Reservoir flood control scheduling method based on prediction model

Abstract

The invention discloses a reservoir flood control scheduling method based on a prediction model, and relates to the technical field of reservoir flood control. The reservoir capacity, the power generation water consumption per unit time and the water supply per unit time are obtained from the reservoir operation data, the flood-fighting capacity of the reservoir is calculated in combination with rainfall runoff, the first flood-fighting capacity risk coefficient of the reservoir is further calculated, and a drainage analysis instruction is sent outwards; and after a discharge analysis instruction is received, the real-time water level and flow velocity of a downstream river channel are measured, the maximum discharge flow of the reservoir is calculated, a second flood-fighting capacity risk coefficient of the reservoir is calculated according to the flood-fighting capacity of the reservoir and the maximum discharge flow of the reservoir, and a reservoir flood-fighting early warning is sent outwards. After the reservoir flood-fighting early warning is received, the reservoir flood-fighting capacity and the maximum discharge flow are obtained, and the maximum flood-fighting duration of the reservoir is calculated. The flood conditions of the reservoir and the downstream river channel can be known in real time, and a scientific basis is provided for making effective flood control measures.

Description

Technical Field

[0001] The invention relates to the technical field of reservoir flood control, and in particular to a reservoir flood control dispatching method based on a prediction model. Background Art

[0002] Reservoir flood control dispatching is to regulate floods in a planned manner by controlling the water storage and release process of the reservoir to achieve the purpose of reducing flood disasters. The prediction model is an important part of the decision support system (DSS) and is one of the key technologies in the basic theoretical research and applied basic research of flood control and power production dispatching and decision-making. The accuracy of prediction directly affects the decision analysis of optimal dispatching. Quantitative prediction methods have been widely used in reservoir flood control dispatching. In the late 1930s, the scientific basis of quantitative prediction began to be established. From the 1950s to the 1970s, quantitative prediction methods developed rapidly in the direction of econometrics, filtering methods, autoregressive moving average models, time series analysis methods and exponential smoothing techniques. In order to improve the accuracy and efficiency of reservoir flood control dispatching, modern research continues to introduce new prediction technologies and optimization algorithms. For example, optimization algorithm theories such as artificial neural networks, genetic algorithms, and simulated annealing algorithms have been applied to hydrological forecasting to solve some complex problems that are difficult to solve with existing methods. These new technologies may open up a new research idea and research path.

[0003] In the Chinese invention application with application publication number CN110930016A, a method for random optimization scheduling of cascade reservoirs based on deep Q learning is disclosed, including describing the reservoir inflow process: establishing a Markov decision process MDPS model; establishing a probability transfer matrix: establishing a random optimization scheduling model for cascade reservoirs; determining the constraint function of the model: introducing a deep neural network to extract the runoff state characteristics of the cascade reservoirs, and at the same time achieving approximate representation and optimization of the target value function of the scheduling model; applying reinforcement learning to the random optimization scheduling of reservoirs; establishing a DQN model; and using a deep reinforcement learning algorithm to solve the random optimization scheduling model for cascade reservoirs.

[0004] In the above invention application, the random optimization scheduling method of cascade reservoirs based on deep Q learning realizes the random optimization scheduling of cascade reservoirs, so that the generator sets are fully utilized during the scheduling period, the electricity demand and various constraints are met, and the annual average power generation revenue is maximized. However, it focuses on making full use of the generator sets during the scheduling period and cannot effectively reduce the loss of life and property caused by floods.

[0005] To this end, the present invention provides a reservoir flood control dispatching method based on a prediction model. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In view of the deficiencies in the prior art, the present invention provides a reservoir flood control scheduling method based on a prediction model. The present invention measures the real-time water level and flow velocity of the downstream river channel, calculates the maximum downstream flow of the reservoir, calculates the second flood control capacity risk coefficient Ekf of the reservoir according to the reservoir flood control capacity and the maximum downstream flow of the reservoir, and sends out a reservoir flood control warning, which helps to understand the flood conditions of the reservoir and the downstream river channel in real time, provides a scientific basis for formulating effective flood control measures, further refines the flood control risk assessment, and helps to more accurately grasp the flood control capacity of the reservoir, thereby solving the technical problems recorded in the background technology.

