Method for real-time automatic measurement of flow under influence of river and canal ice cover

Abstract

This invention provides a real-time automatic flow measurement method for rivers and canals under the influence of ice cover, belonging to the field of river and canal flow measurement technology. This invention collects ice thickness, hydraulic parameters, and environmental parameters in real time from multiple monitoring zones at a cross-section. It combines the average value and fluctuation coefficient of ice cover thickness to determine the ice cover status, enabling dynamic adjustment of the number and layout of monitoring equipment, ensuring the comprehensiveness and accuracy of data collection. It also performs time-series prediction of ice thickness and hydraulic parameters, improving the continuity and reliability of the data. The overall solution integrates measured data and model prediction, not only considering the impact of ice cover thickness changes on flow calculation results but also effectively reducing the risks of on-site operations during ice melting periods. It achieves a balance between accuracy and safety, extending its applicability beyond the cold season and greatly enhancing the overall adaptability of the solution.

Description

Technical Field

[0001] This invention relates to the field of river and canal flow measurement technology, specifically a method for real-time automatic flow measurement under the influence of river and canal ice cover. Background Technology

[0002] Accurate measurement of river and canal flow is fundamental for water resource management, flood control, and ecological protection. Traditional methods for measuring river and canal flow under the influence of ice cover generally approximate a uniform distribution of the ice cover, obtaining the average thickness of the ice cover by taking measurements at multiple points and then averaging the results. The flow rate at the canal cross-section is then calculated based on this average thickness. While this method is suitable for harsh winter conditions, as the weather warms, the ice cover thickness varies across the cross-section due to the combined effects of water erosion and environmental changes. Ignoring thickness variations caused by changes in ice cover shape leads to reduced accuracy in flow calculations. Conversely, measuring relative to the ice cover thickness during thinning operations poses certain safety risks.

[0003] In the prior art, CN116929464A discloses a method for real-time automatic measurement of flow under the influence of ice cover in a river channel, including: measuring the vertical velocity distribution through ice cover boreholes; measuring the riverbed topography and water depth; obtaining the flow rate per unit width; dividing the flow cross-section; calculating the distribution of the flow rate per unit width along the width direction; calculating the flow rate per unit width distribution calculated by the first ice borehole; calculating the flow rate per unit width distribution calculated by the second ice borehole; calculating the average value of the flow rate per unit width at each point; and calculating the total flow rate of the cross-section. The method described in this invention utilizes two fixed ice boreholes to measure the vertical velocity distribution at the locations of the ice boreholes under the ice cover in the river channel, then uses the Einstein cross-section partitioning method to divide the influence zone between the ice cover and the riverbed, and uses the flow tube element method to calculate the distribution of the flow rate per unit width along the river width under different ice thicknesses and riverbed elevations, thereby automatically obtaining the total flow rate under the influence of the ice cover at a fixed cross-section in real time. Although this method can calculate the flow of rivers and canals under the influence of ice sheets, it relies on limited point measurements and does not consider the impact of ice sheet thickness on the calculation results, nor does it consider the safety risks of working on thinning ice sheets. It cannot achieve a balance between accuracy and safety, so it is only suitable for the cold season and has poor adaptability.

[0004] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] The purpose of this invention is to provide a method for real-time automatic measurement of flow rate under the influence of ice cover in rivers and canals, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for real-time automatic measurement of flow rate under the influence of ice cover in river channels, including the following steps: S1: Collect cross-sectional parameters of the river channel, and divide the ice cover into morphological stages based on the physical properties of the ice cover. Determine the corresponding detection equipment layout scheme according to the initial morphological stage of the ice cover, so as to collect ice cover thickness, hydraulic parameters and environmental parameters at different locations on the ice cover. S2: Based on the ice thickness at different locations on the ice sheet, generate the average thickness and fluctuation coefficient of the ice sheet, and then compare the average thickness and fluctuation coefficient with the preset judgment thresholds. Based on the comparison results, determine whether the ice sheet morphology has changed, and simultaneously adjust the layout of the detection equipment. S3: Perform time-series analysis on ice cover thickness, hydraulic parameters and environmental parameters, and generate predicted values ​​of ice cover thickness under the combined influence of hydraulic parameters and environmental parameters. Then, select measured or predicted values ​​of ice cover thickness and hydraulic parameters according to different layout schemes to calculate flow rate. S4: Based on cross-sectional parameters, ice cover thickness, and hydraulic parameters, the water depth distribution and velocity distribution of the cross-section are obtained together, and the flow rate at the cross-section is calculated by combining the two distributions.

