Dynamic prediction method and system for overflow flow of weir gate

Abstract

The invention discloses a dynamic weir gate overflowing flow prediction method and system. The dynamic weir gate overflowing flow prediction method comprises the steps that real-time data of all stations are acquired and processed; the future submerging outflow coefficient is predicted through the submerging outflow coefficient prediction model, the input of the submerging outflow coefficient prediction model comprises the water level above the sluice, the water level below the sluice and the trend change of each parameter under each time step, and the output of the submerging outflow coefficient prediction model is the submerging outflow coefficient at a certain moment in the future; the submerging outflow coefficient prediction model adopts a machine learning model; and calculating by adopting a weir flow formula according to the submerging outflow coefficient at a certain moment in the future output by the submerging outflow coefficient prediction model to obtain and output the predicted weir gate overflowing flow. According to the invention, dynamic prediction of the overflow flow of the weir gate can be realized, and the prediction precision of the overflow flow of the weir gate is improved.

Description

Technical Field

[0001] The invention relates to a method for predicting a flow rate of a weir gate, and in particular to a method and system for dynamically predicting a flow rate of a weir gate. Background Art

[0002] In water conservancy projects, in order to consider the comprehensive requirements of flood control, irrigation, power generation, etc., overflow dams and sluices are often built to control and regulate flow. In hydraulics, a wet water structure with overflow on the top is called a weir. The water flow over the weir is called weir flow when it is not controlled by the gate. Therefore, weir flow refers to the phenomenon of continuous and smooth drop of the water surface when the water flows through the discharge structure. For example, the water flow phenomenon passing through the overflow dam top, bridge hole, and the entrance of the pressureless tunnel is all weir flow. When the water flow over the weir is controlled by the gate, it is the gate hole outflow, referred to as hole flow. The gate hole outflow is closely related to the weir flow. When the gate opening height is greater than a certain value, the bottom edge of the gate does not constrain the upper edge of the water flow, and the gate hole outflow turns into weir flow.

[0003] In plain areas, the flow state of weir gates is generally submerged outflow, so the submerged outflow formula is used for weir gate flow calculation. When using this formula for flow calculation, fixed parameters are used. This value is suitable for steady flow, but in actual situations, the flow is complex and fixed parameters are used. This will reduce the accuracy of the calculation results and cause a large gap between the actual measured flow value. Summary of the invention

[0004] Purpose of the invention: In view of the above problems, the present invention proposes a method and system for dynamic prediction of weir flow rate, which can realize dynamic prediction of weir flow rate and improve the prediction accuracy of weir flow rate.

[0005] Technical solution: The technical solution adopted in the present invention is a method for dynamically predicting the flow rate of a weir gate, comprising: acquiring and processing real-time data from each site; predicting the future submerged outflow coefficient through a submerged outflow coefficient prediction model based on the processed real-time data, wherein the input of the submerged outflow coefficient prediction model includes the water level above the gate, the water level below the gate and the trend changes of each parameter at each time step, and the output of the submerged outflow coefficient prediction model is the submerged outflow coefficient at a certain moment in the future; the submerged outflow coefficient prediction model adopts a machine learning model; according to the submerged outflow coefficient at a certain moment in the future output by the submerged outflow coefficient prediction model, the weir flow formula is used to calculate, and the predicted weir gate flow rate is obtained and output.

[0006] As a preferred solution of the method for dynamically predicting the flow rate of the weir gate, real-time data of each station is obtained, including the time series of the water level above the gate, the water level below the gate and the flow rate of the weir gate; processing the real-time data of each station includes: calculating the trend changes of various parameters at each time step according to the real-time data of each station, including the rate of change of the water level above the gate over time, the rate of change of the water level below the gate over time, the rate of change of the difference between the water level above the gate and the water level below the gate over time, and the rate of change of the flow rate of the weir gate over time.

[0007] As a preferred solution of the dynamic prediction method of the weir flow rate, the submerged outflow coefficient prediction model adopts a BP neural network model or a RNN neural network model.

