the embankment operation, and it is represented by the minimum sum of the squares of the water levels at the flood control station in each period; the flood diversion and inundation losses cover both economic and social losses. The social losses are represented by the number of transferred people, and the economic losses are represented by the reduction in regional GDP.

[0052] The objective function of the joint flood control optimization scheduling model of the flood storage and detention area project group includes: the minimum sum of the squares of the water levels at the flood control station in each period , the minimum number of transferred population , where n is the number of activated flood storage areas, P j Flood storage area j The number of people involved in the activation will be relocated, and the economic losses from flood storage and detention area flooding will be minimal. , where GDP j Flood storage area j The reduction in regional GDP after flooding.

[0053] The constraints of the joint flood control optimization scheduling model for flood storage and detention area project groups include:

[0054] Water balance constraints: S j,τ+1 = S j,τ + W j,τ , where S j,τ 、 S j,τ+1 Periods τ Initial and final flood storage areas j The flood storage capacity, W j,τ for τ Periodic flood storage area j Flood diversion capacity;

[0055] Flood flow constraint for flood storage and detention areas: 0≤ Q j,τ ≤ Q j,max, where Q j,τ Flood storage area j Time τ The diversion flow, Q j,max Flood storage area j The maximum permissible flood diversion flow rate;

[0056] Flood storage capacity constraint of flood storage area: 0≤ S j,τ ≤ S j,max , where S j,max Flood storage area j effective flood storage capacity;

[0057] Conditions for activation of flood storage and detention areas: Z c,max,before ≥ Z c,safe , where Z c,max,before The highest water level of the associated flood control station in the forecast period when the flood storage area of ​​this group does not divert floodwaters. Z c,safe Maintaining water levels for associated flood control stations;

[0058] Constraints on the number of flood storage areas in this group: n ≤ b , where n is the number of activated flood storage areas, b is the total number of flood storage and detention areas in this group;

[0059] The activation order constraints of various flood storage and detention area groups: The activation order of various flood storage and detention areas is to activate important flood storage and detention areas first, then general flood storage and detention areas, and finally reserved flood storage and detention areas;

[0060] Coordination constraints for flood control safety of main and tributary rivers: If the water level of the tailwater control station exceeds its control level, the corresponding flood storage area of ​​the tailwater will be used first; otherwise, the corresponding flood storage area of ​​the main stream will be used first;

[0061] Basin flood control constraints: Z c,max,after ≥ Z c,safe , where Z c,max,after This is the highest water level during the foreseeable period at the flood control station after the flood storage and detention area is activated to divert flood water.

[0062] In step 3), when extracting the knowledge of the impact of flood storage and detention area operation, the impact of flood storage and detention area operation is represented by the amplitude of the change in water level and flow at the flood control station before and after the operation of the flood storage and detention area;

[0063] When the impact of a single flood storage and detention area on the hydrological conditions of a flood control station is insensitive to water regime changes, a knowledge base on the impact of flood storage and detention area application is constructed. Using a calibrated and verified one-dimensional hydrodynamic model, the changing process of the hydrological conditions of the flood control station corresponding to a typical engineering application strategy is simulated and generated. In real-time applications, the corresponding knowledge base can be retrieved based on a given flood storage and detention area application strategy to obtain the corresponding changing process of the hydrological conditions of the flood control station.

[0064] When the impact of the use of a single flood storage and detention area on the hydrological conditions of the flood control station is sensitive to changes in water conditions, a hydrodynamic proxy model is constructed. Based on the one-dimensional hydrodynamic model, the impact samples of the use of a single flood storage and detention area under different hydrological boundary conditions and engineering application strategies are simulated and generated. Based on the samples, the artificial neural network data-driven model is used to construct the data relationship between the hydrological boundary conditions, the use strategy of the flood storage and detention area and the application impact, forming the knowledge of the impact of the use of a single flood storage and detention area. In actual applications, according to the real-time hydrological boundary conditions and the use strategy of the flood storage and detention area, the corresponding hydrological condition change process of the flood control station is obtained from this data relationship. When there is sufficient historical flood data, different historical flood processes can be directly used as hydrological boundary conditions. If the historical flood data is insufficient, the probability distribution characteristics of historical floods are analyzed, and the multi-station runoff random simulation method is used to generate artificial flood processes.

