Method and device for predicting water level of forebay of pump station

Abstract

The invention relates to the technical field of pump station forebay water level prediction, and discloses a pump station forebay water level prediction method and device, and the method comprises the steps: obtaining the flow data of a target pump station, carrying out the cleaning of the flow data of the target pump station, obtaining the corrected inflow flow data of the target pump station, constructing a water level-storage capacity function, and obtaining the corrected inflow flow data of the target pump station; and based on the corrected inflow data of the target pump station, predicting the water level of the forebay of the target pump station by using a water level-storage capacity function to obtain an initial water level prediction sequence value, and finally optimizing the initial water level prediction sequence value to obtain a water level prediction sequence value of the forebay of the pump station. The accuracy of water level prediction is guaranteed, and data support is provided for the reasonability of a pump station forebay water level control decision.

Description

Technical Field

[0001] This invention relates to the field of prediction technology for the water level of a pumping station forebay, and specifically to a method and apparatus for predicting the water level of a pumping station forebay. Background Technology

[0002] Rapid urbanization and population growth have put immense pressure on urban drainage systems. As a crucial component of the drainage system, the stability of the forebay water level directly impacts the safe and efficient operation of the drainage system.

[0003] However, the prediction accuracy of the relevant pump station forebay water level prediction methods is low, which affects the rationality of the pump station forebay water level control decision. Summary of the Invention

[0004] This invention provides a method and apparatus for predicting the water level of a pumping station forebay, in order to solve the problem that the prediction accuracy of related pumping station forebay water level prediction methods is low, which affects the rationality of pumping station forebay water level control decisions.

[0005] In a first aspect, the present invention provides a method for predicting the water level of a pumping station forebay, the method comprising: Acquire the flow data of the target pumping station, clean the flow data of the target pumping station, and obtain the corrected inflow flow data of the target pumping station; Construct the water level-reservoir capacity function; Based on the corrected inflow data of the target pumping station, the water level in the forebay of the target pumping station is predicted using the water level-storage capacity function to obtain the initial water level prediction sequence value. The initial water level prediction sequence values ​​are optimized to obtain the water level prediction sequence values ​​of the pump station forebay.

[0006] The method for predicting the water level of the forebay of a pumping station provided in this embodiment cleanses the flow data of the target pumping station to ensure the authenticity and reliability of the corrected inflow flow data. A water level-storage capacity function is constructed, and the water level of the forebay of the target pumping station is predicted using the water level-storage capacity function based on the corrected inflow flow data of the target pumping station to obtain an initial water level prediction sequence value. Finally, the initial water level prediction sequence value is optimized to eliminate the deviation between the initial water level prediction sequence value and the actual situation, so as to ensure the accuracy of water level prediction and provide data support for the rationality of the water level control decision of the pumping station forebay.

[0007] In one optional implementation, the flow data of the target pumping station is cleaned to obtain corrected inflow flow data of the target pumping station, including: Obtain the reservoir capacity change and flow loss per unit time of the target pumping station. Based on the reservoir capacity change, flow loss per unit time, total inflow and total outflow per unit time of the target pumping station's flow data, calculate the water transfer loss rate for any two time periods. Calculate the dimensionless water loss rate based on the water loss rate and the total outflow per unit time for any two time periods; The system obtains the real-time outflow rate of the upstream pumping station, the minimum sampling interval time of the target pumping station, and the longest sequence interval time of the target pumping station. Based on the dimensionless water transfer loss rate, the loss flow rate per unit time, the real-time outflow rate of the upstream pumping station, the minimum sampling interval time of the target pumping station, and the longest sequence interval time of the target pumping station, the system calculates the corrected inflow rate data of the target pumping station.

[0008] The method for predicting the water level in the forebay of a pumping station provided in this embodiment calculates the water transfer loss rate for any two time periods based on the reservoir capacity change of the target pumping station, the loss flow per unit time, and the total inflow and total outflow per unit time in the flow data of the target pumping station. This directly quantifies the proportion of loss in the actual water transfer process within the two time periods, providing basic loss data for subsequent analysis. Based on the water transfer loss rate and the total outflow per unit time for any two time periods, a dimensionless water transfer loss rate is calculated, eliminating the influence of differences in the scale of total outflow and allowing for horizontal comparison of water transfer loss rates at different flow levels. Finally, based on the dimensionless water transfer loss rate, the loss flow per unit time, the real-time outflow of the upstream pumping station, the minimum sampling interval time and the longest sequence interval time of the target pumping station, the corrected inflow flow data of the target pumping station is calculated, correcting the deviation of the original inflow data and improving the accuracy and practicality of the inflow flow data.

[0009] In one optional implementation, based on the corrected inflow data of the target pumping station, the water level in the forebay of the target pumping station is predicted using a water level-storage capacity function to obtain an initial water level prediction sequence value, including: Inverting the water level-reservoir capacity function yields the water level prediction function; Obtain the current reservoir capacity of the target pumping station, and calculate the predicted reservoir capacity based on the corrected inflow data of the target pumping station and the current reservoir capacity of the target pumping station. The reservoir capacity prediction value is input into the water level prediction function to obtain the initial water level prediction sequence value.

[0010] The method for predicting the water level of the pump station forebay provided in this embodiment obtains a water level prediction function by inverting the water level-storage capacity function, establishing a direct mapping relationship between storage capacity and water level, and providing a core conversion tool for subsequent water level prediction. Based on the corrected inflow data of the target pump station and the current storage capacity of the target pump station, the predicted storage capacity value is calculated, ensuring that the predicted storage capacity value is closer to the actual change trend and improving the accuracy of storage capacity prediction. Finally, the predicted storage capacity value is input into the water level prediction function to obtain the initial water level prediction sequence value, providing clear data support for actual operations such as pump station water level regulation and flood control scheduling.

[0011] In one optional implementation, the initial water level prediction sequence values ​​are optimized to obtain the water level prediction sequence values ​​for the pump station forebay, including: Abrupt changes are determined in the initial water level prediction sequence values, local maxima are identified and removed, and the water level prediction sequence values ​​after removing local maxima are obtained. After removing local maxima, the water level prediction sequence values ​​are smoothed to obtain the water level prediction sequence values ​​of the pump station forebay.