[0008] (II) Technical solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions: A reservoir flood control dispatching method based on a prediction model comprises the following steps:

[0010] Measure the rainfall P in the reservoir area during the detection period t , calculate the watershed soil moisture content P a,t+1 , according to the rainfall P t and the soil moisture content in the basin P a,t+1 , calculate the rainfall runoff Ju t+1 ;

[0011] The reservoir capacity Q, the water consumption per unit time for power generation Fd, and the water supply per unit time Gs are obtained from the reservoir operation data, and the rainfall runoff Ju t+1 , calculate the reservoir flood control capacity Kr t+1 , and further calculate the reservoir's first flood control capacity risk factor Dkf, and send out discharge analysis instructions;

[0012] After receiving the discharge analysis instruction, measure the real-time water level Sw of the downstream river t and flow rate Ls t , calculate the maximum discharge flow of the reservoir Xa t , based on the reservoir flood control capacity Kr t+1 and the maximum discharge flow of the reservoir Xa t , calculate the reservoir's second flood control capacity risk coefficient Ekf, and send out reservoir flood control warnings;

[0013] After receiving the reservoir flood warning, obtain the reservoir flood capacity Kr t+1 and maximum downstream flow rate Xa t , calculate the maximum flood-fighting duration Sc of the reservoir, and record it as the countdown for the flooding of the downstream river channel.

[0014] Furthermore, a rain gauge is used to periodically measure the rainfall P in the reservoir area during the detection period. t , calculate the watershed soil moisture content P a,t+1 :

[0015] P a,t+1 = k(P a,t +P t )

[0016] Among them, we take the rainfall process with large precipitation and runoff after a long drought as an example from the historical basin measured rainfall and flood data, and assume that the soil moisture content in the early stage of this rainfall is close to zero, that is, P a,0 =0, k is the soil moisture decay coefficient, and the calculation formula of k is as follows:

[0017]

[0018] Among them, I M is the maximum soil water shortage measured. Due to different seasons, E P and k values ​​are variables, among which E P is the monthly evaporation capacity, and the k value is a constant that is different every month. P The specific k value is as follows Figure 2 .

[0019] Further, obtain the rainfall P t and the soil moisture content in the basin P a,t+1 , calculate the rainfall runoff Ju t+1 :

[0020]

[0021] Among them, I N Indicates the maximum soil moisture content in the basin.

[0022] Furthermore, the reservoir capacity Q, the power generation water consumption per unit time Fd, and the water supply per unit time Gs are obtained from the reservoir operation data, and the real-time water level of the reservoir is measured using a water level gauge, and the real-time water volume Rs of the reservoir is calculated based on the real-time water level. t , combined with rainfall runoff Ju t+1 , calculate the reservoir flood control capacity Kr t+1 :

[0023] K t+1 =Q-Fd*T-Gs*T-Rs t -(Rk t +Ju t+1 )

[0024] Among them, Rk t is the inflow volume:

[0025] R t =Rs t -Rs t-1

[0026] t is the detection time sequence number of the data.

[0027] Furthermore, the flood control capacity of the reservoir Kr is obtained. t+1 , calculate the reservoir's first flood control capacity risk coefficient Dkf:

[0028]

[0029] Among them, Kr b It is the flood control capacity warning value of the reservoir, which is 0.2 times of the reservoir capacity Q.

[0030] Furthermore, when the reservoir's first flood control capacity risk factor Dkf exceeds When the leakage analysis instruction is sent out, It is the average of the first flood control capacity risk coefficient Dkf of all reservoirs in history.