[0065] When predicting the impact of combined use of flood storage and detention areas, the impact knowledge of single flood storage and detention area use is iteratively applied to obtain the impact of combined use of flood storage and detention areas. Figure 2 As shown in the figure, the knowledge of the impact of the use of a single flood storage and detention area is applied one by one to predict the changes in the hydrological conditions of the flood control stations during the forecast period after each flood storage and detention area is used according to the set activation plan and flood diversion strategy. The hydrological condition process of the flood control station is updated and the initial boundary of the next flood storage and detention area use prediction is set accordingly until the impact of the use of all flood storage and detention areas has been calculated.

[0066] In step 4), when optimizing the combined activation plan and gate control strategy of the flood storage and detention areas based on the multi-objective coordination mechanism, compared with other multi-objective optimization scheduling problems with unclear objective weights, the optimization objectives of basin flood control safety and flood diversion damage control in this problem have clear priorities. The water level of the flood control station is lower than the guaranteed water level as a hard constraint to ensure the overall flood control safety of the basin. In accordance with the principle of people first and life first, the primary goal of disaster damage control is to minimize the number of people transferred. On this basis, the economic losses caused by the flood diversion of flood storage projects are further controlled.

[0067] When optimizing the combined activation plan of flood storage and detention areas, the combined activation plan of flood storage and detention areas is determined by the activation objects and activation order in various flood storage and detention area project groups. Considering that most flood storage and detention areas in my country have not yet achieved block-by-block construction, once activated, all the population in the relocated areas will be sorted from the smallest to the largest according to the number of people in the relocated areas. Based on the water inflow situation, the flood storage and detention areas will be activated one by one in order until the water level at the flood control station is lower than the guaranteed water level. In this way, the flood storage projects that need to be activated and the activation order can be determined.

[0068] When optimizing the flood control and diversion strategy of the flood storage and detention area, the flood control and diversion strategy of the flood storage and detention area is determined by three decision variables: the activation time, the diversion flow and the number of diversion days. The knowledge of the impact of the use of the flood storage and detention area is used to simulate the water level of the flood control point after the flood storage and detention area is activated according to a certain flood diversion strategy. If the flood control constraints of the flood storage and detention area are satisfied, that is, the water level of the flood control station does not exceed the protection level and the levee is within the safe operation range, which can ensure the flood control safety of the basin, the flood diversion strategy is optimized with the goal of minimizing the economic loss of the flood diversion operation of the flood storage and detention area. Otherwise, the flood diversion strategy is optimized with the goal of minimizing the sum of the squares of the water levels of the flood control stations in each period to maximize the protection of the flood control safety of the basin.

[0069] Taking a certain section of a river as an example, basic data are first collected, including the activation conditions of each flood storage and detention area, effective flood storage capacity, flood diversion conditions, the relationship curve between flood diversion volume and flooding loss, social and economic related information, and the flood process forecast and flood control characteristic water level of the key flood control control stations (S1 station, S2 station, S3 station, S4 station) on the main stream of a river above the R1 river section. Flood scheduling plans, flood prevention plans, flood risk maps of each flood storage and detention area, and the correlation between the flood inflow of each flood storage and detention area and the flood loss rate of various assets are collected.

[0070] Then, the joint flood control model of the flood storage and detention area project group was generalized. According to the flood control control target of the main stream of a river above the R1 river section in the flood scheduling plan and flood defense plan, the research area was divided into R0 section and R1 section with S1 station and S4 station as nodes. The prominent contradiction of flood control of a river was mainly concentrated in the R1 section. With S4 station as the flood control control target, according to the classification and batch utilization of flood storage and detention areas, with important (key) flood storage and detention areas as the objects, the flood control model was generalized. The scheduling objects included F1 flood storage and detention area, F2 flood storage and detention area, F3 flood storage and detention area, F4 flood storage and detention area, F5 flood storage and detention area, F6 flood storage and detention area, F7 flood storage and detention area, F8 flood storage and detention area, and F9 flood storage and detention area. Among them, F1 flood storage and detention area, F2 flood storage and detention area, F3 flood storage and detention area, and F4 flood storage and detention area are responsible for ensuring the safety of key areas of the main stream, while F5 flood storage and detention area, F6 flood storage and detention area, F7 flood storage and detention area, F8 flood storage and detention area, and F9 flood storage and detention area need to ensure the safety of key areas of the main stream and flood control safety of the tail area.