[0008] Furthermore, the training method of the submerged outflow coefficient prediction model comprises the following steps:

[0009] (21) Obtain real-time data from each station, including the water level Z at the gate u , water level below the gate Z d The time series of the weir flow rate Q; according to the data of each station, the flooding outflow coefficient is calculated by the weir flow formula time series; based on the data from each station, calculate the trend changes of each parameter at each time step, including the rate of change of the water level at the gate over time Change rate of water level below the sluice gate over time The rate of change of the difference between the water level above the gate and the water level below the gate over time The rate of change of the flow rate of the weir gate over time

[0010] (22) Set the water level above the gate to Z u , water level below the gate Z d , the rate of change of water level above the gate over time Change rate of water level below the sluice gate over time The rate of change of the difference between the water level above the gate and the water level below the gate over time The rate of change of weir flow rate over time The corresponding flooding outflow coefficient at a certain subsequent moment Packaged as a dataset for training machine learning models.

[0011] The weir flow formula is:

[0012]

[0013] Where Q is the predicted weir flow rate, is the flooding outflow coefficient, B is the total width of the gate opening, H S is the water depth above the sill below the gate, Z u is the water level above the gate, Z dis the water level below the gate, and g is the acceleration due to gravity.

[0014] The present invention proposes a weir gate flow rate dynamic prediction system, which includes a data processing module, a submerged outflow coefficient prediction model module and a weir gate flow rate prediction module;

[0015] Data processing module, used to obtain and process real-time data from each site;

[0016] The submerged outflow coefficient prediction model module is used to: predict the future submerged outflow coefficient through the submerged outflow coefficient prediction model according to the processed real-time data of each station, wherein the input of the submerged outflow coefficient prediction model includes the water level above the gate, the water level below the gate and the trend changes of each parameter at each time step, and the output of the submerged outflow coefficient prediction model is the submerged outflow coefficient at a certain moment in the future; the submerged outflow coefficient prediction model adopts a machine learning model;

[0017] The weir gate flow prediction module is used to calculate the submerged outflow coefficient at a certain moment in the future output by the submerged outflow coefficient prediction model, adopt the weir flow formula to obtain the predicted weir gate flow and output it.

[0018] As a preferred solution of the dynamic prediction system for weir flow rate, real-time data of each station is obtained, including the time series of the water level above the gate, the water level below the gate and the weir flow rate; processing the real-time data of each station includes: calculating the trend changes of various parameters at each time step according to the real-time data of each station, including the rate of change of the water level above the gate over time, the rate of change of the water level below the gate over time, the rate of change of the difference between the water level above the gate and the water level below the gate over time, and the rate of change of the weir flow rate over time.

[0019] The present invention proposes a computer device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for dynamically predicting the flow rate of a weir gate when executing the computer program.

[0020] The present invention provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the method for dynamically predicting the flow rate of a weir gate is implemented.

[0021] The present invention provides a computer program product, comprising a computer program and / or instructions, which implement the method for dynamically predicting the flow rate of a weir gate when the computer program and / or instructions are executed by a processor.

[0022] Beneficial effects: Compared with the prior art, the present invention has the following advantages: The present invention simulates and predicts the submerged outflow coefficient in the weir flow formula based on the dynamically changing time series trend. By predicting the submerged outflow coefficient in real time, the influence of non-steady flow on the submerged outflow coefficient in the weir flow formula is taken into account, thereby improving the prediction accuracy of the weir gate flow rate. The present invention adopts the BP neural network model in machine learning, combined with the real-time changing water level above the gate, the water level below the gate and the flow rate, to predict the submerged outflow coefficient, which can further improve the prediction accuracy of the weir gate flow rate. Comparing the existing static method with the dynamic prediction method described in the present invention, it can be seen that the correlation coefficient R between the prediction result and the measured value of the weir gate flow rate is 2 The relative error of the data has been greatly improved from 5.52% to 3.04%. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a generalized schematic diagram of the weir;

[0024] Figure 2 It is the neural network model data training interface;

[0025] Figure 3 It is a comparison chart of the correlation between the calculated flow rate and the measured flow rate in this embodiment. DETAILED DESCRIPTION

