Real-time monitoring and early warning method for sediment curtain interception efficiency

Abstract

The application discloses a kind of real-time monitoring and early warning method of silt curtain interception efficiency, belong to water conservancy engineering silt monitoring technical field.Currently silt curtain monitoring is difficult to measure two kinds of silt flux synchronously, cannot determine interception effect in real time, early warning response lags behind, cannot promptly investigate operation hidden danger.The application is arranged with multiple types of sensing devices on the upstream and downstream sections of silt curtain, continuously collects water flow silt and water pressure data, sequentially calculates the upstream and downstream bed load, suspended load and total silt flux with the aid of preset calibration relationship and water depth-area curve, and then calculates the interception efficiency and water pressure difference rate.According to the efficiency decline speed, silt flux and water pressure difference rate, multiple levels of determination conditions are drawn, and after meeting the standard, the corresponding grade early warning is triggered, which is simultaneously pushed to the remote monitoring terminal and the on-site sound and light alarm is started.This scheme can accurately control the real-time interception working condition of silt curtain, and can timely warn abnormal risks, and is suitable for daily monitoring and control of river silt interception facilities.

Description

Technical Field

[0001] This invention relates to the field of sediment monitoring technology in water conservancy projects. More specifically, this invention relates to a method for real-time monitoring and early warning of sediment curtain interception efficiency. Background Technology

[0002] In river sediment control projects, sediment curtains are commonly used sediment interception facilities. During project operation, it is necessary to monitor their interception effectiveness in real time to ensure stable operation. Currently, the industry's monitoring methods for sediment curtain interception effectiveness have several practical technical shortcomings.

[0003] Current monitoring methods typically only measure suspended sediment parameters, generally failing to effectively monitor bedload sediment in the riverbed. River sediment transport consists of both suspended and bedload sediment; statistically analyzing only one type of sediment cannot yield the total sediment transport volume across a complete cross-section, resulting in inherent biases in basic calculation data and failing to accurately reflect the actual sediment load borne by the sediment curtain.

[0004] Conventional monitoring makes it difficult to achieve simultaneous sampling at upstream and downstream sections. There is a time lag in the transport of water and sediment along the way. The time-disordered monitoring data will further reduce the accuracy of sediment flux calculation and make it impossible to accurately convert the actual sediment curtain interception efficiency.

[0005] Existing technologies lack a multi-parameter collaborative evaluation system and fail to comprehensively determine operating conditions by combining changes in cross-sectional water pressure and sediment flux. Changes in water pressure can directly reflect the flow and interception status of sediment curtains, but such effective parameters have not been effectively utilized.

[0006] Meanwhile, the existing monitoring system lacks a standardized hierarchical early warning logic, making it difficult to identify changes in operating conditions such as a continuous decline in interception efficiency and an abnormal increase in sediment inflow. The sediment flow and sediment concentration in the river channel are constantly fluctuating dynamically, and using fixed judgment standards is difficult to adapt to the changing field conditions, making it impossible to detect in a timely manner conditions such as the degradation of sediment curtain performance and abnormal operation.

[0007] Due to the multi-form movement characteristics of sediment and the time lag of water flow, previous attempts to optimize monitoring methods have made it difficult to unify the time sequence of upstream and downstream data, to fully encompass the sediment transport of the two types of sediment, to accurately quantify the interception effect, and to provide timely warnings of abnormal operating conditions, thus failing to meet the actual needs of routine monitoring and control of the project. Summary of the Invention

[0008] Another objective of this invention is to provide a real-time monitoring and early warning method for sediment curtain interception efficiency. Existing sediment curtain monitoring systems cannot simultaneously collect data on upstream and downstream cross-section water flow, suspended sediment, bedload, and water pressure, cannot fully calculate the total sediment flux, and are difficult to accurately calculate interception efficiency and water pressure difference rate. They lack standardized tiered early warning criteria, cannot identify abnormal operating conditions based on efficiency changes, sediment flux, and water pressure status, and cannot simultaneously issue early warnings and trigger on-site alarms, making it difficult to monitor the real-time operation status of sediment curtain interception.

[0009] Conventional methods are insufficient for accurately calculating bedload flux. They fail to incorporate vibration signal characteristics and water flow shear force analysis, and their fixed calibration relationships cannot adapt to dynamic changes in suspended sediment concentration. Calculating bedload transport rate using a single reference index results in significant errors and fails to reflect actual on-site sediment impact conditions, leading to inaccurate bedload flux calculations.

[0010] There is a time difference in sediment propagation between upstream and downstream areas, resulting in temporal discrepancies in sediment flux data. Directly comparing and calculating interception efficiency will produce significant errors. A single time-delay calculation method cannot take into account the lag characteristics of water flow, suspended sediment, and bedload, and cannot correct for calculation errors caused by temporal misalignment, thus compromising the accuracy of interception efficiency measurements.

[0011] Sensors operating in the field for extended periods are prone to malfunctions such as blockages and structural damage. These malfunctions result in distorted data that cannot be autonomously identified. Without corresponding methods to compensate for failed data, the continued use of abnormal data in calculations significantly reduces monitoring accuracy and fails to promptly alert staff for equipment maintenance, leading to low reliability of the monitoring system.

[0012] Applying empirical formulas directly to calculate bedload flux after a malfunction of the impact plate lacks a basis for judgment and fails to incorporate the constraints of shear force and critical sediment initiation conditions. Calculating bedload flux even when flow intensity is insufficient will produce invalid data that cannot objectively reflect the actual bedload transport situation at the cross-section.

