Integrated method of reservoir forecasting and operation

Abstract

The present invention provides an integrated method for reservoir forecasting and scheduling, comprising dividing the reservoir catchment area into catchment units based on DEM terrain data; using a hydrological model to simulate the hydrological processes of each catchment unit to obtain the flow output of each catchment unit; constructing a topological relationship between the upstream river channel of the reservoir and the reservoir area, using a hydrodynamic model to simulate based on flow output to obtain flow inflow data; constructing a flood discharge model for flood gates and weirs, setting a reservoir scheduling plan, and calculating the reservoir discharge; constructing a reservoir storage model to obtain the reservoir water level during the forecast period; and promoting the reservoir scheduling plan based on the reservoir water level during the forecast period. The method of the present invention achieves accurate simulation of reservoir forecasting and scheduling, provides an effective tool for flood prevention and disaster reduction and water resources optimization and scheduling, and has important practical value.

Description

Technical Field

[0001] The present invention relates to the field of information technology, and in particular to a reservoir forecasting and dispatching integration method. Background Art

[0002] Reservoirs play an important role in flood control, water supply security, irrigation, power generation, ecological protection and landscape value.

[0003] However, traditional reservoir water space management methods have problems such as incomplete coverage, insufficient scheduling and operation decision-making capabilities, and waste of human resources.

[0004] In order to ensure the safe and stable operation of the reservoir and improve the intelligence level and operation and maintenance efficiency of the reservoir management, the present invention provides an integrated reservoir forecasting and scheduling model to form an overall solution for reservoir space management and operation guarantee. Summary of the Invention

[0005] The purpose of the present invention is to propose an integrated reservoir forecasting and dispatching method for the problem of reservoir space management and operation guarantee.

[0006] The technical solution of the present invention is:

[0007] The present invention provides a reservoir forecasting and dispatching integrated method, comprising:

[0008] S1. Divide the water catchment area of ​​the reservoir into water catchment units based on DEM terrain data;

[0009] S2. Using a hydrological model to simulate the hydrological process of each catchment unit to obtain the runoff of each catchment unit;

[0010] S3. Construct the topological relationship between the upstream river channel and the reservoir area, use the hydrodynamic model to simulate the flow production, and obtain the flow inflow data;

[0011] S4, construct the flood discharge model of the floodgate and weir, set the reservoir operation plan, and calculate the reservoir discharge flow;

[0012] S5. Construct a reservoir storage model to obtain the reservoir water level during the forecast period;

[0013] S6. Based on the reservoir water level during the forecast period, the reservoir scheduling plan is pushed.

[0014] Furthermore, the division of the reservoir catchment area into catchment units according to the DEM terrain data in S1 includes:

[0015] The DEM data were preprocessed, and the flow direction was analyzed using the D8 algorithm. The catchment area was calculated through cumulative flow analysis, and the catchment units were divided according to the set thresholds. The topographic and geometric feature data of each unit were extracted.

[0016] Furthermore, in S2, a hydrological model is used to simulate the runoff generation and confluence process of each catchment unit to obtain the flow inflow data of the upstream river channel and the reservoir area of ​​the reservoir, including:

[0017] S21. Based on land use types, the soil underlying surface is divided into water surface, building land and dry land;

[0018] S22. The underlying surface type is dry land (permeable, including woodland, grassland or other dry land);

[0019] If the soil is in an unsaturated seepage state, the following formula is used to calculate the soil infiltration rate:

[0020] ; middle I is the rainfall intensity (mm / hr);

[0021] If the soil is saturated and stable seepage is formed, the soil infiltration rate is calculated and the soil infiltration amount at each time step is obtained using the following formula:

[0022]

[0023] in: f p is the infiltration capacity, unit is mm / hr, K s is the infiltration conductivity, in mm / hr, M d is the initial saturation difference (volume / volume), S av is the average soil matrix suction (mm), F is the initial soil infiltration (mm), t To calculate the duration.

[0024] S23. When the underlying surface type is water surface or building land, the fixed runoff coefficient is used to calculate the runoff:

[0025] ;

[0026] Where: P is a fixed runoff coefficient, S 0 Initial losses;

[0027] When the underlying surface type is dry land, the runoff is calculated using the following formula:

[0028]

[0029] Where: P is the rainfall, E is the evaporation amount, S For loss,q is the flow rate, and the unit is mm.