[0026] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0027] Example 1

[0028] The generalized schematic diagram of the weir is as follows Figure 1 As shown in the figure, the gate bottom elevation Z0 is the elevation of the bottom of the weir gate, and the water level above the gate Z u is the water depth upstream of the weir gate, and the water level Z below the gate d is the water depth downstream of the weir gate, and the water depth above the gate sill H0 is the water level Z above the gate u Subtract the gate bottom elevation Z0, the water depth above the gate sill is H S Z is the water level below the gate d Subtracting the gate bottom elevation Z0, the weir gate flow rate Q is the submerged outflow flow from the upstream of the gate to the downstream of the gate. In the actual production application process, considering the difficulty of data observation, the submerged outflow is generally calculated using the weir flow formula (submerged outflow), and the calculation formula is as follows:

[0029]

[0030] in: is the flooding outflow coefficient, which is generally a fixed constant between 1.0 and 1.18; B is the total width of the gate opening, in meters / m; H S Z is the water depth above the sill below the gate, in meters / m; uZ is the water level upstream of the gate, in meters / m; d It is the water level downstream of the sluice gate, in meters / m.

[0031] The present invention simulates and predicts the submerged outflow coefficient in the weir flow formula based on the dynamically changing time series trend. Taking the measured data of a certain measuring station as an example, the dynamic prediction method of the weir flow rate described in the present invention includes the following steps:

[0032] (1) Acquisition and processing of data, including the water level Z above the gate u , water level below the gate Z d Time series of the weir flow rate Q.

[0033] Obtain the total gate opening width B = 80m, gate bottom elevation Z0 = -3m, g = 9.8m / s 2 , obtain relevant data of each station every day, hourly or more detailed, including time t, water level Z on the gate u , water level below the gate Z d And the weir gate flow rate Q;

[0034] (2) According to the water level Z above the gate u , water level below the gate Z d , the time series of the weir flow rate Q, using the weir flow formula to calculate the flooding outflow coefficient The data excerpt is shown in Table 1.

[0035] Table 1 Collection data and calculation of flooding outflow coefficient

[0036]

[0037]

[0038] (3) According to the water level Z above the gate u , water level below the gate Z d , the time series of the weir flow rate Q, and calculate the trend change at each time step, including the rate of change of the water level on the gate over time Change rate of water level below the sluice gate over time The rate of change of the difference between the water level above the gate and the water level below the gate over time The rate of change of the flow rate of the weir gate over time The data excerpt is shown in Table 2.

[0039] Table 2 Calculation input elements table

[0040]

[0041]

[0042] (4) Set the water level above the gate to Z u, water level below the gate Z d , the rate of change of water level above the gate over time Change rate of water level below the sluice gate over time The rate of change of the difference between the water level above the gate and the water level below the gate over time The rate of change of weir flow rate over time The corresponding flooding outflow coefficient for the next time step The data are packaged into a data set, and the machine learning model is trained through the data set. After the training is completed, the flooding outflow coefficient prediction model is obtained. u , water level below the gate Z d , the rate of change of water level above the gate over time Change rate of water level below the sluice gate over time The rate of change of the difference between the water level above the gate and the water level below the gate over time The rate of change of weir flow rate over time As the input of the machine learning model, the inundation outflow coefficient for the next time step is used as the output of the machine learning model.

[0043] The machine learning model may adopt a BP neural network model or a RNN neural network model (taking time series data as input). Any other machine learning model that can complete data learning may replace the above neural network model.

[0044] This embodiment uses the BP neural network model. The simulation results can be obtained by applying "Automatically train network and parameters". You can click again according to your needs to continue training on the result. The interface is as follows Figure 2 shown.

[0045] Those skilled in the art can foresee that the flooding outflow coefficient data of a certain time step in the future can also be used to train the BP neural network model, so that the BP neural network model outputs the flooding outflow coefficient at a certain moment in the future as the predicted flooding outflow coefficient.