[0013] To achieve these objectives and other advantages according to the present invention, a method for real-time monitoring and early warning of sediment curtain interception efficiency is provided, comprising the following steps: A first current meter, a first suspended sediment concentration meter, a first bedload flux measurement module, and a first pressure sensor are installed at a first cross-section upstream of the sediment curtain. A second current meter, a second suspended sediment concentration meter, a second bedload flux measurement module, and a second pressure sensor are installed at a second cross-section downstream of the sediment curtain. A first water depth-area relationship curve for the first cross-section and a second water depth-area relationship curve for the second cross-section are pre-acquired. The first bedload flux measurement module includes a first impact plate fixedly installed on the riverbed surface of the first cross-section and a first accelerometer fixed to the back of the first impact plate. The second bedload flux measurement module includes a second impact plate fixedly installed on the riverbed surface of the second cross-section and a second accelerometer fixed to the back of the second impact plate. The system continuously collects the first cross-sectional average flow velocity output by the first flow meter, the first cross-sectional average suspended sediment concentration output by the first suspended sediment concentration meter, the first vibration signal output by the first accelerometer, and the first water pressure value output by the first pressure sensor, as well as the second cross-sectional average flow velocity output by the second flow meter, the second cross-sectional average suspended sediment concentration output by the second suspended sediment concentration meter, the second vibration signal output by the second accelerometer, and the second water pressure value output by the second pressure sensor. Based on the first vibration signal, the total bedload flux of the first cross section is calculated through a pre-established first calibration relationship, wherein the first calibration relationship is the correspondence between the characteristic values ​​of the vibration signal and the bedload transport rate; based on the second vibration signal, the total bedload flux of the second cross section is calculated through a pre-established second calibration relationship, wherein the second calibration relationship is the correspondence between the characteristic values ​​of the vibration signal and the bedload transport rate; The water depth of the first cross-section is calculated based on the first water pressure value, and the real-time area of ​​the first cross-section is calculated according to the first water depth-area relationship curve; the water depth of the second cross-section is calculated based on the second water pressure value, and the real-time area of ​​the second cross-section is calculated according to the second water depth-area relationship curve. The first suspended sediment flux is obtained by multiplying the real-time first cross-sectional area, the average flow velocity of the first cross-section, and the average suspended sediment concentration of the first cross-section. This first suspended sediment flux is then added to the total bedload flux of the first cross-section to obtain the first total sediment flux. Similarly, the second suspended sediment flux is obtained by multiplying the real-time second cross-sectional area, the average flow velocity of the second cross-section, and the average suspended sediment concentration of the second cross-section. This second suspended sediment flux is then added to the total bedload flux of the second cross-section to obtain the second total sediment flux. The interception efficiency is calculated by dividing the difference between the first and second total sediment flux by the first total sediment flux and then multiplying by 100%. The water pressure difference rate is obtained by dividing the difference between the first and second water pressure values ​​by the first water pressure value and then multiplying by 100%. When the interception efficiency decreases at a rate exceeding 2% per second for 5-60 consecutive seconds, a first-level warning signal is triggered; when the interception efficiency falls below the first warning threshold in the range of 40% to 60%, a second-level warning signal is triggered; when the second total sediment flux exceeds the second warning threshold in the range of 75% to 90% of the first total sediment flux, or when the water pressure difference rate is less than the third warning threshold in the range of 5% to 10%, a third-level warning signal is triggered. The first-level, second-level, and third-level early warning signals are sent to the remote monitoring terminal, and the on-site audible and visual alarms are activated simultaneously.

[0014] Preferably, the step of calculating the total bedload flux of the first cross-section based on the first vibration signal includes: Wavelet packet decomposition is performed on the first vibration signal output by the first accelerometer to extract the energy component corresponding to the impact characteristic frequency band of the bedload particles and obtain the first impact energy characteristic value; at the same time, the first water pressure value output by the first pressure sensor and the first cross-sectional average flow velocity output by the first flow meter are used to calculate the water flow shear force at the first cross-section. The pre-established first calibration relationship is: the piecewise linear relationship between the first impact energy characteristic value and the dimensionless ratio of the water flow shear force at the first cross-section and the bedload transport rate at the first cross-section; the segmentation point of this piecewise linear relationship is automatically adjusted according to the rate of change of the average suspended sediment concentration at the first cross-section output by the first suspended sediment concentration meter. Calculate the total bedload flux of the first cross section based on the first calibration relationship; The steps for calculating the total bedload flux of the second cross section based on the second vibration signal include: Wavelet packet decomposition is performed on the second vibration signal output by the second accelerometer to extract the energy component corresponding to the impact characteristic frequency band of the bedload particles, and the second impact energy characteristic value is obtained. At the same time, based on the second water pressure value output by the second pressure sensor and the average flow velocity of the second cross section output by the second flow meter, the shear force of the water flow in the second cross section is calculated. The pre-established second calibration relationship is: the piecewise linear relationship between the second impact energy characteristic value and the dimensionless ratio of the water flow shear force at the second cross-section and the bedload transport rate at the second cross-section; the segmentation point of this piecewise linear relationship is automatically adjusted according to the rate of change of the average suspended sediment concentration at the second cross-section output by the second suspended sediment concentration meter. The total bedload flux of the second section is calculated based on the second calibration relationship.

[0015] Preferably, before calculating the interception efficiency, a time delay compensation is applied to the second total sediment flux based on the water flow propagation time between the upstream first cross-section and the downstream second cross-section. Specifically, this includes: The Froude number of the first cross-section is calculated based on the average flow velocity and water depth of the first cross-section, and the theoretical water flow propagation delay is calculated based on the river length from the first cross-section to the second cross-section and the average wave velocity. At the same time, the peak delay of the suspended sediment concentration time series of the first cross-section and the second cross-section, as well as the peak delay of the vibration signal energy series of the first accelerometer and the second accelerometer, are calculated using cross-correlation analysis. The vibration signal energy series is used to characterize the hysteresis characteristics of bedload motion. The theoretical water flow propagation delay, the cross-correlation delay of suspended sediment concentration, and the cross-correlation delay of vibration signal energy are weighted and fused to obtain the final total sediment flux compensation delay. The time series of the second total sediment flux is shifted forward by the final compensation delay, and then the shifted second total sediment flux is compared with the first total sediment flux to calculate the interception efficiency.

[0016] Preferably, it also includes a sensor data validity self-check and repair step: The average suspended mass concentration of the first section output by the first suspended mass concentration meter and the average suspended mass concentration of the second section output by the second suspended mass concentration meter are continuously compared. When the difference between the two exceeds the preset range and the duration exceeds the first set time, and the absolute value of the average suspended mass concentration of the first section does not change abruptly, the ratio of suspended mass flux to bed mass flux of the first section and the ratio of suspended mass flux to bed mass flux of the second section are calculated respectively. If the ratio of a certain section deviates from the historical average of that section by more than 3 times the standard deviation, it is determined that the suspended sediment concentration meter of that section is blocked. At this time, the water pressure difference rate of that section and the upstream sediment flux are used to estimate the suspended sediment concentration replacement value through a pre-trained neural network model and trigger the sensor maintenance reminder. Meanwhile, by comparing the vibration signal characteristics of the first accelerometer and the second accelerometer, if the characteristic energy of one of the accelerometers is continuously lower than the preset lower energy threshold and the average flow velocity of the corresponding section is higher than the bedload initiation velocity, it is determined that the impact plate of the section is buried or damaged. In this case, the suspended sediment concentration of the same section and the empirical bedload transport formula based on the shear force of the water flow are used to estimate the bedload flux of the section.