[0007] Preferably, the cross-sectional parameters include the riverbed morphology and cross-sectional width at the cross-section; the hydraulic parameters include the water flow velocity at different locations at the cross-section; the environmental parameters include ambient temperature, ambient humidity, and ambient wind speed; and the detection equipment includes a thickness detector, a temperature and humidity sensor, a wind speed sensor, and a flow velocity sensor.

[0008] Preferably, the morphological stages of the ice sheet include the freezing period, the thawing period, and the melting period, with the freezing period being regarded as the initial morphological stage of the ice sheet; When determining the layout of the testing equipment, the transverse section of the ice sheet is first divided into several testing zones. Then, the center of each testing zone is defined as a reserve point. Finally, based on the morphological stage of the ice sheet, the setting relationship between the testing equipment and the reserve points is determined. The setting relationship satisfies the following definition: The number of testing devices is the highest when the ice sheet is in a state of severe freezing, and testing devices are set up at all preparation points. When the ice sheet is thawing, the number of testing devices is reduced, and testing devices are set up at 10% to 30% of the reserve points; When the ice sheet is melting, there are zero testing devices, and no testing equipment is set up at any of the preparation points.

[0009] Preferably, the logic for determining whether the ice sheet morphology has changed during a particular stage is as follows: The mean and standard deviation of the ice cover thickness in all detection areas were calculated and defined as the average thickness and fluctuation coefficient of the ice cover, respectively. The judgment threshold includes a first threshold and a second threshold. The average thickness and fluctuation coefficient are compared with the two respectively. Based on the comparison results, the current morphological stage of the ice sheet is identified, and it is continuously compared with the morphological stage of the previous moment to determine whether the morphological stage of the ice sheet has changed. The process of identifying the morphological stages of the ice sheet follows the logic below: When the average thickness is not less than the first threshold and the fluctuation coefficient is less than the second threshold, the ice sheet is considered to be in a period of severe freezing. When the average thickness is not lower than the first threshold and the fluctuation coefficient is not lower than the second threshold, the ice sheet is considered to be in the thawing period. If the average depth is below the first threshold, and the magnitude of the fluctuation coefficient is not considered, the ice sheet is considered to be in the melting period.

[0010] Preferably, a multi-output LSTM model is used when performing time-series analysis on ice cover thickness, hydraulic parameters, and environmental parameters. At the same time, the output layer of the LSTM model is set as a dual-neuron node, with the two nodes corresponding to the predicted values ​​of ice cover thickness and hydraulic parameters, respectively.

[0011] The preferred logic for predicting ice cover thickness and hydraulic parameters in each detection area is as follows: When the ice sheet is in a period of severe freezing, the time series data of ice sheet thickness, hydraulic parameters, and environmental parameters are used as the training set of the LSTM model, so that it learns the mapping relationship from the input vector to the ice sheet thickness. Ice sheet thickness is defined as the main variable, and water parameters and environmental parameters are defined as conditional variables. The time series data of the three are then concatenated into vectors to form the input vector of the LSTM model and substituted into the LSTM model. The mean square error is used as the loss function of the LSTM model. The predicted values ​​of ice cover thickness and hydraulic parameters are compared with the measured values. The mean square error of the two values ​​is calculated separately, and the two sets of mean square errors are weighted and summed to obtain the weighted error. Using the minimum weighted error as the optimization objective, the gradient descent algorithm is used to optimize the parameters of the LSTM model, and the predicted values ​​of ice cover thickness and hydraulic parameters are continuously output after optimization.