[0012] The method for predicting the water level of the pump station forebay provided in this embodiment determines and removes local maxima by performing abrupt change judgment on the initial water level prediction sequence values. This identifies and removes unreasonable abrupt peaks in the initial prediction sequence, eliminates the interference of abnormal data on subsequent analysis, and ensures the rationality of the water level prediction sequence. Furthermore, the method performs data smoothing processing on the water level prediction sequence values ​​after removing local maxima, which weakens the random fluctuations and small errors in the sequence, making the water level prediction sequence more closely match the stable trend of actual water level changes, and improving the accuracy and usability of the pump station forebay water level prediction.

[0013] In one optional implementation, abrupt changes are determined in the initial water level prediction sequence values, local maxima are identified and removed, and the sequence values ​​after removing local maxima are obtained, including: The second derivative of the initial water level prediction sequence value is obtained by taking the second derivative value at the current time. The second derivative value at the current time is compared with a preset threshold. If the second derivative value at the current time is greater than the preset threshold, the second derivative value at the current time is compared with the second derivative value at the adjacent time. If the second derivative value at the current time is greater than the second derivative value at the adjacent time, then the water level prediction value corresponding to the second derivative value is taken as a local maximum, and the local maximum value is removed to obtain the sequence value after removing the local maximum value.

[0014] The method for predicting the water level of the pump station forebay provided in this embodiment accurately captures potential abrupt inflection points in the sequence by taking the second derivative of the initial water level prediction sequence value. This provides a mathematical basis for subsequent determination of local maxima. The second derivative value at the current moment is compared with a preset threshold. If the second derivative value at the current moment is greater than the preset threshold, the second derivative value at the current moment is compared with the second derivative value at the adjacent moment, reducing the probability of misjudgment and improving the targeting of the judgment. Finally, when the second derivative value at the current moment is greater than the second derivative value at the adjacent moment, the water level prediction value corresponding to the second derivative value is taken as a local maximum, and the local maximum value is removed, resulting in a sequence value after removing local maxima that is more consistent with the actual trend.

[0015] In one optional implementation, the water level prediction sequence values ​​after removing local maxima are subjected to data smoothing processing to obtain the water level prediction sequence values ​​of the pump station forebay, including: The water level prediction sequence after removing local maxima is decomposed to obtain high- and low-frequency split water level data. Attention enhancement is applied to the high- and low-frequency split water level data to obtain attention-enhanced high- and low-frequency split water level data. The high- and low-frequency split water level data after attention enhancement were processed by moving regression averaging to obtain the predicted water level sequence value of the pump station forebay.

[0016] The method for predicting the water level of the pump station forebay provided in this embodiment decomposes the water level prediction sequence value after removing local maxima to obtain high- and low-frequency split water level data, which provides a basis for subsequent targeted processing of data with different characteristics and avoids accuracy deviation caused by mixed data processing. Attention enhancement is applied to the high- and low-frequency split water level data to obtain attention-enhanced high- and low-frequency split water level data, which weakens irrelevant noise interference and improves the recognition of core data information. Finally, the attention-enhanced high- and low-frequency split water level data is processed by moving regression averaging to finally output a water level prediction sequence value of the pump station forebay that has both stability and accuracy.

[0017] Secondly, the present invention provides a device for predicting the water level of a pumping station forebay, the device comprising: The cleaning module is used to acquire the flow data of the target pumping station, clean the flow data of the target pumping station, and obtain the corrected inflow flow data of the target pumping station. The module is used to construct the water level-reservoir capacity function; The prediction module is used to predict the water level in the forebay of the target pumping station based on the corrected inflow data of the target pumping station and the water level-storage capacity function, so as to obtain the initial water level prediction sequence value. The optimization module is used to optimize the initial water level prediction sequence values ​​to obtain the water level prediction sequence values ​​of the pump station forebay.

[0018] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method for predicting the water level of the pump station forebay as described in the first aspect or any corresponding embodiment.

[0019] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the method for predicting the water level of the pump station forebay as described in the first aspect or any corresponding embodiment thereof.

[0020] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the method for predicting the water level of the pump station forebay as described in the first aspect or any corresponding embodiment. Attached Figure Description

[0021] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of an application scenario according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the first process of a method for predicting the water level of a pump station forebay according to an embodiment of the present invention. Figure 3 This is a top view of the target pumping station forebay according to an embodiment of the present invention; Figure 4 This is a cross-sectional view of the forebay of the target pumping station according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the second process of the method for predicting the water level of the pump station forebay according to an embodiment of the present invention; Figure 6 This is a schematic diagram of the third process of the method for predicting the water level of the pump station forebay according to an embodiment of the present invention; Figure 7 This is a schematic diagram of the fourth process of the method for predicting the water level of the pump station forebay according to an embodiment of the present invention; Figure 8 This is a schematic diagram of the workflow of the method for predicting the water level of the pump station forebay according to an embodiment of the present invention; Figure 9 This is a structural block diagram of a device for predicting the water level of a pump station forebay according to an embodiment of the present invention. Figure 10 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0025] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0026] As an optional application scenario of this invention, such as Figure 1 As shown, the device for predicting the water level in the forebay of the pumping station may include at least one terminal device and at least one server. Figure 1 The system is illustrated in the example, which includes a computer 101, a mobile terminal 102, and a server 103, and the terminal devices such as the computer 101 and the mobile terminal 102 are connected to the server 103 through a network 110.

[0027] Specifically, the terminal device can be a smartphone, tablet, laptop, PDA, desktop computer, game console, smart TV, smart wearable device, in-vehicle terminal, VR (Virtual Reality) device, AR (Augmented Reality) device, etc. Server 103 can be a standalone physical server, a server cluster, a distributed system, or a cloud server providing cloud services. Network 110 can be a wired or wireless network, examples of which include, but are not limited to, the Internet, corporate intranet, local area network, wide area network, mobile communication network, and combinations thereof.

[0028] With the acceleration of urbanization and the continuous growth of population, urban drainage systems are facing increasing pressure. As an important part of the drainage system, the operating efficiency of pumping stations directly affects the city's drainage capacity. The main function of the pumping station forebay is to ensure that the water flow from the diversion channel to the inlet pipe can spread smoothly, thereby providing good inflow conditions for the unit inlet. Therefore, the stability of the water level in the pumping station forebay is a prerequisite for the safe and efficient operation of the pumping station.