[0046] (5) The real-time collected water level Z on the gate u , water level below the gate Z d , the rate of change of water level above the gate over time Change rate of water level below the sluice gate over time The rate of change of the difference between the water level above the gate and the water level below the gate over time The rate of change of weir flow rate over time Through the flooding outflow coefficient prediction model, the flooding outflow coefficient of the next time step is output, that is, the predicted flooding outflow coefficient

[0047] (6) The predicted flooding outflow coefficient is obtained The weir flow formula is used to calculate the predicted weir gate flow rate Q, and the weir gate flow rate at future times can be calculated.

[0048] like Figure 3 The figure shows the correlation between the calculated flow rate and the measured flow rate by the static method and the present embodiment. 2 From 0.7939 to 0.9795, it shows that the simulation effect between the actual flow and the calculated flow is better when the dynamic prediction method of weir flow rate described in the present invention is used. Further analysis of the data shows that when the existing static method is used, the correlation coefficient between the actual flow and the calculated flow is 0.89, and the relative error of the data is 5.52%. At this time, the mean square error of the error is 42.88, which is at a level with large fluctuations above and below the average value; after using the new trend change method (dynamic prediction method of weir flow rate), the correlation coefficient between the actual flow and the calculated flow increases to 0.99, and the relative error decreases to 3.04%. The data accuracy increases significantly, which is reflected in the mean square error of the error decreasing to 19.14.

[0049] Example 2

[0050] The weir gate flow rate dynamic prediction system of the present invention comprises a data processing module, a submerged outflow coefficient prediction model module and a weir gate flow rate prediction module.

[0051] The data processing module is used to obtain and process the real-time data of each station. Specifically, the real-time data of each station is obtained, including the time series of the water level above the gate, the water level below the gate and the flow rate of the weir gate; the real-time data of each station is processed, including: according to the real-time data of each station, the trend change of each parameter at each time step is calculated, including the change rate of the water level above the gate over time, the change rate of the water level below the gate over time, the change rate of the difference between the water level above the gate and the water level below the gate over time, and the change rate of the flow rate of the weir gate over time.

[0052] The flood outflow coefficient prediction model module adopts a machine learning model architecture. The input of the machine learning model includes the water level Z u , water level below the gate Z d , the rate of change of water level above the gate over time Change rate of water level below the gate over time The rate of change of the difference between the water level above the gate and the water level below the gate over time The rate of change of the flow rate of the weir gate over time The output of the machine learning model is the flooding outflow coefficient for the next time step

[0053] Set the water level Z u , water level below the gate Z d , the rate of change of water level above the gate over time Change rate of water level below the gate over time The rate of change of the difference between the water level above the gate and the water level below the gate over time The rate of change of weir flow rate over time The corresponding flooding outflow coefficient for the next time step The data are packaged into a data set for training machine learning models. The specific data set acquisition process is as follows: obtain real-time data from each station, including the water level Z on the gate. u , water level below the gate Z d The time series of the weir flow rate Q; according to the data of each station, the flooding outflow coefficient is calculated by the weir flow formula time series; based on the data from each station, calculate the trend changes of each parameter at each time step, including the rate of change of the water level at the gate over time Change rate of water level below the sluice gate over time The rate of change of the difference between the water level above the gate and the water level below the gate over time The rate of change of the flow rate of the weir gate over time

[0054] The weir flow prediction module is used to convert the predicted flooding outflow coefficient output by the flooding outflow coefficient prediction model module into the predicted flooding outflow coefficient. Bring back the weir flow formula to calculate and obtain the predicted weir gate flow rate Q output, that is, the weir gate flow rate at the future moment.

[0055] Example 3

[0056] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for dynamically predicting the flow rate of a weir gate when executing the computer program.

[0057] Example 4

[0058] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the method for dynamically predicting the flow rate of a weir gate is implemented.

[0059] Example 5

[0060] In one embodiment, a computer program product is provided, including a computer program / instruction, which implements the method for dynamically predicting the flow rate of a weir gate when executed by a processor.

[0061] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0062] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0063] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0064] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

Claims

1. A method for dynamic prediction of flow rate of weir gate, characterized in that: include: Obtain and process real-time data from each site; According to the processed real-time data, the future submerged outflow coefficient is predicted by the submerged outflow coefficient prediction model, wherein the input of the submerged outflow coefficient prediction model includes the water level above the gate, the water level below the gate and the trend changes of various parameters at each time step, and the output of the submerged outflow coefficient prediction model is the submerged outflow coefficient at a certain moment in the future; The flood outflow coefficient prediction model uses a machine learning model; According to the submerged outflow coefficient at a certain moment in the future output by the submerged outflow coefficient prediction model, the weir flow formula is used to calculate and output the predicted weir gate flow rate.