[0030] Furthermore, the S3 includes:

[0031] S31. Establish a topological connection relationship by superimposing the river network vector data with the watershed unit boundary;

[0032] S32. Taking the flow rate of each water-collecting unit as lateral inflow or upstream inflow, and obtaining flow inflow data through a one-dimensional hydrodynamic model.

[0033] Furthermore, the S32 specifically includes:

[0034] If there is a river network inside the catchment unit or it is located in the reservoir area, the flow produced by the catchment unit will be regarded as the lateral inflow;

[0035] If there is no river network inside the catchment unit or it is outside the reservoir area, then find the downstream unit connected to the catchment unit, and use the flow rate of the catchment unit as the upstream inflow of the downstream catchment unit according to the topological connection relationship;

[0036] The inflow conditions of each catchment unit are used as boundary conditions, and the flow inflow data of the upstream river channel and reservoir area are obtained using a one-dimensional hydrodynamic model.

[0037] Furthermore, the S4 includes:

[0038] Obtain floodgate parameter information and determine whether the flow state is orifice flow or weir flow;

[0039] Calculate the flow rate, perform weighted averaging on the flow rates calculated from the orifice flow and weir flow, and obtain the final discharge flow rate of the flood discharge gate.

[0040] Furthermore, the obtained floodgate and weir parameter information is used to determine whether the flow state is orifice flow or weir flow:

[0041] Get the water depth above the lock H u , total water head on the gate H 0. Water depth under the gate H d , gate opening e Parameter information;

[0042] When the reservoir discharge is not controlled by gates, it is judged as free weir flow;

[0043] when e / H u When it is greater than 0.65, it is judged as weir flow; otherwise, it is hole flow;

[0044] when H d / H When 0>0.8, it is judged as submerged outflow; otherwise it is free outflow;

[0045] when e / H u When it is greater than 0.60, the flow rates calculated from the orifice flow and the weir flow are weighted averaged so that the outflow from the gate orifice reaches the critical point of the weir outflow, which is used as the final discharge flow from the flood discharge gate.

[0046] Through the above judgment results, the specific flow state of the sluice is determined, which is divided into five working conditions: free hole flow, free weir flow, submerged hole flow, submerged weir flow and no flow, and the flow rate is calculated respectively.

[0047] Furthermore, the S5 includes:

[0048] Obtain terrain elevation information of the reservoir area;

[0049] Establish a reservoir storage model, solve the storage differential equation, and obtain the reservoir water level during the forecast period Z :

[0050] Where: Q is the algebraic sum of the inflow and outflow of the storage area, Q o represents the reservoir discharge, Q i Represents the inbound flow, Q e represents the loss flow, including evapotranspiration factors; A ( Z ) is the area of ​​the storage region corresponding to the water level; t For time.

[0051] Furthermore, S6 includes:

[0052] Based on the reservoir water level during the forecast period, determine whether the flood gate discharge needs to be adjusted. If necessary, the optimal scheduling solution is matched based on cluster analysis and Euclidean distance and pushed;

[0053] Specifically: K-means clustering algorithm is used to cluster the historical reservoir operation plans to obtain different types of typical operation plans;

[0054] The reservoir water level in the current forecast period is compared with the water levels of each typical scheduling scheme, and the alternative scheduling scheme closest to the current water level is found by calculating the Euclidean distance.

[0055] Furthermore, the method also includes: selecting observation data of rainfall, water level and flow of typical flood events to calibrate the parameters of the hydrological model, including: using the Nash-Sutcliffe coefficient and the coefficient of determination R2 as evaluation indicators, and adjusting the hydrological model's runoff generation and confluence parameters and river channel roughness parameters to achieve the best degree of agreement between the simulated values ​​and the measured values.

[0056] Beneficial effects of the present invention:

[0057] This invention discloses an integrated method for reservoir forecasting and scheduling. This method divides watershed units based on DEM terrain data; considers the spatial distribution differences of land use types and uses a hydrological model to simulate runoff generation and confluence; constructs a topological relationship between the upstream river channel and the reservoir area, and uses a hydrodynamic model to obtain flow inflow data; and constructs a reservoir regulation model to simulate reservoir water level changes and develop reservoir scheduling plans. This method achieves precise simulation of reservoir forecasting and scheduling, providing an effective tool for flood prevention and disaster reduction and optimized water resource scheduling, and possesses significant practical value.

[0058] The present invention tightly couples the runoff generation and confluence model with the hydrodynamic model to simulate the entire rainfall-runoff generation and confluence-reservoir scheduling process; key parameters are calibrated using data from typical flood events, and indicators such as the Nash coefficient are used to evaluate and optimize model accuracy.