[0031] Furthermore, after receiving the discharge analysis instruction, the real-time water level Sw of the downstream river is measured. t and flow rate Ls t , according to the historical hydrological data of the downstream river, obtain the cross-sectional data and warning water level Jw corresponding to each water level in the downstream river, and calculate the maximum discharge flow Xa of the reservoir t :

[0032]

[0033] Among them, Hm Jw Indicates the cross-sectional area of ​​the downstream river corresponding to the warning water level, Indicates the cross-sectional area of ​​the downstream river corresponding to the real-time water level of the downstream river.

[0034] Furthermore, the flood control capacity of the reservoir Kr is obtained. t+1 and the maximum discharge flow of the reservoir Xa t , calculate the reservoir's second flood control capacity risk coefficient Ekf:

[0035]

[0036] When the second flood resistance capacity risk factor Ekf exceeds When a reservoir flood warning is issued, it will be issued.

[0037] Furthermore, after receiving the reservoir flood warning, the reservoir flood control capacity Kr is obtained. t+1 and maximum downstream flow rate Xa t , calculate the maximum flood-fighting duration Sc of the reservoir:

[0038]

[0039] Among them, [] is the floor function.

[0040] Furthermore, the maximum flood fighting time Sc of the output reservoir is recorded as the countdown for the flooding of the downstream river channel, and evacuation is carried out according to the countdown for the flooding of the downstream river channel.

[0041] (III) Beneficial effects

[0042] The present invention provides a reservoir flood control dispatching method based on a prediction model, which has the following beneficial effects:

[0043] 1. Measure the rainfall P in the reservoir area during the detection period t , calculate the watershed soil moisture content P a,t+1 , according to the rainfall P t and the soil moisture content in the basin P a,t+1 , calculate the rainfall runoff Ju t+1 , which helps to predict the possibility and scale of floods, ensure the accuracy of data, and provide a reliable basis for subsequent analysis and calculations, so as to take flood prevention measures in time and reduce the losses caused by flood disasters.

[0044] 2. Obtain the reservoir capacity Q, power generation water consumption per unit time Fd, and water supply per unit time Gs from the reservoir operation data, and combine the rainfall runoff Ju t+1 , calculate the reservoir flood control capacity Kr t+1 , and further calculate the reservoir flood control capacity risk coefficient Kf, and send out discharge analysis instructions, which can intuitively reflect the risk level of the reservoir when facing floods, more accurately evaluate the reliability of the reservoir's flood control capacity, and provide a scientific basis for decision makers.

[0045] 3. After receiving the discharge analysis instruction, measure the real-time water level Sw of the downstream river t and flow rate Ls t , calculate the maximum discharge flow of the reservoir Xa t , based on the reservoir flood control capacity Kr t+1 and the maximum discharge flow of the reservoir Xa t , calculate the second flood control capacity risk coefficient Ekf of the reservoir, and send out reservoir flood control warnings, which helps to understand the flood conditions of the reservoir and downstream rivers in real time, provide a scientific basis for formulating effective flood control measures, further refine the flood control risk assessment, and help to grasp the flood control capacity of the reservoir more accurately.

[0046] 4. After receiving the reservoir flood warning, obtain the reservoir flood capacity Kr t+1 and maximum downstream flow rate Xa t , calculate the maximum flood resistance time Sc of the reservoir, and record it as the countdown for the flooding of the downstream river. This can help us intuitively understand the urgency of the impending flood, thereby increasing our vigilance. We can make time arrangements based on the countdown for the flooding of the downstream river, make preparations for evacuation and disaster prevention in advance, and reduce the losses caused by disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 A schematic diagram of a flow chart of a reservoir flood control dispatching method based on a prediction model according to the present invention;

[0048] Figure 2 The invention discloses a method for flood control of reservoirs based on a prediction model for the Shenzhen Basin. P And k value. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] See also Figure 1-Figure 2 The present invention provides a reservoir flood control dispatching method based on a prediction model, comprising the following steps:

[0051] Step 1: Measure the rainfall P in the reservoir area during the detection period t , calculate the watershed soil moisture content P a,t+1 , according to the rainfall P t and the soil moisture content in the basin P a,t+1 , calculate the rainfall runoff Ju t+1 .