[0029] Accurate prediction of the forebay water level can provide effective data support for the constant water level automatic control system, help optimize the pump group control scheme, and enable the control commands to be generated and sent to the constant water level automatic control system in advance before the water level fluctuates significantly. This allows the forebay water level to be stabilized within a narrow range, thereby improving drainage efficiency, extending equipment life and reducing operating costs, while avoiding water level deviations caused by data lag and control delay. This brings new ideas and logical systems to the construction of future smart water network projects.

[0030] The relevant methods for predicting the water level in the forebay of pumping stations are mainly based on hydrodynamic models or statistical time series models. Among them, the hydrodynamic model usually includes one-dimensional or two-dimensional fluid dynamics equations. Combined with the structural parameters and operating conditions of the pumping station, the fluid motion equations are solved by numerical methods (such as the finite difference method or the finite element method) to obtain the change of water level over time. The water level prediction method based on the hydrodynamic model has clear physical meaning and can consider the influence of multiple factors on the water level.

[0031] Water level prediction methods based on statistical time series models require the establishment of water level time series models through statistical analysis of historical water level data. Water level time series prediction methods include ARIMA (Autoregressive Integrated Moving Average) and exponential smoothing. Water level prediction methods based on statistical time series models predict future water levels by statistically analyzing historical water level data and then establishing mathematical expressions. They have the advantages of being simple to calculate, easy to implement, and not requiring high-quality data.

[0032] With the rapid development of artificial intelligence, the method for predicting the water level of the pump station forebay has incorporated the ideas of artificial intelligence algorithms. These algorithms mainly include: 1) Long Short-Term Memory (LSTM) neural networks, ensemble learning methods, and feature extraction methods based on deep learning. LSTM is a special type of recurrent neural network. By constructing an LSTM model and training it with historical water level data, it can predict future water levels. Furthermore, the LSTM model can effectively remember long-term dependent information and capture complex nonlinear relationships, thereby improving prediction accuracy. 2) Ensemble learning methods can effectively reduce overfitting and improve the robustness and accuracy of water level prediction by training multiple base models and fusing their predictions through weighted averaging or voting. Ensemble learning methods include random forests and gradient boosting. 3) Feature extraction methods based on deep learning use deep neural networks, such as CNN (Convolutional Neural Network) and RNN (Recurrent Neural Network). Models such as recurrent neural networks (RNNs) are used to extract features and model raw hydrological data. By automatically extracting features from the data through multi-layer neural networks, water level prediction models are established to predict water levels. Deep learning-based feature extraction methods can automatically extract complex features and have good prediction accuracy and robustness.

[0033] Water level prediction methods based on machine learning and deep learning models have significant advantages in predicting water levels in the forebay of pumping stations, but they also have some limitations: 1) Water level prediction based on deep learning models is highly dependent on the quality and volume of historical data. Insufficient or poor-quality data will affect the accuracy of model prediction; 2) It is difficult to tune and optimize complex models. For example, training a deep LSTM model requires a long time, which affects real-time prediction; 3) Model parameters such as the number of layers and units need to be tuned. Insufficient parameter accuracy may affect the model prediction performance. However, in actual engineering deployment, using similar deep learning models requires powerful computing resources and hardware support, which increases the implementation cost.

[0034] According to an embodiment of the present invention, a method for predicting the water level of a pump station forebay is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0035] This embodiment provides a method for predicting the water level in the forebay of a pumping station, which can be used in the aforementioned electronic equipment. Figure 2This is a flowchart of a method for predicting the water level of a pumping station forebay according to an embodiment of the present invention, as shown below. Figure 2 As shown, the process includes the following steps: Step S201: Obtain the flow data of the target pumping station, clean the flow data of the target pumping station, and obtain the corrected inflow flow data of the target pumping station.

[0036] Specifically, the target pumping station observes hydrological data through hydrological measurement equipment such as water level gauges and flow meters to obtain the flow data of the target pumping station. This data is then transmitted in real-time to the lower-level programmable logic controller (PLC) using a 4G (4th Generation Mobile Communication Technology) signal card and an RTU (Remote Terminal Unit) module. The status information acquisition module has both Modbus (an industrial automation communication protocol) and OPCUA (another industrial automation communication protocol) communication program clients and servers. The Modbus communication protocol is mainly used for communication between PLCs and is widely used in industrial automation systems due to its simplicity and stability. The OPCUA communication protocol offers more functions and higher security, supporting complex data structures and multiple data transmission modes. The SCADA (Supervisory Control and Data Acquisition) system connects to the constant water level control program server through the client. During status acquisition, the server periodically reads data from the client devices. After being aggregated by the intelligent integrated platform, the data is finally synchronously written into a time-series database to form complete flow data for the target pumping station.

[0037] Furthermore, before writing the flow data of the target pumping station into the time series database, a data quality control process is required to ensure the reliability and stability of the acquired data and prevent water level prediction deviations or erroneous command issuance due to abnormal data. The specific process includes: 1) The aggregated data is forwarded to a high-speed data transmission channel, such as the DDC (Direct Digital Control) data pipeline, via Kafka (a distributed stream processing platform). After being filtered by the data quality control module in the DDC pipeline, abnormal data (such as jumps, exceeding limits, etc.) will be written into the time series database in a tagged format. In the event of data interruption or missing data, the data quality control module will fill in the missing values ​​to prevent signal delays that could lead to untimely water level regulation. Since the order of magnitude of the input data sometimes differs significantly, a deviation standardization method is used to normalize the input data; the normalization calculation formula is as follows: (1) in, For the normalized data, X This is the original data. X max The maximum value of the original data. X min This represents the minimum value of the original data.

[0038] Step S202: Construct the water level-reservoir capacity function.

[0039] Specifically, the water level-storage capacity relationship curve is calculated based on the physical model of the pump station forebay. The specific water level-storage capacity relationship curve needs to be determined according to the actual engineering situation. Figure 3 This is a top view of the pump station forebay. Figure 4 The diagram shows a cross-sectional view of the forebay of the pumping station. Because the forebay of the target pumping station has a bottom slope, meaning there is a height difference between the left and right reservoirs, and because the water level-storage capacity function is a piecewise function, and the base height should be subtracted during water level calculation, the water level-storage capacity function can be obtained. The expression is: (2) in, The bottom area of ​​the main reservoir The bottom area of ​​the grating-equipped reservoir (right side of the main reservoir) is... This is the current absolute water level elevation. The base elevation, The height difference between the main reservoir and the reservoir with a screen.

[0040] Step S203: Based on the corrected inflow data of the target pumping station, the water level of the forebay of the target pumping station is predicted using the water level-storage capacity function to obtain the initial water level prediction sequence value.