[0017] Preferably, before using the suspended sediment concentration and empirical bedload transport formula based on water flow shear force to estimate the bedload flux of the cross section, the real-time water flow shear force of the cross section is calculated based on the water pressure and average flow velocity of the cross section. Then, it is determined whether the real-time water flow shear force of the cross section is greater than the preset bedload initiation critical shear force. Only when the real-time water flow shear force of the cross section is greater than the bedload initiation critical shear force is the empirical bedload transport formula allowed for estimation. If the real-time water flow shear force of the cross section is less than or equal to the bedload initiation critical shear force, the bedload flux of the cross section is set to zero, and an insufficient water flow intensity reminder message is sent to the remote monitoring terminal.

[0018] Preferably, the first, second, and third warning thresholds are dynamically adjusted based on upstream sediment inflow conditions. The sliding average value and coefficient of variation of the total sediment flux at the first cross section are calculated in real time. When the coefficient of variation is higher than the preset value, the second warning threshold is raised. When the fluctuation range of the water pressure differential exceeds the upper limit of the historical fluctuation range, the third warning threshold is lowered; and the threshold for the rate of decrease in interception efficiency corresponding to the first warning signal is adjusted according to the noise level of the interception efficiency, which is characterized by the root mean square of the high-frequency component of the interception efficiency.

[0019] Preferably, after sending the first-level, second-level, and third-level warning signals to the remote monitoring terminal and simultaneously activating the on-site audible and visual alarms, interlocking and priority management of the warning signals are also performed: When both the first-level and second-level warning signals are triggered simultaneously, only the alarm level corresponding to the second-level warning signal is output, along with the feature code of the first-level warning signal. When a Level 3 warning signal exists simultaneously with a Level 1 or Level 2 warning signal, the Level 3 warning signal will be elevated to the highest priority alarm, and the on-site audible and visual alarms will be forcibly triggered to distinguish them by different flashing colors. An emergency shutdown command will be sent to the remote monitoring terminal when any of the following combinations of conditions are met, and it will be recommended to shut down the upstream water supply or activate the emergency sediment flushing facilities: The sediment curtain is damaged or fails when the interception efficiency is lower than the lower limit of the second warning threshold and the water pressure difference rate is lower than the preset low water pressure difference emergency threshold. The lower limit of the second warning threshold is preset in the range of 40% to 60%, and the low water pressure difference emergency threshold is preset in the range of 5% to 15%. Severe blockage by sediment curtain: The interception efficiency is higher than the preset high interception efficiency threshold, and the water pressure difference rate is higher than the preset high water pressure difference emergency threshold; wherein, the high interception efficiency threshold is preset within the range of 70% to 90%, and the high water pressure difference emergency threshold is preset within the range of 20% to 30%. Furthermore, when the conditions of mud curtain damage or failure and severe mud curtain blockage are triggered simultaneously, it is determined that the sensor data is abnormal or the cross-sectional shape has changed drastically. The automatic shutdown command is suspended, the data self-check level is upgraded, and a prompt message requiring manual review is sent to the remote monitoring terminal.

[0020] Preferably, before calculating the interception efficiency by dividing the difference between the first total sediment flux and the second total sediment flux by the first total sediment flux and then multiplying by 100%, a low flux determination and processing step is performed first: The first total sediment flux is compared with the preset minimum effective flux threshold, which is dynamically determined based on the first cross-sectional area, the average flow velocity of the first cross-section, and the reference sediment transport rate corresponding to the critical velocity for bedload initiation. If the first total sediment flux is less than or equal to the minimum effective flux threshold, the interception efficiency will not be calculated, and no level of warning signal (first, second, or third level) will be triggered. At the same time, a status message indicating "insufficient sediment transport from the inflow, interception efficiency assessment is suspended" will be sent to the remote monitoring terminal. If the first total sediment flux is greater than the minimum effective flux threshold for three consecutive judgment periods, then the normal interception efficiency calculation and early warning logic will be restored. During the pause, the on-site audible and visual alarm only indicates that the system is in standby monitoring mode by flashing green light, without emitting any sound alarm.

[0021] The present invention has at least the following beneficial effects: This invention comprehensively collects monitoring data from multiple upstream and downstream sensors, and combines curve relationships with flux calculation formulas to fully calculate the total amount of two types of sediment, and stably calculate the interception efficiency and water pressure difference rate. Based on multi-layered judgment criteria and graded triggering of early warnings, it can promptly detect abnormalities in the operation of the sediment curtain. Early warning signals are simultaneously uploaded to the terminal and audible and visual alarms are activated, achieving real-time, 24 / 7 monitoring of the sediment curtain's interception effect and providing fundamental data support for engineering management.

[0022] This invention employs wavelet packet decomposition to extract sediment impact vibration characteristics, combines this with water flow shear force to construct a dimensionless calibration relationship, and automatically adjusts segment nodes based on changes in suspended sediment concentration. This allows for the calculation of bedload flux that closely matches the actual sediment transport patterns in the field. It effectively reduces the calculation deviation of bedload transport rate under complex operating conditions, improves the accuracy of cross-sectional bedload flux calculation, and makes the total sediment volume statistics more consistent with the actual sediment transport status of the river channel.

[0023] This invention integrates multi-dimensional delay parameters to obtain the compensation duration, and performs time-series shift correction on downstream sediment flux data, which can eliminate the time-series errors caused by the lag in water flow and the movement of sediment in different forms. After correction, the two sets of cross-sectional sediment data can achieve time-series matching, effectively improving the accuracy of interception efficiency calculation and ensuring that the monitoring results can objectively reflect the true interception capability of the sediment curtain.

[0024] This invention can autonomously identify common sensor faults such as concentration meter blockage and impact plate damage, and promptly determine abnormal equipment status. By using a neural network model and empirical formulas to compensate for distorted monitoring data, it ensures continuous and normal system operation, while simultaneously pushing equipment maintenance reminders to reduce the interference of fault data on monitoring results and improve the long-term operational stability of the entire monitoring system.