[0059] Other features and advantages of the present invention will be described in detail in the following detailed description. DETAILED DESCRIPTION

[0060] The preferred embodiments of the present invention will be described in more detail below. It should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0061] The present invention provides a reservoir forecasting and dispatching integrated method, comprising:

[0062] S1. Preprocess the DEM data, use the D8 algorithm to perform flow direction analysis, calculate the catchment area through cumulative flow analysis, divide the catchment unit according to the set threshold, and extract the topographic and geometric feature data of each unit.

[0063] S2. Use a hydrological model to simulate the hydrological process of each catchment unit to obtain the runoff of each catchment unit; specifically, the following steps are involved:

[0064] S21. Based on land use types, the soil underlying surface is divided into water surface, building land and dry land;

[0065] S22. When the underlying surface type is dry land, the dry land is permeable and includes woodland, grassland or other dry land;

[0066] If the soil is in an unsaturated seepage state, the following formula is used to calculate the soil infiltration rate:

[0067] ; middle I is the rainfall intensity (mm / hr);

[0068] If the soil is saturated and stable seepage is formed, the soil infiltration rate is calculated and the soil infiltration amount at each time step is obtained using the following formula:

[0069] ;

[0070] in: f p is the infiltration capacity, unit is mm / hr, K s is the infiltration conductivity, in mm / hr, M d is the initial saturation difference (volume / volume), S av is the average soil matrix suction (mm), F is the initial soil infiltration (mm), t To calculate the duration.

[0071] S23. When the underlying surface type is water surface or building land, a fixed runoff coefficient is used to calculate the runoff yield;

[0072]

[0073] Where: P is a fixed runoff coefficient, S 0 Initial losses;

[0074] When the underlying surface type is dry land, the runoff is calculated using the following formula:

[0075]

[0076] Where: P is the rainfall, E is the evaporation amount, S For loss, q is the flow rate, and the unit is mm.

[0077] S3. Construct the topological relationship between the upstream river channel and the reservoir area, use the hydrodynamic model to simulate the flow production, and obtain the flow inflow data, including:

[0078] S31. Establish a topological connection relationship by superimposing the river network vector data with the watershed unit boundary;

[0079] S32, using the flow rate of each water-collecting unit as lateral inflow or upstream inflow, and obtaining flow inflow data through a one-dimensional hydrodynamic model;

[0080] If there is a river network inside the catchment unit or it is located in the reservoir area, the flow produced by the catchment unit will be regarded as the lateral inflow;

[0081] If there is no river network inside the catchment unit or it is outside the reservoir area, then find the downstream unit connected to the catchment unit, and use the flow rate of the catchment unit as the upstream inflow of the downstream catchment unit according to the topological connection relationship;

[0082] The inflow conditions of each catchment unit are used as boundary conditions, and the flow inflow data of the upstream river channel and reservoir area are obtained using a one-dimensional hydrodynamic model.

[0083] S4. Obtain parameter information of the flood discharge gate and determine whether the flow state is orifice flow or weir flow; calculate the flow rate, perform weighted averaging on the flow rates calculated for orifice flow and weir flow, and obtain the final discharge flow rate of the flood discharge gate.

[0084] S5. Construct a reservoir regulation model to obtain the reservoir water level during the forecast period; including:

[0085] Obtain terrain elevation information of the reservoir area;

[0086] Establish a reservoir storage model, solve the storage differential equation, and obtain the reservoir water level during the forecast period Z :

[0087]

[0088] Where: Q is the algebraic sum of the inflow and outflow of the storage area, Q o represents the reservoir discharge, Q i Represents the inbound flow, Q e represents the loss flow, including evapotranspiration factors; A ( Z ) is the area of ​​the storage region corresponding to the water level; t For time.

[0089] S6. Based on the reservoir water level during the forecast period, the reservoir scheduling plan is pushed.

[0090] In one example, the acquired floodgate and weir parameter information is used to determine whether the flow state is orifice flow or weir flow as follows:

[0091] Get the water depth above the lock H u , total water head on the gate H 0. Water depth under the gateH d , gate opening e Parameter information;

[0092] When the reservoir discharge is not controlled by gates, it is judged as free weir flow;

[0093] when e / H u When it is greater than 0.65, it is judged as weir flow; otherwise, it is hole flow;

[0094] when H d / H When 0>0.8, it is judged as submerged outflow; otherwise it is free outflow;

[0095] when e / H u When it is greater than 0.60, the flow rates calculated from the orifice flow and the weir flow are weighted averaged so that the outflow from the gate orifice reaches the critical point of the weir outflow, which is used as the final discharge flow from the flood discharge gate.