[0041] Step S204: Optimize the initial water level prediction sequence value to obtain the water level prediction sequence value of the pump station forebay.

[0042] The method for predicting the water level of the forebay of a pumping station provided in this embodiment cleanses the flow data of the target pumping station to ensure the authenticity and reliability of the corrected inflow flow data. A water level-storage capacity function is constructed, and the water level of the forebay of the target pumping station is predicted using the water level-storage capacity function based on the corrected inflow flow data of the target pumping station to obtain an initial water level prediction sequence value. Finally, the initial water level prediction sequence value is optimized to eliminate the deviation between the initial water level prediction sequence value and the actual situation, so as to ensure the accuracy of water level prediction and provide data support for the rationality of the water level control decision of the pumping station forebay.

[0043] This embodiment provides a method for predicting the water level in the forebay of a pumping station, which can be used in the aforementioned electronic equipment. Figure 5 This is a flowchart of a method for predicting the water level of a pumping station forebay according to an embodiment of the present invention, as shown below. Figure 5 As shown, the process includes the following steps: Step S501: Obtain the flow data of the target pumping station, clean the flow data of the target pumping station, and obtain the corrected inflow flow data of the target pumping station.

[0044] Specifically, step S501 includes: Step S5011: Obtain the reservoir capacity change and the loss flow per unit time of the target pumping station. Based on the reservoir capacity change, the loss flow per unit time of the target pumping station, and the total inflow and total outflow per unit time of the target pumping station's flow data, calculate the water transfer loss rate for any two time periods.

[0045] Specifically, the flow data of the target pumping station is arranged in a time series to represent the station's operating status at different times. Arranging it in a spatial series represents the relationship between the station's operating status at different times and the current water volume in the canal section. To ensure the logical consistency and reliability of the flow data, it is necessary to clean the flow data of the target pumping station at both the time and spatial scales, and to calculate the water transfer loss rate for any two time periods based on the principle of dynamic water balance. The water transfer loss rate for any two time periods... The calculation formula is: (3) in, This refers to the flow loss per unit time (such as evaporation, leakage, etc.). For time intervals, The change in reservoir capacity of the target pumping station. and These represent the total inflow and total outflow per unit time in the flow data of the target pumping station, respectively.

[0046] Step S5012: Calculate the dimensionless water loss rate based on the water loss rate and the total outflow per unit time for any two time periods.

[0047] Specifically, after calculating the water conveyance loss rate for any two time periods, the dimensionless water conveyance loss rate is obtained by dividing the water conveyance loss flow by the inflow flow. The calculation formula is: (4) Step S5013: Obtain the real-time outflow rate of the upstream pumping station, the minimum sampling interval time of the target pumping station, and the longest sequence interval time of the target pumping station. Based on the dimensionless water transfer loss rate, the loss flow rate per unit time, the real-time outflow rate of the upstream pumping station, the minimum sampling interval time of the target pumping station, and the longest sequence interval time of the target pumping station, calculate the corrected inflow rate data of the target pumping station.

[0048] Specifically, the water level prediction program searches for the longest sequence in the time series database based on the prediction time length, and then uses the principle of dynamic water balance to clean the flow data within the interval (i.e., the inflow flow data of the target pumping station) to obtain the corrected inflow flow data of the target pumping station; wherein, the calculation formula for the corrected inflow flow data of the target pumping station is... for: (5) in, This refers to the real-time outflow rate of the upstream pumping station. The minimum sampling interval for the target pumping station. L s The longest sequence interval time for the target pumping station refers to the interval time for obtaining the longest sequence.

[0049] Furthermore, the time series database data is stored in its original form according to the water volume model number, IEC 61850 (a type of data measurement point number) standard data measurement point number, timestamp, and data volume. The data is then compiled in time intervals of 5 minutes, half an hour, etc., through the background program. In order to ensure the accuracy of the prediction, when sampling the flow data of the target pumping station, a data sliding window needs to be created according to the required prediction duration (1h, 3h, and 6h) to select the sequence with the most data in the original data. This is because the original data of the pumping station water level monitoring is uploaded through PLC integration, and the upload interval is reported through multicast, so the time interval is not fixed.

[0050] Step S502: Construct the water level-reservoir capacity function. For details, please refer to [link to relevant documentation]. Figure 2 Step S202 of the illustrated embodiment will not be described again here.

[0051] Step S503: Based on the corrected inflow data of the target pumping station, the water level in the forebay of the target pumping station is predicted using the water level-storage capacity function to obtain the initial water level prediction sequence values. For details, please refer to [link to relevant documentation]. Figure 2 Step S203 of the illustrated embodiment will not be described again here.

[0052] Step S504: Optimize the initial water level prediction sequence values ​​to obtain the water level prediction sequence values ​​for the pump station forebay. For details, please refer to [link to details]. Figure 2 Step S204 of the illustrated embodiment will not be described again here.

[0053] The method for predicting the water level in the forebay of a pumping station provided in this embodiment calculates the water transfer loss rate for any two time periods based on the reservoir capacity change of the target pumping station, the loss flow per unit time, and the total inflow and total outflow per unit time in the flow data of the target pumping station. This directly quantifies the proportion of loss in the actual water transfer process within the two time periods, providing basic loss data for subsequent analysis. Based on the water transfer loss rate and the total outflow per unit time for any two time periods, a dimensionless water transfer loss rate is calculated, eliminating the influence of differences in the scale of total outflow and allowing for horizontal comparison of water transfer loss rates at different flow levels. Finally, based on the dimensionless water transfer loss rate, the loss flow per unit time, the real-time outflow of the upstream pumping station, the minimum sampling interval time and the longest sequence interval time of the target pumping station, the corrected inflow flow data of the target pumping station is calculated, correcting the deviation of the original inflow data and improving the accuracy and practicality of the inflow flow data.

[0054] This embodiment provides a method for predicting the water level in the forebay of a pumping station, which can be used in the aforementioned electronic equipment. Figure 6 This is a flowchart of a method for predicting the water level of a pumping station forebay according to an embodiment of the present invention, as shown below. Figure 6 As shown, the process includes the following steps: Step S601: Obtain the flow data of the target pumping station, clean the flow data of the target pumping station, and obtain the corrected inflow flow data of the target pumping station. For details, please refer to [link to relevant documentation]. Figure 5 Step S501 of the illustrated embodiment will not be described again here.