[0052] The step 1 comprises the following steps:

[0053] Step 101: Use a rain gauge to periodically measure the rainfall P in the reservoir area during the detection period. t , calculate the watershed soil moisture content P a,t+1 :

[0054] P a,t+1 = k(P a,t +P t )

[0055] Among them, we take the rainfall process with large precipitation and runoff after a long drought as an example from the historical basin measured rainfall and flood data, and assume that the soil moisture content in the early stage of this rainfall is close to zero, that is, P a,0 =0, k is the soil moisture decay coefficient, and the calculation formula of k is as follows:

[0056]

[0057] Among them, I M is the maximum soil water shortage measured. Due to different seasons, E Pand k values ​​are variables, among which E P is the monthly evaporation capacity, and the k value is a constant that is different every month. P The specific k value is as follows Figure 2 .

[0058] Step 102: Obtain rainfall P t and the soil moisture content in the basin P a,t+1 , calculate the rainfall runoff Ju t+1 :

[0059]

[0060] Among them, I N Indicates the maximum soil moisture content in the basin.

[0061] When using, combine the contents in steps 101 and 102:

[0062] Measure the rainfall P in the reservoir area during the detection period t , calculate the watershed soil moisture content P a,t+1 , according to the rainfall P t and the soil moisture content in the basin P a,t+1 , calculate the rainfall runoff Ju t+1 , which helps to predict the possibility and scale of floods, ensure the accuracy of data, and provide a reliable basis for subsequent analysis and calculations, so as to take flood prevention measures in time and reduce the losses caused by flood disasters.

[0063] Step 2: Obtain the reservoir capacity Q, power generation water consumption per unit time Fd, and water supply per unit time Gs from the reservoir operation data, and combine the rainfall runoff Ju t+1 , calculate the reservoir flood control capacity Kr t+1 , and further calculate the reservoir's first flood control capacity risk coefficient Dkf, and send out discharge analysis instructions.

[0064] The step 2 includes the following contents:

[0065] Step 201: Obtain the reservoir capacity Q, the power generation water consumption per unit time Fd, and the water supply per unit time Gs from the reservoir operation data, and use a water level meter to measure the real-time water level of the reservoir, and calculate the real-time water volume Rs of the reservoir based on the real-time water level. t , combined with rainfall runoff Ju t+1 , calculate the reservoir flood control capacity Kr t+1 :

[0066] K t+1 =Q-Fd*T-Gs*T-Rs t -(Rk t +Ju t+1 )

[0067] Among them, Rk t is the inflow volume:

[0068] R t =Rs t -Rs t-1

[0069] t is the detection time sequence number of the data.

[0070] Step 202: Obtain the flood control capacity Kr of the reservoir t+1 , calculate the reservoir's first flood control capacity risk coefficient Dkf:

[0071]

[0072] Among them, Kr b It is the flood control capacity warning value of the reservoir, which is 0.2 times of the reservoir capacity Q.

[0073] When the reservoir's first flood control capacity risk factor Dkf exceeds When the leakage analysis instruction is sent out, It is the average of the first flood control capacity risk coefficient Dkf of all reservoirs in history.

[0074] When using, combine the contents in steps 201 and 202:

[0075] The reservoir capacity Q, the water consumption per unit time for power generation Fd, and the water supply per unit time Gs are obtained from the reservoir operation data, and the rainfall runoff Ju t+1 , calculate the reservoir flood control capacity Kr t+1 , and further calculate the reservoir's first flood control capacity risk coefficient Dkf, and send out discharge analysis instructions, which can intuitively reflect the risk level of the reservoir when facing a flood, more accurately evaluate the reliability of the reservoir's flood control capacity, and provide a scientific basis for decision makers.

[0076] Step 3: After receiving the discharge analysis instruction, measure the real-time water level Sw of the downstream river t and flow rate Ls t , calculate the maximum discharge flow of the reservoir Xa t , based on the reservoir flood control capacity Kr t+1 and the maximum discharge flow of the reservoir Xa t , calculate the reservoir’s second flood control capacity risk coefficient Ekf, and send out reservoir flood control warnings.