[0012] Preferably, when performing flow rate calculations, the ice cover thickness and hydraulic parameters used in the flow rate calculation are defined as calculated values, and the selection logic for the two types of calculated values ​​is as follows: When the ice sheet is in a state of severe freezing, all calculated values ​​for the detection areas are selected from the measured values; When the ice sheet is thawing, for the detection area equipped with detection equipment, the calculated value is the measured value; for the detection area without detection equipment, the calculated value is the predicted value. When the ice sheet is in the melting phase, the calculated values ​​for all detection areas are selected as predicted values.

[0013] Preferably, the selection of ice cover thickness and hydraulic parameters is achieved by combining binarized labels and weighted summation. The specific logic is as follows: Based on the setting of the detection equipment, a binary label is added to each detection area in sequence. When the label value is 0, it means that no detection equipment is set in the detection area, and when the label value is 1, it means that the detection area is set with detection equipment. The measured value and the predicted value are weighted and summed to obtain the calculated value, with the weights being the label value and one minus the label value, respectively.

[0014] The preferred logic for traffic calculation is as follows: Based on the riverbed morphology and ice sheet thickness, the water flow depth in each detection area is calculated, and the water depth distribution within the cross section is obtained accordingly. At the same time, the flow velocity distribution is obtained based on the water flow velocity at different locations in the cross section. The hydraulic area of ​​the cross section is obtained based on the width and water depth distribution of the detection zone. Then, it is combined with the velocity distribution at the cross section to obtain the total flow rate at the cross section in integral form.

[0015] Compared with the prior art, the beneficial effects of the present invention are: This invention collects ice thickness, hydraulic parameters, and environmental parameters in multiple monitoring zones across a cross-section in real time. It then combines the average and fluctuation coefficients of ice cover thickness to determine the ice cover condition, enabling dynamic adjustment of the number and layout of monitoring equipment and ensuring the comprehensiveness and accuracy of data collection. Furthermore, it performs time-series predictions of ice thickness and hydraulic parameters, improving data continuity and reliability. The overall solution integrates measured data and model predictions, not only considering the impact of ice cover thickness variations on flow calculations but also effectively reducing the risks of on-site operations during ice melt periods. This achieves a balance between accuracy and safety, extending its applicability beyond the frigid season and significantly enhancing the overall adaptability of the solution. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the overall method flow of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0018] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0019] Example: Please see Figure 1 The present invention provides a technical solution: A method for real-time automatic measurement of flow rate under the influence of ice cover in river channels, including the following steps: S1: Collect cross-sectional parameters of the river channel, and divide the ice cover into morphological stages based on the physical properties of the ice cover. Determine the corresponding detection equipment layout scheme according to the initial morphological stage of the ice cover, so as to collect ice cover thickness, hydraulic parameters and environmental parameters at different locations on the ice cover.

[0020] The cross-sectional parameters include the riverbed morphology and cross-sectional width at the cross-section; the hydraulic parameters include the water flow velocity at different locations at the cross-section; the environmental parameters include the ambient temperature, ambient humidity, and ambient wind speed; and the detection equipment includes a thickness gauge, temperature and humidity sensor, wind speed sensor, and flow velocity sensor.

[0021] Specifically, the riverbed morphology and cross-sectional width can be detected in advance using underwater sonar scanners, laser rangefinders, and other equipment. The thickness detector can use an ultrasonic thickness sensor, which can perform non-contact measurement and is more suitable for automated deployment. The temperature and humidity sensor is digital, with low power consumption and good stability. The wind speed sensor can use an ultrasonic anemometer, which has high accuracy, no mechanical friction, and is more suitable for long-term outdoor use. The flow velocity sensor can use a Doppler current meter (ADCP), which can quickly measure the water flow velocity at different water depths, thereby quickly obtaining the flow velocity distribution of the entire cross-section. Flow meters can also be set up at different water depths to detect the water flow velocity.