[0055] Step S602: Construct the water level-reservoir capacity function. For details, please refer to [link to relevant documentation]. Figure 5 Step S502 of the illustrated embodiment will not be described again here.

[0056] Step S603: Based on the corrected inflow data of the target pumping station, the water level of the forebay of the target pumping station is predicted using the water level-storage capacity function to obtain the initial water level prediction sequence value.

[0057] Specifically, step S603 includes: Step S6031: Invert the water level-reservoir capacity function to obtain the water level prediction function.

[0058] Specifically, the inflow rate of the target pumping station (i.e., the total inflow and total outflow per unit time in the flow data of the target pumping station) is the main influencing factor on the water level of the forebay of the pumping station. If the inflow rate is continuously input, the reservoir capacity changes over time as a function: (6) Furthermore, by combining the water level-storage capacity function, the inverse function of the water level-storage capacity function is calculated to obtain the water level prediction function; where the water level prediction function... The expression is: (7) Furthermore, by differentiating the water level prediction function, we can obtain the relationship between water level and time. The expression for the function of water level changing with time is: (8) Step S6032: Obtain the current reservoir capacity of the target pumping station, and calculate the predicted reservoir capacity based on the corrected inflow data of the target pumping station and the current reservoir capacity of the target pumping station.

[0059] Specifically, the corrected inflow data of the target pumping station directly affects the reservoir capacity of the target pumping station. The predicted reservoir capacity can be calculated based on the corrected inflow data of the target pumping station and the current reservoir capacity of the target pumping station.

[0060] Step S6033: Input the reservoir capacity prediction value into the water level prediction function to obtain the initial water level prediction sequence value.

[0061] Specifically, by substituting the reservoir capacity prediction value into the calculation formula (7), the initial water level prediction sequence value can be obtained; where, the initial water level prediction sequence value The calculation formula is: (9) Step S604: Optimize the initial water level prediction sequence values ​​to obtain the water level prediction sequence values ​​for the pump station forebay. For details, please refer to [link to details]. Figure 5 Step S504 of the illustrated embodiment will not be described again here.

[0062] The method for predicting the water level of the pump station forebay provided in this embodiment obtains a water level prediction function by inverting the water level-storage capacity function, establishing a direct mapping relationship between storage capacity and water level, and providing a core conversion tool for subsequent water level prediction. Based on the corrected inflow data of the target pump station and the current storage capacity of the target pump station, the predicted storage capacity value is calculated, ensuring that the predicted storage capacity value is closer to the actual change trend and improving the accuracy of storage capacity prediction. Finally, the predicted storage capacity value is input into the water level prediction function to obtain the initial water level prediction sequence value, providing clear data support for actual operations such as pump station water level regulation and flood control scheduling.

[0063] This embodiment provides a method for predicting the water level in the forebay of a pumping station, which can be used in the aforementioned electronic equipment. Figure 7 This is a flowchart of a method for predicting the water level of a pumping station forebay according to an embodiment of the present invention, as shown below. Figure 7 As shown, the process includes the following steps: Step S701: Obtain the flow data of the target pumping station, clean the flow data of the target pumping station, and obtain the corrected inflow flow data of the target pumping station. For details, please refer to [link to relevant documentation]. Figure 6Step S601 of the illustrated embodiment will not be described again here.

[0064] Step S702: Construct the water level-reservoir capacity function. For details, please refer to [link to relevant documentation]. Figure 6 Step S602 of the illustrated embodiment will not be described again here.

[0065] Step S703: Based on the corrected inflow data of the target pumping station, the water level in the forebay of the target pumping station is predicted using the water level-storage capacity function to obtain the initial water level prediction sequence values. For details, please refer to [link to relevant documentation]. Figure 6 Step S603 of the illustrated embodiment will not be described again here.

[0066] Step S704: Optimize the initial water level prediction sequence value to obtain the water level prediction sequence value of the pump station forebay.

[0067] Specifically, step S704 above includes: Step S7041: Perform a mutation determination on the initial water level prediction sequence value, identify and remove local maxima, and obtain the water level prediction sequence value after removing local maxima.

[0068] In some optional implementations, step S7041 above includes: Step a1: Take the second derivative of the initial water level prediction sequence value to obtain the second derivative value at the current time.

[0069] Specifically, after obtaining the water level prediction sequence values, it is necessary to use the second derivative to determine abrupt changes and remove local maxima to prevent the water level prediction values ​​from deviating locally from the actual values. Specifically, the finite difference method is used to approximate the second derivative value at the current moment; the formula for calculating the second derivative value is as follows: (10) in, This is the value of the second derivative at the current moment.

[0070] Step a2: Compare the second derivative value at the current time with a preset threshold. If the second derivative value at the current time is greater than the preset threshold, then compare the second derivative value at the current time with the second derivative value at the adjacent time.

[0071] Specifically, a threshold for the second derivative is set. θ (i.e., a preset threshold) If the absolute value of the second derivative at the current moment exceeds the preset threshold, then the current moment is considered to be a potential point of change. θ The selection of the threshold needs to be adjusted based on the specific data situation. It can be determined through statistical methods (such as standard deviation) or expert experience. In this application, the preset threshold is set as the average of the second derivative sequence plus several times the standard deviation. The formula for calculating the preset threshold is as follows: (11) in, k This is a natural constant, usually taken as 2-3, and in this application, it is taken as... k =2.

[0072] Step a3: If the second derivative value at the current time is greater than the second derivative value at the adjacent time, then the water level prediction value corresponding to the second derivative value is taken as a local maximum value, and the local maximum value is removed to obtain the sequence value after removing the local maximum value.

[0073] Specifically, if the second derivative value at the current moment is [value], it is further checked whether the water level prediction value corresponding to the second derivative value is a local maximum. If the water level prediction value corresponding to the second derivative value is a local maximum, then the water level prediction value corresponding to the second derivative value is a local maximum. The expression for the criterion of whether the water level prediction value corresponding to the second derivative value is a local maximum is: (12) Furthermore, for the detected local maxima, spline interpolation can be used to correct the data or remove the mutation points, thereby obtaining the water level prediction sequence value after removing the local maxima.

[0074] Step S7042: Perform data smoothing on the water level prediction sequence after removing local maxima to obtain the water level prediction sequence of the pump station forebay.