[0025] This invention uses critical shear force as the activation condition for bedload flux estimation, and only performs calculations when the sediment initiation conditions are met; if the flow intensity is insufficient, the result is directly reset to zero. This avoids the problem of invalid data calculation, ensures that the bedload flux measurement closely matches actual sediment movement conditions, and simultaneously feeds back the flow conditions, making the sediment statistics results have practical reference value.

[0026] This invention dynamically adjusts early warning thresholds at various levels based on sediment fluctuations, water pressure changes, and monitoring noise, adapting to the variable sediment and flow conditions of rivers. It reduces the probability of misjudgments and missed judgments caused by fixed thresholds, ensuring that the early warning judgment scale aligns with real-time on-site conditions, improving the rationality of early warning results, and accurately identifying various abnormal operating conditions of sediment curtains.

[0027] This invention establishes interlocking and priority management rules for early warning signals, systematically distinguishes alarm levels, and identifies fault types based on differentiated audible and visual prompts. Emergency commands are issued for curtain wall damage or blockage conditions. When multiple anomalies overlap, automatic operation is paused and manual verification is prompted, standardizing anomaly handling procedures and reducing operational risks associated with extreme conditions.

[0028] This invention automatically pauses efficiency calculations and early warning actions upon identifying low effective sediment transport flux conditions, avoiding invalid calculations and erroneous warnings. A dedicated flashing green indicator clearly distinguishes the standby monitoring status, reducing unnecessary system losses and ensuring reliable judgment results for effective monitoring periods, thus adapting to natural low sediment transport scenarios in river channels.

[0029] Other advantages, objectives and features of the present invention will become apparent in part from the following description, and in part from those skilled in the art through study and practice of the invention. Detailed Implementation

[0030] The present invention will be further described in detail below with reference to embodiments, so that those skilled in the art can implement it based on the description.

[0031] This invention provides a method for real-time monitoring and early warning of sediment curtain interception efficiency, comprising the following steps: A first current meter, a first suspended sediment concentration meter, a first bedload flux measurement module, and a first pressure sensor are installed at a first cross-section upstream of the sediment curtain. A second current meter, a second suspended sediment concentration meter, a second bedload flux measurement module, and a second pressure sensor are installed at a second cross-section downstream of the sediment curtain. A first water depth-area relationship curve for the first cross-section and a second water depth-area relationship curve for the second cross-section are pre-acquired. The first bedload flux measurement module includes a first impact plate fixedly installed on the riverbed surface of the first cross-section and a first accelerometer fixed to the back of the first impact plate. The second bedload flux measurement module includes a second impact plate fixedly installed on the riverbed surface of the second cross-section and a second accelerometer fixed to the back of the second impact plate. The system continuously collects the first cross-sectional average flow velocity output by the first flow meter, the first cross-sectional average suspended sediment concentration output by the first suspended sediment concentration meter, the first vibration signal output by the first accelerometer, and the first water pressure value output by the first pressure sensor, as well as the second cross-sectional average flow velocity output by the second flow meter, the second cross-sectional average suspended sediment concentration output by the second suspended sediment concentration meter, the second vibration signal output by the second accelerometer, and the second water pressure value output by the second pressure sensor. Based on the first vibration signal, the total bedload flux of the first cross section is calculated through a pre-established first calibration relationship, wherein the first calibration relationship is the correspondence between the characteristic values ​​of the vibration signal and the bedload transport rate; based on the second vibration signal, the total bedload flux of the second cross section is calculated through a pre-established second calibration relationship, wherein the second calibration relationship is the correspondence between the characteristic values ​​of the vibration signal and the bedload transport rate; The water depth of the first cross-section is calculated based on the first water pressure value, and the real-time area of ​​the first cross-section is calculated according to the first water depth-area relationship curve; the water depth of the second cross-section is calculated based on the second water pressure value, and the real-time area of ​​the second cross-section is calculated according to the second water depth-area relationship curve. The first suspended sediment flux is obtained by multiplying the real-time first cross-sectional area, the average flow velocity of the first cross-section, and the average suspended sediment concentration of the first cross-section. This first suspended sediment flux is then added to the total bedload flux of the first cross-section to obtain the first total sediment flux. Similarly, the second suspended sediment flux is obtained by multiplying the real-time second cross-sectional area, the average flow velocity of the second cross-section, and the average suspended sediment concentration of the second cross-section. This second suspended sediment flux is then added to the total bedload flux of the second cross-section to obtain the second total sediment flux. The interception efficiency is calculated by dividing the difference between the first and second total sediment flux by the first total sediment flux and then multiplying by 100%. The water pressure difference rate is obtained by dividing the difference between the first and second water pressure values ​​by the first water pressure value and then multiplying by 100%. When the interception efficiency decreases at a rate exceeding 2% per second for 5-60 consecutive seconds, a first-level warning signal is triggered; when the interception efficiency falls below the first warning threshold in the range of 40% to 60%, a second-level warning signal is triggered; when the second total sediment flux exceeds the second warning threshold in the range of 75% to 90% of the first total sediment flux, or when the water pressure difference rate is less than the third warning threshold in the range of 5% to 10%, a third-level warning signal is triggered. The first-level, second-level, and third-level early warning signals are sent to the remote monitoring terminal, and the on-site audible and visual alarms are activated simultaneously.

[0032] In the aforementioned technical solution, a general-purpose river current meter, a suspended sediment concentration meter, an impactor bedload measurement assembly, and a water pressure sensor are installed inside the water body and on the riverbed surface at the upstream monitoring section of the sediment curtain. The impactor bedload measurement assembly consists of an impact plate (which can be made of ordinary carbon steel) and an accelerometer fixed to its back; the impact plate is laid flat and fixed to the riverbed surface. At the same elevation and riverbed location at the downstream monitoring section of the sediment curtain, a second set of the same type of current meter, suspended sediment concentration meter, impactor bedload measurement assembly, and pressure sensor are installed in a completely corresponding manner. Furthermore, the water depth-flow area relationship curves for the first and second sections are obtained in advance through cross-sectional water depth measurement experiments and stored in the system database.