[0022] Understandably, the shape of ice sheets changes with the seasons, and correspondingly, the risks of conducting inspections on ice sheets also change. For example, during the colder winter months, ice sheets are more solid and thicker, making the installation of various inspection devices on them less risky. The impact of the equipment on the ice sheet's surface morphology is negligible and won't affect flow rate calculations. However, in early spring, as temperatures begin to rise, some areas of the ice sheet will gradually melt, but the overall thickness won't change much. At this point, some inspection equipment needs to be removed in advance to avoid the risks of operating on the ice sheet as it continues to thin. Furthermore, the impact of the inspection equipment on the ice sheet's surface morphology can no longer be ignored; too much equipment can interfere with the melting process, leading to a decrease in the accuracy of flow rate calculations. As temperatures continue to rise, the entire ice sheet will begin to melt, significantly increasing the risks of operating on it.

[0023] Therefore, different detection equipment layout schemes need to be set up for ice sheets with different morphologies. The construction logic of the detection equipment layout scheme is as follows: The ice sheet is divided into three different morphological stages: the freezing period, the thawing period, and the melting period. The freezing period is considered the initial morphological stage of the ice sheet because, in order to ensure safety, the initial operations are usually carried out during the colder winter months. The transverse section of the ice sheet is divided into several groups of testing zones, and the center of each testing zone is defined as the preparatory point. Three sets of testing equipment are arranged to correspond one-to-one with the morphological stages of the ice sheet, and each scheme satisfies the following: The number of testing devices is the highest when the ice sheet is in a state of severe freezing, and testing devices are set up at all preparation points. When the ice sheet is thawing, the number of testing devices is reduced, and testing devices are set up at 10% to 30% of the reserve points; When the ice sheet is melting, there are zero testing devices, and no testing equipment is set up at any of the preparation points.

[0024] When dividing the testing areas, the more testing areas there are and the smaller the width of each testing area, the more accurate the measured data will be. However, the more testing equipment is required. Under normal circumstances, the number should not be less than 10 sets to ensure the accuracy of the data, but the actual number can be adjusted according to the project budget and expert experience.

[0025] In this step, by constructing three sets of detection equipment layout schemes, it is possible to dynamically adapt to different physical morphological stages of the ice sheet, avoiding a one-size-fits-all fixed detection strategy, effectively balancing detection accuracy, equipment cost and on-site safety risks, and also providing a more accurate data foundation for subsequent flow calculation.

[0026] S2: Based on the ice thickness at different locations on the ice sheet, generate the average thickness and fluctuation coefficient of the ice sheet. Then compare the average thickness and fluctuation coefficient with preset judgment thresholds. Based on the comparison results, determine whether the ice sheet morphology has changed and simultaneously adjust the layout of the detection equipment.

[0027] The logic for determining whether the ice sheet morphology has changed during a particular stage is as follows: The mean and standard deviation of the ice cover thickness in all tested areas are calculated and defined as the average thickness and fluctuation coefficient of the ice cover, respectively. The calculation formula is as follows: In the formula , These represent the average thickness and fluctuation coefficient of the ice sheet, respectively. Indicates the first Ice thickness in each testing area, subscript Indicates the index of the detection area. This indicates the total number of areas tested. The fluctuation coefficient reflects the overall mechanical stability of the ice sheet; excessive fluctuation indicates that the ice sheet is beginning to show signs of localized weakness, bulging, depression, or premature melting.

[0028] The judgment thresholds include a first threshold and a second threshold. The average thickness and fluctuation coefficient are compared with the two thresholds respectively. Based on the comparison results, the current morphological stage of the ice sheet is identified, and it is continuously compared with the morphological stage of the previous moment to determine whether the morphological stage of the ice sheet has changed.

[0029] when: Average thickness not less than the first threshold And when the volatility coefficient is lower than the second threshold ,Right now: At this point, the overall thickness of the ice sheet met the standard, and the surface thickness of each area was relatively uniform, so the ice sheet was considered to be in a period of severe freezing. When the average thickness is not lower than the first threshold and the fluctuation coefficient is lower than the second threshold, that is: At this point, the overall thickness of the ice sheet meets the standard, but the surface thickness of different areas begins to become uneven, which means that some areas have begun to melt. Therefore, the ice sheet is considered to be in the thawing period, and the number of testing devices needs to be reduced. The average depth is below the first threshold, i.e. At this point, the overall thickness of the ice sheet has decreased, and there would be certain risks in continuing to work on it. Therefore, regardless of the magnitude of the fluctuation coefficient, it is assumed that the ice sheet is in the melting period and all testing equipment needs to be removed.