[0075] Specifically, to obtain a more accurate predicted water level sequence for the pump station forebay, data smoothing is required after removing local maxima. First, wavelet transform is used to decompose the data sequence, separating high- and low-frequency water level data. Then, a weighted moving average is applied to the high- and low-frequency water level data. Finally, the processed data is reconstructed using wavelet transform to obtain the smoothed water level sequence (i.e., the predicted water level sequence for the pump station forebay). Compared to a simple moving average, a weighting factor is introduced, avoiding the unsatisfactory smoothing caused by data reception lag. The formula for calculating the predicted water level sequence for the pump station forebay is as follows: (13) in, N The degree of smoothness is determined by the window size (number of pixels). H [ t [] is the input signal (i.e., the water level prediction sequence value after removing local maxima). Y [ t [This represents the predicted sequence value of the water level in the forebay of the pumping station (filtered data)]. w [ k Different weights are assigned to data at different times, with data closer to time t receiving higher weights. When using linear weighting, w [k ]= N k .

[0076] In some optional implementations, step S4042 above includes: Step b1 involves decomposing the water level prediction sequence after removing local maxima to obtain high- and low-frequency split water level data.

[0077] Furthermore, the water level prediction sequence values ​​after removing local maxima exhibit different frequency characteristics due to various influencing factors. These factors include: 1) Short-cycle high-frequency influences: Rainfall intensity: Short-duration heavy rainfall directly leads to a surge in inflow, affecting water level fluctuations; 2) Pump start-up and shutdown frequency: Pump operating cycles (such as the number of times they start and stop per hour) directly change the outflow rate of the forebay; 3) Instantaneous sewage load: High volatility in industrial wastewater discharge (such as peak ammonia nitrogen concentration) may trigger adjustments to treatment processes, indirectly affecting water levels; 4) Long-cycle low-frequency influences: Seasonal hydrological changes: Interannual fluctuations in river baseflow and groundwater levels affect long-term drainage trends; 5) Urban water management: Water diversion plans for industrial, production, and domestic use in the water-receiving and water-using areas affect the forebay water level through periodic water replenishment and discharge.

[0078] Step b2 involves performing attention enhancement on the high- and low-frequency split water level data to obtain attention-enhanced high- and low-frequency split water level data.

[0079] Specifically, a basic attention mechanism is introduced into this high- and low-frequency split water level data. By calculating the correlation weights between the high- and low-frequency split water level data, different importance is dynamically assigned to focus on key information. That is, the input element is the high- and low-frequency split water level data. First, the query vector, key vector, and value vector need to be determined based on the high- and low-frequency split water level data. The dot product of the query vector and the key vector is calculated to obtain the unnormalized attention score. Then, scaling is used to prevent the dot product from being too large and causing gradient vanishing. Finally, the score matrix is ​​normalized and multiplied with the value vector to obtain the attention-enhanced high- and low-frequency split water level data, so as to realize the noise reduction of the unnormalized data after wavelet transform. The expression for attention enhancement of high- and low-frequency split water level data is: (14) in, The query vector is determined based on the length and amplitude of the high- and low-frequency split water level data. , The key vector is determined based on the peak, trough, and mean values ​​of the high- and low-frequency water level data. , It is a value vector, determined based on the peak, trough, and mean values ​​of the high- and low-frequency water level data. , Let be the dimension of the vector.

[0080] Furthermore, Calculate the dot product of the query vector and the key vector to obtain the unnormalized attention score, then divide by... Scaling is performed to prevent the gradient from vanishing due to excessively large dot products. The normalized attention score matrix is ​​then normalized using softmax (an activation function in deep learning) to obtain the weights. αij Later and V The value vectors are multiplied and weighted to obtain the attention-enhanced high and low frequency water level data.

[0081] Step b3: Perform moving regression averaging on the high and low frequency split water level data after attention enhancement to obtain the predicted water level sequence value of the pump station forebay.

[0082] Specifically, the high- and low-frequency water level data after attention enhancement are processed using a moving regression average. Compared to a moving average (which only takes the mean), moving regression can capture local linear trends within the window and is more robust to non-stationary sequences. The specific steps of the moving regression average processing include: 1) dividing the sliding window, i.e., at time... t Take the window N Data points, g If the index is the observation point, then the expression after the sliding window is divided is: (15) 2) Local linear regression fitting: The formula for calculating the linear model fitted within the window is as follows: (16) in, β 0 is the intercept. β 1 represents the slope, and the regression coefficients are solved using the least squares method (i.e., β 0 and β 1) The expression for the solution process is: (17) 3) Using the predicted value of the fitted regression line at the center point of the window (or the next time step) as the smoothing result, the smoothed frequency components are linearly weighted and reconstructed. The weights are then used to... wi The final predicted forebay water level data is obtained (i.e., the predicted sequence value of the pump station forebay water level); among which, the predicted value of the center point of the sequence window is... Predicted sequence values ​​of water level in the forebay of the pumping station and weight wi The calculation formulas are as follows: (18) (19) (20) Furthermore, based on the predicted sequence value of the water level in the forebay of the pumping station, and in conjunction with the SCADA system of the water conservancy project, on-site monitoring parameter measurement points are constructed, and data is read from the measurement points. Through the above water level prediction algorithm, the predicted water level curve of the forebay of the pumping station is obtained. The input and output of the data are visualized and finally displayed on the distributed drainage control page.

[0083] The method for predicting the water level of the pump station forebay provided in this embodiment determines and removes local maxima by performing abrupt change judgment on the initial water level prediction sequence values. This identifies and removes unreasonable abrupt peaks in the initial prediction sequence, eliminates the interference of abnormal data on subsequent analysis, and ensures the rationality of the water level prediction sequence. Furthermore, the method performs data smoothing processing on the water level prediction sequence values ​​after removing local maxima, which weakens the random fluctuations and small errors in the sequence, making the water level prediction sequence more closely match the stable trend of actual water level changes, and improving the accuracy and usability of the pump station forebay water level prediction.

[0084] The following specific embodiment illustrates the detailed steps of a method for predicting the water level in a pump station forebay.

[0085] Example 1: Current methods for predicting the water level in the forebay of pumping stations rely on model learning. However, the development of underlying algorithms, especially data processing algorithms, is incomplete. The water level prediction method for the forebay proposed in this application aims to address the shortcomings of existing control methods, such as over-reliance on empirical models and historical data, and high requirements for hardware and software configurations. Figure 8 As shown, the specific steps of the method for predicting the water level in the pump station forebay include: 1) Obtain the real-time operation data of the target pumping station, perform data quality control and normalization on the real-time operation data of the target pumping station, and obtain the flow data of the target pumping station.