[0077] The step three includes the following contents:

[0078] Step 301: After receiving the downstream analysis instruction, measure the real-time water level Sw of the downstream river. t and flow rate Ls t, according to the historical hydrological data of the downstream river, obtain the cross-sectional data and warning water level Jw corresponding to each water level in the downstream river, and calculate the maximum discharge flow Xa of the reservoir t :

[0079]

[0080] Among them, Hm Jw Indicates the cross-sectional area of ​​the downstream river corresponding to the warning water level, Indicates the cross-sectional area of ​​the downstream river corresponding to the real-time water level of the downstream river.

[0081] Step 302: Obtain the flood control capacity Kr of the reservoir t+1 and the maximum discharge flow of the reservoir Xa t , calculate the reservoir's second flood control capacity risk coefficient Ekf:

[0082]

[0083] When the second flood resistance capacity risk factor Ekf exceeds When a reservoir flood warning is issued, it will be issued.

[0084] When using, combine the contents in steps 301 and 302:

[0085] After receiving the discharge analysis instruction, measure the real-time water level Sw of the downstream river t and flow rate Ls t , calculate the maximum discharge flow of the reservoir Xa t , based on the reservoir flood control capacity Kr t+1 and the maximum discharge flow of the reservoir Xa t , calculate the second flood control capacity risk coefficient Ekf of the reservoir, and send out reservoir flood control warnings, which helps to understand the flood conditions of the reservoir and downstream rivers in real time, provide a scientific basis for formulating effective flood control measures, further refine the flood control risk assessment, and help to grasp the flood control capacity of the reservoir more accurately.

[0086] Step 4: After receiving the reservoir flood warning, obtain the reservoir flood capacity Kr t+1 and maximum downstream flow rate Xa t , calculate the maximum flood-fighting duration Sc of the reservoir, and record it as the countdown for the flooding of the downstream river channel.

[0087] The step 4 includes the following contents:

[0088] Step 401: After receiving the reservoir flood warning, obtain the reservoir flood capacity Kr t+1 and maximum downstream flow rate Xa t , calculate the maximum flood-fighting duration Sc of the reservoir:

[0089]

[0090] Among them, [] is the floor function.

[0091] Step 402: Output the maximum flood fighting time Sc of the reservoir, record it as the countdown of downstream river flooding, and carry out evacuation according to the countdown of downstream river flooding.

[0092] When using, combine the contents in steps 401 and 402:

[0093] After receiving the reservoir flood warning, obtain the reservoir flood capacity Kr t+1 and maximum downstream flow rate Xa t , calculate the maximum flood resistance time Sc of the reservoir, and record it as the countdown for the flooding of the downstream river. This can help us intuitively understand the urgency of the impending flood, thereby increasing our vigilance. We can make time arrangements based on the countdown for the flooding of the downstream river, make preparations for evacuation and disaster prevention in advance, and reduce the losses caused by disasters.

[0094] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0095] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0096] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A reservoir flood control dispatching method based on a prediction model, characterized in that: The steps include: Measure the rainfall P in the reservoir area during the detection period t , calculate the watershed soil moisture content P a，t+1 , according to the rainfall P t and the soil moisture content in the basin P a，t+1 , calculate the rainfall runoff Ju t+1 ; The reservoir capacity Q, the water consumption per unit time for power generation Fd, and the water supply per unit time Gs are obtained from the reservoir operation data, and the rainfall runoff Ju t+1 , calculate the reservoir flood control capacity Kr t+1 , and further calculate the reservoir's first flood control capacity risk factor Dkf, and send out discharge analysis instructions; After receiving the discharge analysis instruction, measure the real-time water level Sw of the downstream river t and flow rate Ls t , calculate the maximum discharge flow of the reservoir Xa t , based on the reservoir flood control capacity Kr t+1 and the maximum discharge flow of the reservoir Xa t , calculate the reservoir's second flood control capacity risk coefficient Ekf, and send out reservoir flood control warnings; After receiving the reservoir flood warning, obtain the reservoir flood capacity Kr t+1 and maximum downstream flow rate Xa t , calculate the maximum flood-fighting duration Sc of the reservoir, and record it as the countdown for the flooding of the downstream river channel.