[0030] The first threshold can be set by detecting the maximum average thickness of the ice sheet during the freezing period, typically set to 80% to 90% of the maximum average thickness. The maximum average thickness can be obtained by reviewing historical data from previous years or by analyzing data collected during the freezing period. Simply put, a relatively small first threshold is preset during the harsh winter months, and the ice sheet thickness in different regions is monitored in real time. Once the ice sheet thickness stabilizes, it can be considered to have reached the maximum average thickness, at which point the first threshold can be adjusted accordingly. The second threshold is determined based on expert experience and is typically set between 0.15 and 0.2.

[0031] In this step, by using the average thickness and fluctuation coefficient of the ice sheet as indicators for judging the state of the ice sheet, taking into account the overall thickness level and thickness uniformity of the ice sheet, it is possible to objectively reflect its physical change characteristics, thereby realizing the switching of the detection equipment layout scheme from high-density layout to low-density layout, ensuring reasonable equipment investment, accurate measurement and reducing unnecessary waste of resources.

[0032] S3: Perform time-series analysis on ice cover thickness, hydraulic parameters, and environmental parameters, and generate predicted values ​​of ice cover thickness under the combined influence of hydraulic and environmental parameters. Then, select measured or predicted values ​​of ice cover thickness and hydraulic parameters according to different layout schemes to calculate flow.

[0033] The LSTM model was used to perform time series analysis on ice cover thickness, hydraulic parameters and environmental parameters, and the ice cover thickness and hydraulic parameters of each detection area were predicted.

[0034] It is understandable that the change in ice sheet thickness is a dynamic time-series process influenced by various factors, such as ambient temperature, humidity, wind speed, and water flow velocity. These influencing factors have a complex, nonlinear, and multivariate coupled relationship with ice sheet thickness. Therefore, when making time-series predictions of ice sheet thickness, hydraulic parameters, and environmental parameters, the latter two are considered as prerequisites for ice sheet thickness changes. The ice sheet thickness output by the model is essentially a prediction under a specific combination of hydraulic and environmental parameters.

[0035] For the reasons mentioned above, the LSTM (Long Short-Term Memory) model was used for time series analysis of the three. This model, by introducing gating mechanisms such as forget gate, input gate, and output gate, can better handle nonlinear multivariate inputs compared to traditional time series analysis models (that is, ice sheet thickness can be used as the main variable, and environmental parameters and hydraulic parameters as conditional variables to form a multidimensional input sequence). It can also learn complex dynamic mappings and automatically fit the nonlinear mapping relationship between historical ice sheet thickness and environmental parameters on future ice sheet thickness, thereby reflecting the influence of the external environment on the evolution of ice sheet morphology.

[0036] The logic for using an LSTM model for prediction is as follows: When the ice sheet is in a period of severe freezing, the time series data of ice sheet thickness, hydraulic parameters, and environmental parameters are used as the training set of the LSTM model, so that it learns the mapping relationship from the input vector to the ice sheet thickness. Ice sheet thickness is defined as the main variable, and water parameters and environmental parameters are defined as conditional variables. The time series data of the three are then concatenated into vectors to form the input vector of the LSTM model and substituted into the LSTM model. The output layer of the LSTM model is set to a two-neuron node, with the two nodes corresponding to the predicted values ​​of ice sheet thickness and hydraulic parameters, respectively. The predicted values ​​of ice cover thickness and hydraulic parameters are compared with the measured values. The mean squared error is used as the loss function of the LSTM model. At the same time, the gradient descent algorithm is used to optimize the parameters of the LSTM model. After optimization, the predicted values ​​of ice cover thickness and hydraulic parameters are continuously output.