[0086] 2) Based on the normalized processing of the flow data of the target pumping station collected in real time on site, the flow data of the target pumping station is cleaned by introducing parameters such as water loss rate and using the principle of dynamic water balance to obtain the corrected inflow flow data of the target pumping station.

[0087] 3) Based on the corrected inflow data of the target pumping station, the water level in the forebay is predicted in multiple time steps using the water level-storage capacity function to obtain the initial water level prediction sequence value.

[0088] 4) The initial water level prediction sequence values ​​are determined by the finite difference method, and local maxima are removed by spline interpolation.

[0089] 5) Wavelet transform is used to decompose the predicted sequence values ​​after removing local maxima to obtain high and low frequency split water level data. Linear weighting factors are calculated based on the high and low frequency split water level data, and the high and low frequency split water level data are weighted and summed based on the linear weighting factors. Thus, the basic attention mechanism is incorporated into the wavelet transform to obtain high and low frequency split water level data.

[0090] 6) Perform sliding regression processing on the high and low frequency split water level data after attention enhancement to output a smooth water level prediction curve (i.e., the water level prediction sequence value of the pump station forebay). The water level prediction method of the pump station forebay can accurately predict the water level change trend of the pump station forebay within a certain time step in the future.

[0091] The beneficial effects corresponding to the above embodiments are as follows: 1) It can be combined with the constant water level control system of the forebay. Based on the water level prediction method, the constant water level can generate a reasonable scheduling plan according to the predicted water level, and adjust the pump group status in time before the water level fluctuates greatly or deviates from the target water level, thus avoiding the impact of data delay and control lag.

[0092] 2) A linear weighting factor was introduced to avoid the unsatisfactory smoothing caused by the lag in data reception.

[0093] 3) To focus on key water level information, a basic attention mechanism was introduced on the basis of smoothing. By calculating the correlation weights between the input water level data, different importance was dynamically allocated to focus on key information.

[0094] 4) It avoids the inaccuracy of water level prediction caused by external interference factors or occasional data anomalies, and does not require the reliance on complex pump station network models and neural network models. It brings a highly accurate, fast calculation speed and practical engineering approach to urban water network optimization scheduling and pump station forebay water level control technology.

[0095] 5) The predicted sequence value is obtained by using the finite difference method to find the second derivative. The finite difference method is used to determine abrupt changes and remove local extrema to prevent the predicted water level value from deviating from the actual value. When the second derivative exceeds the threshold, it is further detected whether it is a local maximum. For the detected local maximum, the data can be corrected or the abrupt change point can be deleted by spline interpolation.

[0096] 6) Perform sliding regression averaging on each decomposed sequence. This method covers sliding window partitioning, local linear regression fitting, and smoothing output. It can capture local linear trends within the window and is more robust to non-stationary sequences.

[0097] 7) By combining the SCADA system with the water level prediction value and the real-time data system, a constant water level control system for the pump station forebay based on water level prediction will be developed to realize the implementation and systematic application of the method and improve the scientific nature of automatic water level control in the pump station forebay.

[0098] This embodiment also provides a device for predicting the water level of the pump station forebay. This device is used to implement the above embodiments and preferred embodiments, and details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0099] This embodiment provides a device for predicting the water level in the forebay of a pumping station, such as... Figure 9 As shown, it includes: The cleaning module 901 is used to acquire the flow data of the target pumping station, clean the flow data of the target pumping station, and obtain the corrected inflow flow data of the target pumping station.

[0100] Module 902 is used to construct the water level-reservoir capacity function.

[0101] The prediction module 903 is used to predict the water level of the forebay of the target pumping station based on the corrected inflow data of the target pumping station and the water level-storage capacity function, so as to obtain the initial water level prediction sequence value.

[0102] The optimization module 904 is used to optimize the initial water level prediction sequence value to obtain the water level prediction sequence value of the pump station forebay.

[0103] In some alternative implementations, the cleaning module 901 includes: The first calculation unit is used to obtain the reservoir capacity change and the loss flow per unit time of the target pumping station. Based on the reservoir capacity change, the loss flow per unit time of the target pumping station, and the total inflow and total outflow per unit time of the target pumping station's flow data, it calculates the water transfer loss rate for any two time periods.

[0104] The second calculation unit is used to calculate the dimensionless water loss rate based on the water loss rate and the total outflow per unit time for any two time periods.

[0105] The third calculation unit is used to obtain the real-time outflow of the upstream pumping station, the minimum sampling interval of the target pumping station, and the longest sequence interval of the target pumping station. Based on the dimensionless water transfer loss rate, the loss flow per unit time, the real-time outflow of the upstream pumping station, the minimum sampling interval of the target pumping station, and the longest sequence interval of the target pumping station, it calculates the corrected inflow data of the target pumping station.

[0106] In some alternative implementations, the prediction module 903 includes: The inverting unit is used to invert the water level-reservoir capacity function to obtain the water level prediction function.

[0107] The fourth calculation unit is used to obtain the current reservoir capacity of the target pumping station and calculate the predicted reservoir capacity based on the corrected inflow data of the target pumping station and the current reservoir capacity of the target pumping station.

[0108] The input unit is used to input the reservoir capacity prediction value into the water level prediction function to obtain the initial water level prediction sequence value.

[0109] In some alternative implementations, the optimization module 904 includes: The determination unit is used to determine abrupt changes in the initial water level prediction sequence value, identify and remove local maxima, and obtain the water level prediction sequence value after removing local maxima.

[0110] The processing unit is used to perform data smoothing on the water level prediction sequence values ​​after removing local maxima, so as to obtain the water level prediction sequence values ​​of the pump station forebay.

[0111] In some optional implementations, the determination unit includes: The derivative sub-unit is used to perform second-order differentiation on the initial water level prediction sequence value to obtain the second-order derivative value at the current time.

[0112] The comparison sub-unit is used to compare the second derivative value at the current time with a preset threshold. If the second derivative value at the current time is greater than the preset threshold, the second derivative value at the current time is compared with the second derivative value at the adjacent time.

[0113] As a sub-unit, if the second derivative value at the current time is greater than the second derivative value at the adjacent time, the predicted water level value corresponding to the second derivative value is taken as a local maximum, and the local maximum value is removed to obtain the sequence value after removing the local maximum value.