[0071] A real-time optimization model for joint flood control of flood storage and detention area engineering groups was constructed. The 1998 type 100-year flood and the discharge of R1 reservoir under the current engineering construction conditions were used as the boundary to simulate the water level change process of each control station in the middle and lower reaches of a river as the input of the scheduling model. Considering the effective forecast period of the current hydrological forecast in the middle and lower reaches of a river, a joint flood control optimization model for the important flood storage and detention area engineering group in the R1 river section with a scheduling period of 7 days and a scheduling period of 6 hours was constructed. Under this water condition, the S4 station exceeded the guaranteed water level on July 26, 1998. July 24, 1998 to July 31, 1998 was selected as the scheduling period. The water level of S4 station did not exceed the guaranteed water level of 34.4 meters ( Freeze Wusong elevation, the same below) as the flood control target, optimize the use strategy of flood storage and detention areas, coordinate flood control safety of main and tributary rivers according to the flood defense plan, if the water level of the tailwater control station exceeds its control water level, use the corresponding tailwater flood storage and detention area in advance, otherwise, use the corresponding main stream flood storage and detention area in advance, when the water level of S5 station is expected to exceed 39 meters, activate F7 flood storage and detention area to store flood water, when the water level of S6 station is expected to exceed 39 meters, activate F6 flood storage and detention area to store flood water, when the water level of S7 station is expected to exceed 41.5 meters, activate F5 flood storage and detention area to store flood water, when the water level of S8 station is expected to exceed 44 meters, activate F8 flood storage and detention area and F9 flood storage and detention area to store flood water.

[0072] Next, we extract the knowledge of the impact of flood storage and detention area application and predict the impact of combined flood storage and detention area application. The application of important flood storage and detention areas in the R1 river section is not sensitive to water regime changes at the S4 station. A knowledge base of the impact of flood storage and detention area application is constructed. The knowledge base of the impact of the F3 flood storage and detention area application is used as an example to illustrate. Using a calibrated and verified one-dimensional hydrodynamic model, within the maximum diversion flow of 3630m³ / s, with a step size of 1000m³ / s, and within the forecast period of 7 days, with a step size of 1 day, the F3 flood storage and detention area is simulated with different diversion flows and different diversion days. The water level change process of the S4 station in the next 14 days is as follows: Figure 3 As shown in the figure, according to the given flood diversion flow and flood diversion days strategy, the corresponding knowledge base of each important flood storage area in the R1 river section is retrieved to obtain the corresponding water level change process of the control station.

[0073] Finally, the combined activation plan of the flood storage and detention area and the optimization of the gate control and flood diversion strategy based on the multi-objective coordination mechanism are carried out. Figure 4As shown in the figure, the water level at station S4 began to exceed the guaranteed water level at 18:00 on July 26, 1998 before flood diversion, and continued to exceed the guaranteed water level for 4 days during the forecast period, with the highest exceeding the guaranteed water level by 0.31 meters, endangering the safe and stable operation of the levee. The conditions for activating flood diversion in the flood storage and detention areas were met. According to the order of population from small to large, the activation order of the main stream flood storage and detention areas was determined to be: F4 flood storage and detention area, F3 flood storage and detention area, F2 flood storage and detention area, and F1 flood storage and detention area. According to the order of population from small to large, the activation order of the tail bank flood storage and detention areas was determined to be: F5 flood storage and detention area, F9 flood storage and detention area, F8 flood storage and detention area, F7 flood storage and detention area, and F6 flood storage and detention area. The flood storage and detention areas were activated one by one until the solution of the control station water level being lower than the guaranteed water level was found and the hard constraints of the basin flood control were met. In this case, the flood storage and detention areas that needed to be activated were determined to be F4 flood storage and detention area and F3 flood storage and detention area.

[0074] If only opening the F4 flood storage and detention area to divert floods cannot ensure that the water level at the control station does not exceed the protection level, then maximize its use to ensure the flood control safety of the basin. With the goal of minimizing the sum of the squares of the water levels at the control station in each period, optimize the activation strategy of the F4 flood storage and detention area, such as Figure 5 As shown in the figure, from 0:00 on July 24, 1998, the F4 flood storage and detention area was activated for 6.75 days with a flow of 2000 m³ / s to divert flood water. The F3 flood storage and detention area was opened for diverting flood water. There was a strategy to ensure that the water level at the control station did not exceed the guaranteed level. With the goal of minimizing the economic loss of flood diversion, the activation strategy was optimized and determined as follows: Figure 5 As shown, from 18:00 on July 28, 1998, the F3 flood storage area was activated with a flow rate of 3000m³ / s to divert flood water for two days.