[0033] The system continuously collects average flow velocity, average suspended sediment concentration, vibration signals from the accelerometer, and water pressure at the upstream cross-section at a sampling frequency of no less than once per second, while simultaneously collecting the corresponding four data points at the downstream cross-section. A general-purpose accelerometer with a range of ±10g and a frequency response range of 0-500Hz can be selected. During installation, ensure that the back of the impact plate is rigidly connected to the sensor without any loosening. For the vibration signal output by the upstream accelerometer, the system first establishes a calibration relationship between the characteristic values ​​of the vibration signal and the bedload transport rate through an indoor water tank calibration test. Under various water flow intensities, the vibration signal of the impact plate and the measured bedload transport rate of the sampler are recorded simultaneously. The vibration waveform is split to extract the energy values ​​of the characteristic frequency bands, and a linear relationship curve is fitted as the calibration basis. A second calibration relationship is established at the downstream cross-section in the same way. In the field operation, the total bedload flux of the cross-section is calculated by matching the calibration relationship with the real-time vibration signal.

[0034] The system calculates the cross-sectional water depth based on water pressure monitoring data, interpolates the real-time flow area by comparing with pre-stored relationship curves, and calculates the suspended mass flux at the upstream and downstream sections. The total sediment flux is then obtained by superimposing the corresponding bedload flux. The interception efficiency is calculated based on the ratio of the upstream-downstream sediment flux difference to the upstream flux, and the water pressure difference rate is calculated based on the ratio of the upstream-downstream water pressure difference to the upstream water pressure. The first-level warning judgment duration can be selected as 5s, 20s, 40s, or 60s, with a fixed efficiency decrease rate of 2% / s. The second-level warning threshold can be selected as 40%, 50%, or 60%. The third-level sediment flux comparison value can be selected as 75%, 80%, or 90%, and the water pressure difference judgment threshold can be selected as 5%, 8%, or 10%. Once the judgment conditions are met, the system transmits a warning signal to the remote monitoring terminal via a communication link and can simultaneously activate the on-site multi-color audible and visual alarm. The entire monitoring and warning process runs cyclically according to the judgment cycle.

[0035] Using this technical solution, the present invention can comprehensively collect various hydrological and sediment monitoring parameters from upstream and downstream sections, fully summarize the total transport volume of both types of sediment, and identify various abnormal conditions in facility operation based on multi-level judgment standards. Monitoring information can be simultaneously uploaded to the control terminal and accompanied by on-site audio-visual prompts, enabling continuous monitoring of the actual operating status of the sediment control system and providing fundamental data support for river sediment management and daily facility operation and maintenance.

[0036] In another technical solution, the step of calculating the total bedload flux of the first cross-section based on the first vibration signal includes: Wavelet packet decomposition is performed on the first vibration signal output by the first accelerometer to extract the energy component corresponding to the impact characteristic frequency band of the bedload particles and obtain the first impact energy characteristic value; at the same time, the first water pressure value output by the first pressure sensor and the first cross-sectional average flow velocity output by the first flow meter are used to calculate the water flow shear force at the first cross-section. The pre-established first calibration relationship is: the piecewise linear relationship between the first impact energy characteristic value and the dimensionless ratio of the water flow shear force at the first cross-section and the bedload transport rate at the first cross-section; the segmentation point of this piecewise linear relationship is automatically adjusted according to the rate of change of the average suspended sediment concentration at the first cross-section output by the first suspended sediment concentration meter. Calculate the total bedload flux of the first cross section based on the first calibration relationship; The steps for calculating the total bedload flux of the second cross section based on the second vibration signal include: Wavelet packet decomposition is performed on the second vibration signal output by the second accelerometer to extract the energy component corresponding to the impact characteristic frequency band of the bedload particles, and the second impact energy characteristic value is obtained. At the same time, based on the second water pressure value output by the second pressure sensor and the average flow velocity of the second cross section output by the second flow meter, the shear force of the water flow in the second cross section is calculated. The pre-established second calibration relationship is: the piecewise linear relationship between the second impact energy characteristic value and the dimensionless ratio of the water flow shear force at the second cross-section and the bedload transport rate at the second cross-section; the segmentation point of this piecewise linear relationship is automatically adjusted according to the rate of change of the average suspended sediment concentration at the second cross-section output by the second suspended sediment concentration meter. The total bedload flux of the second section is calculated based on the second calibration relationship.

[0037] In the above technical solution, when calculating the bedload flux, the system first performs wavelet packet decomposition on the original vibration signal output by the accelerometer. A db4 wavelet basis can be used, with a fixed decomposition level of 4 layers, resulting in sixteen different frequency band components. Based on the indoor water tank test results, the effective energy generated by bedload particles impacting the impact plate is mainly concentrated in the 200Hz to 800Hz frequency range. The system sums the signal energy of each component within this frequency band, and then divides the sum of the signal energy within the calculated frequency band by the total energy of the entire frequency band to obtain a dimensionless impact energy characteristic value between 0 and 1.

[0038] Water depth is estimated by combining cross-sectional water pressure data, and flow shear force is calculated using the average flow velocity parameters. The water surface slope is determined by the cross-sectional water level difference and inter-section spacing. The magnitude of the shear force is then verified using the basic physical parameters of the water body, and dimensionless conversion is performed to obtain the corresponding reference parameters. The preset calibration relationship adopts a piecewise linear form, using the ratio of the impact energy characteristic value to the dimensionless shear force parameter as the lateral reference quantity, and the bedload transport rate as the longitudinal reference quantity. The segment nodes can autonomously adjust according to the suspended sediment concentration change rate, and the concentration change range per unit time period can be set to 0.05 kg / (m³). 3 •s) is used as the adjustment boundary standard.

[0039] The adjustment criteria for the segment points of the calibration curves for suspended sediment concentration change rate and bedload transport rate are as follows: The suspended sediment concentration change rate reflects the dynamic trend of the water flow's sediment-carrying capacity. When the suspended sediment concentration rises rapidly (concentration change rate > 0.05 kg / (m³)), the adjustment is adjusted accordingly. 3 The presence of high-frequency components in the impact plate vibration signal indicates an enhanced sediment-carrying capacity of the water flow, with more active initiation and transport of bedload particles. In this case, the proportion of high-frequency components in the vibration signal increases, and the segmentation point of the calibration curve should shift towards the higher energy range. Conversely, when the suspended sediment concentration decreases rapidly (concentration change rate < -0.05 kg / (m³)), the proportion of high-frequency components increases, and the segmentation point of the calibration curve should shift towards the higher energy range. 3 Since bedload transport weakens, the breakpoint should shift towards a lower energy range. The specific adjustment rule is: 0.05 kg / (m³). 3 The baseline change rate is ·s), and for every 0.01 kg / (m³) increase in the concentration change rate... 3 ·s), the energy characteristic value corresponding to the segment point shifts upward by 5%; for every decrease in the concentration change rate of 0.01 kg / (m³), the energy characteristic value shifts upward by 5%. 3The adjustment coefficient and benchmark value are obtained through indoor flume calibration tests. During the tests, the vibration signals of the impact plate and the measured bedload transport rate were recorded simultaneously under different suspended sediment concentration variations. An empirical relationship between the segment point and the concentration change rate was fitted and pre-loaded into the system database as a piecewise linear function. During actual operation, the system calculates the current suspended sediment concentration change rate in real time, substitutes it into the relationship, and calculates the current segment point position, thus achieving online automatic adjustment of the segment nodes. Through this dynamic adjustment, the calibration relationship maintains high calculation accuracy under different sediment inflow conditions.