2. A reservoir flood control dispatching method based on a prediction model according to claim 1, characterized in that: Use a rain gauge to periodically measure the rainfall P in the reservoir area during the detection period t , calculate the watershed soil moisture content P a，t+1 : P a，t+1 =k(P a，t +P t ) Among them, we take the rainfall process with large precipitation and runoff after a long drought as an example from the historical basin measured rainfall and flood data, and assume that the soil moisture content in the early stage of this rainfall is close to zero, that is, P a，0 =0, k is the soil moisture decay coefficient, and the calculation formula of k is as follows: Among them, I M is the maximum soil water shortage measured. Due to different seasons, E P and k values ​​are variables, among which E P is the monthly evaporation capacity, and the k value is a constant.

3. A reservoir flood control dispatching method based on a prediction model according to claim 2, characterized in that: Get the rainfall P t and the soil moisture content in the basin P a，t+1 , calculate the rainfall runoff Ju t+1 : Among them, I N Indicates the maximum soil moisture content in the basin.

4. The reservoir flood control dispatching method based on the prediction model according to claim 3 is characterized by: The reservoir capacity Q, the power generation water consumption per unit time Fd, and the water supply per unit time Gs are obtained from the reservoir operation data, and the real-time water level of the reservoir is measured using a water level gauge. The real-time water volume Rs of the reservoir is calculated based on the real-time water level. t , combined with rainfall runoff Ju t+1 , calculate the reservoir flood control capacity Kr t+1 : Kr t+1 =Q-Fd*T-Gs*T-Rs t -(Rk t +Ju t+1 ) Among them, Rk t is the inflow volume: Rk t =Rs t -Rs t-1 t is the detection time sequence number of the data.

5. The reservoir flood control dispatching method based on the prediction model according to claim 4 is characterized in that: Get the reservoir flood control capacity Kr t+1 , calculate the reservoir's first flood control capacity risk coefficient Dkf: Among them, Krb is the flood control capacity warning value of the reservoir, which is 0.2 times the reservoir capacity Q.

6. A reservoir flood control dispatching method based on a prediction model according to claim 5, characterized in that: When the reservoir's first flood control capacity risk factor Dkf exceeds When the leakage analysis instruction is sent out, It is the average of the first flood control capacity risk coefficient Dkf of all reservoirs in history.

7. A reservoir flood control dispatching method based on a prediction model according to claim 6, characterized in that: After receiving the discharge analysis instruction, measure the real-time water level Sw of the downstream river t and flow rate Ls t , according to the historical hydrological data of the downstream river, obtain the cross-sectional data and warning water level Jw corresponding to each water level in the downstream river, and calculate the maximum discharge flow Xa of the reservoir t : Among them, Hm Jw Indicates the cross-sectional area of ​​the downstream river corresponding to the warning water level, Indicates the cross-sectional area of ​​the downstream river corresponding to the real-time water level of the downstream river.

8. The reservoir flood control dispatching method based on the prediction model according to claim 7 is characterized by: Get the reservoir flood control capacity Kr t+1 and the maximum discharge flow of the reservoir Xa t , calculate the reservoir's second flood control capacity risk coefficient Ekf: When the second flood resistance capacity risk factor Ekf exceeds When a reservoir flood warning is issued, it will be issued.

9. A reservoir flood control dispatching method based on a prediction model according to claim 8, characterized in that: After receiving the reservoir flood warning, obtain the reservoir flood capacity Kr t+1 and maximum downstream flow rate Xa t , calculate the maximum flood-fighting duration Sc of the reservoir: Among them, [] is the floor function.

10. A reservoir flood control dispatching method based on a prediction model according to claim 9, characterized in that: The maximum flood fighting time Sc of the output reservoir is recorded as the countdown to the flooding of the downstream river channel, and evacuation is carried out according to the countdown to the flooding of the downstream river channel.

Images