[0075] Using the knowledge of flood storage and detention area application impact, the water level changes at S4 station during the forecast period are simulated after the flood storage and detention area is activated according to the above activation combination scheme and gate control and flood diversion strategy. Figure 4 As shown in the figure, after the flood storage area is activated according to the model proposed in the present invention, the maximum water level of the S4 station is controlled below the guaranteed water level, effectively avoiding the risk of excessive operation of the levee. At the same time, the maximum water level of the S4 station is close to its guaranteed water level, indicating that the optimization scheme can avoid excessive flood diversion and retain flood storage resources as much as possible.

[0076] The knowledge-driven flood storage and detention area combination activation scheme and gate control flood diversion strategy optimization method of the present invention integrates the knowledge base and data-driven model, mines the knowledge of flood storage and detention area scheduling impact, replaces the hydrodynamic mechanism model, and can quickly and quantitatively determine the impact of flood diversion on the flood discharge level of the river section; by analyzing and formulating the overall flood control safety and flood diversion disaster coordination mechanism of the basin, designing the flood storage and detention area combination activation scheme and gate control flood diversion strategy optimization strategy, automatically comparing different flood storage and detention area combination activation schemes and different flood diversion strategies for the same flood storage and detention area, the calculation is fast and convenient; it fills the gap in the real-time scheduling and application technology of flood storage and detention area engineering groups in my country, and is a short-board work for the real-time flood control scheduling of flood storage and detention areas in the post-engineering period, and has engineering application and promotion value.

[0077] At the same time, it should be noted that the description of the above technical solutions is exemplary, and this specification can be embodied in different forms and should not be construed as being limited to the technical solutions set forth herein. On the contrary, providing these descriptions will make the disclosure of the present invention thorough and complete, and will fully convey the scope disclosed in this specification to those skilled in the art. In addition, the technical solutions of the present invention are limited only by the scope of the claims. The features of the various embodiments of the present invention may be combined or spliced ​​with each other in part or in whole, and may be performed in various different configurations as will be fully understood by those skilled in the art. The embodiments of the present invention may be performed independently of each other, or may be performed together in a mutually dependent relationship.

[0078] For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or replacements can be made without departing from the concept of the present invention. The above structures should be regarded as belonging to the protection scope of the present invention.

Claims

1. A knowledge-driven flood storage and detention area combined activation scheme and gate control flood diversion strategy optimization method, characterized by: The steps include: 1) Collect basic data and generalize the joint flood control model of flood storage and detention area project groups. The collected basic data includes basic data of flood storage and detention areas and flood control stations, as well as basic data of flood control and scheduling in the study area. When generalizing the joint flood control model of flood storage and detention area project groups, determine the flood control objectives and scheduling objects. Starting from the flood control needs of the river basin, with important flood control sections as nodes, each river section in the region is divided into several sections, and flood control and regulation are carried out separately for each river section. On this basis, using the existing results of the classification and batch application of flood storage and detention areas, the flood storage and detention areas are divided into different project groups, including important flood storage and detention area project groups, general flood storage and detention area project groups, and reserved flood storage and detention area project groups. According to the flood control control objectives of each river section, with each flood storage and detention area project group as the scheduling object, generalize the joint flood control model of the flood storage and detention area project groups; 2) Construct a real-time optimization model for joint flood control of flood storage and detention area engineering groups. In the dispatch period t, the water level forecast process Z of each flood control station is used to calculate the water level forecast process Z of each flood control station. τ ,τ=t,t+1,…,t+h-1 are used as model inputs, h is the forecast period, and the model boundary conditions are set according to the approved flood control scheduling rules. A joint flood control optimization scheduling model for a flood storage and detention area project group with a scheduling period of [t,t+h] is constructed. The operation of flood storage and detention area projects should coordinate the two competing control objectives of overall flood control safety of the basin and flood diversion and inundation losses. The joint flood control optimization scheduling model for a flood storage and detention area project group includes an objective function and constraint conditions. 3) Extracting knowledge about the impact of flood storage and detention area operation and predicting the impact of combined flood storage and detention area operation. When extracting knowledge about the impact of flood storage and detention area operation, the impact of flood storage and detention area operation is represented by the amplitude of changes in water level and flow at flood control stations before and after the operation of the flood storage and detention area. 4) Optimization of combined activation scheme of flood storage and detention areas and gate control and flood diversion strategy based on multi-objective coordination mechanism.