[0040] The system matches the corresponding linear calculation rules based on the real-time parameter ratio interval, calculates the bedload transport rate, and then calculates the total bedload flux in combination with the cross-section width. The upstream and downstream cross-sections independently follow the unified calculation logic to complete the data processing work, and the calculation mode can adapt to the actual working conditions of dynamic changes in sediment concentration on site.

[0041] By adopting this technical solution, the present invention combines sediment vibration feedback and water flow force to carry out parameter calculation. The calculation rules can be autonomously adjusted and adapted to changes in sediment concentration on site, effectively reducing the deviation generated in the calculation of bedload flux, making the sediment flux statistics results more consistent with the actual sediment transport situation in the river channel, and improving the practical reference value of total sediment statistics.

[0042] In another technical solution, before calculating the interception efficiency, a time delay compensation is applied to the second total sediment flux based on the water flow propagation time between the upstream first cross-section and the downstream second cross-section. Specifically, this includes: The Froude number of the first cross-section is calculated based on the average flow velocity and water depth of the first cross-section, and the theoretical water flow propagation delay is calculated based on the river length from the first cross-section to the second cross-section and the average wave velocity. At the same time, the peak delay of the suspended sediment concentration time series of the first cross-section and the second cross-section, as well as the peak delay of the vibration signal energy series of the first accelerometer and the second accelerometer, are calculated using cross-correlation analysis. The vibration signal energy series is used to characterize the hysteresis characteristics of bedload motion. The theoretical water flow propagation delay, the cross-correlation delay of suspended sediment concentration, and the cross-correlation delay of vibration signal energy are weighted and fused to obtain the final total sediment flux compensation delay. The time series of the second total sediment flux is shifted forward by the final compensation delay, and then the shifted second total sediment flux is compared with the first total sediment flux to calculate the interception efficiency.

[0043] In the above technical solution, the system eliminates the time lag problem in upstream and downstream monitoring data before calculating the interception efficiency. The Froude number is calculated based on the upstream cross-sectional flow velocity and water depth parameters, and the theoretical water propagation delay is estimated by combining the upstream and downstream channel centerline lengths and water surface wave velocities. Cross-correlation analysis is used to process the two sets of time-series data, a fixed-duration statistical window is defined, and the time difference corresponding to the peak correlation coefficient is obtained through sequence sliding comparison, thus obtaining the measured delay parameters corresponding to the sediment concentration sequence and the vibration energy sequence, respectively.

[0044] The three types of delay parameters can be weighted and fused according to a weight ratio of 0.2, 0.5, and 0.3. The weight values ​​are determined by back-calculation based on historical hydrological measurement samples. The integrated parameters are used as a unified compensation delay for sediment flux. The downstream sediment flux time series data are shifted forward according to the compensation duration to complete the time series alignment of upstream and downstream monitoring data. The aligned parameters are then used in the interception efficiency calculation.

[0045] This processing method can offset the time sequence misalignment caused by water flow propagation and sediment transport, ensure that the monitoring data involved in the calculation are consistent in time sequence, and reduce the interference of time deviation on the calculation results.

[0046] The construction method of the "vibration signal energy sequence" is as follows: The original vibration signal output by the accelerometer is segmented into segments with a fixed time window (the window length is 30 seconds). The vibration signal in each time window is first decomposed by wavelet packet decomposition to extract the energy component corresponding to the impact characteristic frequency band (200Hz-800Hz) of the bedload particles. The total energy of this energy component in the window is taken as the representative energy value of the window. Arranged in time order, the vibration signal energy sequence E(t)=[E1, E2, E3, …] is formed, where Eᵢ represents the total energy of the impact characteristic frequency band in the i-th time window.

[0047] The physical significance of this vibration signal energy sequence lies in the fact that as bedload particles move along the riverbed, their concentration and sediment transport intensity are continuous in both space and time. The bedload sediment transport intensity at a certain moment in the upstream section will generate a corresponding impact response in the downstream section after a certain propagation time. The peak delay of the cross-correlation between the vibration signal energy sequences of the upstream and downstream sections is the average time required for the bedload particle group to propagate from the upstream to the downstream. This delay is physically consistent with the water flow propagation delay and the suspended sediment concentration propagation delay. All three reflect the propagation characteristics of water flow and sediment moving from the upstream to the downstream, mutually corroborating and complementing each other. The specific steps for calculating this delay using the cross-correlation analysis method are as follows: take the vibration signal energy sequences of the upstream and downstream sections within the same time period, calculate the curve of the cross-correlation coefficient between the two as a function of the delay, and the delay corresponding to the peak value of the curve is the characteristic propagation delay of bedload movement.

[0048] By adopting this technical solution, the present invention integrates multiple time deviations to complete the monitoring data correction process, effectively reducing the adverse effects of water flow and sediment transport lag, ensuring that the data involved in the calculation from upstream and downstream remain consistent in time sequence, and enabling the final calculated interception efficiency to objectively reflect the true level of sediment curtain interception.