[0114] In some optional implementations, the processing unit includes: The decomposition sub-unit is used to decompose the water level prediction sequence values ​​after removing local maxima to obtain high- and low-frequency split water level data.

[0115] The enhancement subunit is used to perform attention enhancement on the high- and low-frequency split water level data to obtain attention-enhanced high- and low-frequency split water level data.

[0116] The processing subunit is used to perform moving regression averaging on the attention-enhanced high and low frequency split water level data to obtain the predicted water level sequence value of the pump station forebay.

[0117] The pump station forebay water level prediction device provided in this embodiment of the invention can execute the pump station forebay water level prediction method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as in the corresponding embodiments described above, and will not be repeated here.

[0118] Figure 10 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0119] The following is a detailed reference. Figure 10 This diagram illustrates a suitable structural schematic for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from memory 1008 into random access memory (RAM) 1003. The RAM 1003 also stores various programs and data required for the operation of the electronic device. The processor 1001, ROM 1002, and RAM 1003 are interconnected via a bus 1004. An input / output (I / O) interface 1005 is also connected to the bus 1004.

[0120] Typically, the following devices can be connected to the I / O interface 1005: input devices 1006 including, for example, a touchscreen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 1007 including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; memory devices 1008 including, for example, magnetic tape, hard disk, etc.; and communication devices 1009. Communication device 1009 allows electronic devices to exchange data via wireless or wired communication with other devices. Although Figure 10 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0121] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 1009, or installed from a memory 1008, or installed from a ROM 1002. When the computer program is executed by the processor 1001, it performs the functions defined in the method for predicting the water level of the pump station forebay according to embodiments of the present invention.

[0122] Figure 10The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0123] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that the computer, processor, microprocessor controller, or programmable hardware includes storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the method for predicting the water level of the pump station forebay shown in the above embodiments is implemented.

[0124] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0125] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for predicting the water level in the forebay of a pumping station, characterized in that, The method includes: Acquire the flow data of the target pumping station, clean the flow data of the target pumping station, and obtain the corrected inflow flow data of the target pumping station; Construct the water level-reservoir capacity function; Based on the corrected inflow data of the target pumping station, the water level in the forebay of the target pumping station is predicted using the water level-storage capacity function to obtain the initial water level prediction sequence value. The initial water level prediction sequence value is optimized to obtain the water level prediction sequence value of the pump station forebay.

2. The method according to claim 1, characterized in that, The process of cleaning the flow data of the target pumping station to obtain the corrected inflow flow data of the target pumping station includes: Obtain the reservoir capacity change and the loss flow per unit time of the target pumping station. Based on the reservoir capacity change of the target pumping station, the loss flow per unit time, and the total inflow and total outflow per unit time in the flow data of the target pumping station, calculate the water transfer loss rate for any two time periods. Calculate the dimensionless water loss rate based on the water loss rate of any two time periods and the total outflow per unit time. The real-time outflow rate of the upstream pumping station, the minimum sampling interval time of the target pumping station, and the longest sequence interval time of the target pumping station are obtained. Based on the dimensionless water loss rate, the loss flow rate per unit time, the real-time outflow rate of the upstream pumping station, the minimum sampling interval time of the target pumping station, and the longest sequence interval time of the target pumping station, the corrected inflow rate data of the target pumping station is calculated.

3. The method according to claim 1, characterized in that, The method of predicting the water level in the forebay of the target pumping station based on the corrected inflow data of the target pumping station using the water level-storage capacity function to obtain an initial water level prediction sequence value includes: Inverting the water level-reservoir capacity function yields the water level prediction function; Obtain the current reservoir capacity of the target pumping station, and calculate the predicted reservoir capacity based on the corrected inflow data of the target pumping station and the current reservoir capacity of the target pumping station. The predicted reservoir capacity is input into the water level prediction function to obtain the initial water level prediction sequence value.

4. The method according to claim 1, characterized in that, The optimization of the initial water level prediction sequence value to obtain the water level prediction sequence value of the pump station forebay includes: The initial water level prediction sequence value is subjected to abrupt change determination, local maxima are identified and removed, and the water level prediction sequence value after removing local maxima is obtained. The water level prediction sequence values ​​after removing local maxima are subjected to data smoothing processing to obtain the water level prediction sequence values ​​of the pump station forebay.

5. The method according to claim 4, characterized in that, The step of determining abrupt changes in the initial water level prediction sequence values, identifying and removing local maxima, and obtaining the sequence values ​​after removing local maxima includes: The second derivative of the initial water level prediction sequence value is obtained by taking the second derivative value at the current time. The second derivative value at the current time is compared with a preset threshold. If the second derivative value at the current time is greater than the preset threshold, the second derivative value at the current time is compared with the second derivative value at the adjacent time. If the second derivative value at the current time is greater than the second derivative value at the adjacent time, then the water level prediction value corresponding to the second derivative value is taken as the local maximum value, and the local maximum value is removed to obtain the sequence value after removing the local maximum value.

6. The method according to claim 4, characterized in that, The step of smoothing the water level prediction sequence after removing local maxima to obtain the water level prediction sequence of the pump station forebay includes: The water level prediction sequence value after removing local maxima is decomposed to obtain high and low frequency split water level data; Attention enhancement is applied to the high- and low-frequency split water level data to obtain attention-enhanced high- and low-frequency split water level data; The high and low frequency split water level data after attention enhancement are processed by moving regression averaging to obtain the predicted water level sequence value of the pump station forebay.

7. A device for predicting the water level in the forebay of a pumping station, characterized in that, The device includes: The cleaning module is used to acquire the flow data of the target pumping station, clean the flow data of the target pumping station, and obtain the corrected inflow flow data of the target pumping station. The module is used to construct the water level-reservoir capacity function; The prediction module is used to predict the water level of the forebay of the target pumping station based on the corrected inflow data of the target pumping station and using the water level-storage capacity function to obtain the initial water level prediction sequence value. An optimization module is used to optimize the initial water level prediction sequence value to obtain the water level prediction sequence value of the pump station forebay.

8. An electronic device, characterized in that, include: A memory and a processor are interconnected, the memory storing computer instructions, and the processor executing the computer instructions to perform the method for predicting the water level of the pump station forebay as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to execute the method for predicting the water level of the pump station forebay as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, Includes computer instructions for causing a computer to execute the method for predicting the water level of the pump station forebay as described in any one of claims 1 to 6.

Images