2. The knowledge-driven flood storage and detention area combined activation plan and gate control and flood diversion strategy optimization method according to claim 1 is characterized by: In the step 1), the basic data collected include basic data of flood storage and detention areas and flood control stations, and also basic data of flood control scheduling in the study area. The basic data of flood storage and detention areas include the conditions for activation of the flood storage and detention areas, effective flood storage capacity, a curve of the relationship between flood storage capacity and water storage level in the area, a curve of the relationship between flood storage capacity and flooding loss, and socioeconomic information. The basic data of flood control stations include the flood process predicted by the flood control stations, the relationship between water level and flow, the characteristic water level for flood control, and the safe discharge volume. The basic data of flood control scheduling in the study area include the flood scheduling plan, flood defense plan, flood control manual, and flood risk map of the study area.

3. The knowledge-driven flood storage and detention area combined activation plan and gate control flood diversion strategy optimization method according to claim 1 is characterized by: In order to ensure the overall flood control safety of the basin, the water level at the flood control station is set to be lower than the guaranteed water level as a hard constraint. On this basis, the maximum water level of the flood control station is controlled as much as possible to ensure the safe operation of the embankment, and is represented by the minimum sum of the squares of the water levels at the flood control station in each period; the flood diversion and inundation losses cover both economic and social losses. The social losses are represented by the number of people transferred, and the economic losses are represented by the reduction in regional GDP.

4. The knowledge-driven flood storage and detention area combined activation plan and gate control and flood diversion strategy optimization method according to claim 3 is characterized by: The objective function of the joint flood control optimization scheduling model of the flood storage and detention area project group includes: the minimum sum of the squares of the water levels at the flood control station in each period Minimum number of transferred population Where n is the number of activated flood storage areas, P j The number of people involved in the displacement of flood storage and detention area j when it is activated, and the economic loss of flood storage and detention area flooding is minimized Where, GDP j is the reduction in regional GDP after flood storage area j is flooded.

5. The knowledge-driven flood storage and detention area combined activation plan and gate control and flood diversion strategy optimization method according to claim 3 is characterized by: The constraints of the joint flood control optimization scheduling model for flood storage and detention area project groups include: Water balance constraint: S j,τ+1 =S j,τ +W j,τ , where S j,τ 、S j,τ+1 are the flood storage capacity of flood storage area j at the beginning and end of time period τ, W j,τ is the flood diversion volume of flood storage area j during period τ; Flood flow constraint for flood storage and detention areas: 0≤Q j,τ ≤Q j,max , where Q j,τ is the diversion flow of the flood storage area in period j, Q j,max is the maximum allowable diversion flow of flood storage area j; Flood storage capacity constraint of flood storage area: 0≤S j,τ ≤S j,max , where S j,max is the effective flood storage capacity of flood storage area j; Conditions for activation of flood storage and detention areas: Z c,max,before ≥Z c,safe , where Z c,max,before is the highest water level of the associated flood control station in the forecast period when the flood storage area of ​​this group does not divert floodwaters, Z c,safe Maintaining water levels for associated flood control stations; The number of flood storage and detention areas in this group is constrained as follows: n≤b, where n is the number of activated flood storage and detention areas, and b is the total number of flood storage and detention areas in this group; The activation order constraints of various flood storage and detention area groups: The activation order of various flood storage and detention areas is to activate important flood storage and detention areas first, then general flood storage and detention areas, and finally reserved flood storage and detention areas; Coordination constraints for flood control safety of main and tributary rivers: If the water level of the tailwater control station exceeds its control level, the corresponding flood storage area of ​​the tailwater will be used first; otherwise, the corresponding flood storage area of ​​the main stream will be used first; Basin flood control constraints: Z c,max,after ≥Z c,safe , where Z c,max,after This is the highest water level during the foreseeable period at the flood control station after the flood storage and detention area is activated to divert flood water.