[0049] Another technical solution also includes a sensor data validity self-check and repair step: The average suspended mass concentration of the first section output by the first suspended mass concentration meter and the average suspended mass concentration of the second section output by the second suspended mass concentration meter are continuously compared. When the difference between the two exceeds the preset range and the duration exceeds the first set time, and the absolute value of the average suspended mass concentration of the first section does not change abruptly, the ratio of suspended mass flux to bed mass flux of the first section and the ratio of suspended mass flux to bed mass flux of the second section are calculated respectively. If the ratio of a certain section deviates from the historical average of that section by more than 3 times the standard deviation, it is determined that the suspended sediment concentration meter of that section is blocked. At this time, the water pressure difference rate of that section and the upstream sediment flux are used to estimate the suspended sediment concentration replacement value through a pre-trained neural network model and trigger the sensor maintenance reminder. Meanwhile, by comparing the vibration signal characteristics of the first accelerometer and the second accelerometer, if the characteristic energy of one of the accelerometers is continuously lower than the preset lower energy threshold (this threshold is set according to the environmental background vibration noise level and is taken as 3 times the background vibration energy during the historical still water period) and the average flow velocity of the corresponding section is higher than the bedload initiation velocity, it is determined that the impact plate of the section is buried or damaged. At this time, the suspended sediment concentration of the same section and the empirical bedload transport formula based on the shear force of the water flow are used to estimate the bedload flux of the section.

[0050] In the above technical solution, the system performs periodic self-checks on sensor data during operation. The self-check interval can be set to 10 seconds. The suspended sediment concentration monitoring values ​​at the upstream and downstream sections are compared, and the concentration difference judgment threshold can be set to 0.2 kg / m³. 3 The duration of abnormality detection can be set to 30 seconds, simultaneously verifying the concentration fluctuation status of a single cross-section. The limit for a single fluctuation can be set to 0.05 kg / m³. 3 .

[0051] After identifying data anomalies, the ratio of the two types of sediment flux was calculated, with three times the standard deviation used as the criterion for deviation, thus identifying clogging faults in the suspended sediment concentration meter. A three-layer neural network model can be used to estimate the effective concentration value. The network input parameters are selected as cross-sectional sediment flux, water pressure differential rate, and water flow velocity. The model is trained using two thousand sets of historical normal data, with the training objective being that the root mean square error (RMSE) between the model output value and the measured value is less than 0.01 kg / m³. 3After the fault is determined, valid data is replaced in the calculation, and equipment maintenance prompts are pushed out simultaneously.

[0052] “Normal operating condition” is defined as an operating state that simultaneously meets the following conditions: (1) The difference between the readings of the first suspended mass concentration meter and the second suspended mass concentration meter is within ±0.2 kg / m³. 3 (1) Within; (2) The characteristic energy of the first accelerometer and the second accelerometer are both higher than their respective preset energy lower limit thresholds (the threshold is set according to the environmental background vibration noise level and is taken as 3 times the background vibration energy of the historical still water period); (3) The coefficient of variation of each sensor reading is less than 0.3 in 10 consecutive judgment periods; (4) The system does not trigger any level of early warning signal. The data collected under the above conditions is marked as "normal working condition" data and used for training the neural network model.

[0053] The physical relationship between the input parameters (section sediment flux, water pressure differential, and water flow velocity) and the output (suspended sediment concentration) of the neural network model is as follows: the section sediment flux reflects the total sediment transport intensity of the section and is positively correlated with the suspended sediment concentration; the water pressure differential reflects the flow state and interception degree of the sediment curtain, indirectly affecting the change in downstream suspended sediment concentration; the water flow velocity determines the sediment-carrying capacity of the water flow and is one of the main controlling factors of suspended sediment concentration. These three factors together constitute the set of main influencing factors of suspended sediment concentration.

[0054] To ensure the generalization ability of the model under extreme conditions, the following measures are taken: (1) In addition to normal operating conditions, the training dataset also includes measured data under different hydrological conditions, such as below 50%, 50% to 70% of the highest historical flood level, and above 70% of the highest flood level, to ensure that the training samples cover a wide range of operating conditions from normal water to extreme floods; (2) The model is trained using the Bayesian regularization algorithm, and a weight decay term is introduced in the loss function to suppress overfitting; (3) After the model is deployed, it continues to learn online, and the model is incrementally updated for every 100 sets of new effective measured data; (4) When the deviation between the model output value and the theoretical value calculated by other sensors (such as flow meter and pressure sensor) at the same cross section exceeds 3 times the standard deviation, the system automatically marks the output as low confidence and sends a prompt message to the remote monitoring terminal that "the model output needs to be manually reviewed".

[0055] The system synchronously identifies impact plate burial and damage faults. The vibration energy zeroing judgment time can be set to 1 minute. The bedload initiation velocity can be selected as 0.3 m / s as the comparison standard. After a fault occurs, it switches to the general Meyer-Peter & Müller bedload transport formula for estimation. The basic parameters such as the median particle size and specific gravity of the sediment required in the formula are pre-entered into the system. The system still maintains uninterrupted monitoring and operation in the event of equipment failure.

[0056] By adopting this technical solution, the present invention can automatically identify common faults in field monitoring equipment, promptly replace and compensate for failed monitoring data, ensure that monitoring operations are not interrupted, and remind staff to carry out equipment maintenance in a timely manner, thereby steadily improving the stability of the entire monitoring system for long-term continuous operation.

[0057] In another technical solution, before estimating the bedload flux of a cross-section using the suspended sediment concentration and the empirical bedload transport formula based on the shear force of the water flow, the real-time shear force of the water flow at the cross-section is calculated based on the water pressure and average flow velocity. Then, it is determined whether the real-time shear force of the water flow at the cross-section is greater than the preset critical shear force for bedload initiation. Only when the real-time shear force of the water flow at the cross-section is greater than the critical shear force for bedload initiation is the empirical bedload transport formula allowed for estimation. If the real-time shear force of the water flow at the cross-section is less than or equal to the critical shear force for bedload initiation, the bedload flux at the cross-section is set to zero, and an insufficient flow intensity reminder is sent to the remote monitoring terminal.

[0058] In the above technical solution, before using the empirical sediment transport formula to calculate the bedload flux, the system first determines the initiation dynamic conditions of sediment at the cross-section, and calculates the real-time flow shear force according to the predetermined calculation method. This can be based on the median particle size d of the sediment in the first cross-section. 50 Determine the critical shear force reference value by referring to the Shields curve (for example, for silt with a median particle size of about 0.5 mm, the critical shear force is about 0.05 Pa).

[0059] The system compares real-time shear force parameters with critical parameters. When the measured value exceeds the critical standard, the system uses empirical formulas to estimate the flux. If the measured value does not meet the critical criteria, the bedload flux at that section is recorded as zero, and subsequent flux calculations are not performed. Simultaneously, a flow condition alert is sent to the remote monitoring terminal. The overall calculation logic is set according to the objective laws of sediment initiation, and the calculation mode is divided based on the actual sediment-carrying capacity of the on-site flow to avoid invalid calculations that are detached from actual conditions.