[0037] Since all parameters of the ice sheet are obtained through actual measurements during the freezing period, they can be used to train the model. For example, during the freezing period, all time series data up to the current moment can be used as the training set for the LSTM model, and predicted values ​​for ice sheet thickness and hydraulic parameters at future moments can be generated. These predicted values ​​can then be compared with subsequent measured values ​​to optimize the LSTM model.

[0038] Specifically, for any detection region, the input vector at a certain time... It can be represented as: In the formula express At that moment, the ice thickness of the testing area was... express At any given time, the water conservancy parameters and environmental parameters of the monitoring area are both in vector form. Indicates time (equivalent to an index of time-series data). ,in For the current moment, subscript Index representing time, This represents the total number of time-series data.

[0039] The time-series data of the input vector is used as input to the LSTM model to learn the mapping relationship between the input vector and the ice sheet thickness. The expression is as follows: In the formula The equation representing the mapping from the input vector to the ice sheet thickness is... This represents the time-series data of the input vector. In the future Predicted ice sheet thickness at the current time. This indicates the time step for the forecast. Since water and environmental parameters are only affected by the seasonality of the environment, it is not necessary to construct a mapping relationship when forecasting water parameters; forecasts can be made directly through time series analysis.

[0040] Upon arrival At that time, the predicted ice sheet thickness is compared with the measured value, and the mean square error of the LSTM model for the ice sheet thickness is calculated: In the formula Indicates in The mean square error of the LSTM model for ice sheet thickness at time t is given. This represents the measured value of the ice sheet thickness at that moment.

[0041] Similarly, by comparing the predicted and measured values ​​of the hydraulic parameters, the mean square error of the LSTM model for the hydraulic parameters can be calculated. The weighted sum of the two types of mean square errors (usually set to equal weight) yields the final weighted error.

[0042] By minimizing the weighted error of the model, the parameters of the LSTM model can be optimized using the gradient descent algorithm. The optimization is then carried out iteratively throughout the entire freezing period. When the weighted error is no higher than 10% (this value can also be modified according to expert experience, and is usually set between 5% and 10%), the optimization is considered complete.

[0043] S4: Based on cross-sectional parameters, ice cover thickness, and hydraulic parameters, the water depth distribution and velocity distribution of the cross-section are obtained together, and the flow rate at the cross-section is calculated by combining the two distributions.

[0044] When performing flow rate calculations, the ice cover thickness and hydraulic parameters used in the calculations are defined as calculated values. The selection logic for the two types of calculated values ​​is as follows: When the ice sheet is in a state of severe freezing, all calculated values ​​for the detection areas are selected from the measured values; When the ice sheet is thawing, for the detection area equipped with detection equipment, the calculated value is the measured value; for the detection area without detection equipment, the calculated value is the predicted value. When the ice sheet is in the melting phase, the calculated values ​​for all detection areas are selected as predicted values.

[0045] In simple terms, during the freezing period, water depth distribution and hydraulic parameters are calculated using purely measured values; during the thawing period, a mixed model of measured and predicted values ​​is used; and during the melting period, purely predicted values ​​are used.

[0046] The selection of data for ice sheet thickness and hydraulic parameters is achieved by combining binarized labels and weighted summation. The specific logic is as follows: Based on the setting of the detection equipment, a binary label is added to each detection area in sequence. When the label value is 0, it means that no detection equipment is set in the detection area, and when the label value is 1, it means that the detection area is set with detection equipment. The measured value and the predicted value are weighted and summed to obtain the calculated value, with the weights being the label value and one minus the label value, respectively.

[0047] Taking ice sheet thickness as an example, the expression for its calculation value is as follows: In the formula Indicates in At this moment Calculated values ​​of ice cover thickness at each detection area. Indicates the first The calculation formulas for the binary labels of each detection area and the hydraulic parameters are similar and will not be repeated here.

[0048] The logic for traffic calculation is as follows: Based on the riverbed morphology and ice cover thickness, the water flow depth in each detection area is calculated, i.e.: In the formula Indicates in At this moment The water depth at each testing area, that is, the distance from below the ice sheet to the bottom of the riverbed. Indicates the first The total depth at each testing area, that is, the distance from the top of the ice sheet to the bottom of the riverbed.