6. The knowledge-driven flood storage and detention area combined activation plan and gate control and flood diversion strategy optimization method according to claim 3 is characterized by: In step 3), when extracting the knowledge of the impact of the use of the flood storage and detention area, the impact of the use of the flood storage and detention area is represented by the amplitude of the change in the water level and flow of the flood control station before and after the use of the flood storage and detention area; When the impact of a single flood storage and detention area on the hydrological conditions of a flood control station is insensitive to water regime changes, a knowledge base on the impact of flood storage and detention area application is constructed. Using a calibrated and verified one-dimensional hydrodynamic model, the changing process of the hydrological conditions of the flood control station corresponding to a typical engineering application strategy is simulated and generated. In real-time applications, the corresponding knowledge base can be retrieved based on a given flood storage and detention area application strategy to obtain the corresponding changing process of the hydrological conditions of the flood control station. When the impact of the use of a single flood storage and detention area on the hydrological conditions of the flood control station is sensitive to changes in water conditions, a hydrodynamic proxy model is constructed. Based on the one-dimensional hydrodynamic model, the impact samples of the use of a single flood storage and detention area under different hydrological boundary conditions and engineering application strategies are simulated and generated. Based on the samples, the artificial neural network data-driven model is used to construct the data relationship between the hydrological boundary conditions, the use strategy of the flood storage and detention area and the application impact, forming the knowledge of the impact of the use of a single flood storage and detention area. In actual applications, according to the real-time hydrological boundary conditions and the use strategy of the flood storage and detention area, the corresponding hydrological condition change process of the flood control station is obtained from this data relationship. When there is sufficient historical flood data, different historical flood processes can be directly used as hydrological boundary conditions. If the historical flood data is insufficient, the probability distribution characteristics of historical floods are analyzed, and the multi-station runoff random simulation method is used to generate artificial flood processes. When predicting the impact of the combined use of flood storage and detention areas, the knowledge of the impact of the use of a single flood storage and detention area is iteratively applied to obtain the impact of the combined use of flood storage and detention areas. The knowledge of the impact of the use of a single flood storage and detention area is applied one by one to predict the changes in the hydrological conditions of the flood control stations during the forecast period after each flood storage and detention area is used according to the set activation plan and flood diversion strategy. The hydrological condition process of the flood control station is updated, and the initial boundary of the next flood storage and detention area use prediction is set accordingly until the impact of the use of all flood storage and detention areas has been calculated.

7. The knowledge-driven flood storage and detention area combined activation plan and gate control and flood diversion strategy optimization method according to claim 6, characterized in that: In step 4), when optimizing the combined activation plan and gate control strategy of the flood storage and detention areas based on the multi-objective coordination mechanism, the water level at the flood control station is set to be lower than the guaranteed water level as a hard constraint to ensure the overall flood control safety of the basin, and the primary goal of disaster loss control is to minimize the number of people to be relocated. On this basis, the economic losses caused by the diversion of flood storage projects are further controlled.

8. The knowledge-driven flood storage and detention area combined activation plan and gate control and flood diversion strategy optimization method according to claim 7, characterized in that: When optimizing the combined activation plan for flood storage and detention areas, the combined activation plan for each type of flood storage and detention area project group is determined by the activation objects and activation order. The population of the transfer area is sorted from small to large. Based on the incoming water situation, the flood storage and detention areas are activated one by one in sequence until the water level at the flood control station is lower than the guaranteed water level. At this point, the flood storage projects that need to be activated and the activation order can be determined. When optimizing the flood control and diversion strategy of the flood storage and detention area, the flood control and diversion strategy of the flood storage and detention area is determined by three decision variables: the activation time, the diversion flow and the number of diversion days. The knowledge of the impact of the use of the flood storage and detention area is used to simulate the water level of the flood control point after the flood storage and detention area is activated according to a certain flood diversion strategy. If the flood control constraints of the flood storage and detention area are satisfied, that is, the water level of the flood control station does not exceed the protection level and the levee is within the safe operation range, which can ensure the flood control safety of the basin, the flood diversion strategy is optimized with the goal of minimizing the economic loss of the flood diversion operation of the flood storage and detention area. Otherwise, the flood diversion strategy is optimized with the goal of minimizing the sum of the squares of the water levels of the flood control stations in each period to maximize the protection of the flood control safety of the basin.

Images