[0096] Based on the above judgment results, the specific flow state of the sluice is determined, which is divided into five working conditions: free hole flow, free weir flow, submerged hole flow, submerged weir flow and no flow. The flow rate is calculated using the following formulas respectively.

[0097] (a) No outflow ;

[0098] (b) Free hole flow ;

[0099] (c) Free weir flow ;

[0100] (d) Submerged hole flow ;

[0101] (e) Submerged weir flow ;

[0102] Where: M 1. M 2. M 3. M 4 are the comprehensive discharge coefficients under free hole flow, free weir flow, submerged hole flow and submerged weir flow conditions respectively; B w The total clear width of the gate opening; Q i and Q i+1 They are i and i Flow rate at section +1.

[0103] In one example, the reservoir water level during the forecast period is used to determine whether the discharge of the floodgate needs to be adjusted. If necessary, the optimal scheduling solution is matched based on cluster analysis and Euclidean distance and pushed;

[0104] Specifically: K-means clustering algorithm is used to cluster the historical reservoir operation plans to obtain different types of typical operation plans;

[0105] The reservoir water level in the current forecast period is compared with the water levels of each typical scheduling scheme, and the alternative scheduling scheme closest to the current water level is found by calculating the Euclidean distance.

[0106] In one example, the method further includes: selecting observation data of rainfall, water level, and flow of typical flood events to calibrate the parameters of the hydrological model, including: using the Nash-Sutcliffe coefficient and the coefficient of determination R 2 As evaluation indicators, the runoff parameters and river channel roughness parameters of the hydrological model are adjusted to achieve the best agreement between the simulated values ​​and the measured values.

[0107] According to the relevant specifications for model calibration and verification, the measured rainfall, water level and flow data of more than three typical rainstorms were selected; the Nash-Sutcliffe coefficient NSE and coefficient of determination were used. R 2 As an evaluation index, the effectiveness of the model is assessed; by adjusting the hydrological model parameters such as runoff coefficient, soil infiltration rate and the hydrodynamic model parameters such as river roughness, gate and weir comprehensive flow coefficient in the model, the accuracy of the simulation results is improved; if the simulated value is close to the measured order of magnitude and process trend, the Nash-Sutcliffe coefficient and determination coefficient are determined. R 2 Whether the requirements are met; It is usually stipulated that the maximum water level error between the verification result and the actual flood is no more than 15cm, and the Nash efficiency coefficient of the predicted water level is greater than or equal to 0.7, and the determination coefficient is greater than or equal to 0. R 2 If the value is greater than 0.8 and the relative error is less than or equal to 10%, the model accuracy is high and can be verified through calibration. The accuracy index can be adjusted according to the actual situation of the calculation object and specific business needs. By continuously adjusting and optimizing the model parameters, the degree of agreement between the simulated value and the measured value is optimized, and the model calibration is completed.

[0108] While various embodiments of the present invention have been described above, the above description is intended to be illustrative, not exhaustive, and not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments.