[0049] Then, the water depth distribution within the cross section is obtained, and the flow velocity distribution is obtained based on the water flow velocity at different locations within the cross section. Based on the width of the detection area The cross-sectional hydraulic area is obtained from the water depth distribution. , Then, combining this with the velocity distribution at the cross-section, the total flow rate at the cross-section can be obtained in integral form, which can be expressed as: In the formula express The total flow rate at that cross-section at that moment. express At this moment The depth of each detection zone is The water flow velocity at that location.

[0050] In this step, by flexibly switching between measured and predicted values ​​according to the ice cover morphology stage (severe freezing period, thawing period, melting period), the maximum availability of the calculated data is ensured. Under the premise of ensuring calculation accuracy, the detection equipment can be removed in advance to avoid potential operational risks after the ice cover is thinned. Moreover, the dynamic ice cover thickness can better reflect the impact of water flow erosion and environmental changes on the ice cover, thereby making the calculated water flow depth more accurate and improving the overall calculation precision of the method.

[0051] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0052] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0053] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0054] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for real-time automatic measurement of flow rate under the influence of ice cover in river channels, characterized in that, The specific steps include: S1: Collect cross-sectional parameters of the river channel, and divide the ice cover into morphological stages based on the physical properties of the ice cover. Determine the corresponding detection equipment layout scheme according to the initial morphological stage of the ice cover, so as to collect ice cover thickness, hydraulic parameters and environmental parameters at different locations on the ice cover. S2: Based on the ice thickness at different locations on the ice sheet, generate the average thickness and fluctuation coefficient of the ice sheet, and then compare the average thickness and fluctuation coefficient with the preset judgment thresholds. Based on the comparison results, determine whether the ice sheet morphology has changed, and simultaneously adjust the layout of the detection equipment. S3: Perform time-series analysis on ice cover thickness, hydraulic parameters and environmental parameters, and generate predicted values ​​of ice cover thickness under the combined influence of hydraulic parameters and environmental parameters. Then, select measured or predicted values ​​of ice cover thickness and hydraulic parameters according to different layout schemes to calculate flow rate. S4: Based on cross-sectional parameters, ice cover thickness, and hydraulic parameters, the water depth distribution and velocity distribution of the cross-section are obtained together, and the flow rate at the cross-section is calculated by combining the two distributions.

2. The method for real-time automatic measurement of flow rate under the influence of ice cover in river channels according to claim 1, characterized in that: The cross-sectional parameters include the riverbed morphology and cross-sectional width at the cross-section; the hydraulic parameters include the water flow velocity at different locations at the cross-section; the environmental parameters include ambient temperature, ambient humidity, and ambient wind speed; and the detection equipment includes a thickness detector, a temperature and humidity sensor, a wind speed sensor, and a flow velocity sensor.

3. The method for real-time automatic measurement of flow rate under the influence of ice cover in river channels according to claim 2, characterized in that: The morphological stages of ice sheets include the freezing period, the thawing period, and the melting period, with the freezing period being considered the initial morphological stage of ice sheets. When determining the layout of the testing equipment, the transverse section of the ice sheet is first divided into several testing zones. Then, the center of each testing zone is defined as a reserve point. Finally, based on the morphological stage of the ice sheet, the setting relationship between the testing equipment and the reserve points is determined. The setting relationship satisfies the following definition: The number of testing devices is the highest when the ice sheet is in a state of severe freezing, and testing devices are set up at all preparation points. When the ice sheet is thawing, the number of testing devices is reduced, and testing devices are set up at 10% to 30% of the reserve points; When the ice sheet is melting, there are zero testing devices, and no testing equipment is set up at any of the preparation points.