2. The method for dynamic prediction of weir flow rate according to claim 1 is characterized in that: Acquire real-time data of each station, including the time series of water level above the gate, water level below the gate and flow rate of the weir gate; process the real-time data of each station, including: according to the real-time data of each station, calculate the trend change of each parameter at each time step, including the rate of change of water level above the gate over time, the rate of change of water level below the gate over time, the rate of change of the difference between the water level above the gate and the water level below the gate over time, and the rate of change of flow rate of the weir gate over time.

3. The method for dynamic prediction of weir flow rate according to claim 1 is characterized in that: The submerged outflow coefficient prediction model adopts a BP neural network model or a RNN neural network model.

4. The method for dynamic prediction of weir flow rate according to claim 1 is characterized in that: The training method of the submerged outflow coefficient prediction model comprises the following steps: (1) Obtain real-time data from each station, including the water level Z at the gate u , water level below the gate Z d The time series of the weir flow rate Q; according to the data of each station, the flooding outflow coefficient is calculated by the weir flow formula time series; based on the data from each station, calculate the trend changes of each parameter at each time step, including the rate of change of the water level at the gate over time Change rate of water level below the gate over time The rate of change of the difference between the water level above the gate and the water level below the gate over time The rate of change of the flow rate of the weir gate over time (2) Set the water level above the gate to Z u , water level below the gate Z d , the rate of change of water level above the gate over time Change rate of water level below the gate over time The rate of change of the difference between the water level above the gate and the water level below the gate over time The rate of change of weir flow rate over time The corresponding flooding outflow coefficient at a certain subsequent moment Packaged as a dataset for training machine learning models.

5. The method for dynamic prediction of weir flow rate according to claim 1 is characterized in that: The weir flow formula is: Among them, Q is the predicted flow rate of the weir gate, is the flooding outflow coefficient, B is the total width of the gate opening, H S is the water depth above the sill below the gate, Z u is the water level above the gate, Z d is the water level below the gate, and g is the acceleration due to gravity.

6. A dynamic prediction system for flow rate of weir gate, characterized by: The system includes a data processing module, a flooding outflow coefficient prediction model module and a weir gate flow prediction module; Data processing module, used to obtain and process real-time data from each site; The submerged outflow coefficient prediction model module is used to: predict the future submerged outflow coefficient through the submerged outflow coefficient prediction model, wherein the input of the submerged outflow coefficient prediction model includes the water level above the gate, the water level below the gate and the trend changes of various parameters at each time step, and the output of the submerged outflow coefficient prediction model is the submerged outflow coefficient at a certain moment in the future; the submerged outflow coefficient prediction model adopts a machine learning model; The weir gate flow prediction module is used to calculate the submerged outflow coefficient at a certain moment in the future output by the submerged outflow coefficient prediction model, adopt the weir flow formula to obtain the predicted weir gate flow and output it.

7. The method for dynamic prediction of weir flow rate according to claim 6 is characterized in that: Acquire real-time data of each station, including the time series of water level above the gate, water level below the gate and flow rate of the weir gate; process the real-time data of each station, including: according to the real-time data of each station, calculate the trend change of each parameter at each time step, including the rate of change of water level above the gate over time, the rate of change of water level below the gate over time, the rate of change of the difference between the water level above the gate and the water level below the gate over time, and the rate of change of flow rate of the weir gate over time.

8. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for dynamically predicting the flow rate of a weir gate according to any one of claims 1 to 5 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for dynamically predicting the flow rate of a weir gate according to any one of claims 1 to 5 is implemented.

10. A computer program product comprising a computer program and / or instructions, characterized in that: When the computer program and / or instruction is executed by a processor, the method for dynamically predicting the flow rate of a weir gate as described in any one of claims 1 to 5 is implemented.

Images