Claims

1. A reservoir forecasting and dispatching integrated method, characterized in that: include: S1. Divide the water catchment area of ​​the reservoir into water catchment units based on DEM terrain data; S2. Using a hydrological model to simulate the hydrological process of each catchment unit to obtain the runoff of each catchment unit; S3. Construct the topological relationship between the upstream river channel and the reservoir area, use the hydrodynamic model to simulate the flow production, and obtain the flow inflow data; S4, construct the flood discharge model of the floodgate and weir, set the reservoir operation plan, and calculate the reservoir discharge flow; S5. Construct a reservoir storage model to obtain the reservoir water level during the forecast period; S6. Based on the reservoir water level during the forecast period, the reservoir operation plan is issued; The S4 includes: Obtain floodgate parameter information and determine whether the flow state is orifice flow or weir flow; Calculate the flow rate, perform weighted averaging on the flow rates calculated from the orifice flow and the weir flow, and obtain the final discharge flow rate from the floodgate; The obtained floodgate and weir parameter information is used to determine whether the flow state is orifice flow or weir flow: Get the water depth above the lock H u , total water head on the gate H 0. Water depth under the gate H d , gate opening e Parameter information; When the reservoir discharge is not controlled by gates, it is judged as free weir flow; when e / H u When it is greater than 0.65, it is judged as weir flow; otherwise, it is hole flow; when H d / H When 0>0.8, it is judged as submerged outflow; otherwise it is free outflow; when e / H u When it is greater than 0.60, the calculated flow rates of the orifice flow and the weir flow are weighted averaged so that the outflow from the gate orifice reaches the critical point of the weir outflow, which is used as the final discharge rate of the floodgate; In S2, a hydrological model is used to simulate the flow generation and confluence process of each catchment unit to obtain the flow inflow data of the upstream river channel and reservoir area of ​​the reservoir, including: S21. Based on land use types, the soil underlying surface is divided into water surface, building land and dry land; S22, when the surface type is dry land; If the soil is in an unsaturated seepage state, the following formula is used to calculate the soil infiltration rate: ； If the soil is saturated and stable seepage is formed, the soil infiltration rate is calculated and the soil infiltration amount at each time step is obtained using the following formula: ； in: f p is the infiltration capacity, unit is mm / hr, K s is the infiltration conductivity, in mm / hr, M d is the initial saturation difference, S av is the average soil matrix suction, unit is mm, F is the initial soil infiltration volume, unit is mm, t To calculate the duration; S23. When the underlying surface type is water surface or building land, the fixed runoff coefficient is used to calculate the runoff: ； Where: Ψ is a fixed runoff coefficient, S 0 Initial losses; When the underlying surface type is dry land, the runoff is calculated using the following formula: ； Where: P is the rainfall, E is the evaporation amount, S For loss, q is the flow rate, the unit is mm; The S5 includes: Obtain terrain elevation information of the reservoir area; Establish a reservoir storage model, solve the storage differential equation, and obtain the reservoir water level during the forecast period Z : ； Where: Q is the algebraic sum of the inflow and outflow of the storage area, Q o represents the reservoir discharge, Q i Represents the inbound flow, Q e represents the loss flow, including evapotranspiration factors; A ( Z ) is the area of ​​the storage region corresponding to the water level; t For time.

2. The reservoir forecast and dispatch integration method according to claim 1, characterized in that: In S1, the water catchment area of ​​the reservoir is divided into water catchment units according to the DEM terrain data, including: The DEM data were preprocessed, and the flow direction was analyzed using the D8 algorithm. The catchment area was calculated through cumulative flow analysis, and the catchment units were divided according to the set thresholds. The topographic and geometric feature data of each unit were extracted.

3. The reservoir forecasting and dispatching integrated method according to claim 1 is characterized in that The S3 includes: S31. Establish a topological connection relationship by superimposing the river network vector data with the watershed unit boundary; S32. Taking the flow rate of each water-collecting unit as lateral inflow or upstream inflow, and obtaining flow inflow data through a one-dimensional hydrodynamic model.

4. The reservoir forecasting and dispatching integrated method according to claim 3 is characterized in that The S32 specifically includes: If there is a river network inside the catchment unit or it is located in the reservoir area, the flow produced by the catchment unit will be regarded as the lateral inflow; If there is no river network inside the catchment unit or it is outside the reservoir area, then find the downstream unit connected to the catchment unit, and use the flow rate of the catchment unit as the upstream inflow of the downstream catchment unit according to the topological connection relationship; The inflow conditions of each catchment unit are used as boundary conditions, and the flow inflow data of the upstream river channel and reservoir area are obtained using a one-dimensional hydrodynamic model.

5. The reservoir forecasting and dispatching integrated method according to claim 1 is characterized in that S6 include: Based on the reservoir water level during the forecast period, determine whether the flood gate discharge needs to be adjusted. If necessary, the optimal scheduling solution is matched based on cluster analysis and Euclidean distance and pushed; Specifically: K-means clustering algorithm is used to cluster the historical reservoir operation plans to obtain different types of typical operation plans; The reservoir water level in the current forecast period is compared with the water levels of each typical scheduling scheme, and the alternative scheduling scheme closest to the current water level is found by calculating the Euclidean distance.

6. The reservoir forecast and dispatch integration method according to claim 1, characterized in that: The method also includes: selecting observation data of rainfall, water level and flow of typical flood events to calibrate the parameters of the hydrological model, including: using the Nash-Sutcliffe coefficient and the coefficient of determination R 2 As evaluation indicators, the runoff parameters and river channel roughness parameters of the hydrological model are adjusted to achieve the best agreement between the simulated values ​​and the measured values.

Images