4. The method for real-time automatic measurement of flow rate under the influence of ice cover in river channels according to claim 3, characterized in that: The logic for determining whether the ice sheet morphology has changed during a particular stage is as follows: The mean and standard deviation of the ice cover thickness in all detection areas were calculated and defined as the average thickness and fluctuation coefficient of the ice cover, respectively. The judgment threshold includes a first threshold and a second threshold. The average thickness and fluctuation coefficient are compared with the two respectively. Based on the comparison results, the current morphological stage of the ice sheet is identified, and it is continuously compared with the morphological stage of the previous moment to determine whether the morphological stage of the ice sheet has changed. The process of identifying the morphological stages of the ice sheet follows the logic below: When the average thickness is not lower than the first threshold and the fluctuation coefficient is lower than the second threshold, the ice sheet is considered to be in a period of severe freezing. When the average thickness is not lower than the first threshold and the fluctuation coefficient is not lower than the second threshold, the ice sheet is considered to be in the thawing period. If the average depth is below the first threshold, and the magnitude of the fluctuation coefficient is not considered, the ice sheet is considered to be in the melting period.

5. The method for real-time automatic measurement of flow rate under the influence of ice cover in river channels according to claim 4, characterized in that: When performing time-series analysis on ice sheet thickness, hydraulic parameters, and environmental parameters, a multi-output LSTM model is used. The output layer of the LSTM model is set as a two-neuron node, with the two nodes corresponding to the predicted values ​​of ice sheet thickness and hydraulic parameters, respectively.

6. The method for real-time automatic measurement of flow rate under the influence of ice cover in river channels according to claim 5, characterized in that: The logic for predicting ice cover thickness and hydraulic parameters in each monitoring area is as follows: When the ice sheet is in a period of severe freezing, the time series data of ice sheet thickness, hydraulic parameters, and environmental parameters are used as the training set of the LSTM model, so that it learns the mapping relationship from the input vector to the ice sheet thickness. Ice sheet thickness is defined as the main variable, and water parameters and environmental parameters are defined as conditional variables. The time series data of the three are then concatenated into vectors to form the input vector of the LSTM model and substituted into the LSTM model. The mean square error is used as the loss function of the LSTM model. The predicted values ​​of ice cover thickness and hydraulic parameters are compared with the measured values. The mean square error of the two values ​​is calculated separately, and the two sets of mean square errors are weighted and summed to obtain the weighted error. Using the minimum weighted error as the optimization objective, the gradient descent algorithm is used to optimize the parameters of the LSTM model, and the predicted values ​​of ice cover thickness and hydraulic parameters are continuously output after optimization.

7. The method for real-time automatic measurement of flow rate under the influence of ice cover in river channels according to claim 4, characterized in that: When performing flow rate calculations, the ice cover thickness and hydraulic parameters used in the calculations are defined as calculated values. The selection logic for the two types of calculated values ​​is as follows: When the ice sheet is in a state of severe freezing, all calculated values ​​for the detection areas are selected from the measured values; When the ice sheet is thawing, for the detection area equipped with detection equipment, the calculated value is the measured value; for the detection area without detection equipment, the calculated value is the predicted value. When the ice sheet is in the melting phase, the calculated values ​​for all detection areas are selected as predicted values.

8. The method for real-time automatic measurement of flow rate under the influence of ice cover in river channels according to claim 7, characterized in that: The selection of data for ice sheet thickness and hydraulic parameters is achieved by combining binarized labels and weighted summation. The specific logic is as follows: Based on the setting of the detection equipment, a binary label is added to each detection area in sequence. When the label value is 0, it means that no detection equipment is set in the detection area, and when the label value is 1, it means that the detection area is set with detection equipment. The measured value and the predicted value are weighted and summed to obtain the calculated value, with the weights being the label value and one minus the label value, respectively.

9. The method for real-time automatic measurement of flow rate under the influence of ice cover in river channels according to claim 8, characterized in that: The logic for traffic calculation is as follows: Based on the riverbed morphology and ice sheet thickness, the water flow depth in each detection area is calculated, and the water depth distribution within the cross section is obtained accordingly. At the same time, the flow velocity distribution is obtained based on the water flow velocity at different locations in the cross section. The hydraulic area of ​​the cross section is obtained based on the width and water depth distribution of the detection zone. Then, it is combined with the velocity distribution at the cross section to obtain the total flow rate at the cross section in integral form.

Images