[0060] By adopting this technical solution, the present invention constrains the overall calculation process with the critical condition for sediment initiation, eliminates invalid calculation steps that do not conform to actual working conditions, strictly follows the inherent laws of sediment movement in the river channel to collect flux data, and ensures that the statistical data can accurately reflect the actual sediment transport status at the cross-section.

[0061] In another technical solution, the first, second, and third warning thresholds are dynamically adjusted based on upstream sediment inflow conditions. The sliding average value and coefficient of variation of the total sediment flux at the first cross section are calculated in real time. When the coefficient of variation is higher than the preset value, the second warning threshold is raised. When the fluctuation range of the water pressure differential exceeds the upper limit of the historical fluctuation range, the third warning threshold is lowered; and the threshold for the rate of decrease in interception efficiency corresponding to the first warning signal is adjusted according to the noise level of the interception efficiency, which is characterized by the root mean square of the high-frequency component of the interception efficiency.

[0062] In the above technical solution, the system's various early warning thresholds are dynamically adjusted according to the on-site operating conditions, with a threshold update operation performed every 10 minutes. The statistical duration can be set to 1 hour, and the abnormal threshold value for the flux variation coefficient can be set to 0.5 or 0.7, with the secondary early warning threshold being increased accordingly based on the degree of fluctuation. The statistical duration for water pressure data can be set to 30 minutes, and the historical reference statistical period can be set to 24 hours, with the water pressure-related judgment thresholds being decreased according to changes in the fluctuation amplitude. For every 0.1 increase in the variation coefficient, the secondary early warning threshold is increased by 1 to 3 percentage points.

[0063] A high-pass digital filter (cutoff frequency set to 0.1Hz) is applied to the interception efficiency time series to remove low-frequency trends and obtain high-frequency components. The root mean square value of this high-frequency component is then calculated as the data noise level. The noise threshold can be set to 0.5%, and the efficiency decline rate threshold is adjusted synchronously with the noise value. Various criteria parameters are adapted to changes in river sediment load, water pressure fluctuations, and data interference to flexibly meet the ever-changing on-site hydrological operating conditions.

[0064] By adopting this technical solution, the present invention can flexibly adjust the judgment threshold according to the river sediment conditions, water pressure fluctuations and monitoring data noise, adapt to complex and ever-changing on-site hydrological conditions, reduce judgment errors caused by fixed judgment standards, reduce the probability of false and missed judgments in early warnings, and improve the rationality of abnormal condition identification.

[0065] In another technical solution, after sending the first-level, second-level, and third-level warning signals to the remote monitoring terminal and simultaneously activating the on-site audible and visual alarms, interlocking and priority management of the warning signals are also performed: When both the first-level and second-level warning signals are triggered simultaneously, only the alarm level corresponding to the second-level warning signal is output, along with the feature code of the first-level warning signal. When a Level 3 warning signal exists simultaneously with a Level 1 or Level 2 warning signal, the Level 3 warning signal will be elevated to the highest priority alarm, and the on-site audible and visual alarms will be forcibly triggered to distinguish them by different flashing colors. An emergency shutdown command will be sent to the remote monitoring terminal when any of the following combinations of conditions are met, and it will be recommended to shut down the upstream water supply or activate the emergency sediment flushing facilities: The sediment curtain is damaged or fails when the interception efficiency is lower than the lower limit of the second warning threshold and the water pressure difference rate is lower than the preset low water pressure difference emergency threshold. The lower limit of the second warning threshold is preset in the range of 40% to 60%, and the low water pressure difference emergency threshold is preset in the range of 5% to 15%. Severe blockage by sediment curtain: The interception efficiency is higher than the preset high interception efficiency threshold, and the water pressure difference rate is higher than the preset high water pressure difference emergency threshold; wherein, the high interception efficiency threshold is preset within the range of 70% to 90%, and the high water pressure difference emergency threshold is preset within the range of 20% to 30%. Furthermore, when the conditions of mud curtain damage or failure and severe mud curtain blockage are triggered simultaneously, it is determined that the sensor data is abnormal or the cross-sectional shape has changed drastically. The automatic shutdown command is suspended, the data self-check level is upgraded, and a prompt message requiring manual review is sent to the remote monitoring terminal.

[0066] In the above technical solution, when multiple sets of early warning signals are triggered simultaneously, the system executes hierarchical interlocking and priority control logic. When level one and level two early warnings coexist, the level two early warning output level is retained, and the associated operating condition characteristic identifier is transmitted to the monitoring terminal. Level three early warnings have the highest response priority, and differentiated fault levels are distinguished by different light colors and alarm frequencies. Yellow, orange, and red lights correspond to level one, level two, and level three early warning states, respectively. When a level three early warning is triggered simultaneously with other levels, it is forced to flash rapidly in red.

[0067] The threshold values ​​for curtain wall damage can be selected as 40%, 50%, and 60%; the emergency threshold for low water pressure can be selected as 5%, 10%, and 15%; the threshold values ​​for blockage can be selected as 70%, 80%, and 90%; and the emergency threshold for high water pressure can be selected as 20%, 25%, and 30%. If extreme conditions are met, shutdown and sand removal control recommendations will be pushed out. When two types of extreme condition signals are triggered simultaneously, the system increases the self-check frequency, with the self-check interval adjustable to 1 minute, pauses automatic control commands, and simultaneously issues a manual review reminder.

[0068] By adopting this technical solution, the present invention systematically divides the response levels and handling sequence of early warning signals, distinguishes different fault types by using differentiated audio-visual prompts, outputs corresponding handling suggestions for extreme working conditions such as curtain damage and mud and sand blockage, and reasonably responds to scenarios where multiple anomalies occur simultaneously, effectively reducing the potential safety risks during the operation of the mud and sand curtain.

[0069] In another technical solution, before calculating the interception efficiency by dividing the difference between the first and second total sediment flux by the first total sediment flux and then multiplying by 100%, a low flux determination and processing step is first performed: The first total sediment flux is compared with the preset minimum effective flux threshold, which is dynamically determined based on the first cross-sectional area, the average flow velocity of the first cross-section, and the reference sediment transport rate corresponding to the critical velocity for bedload initiation. If the first total sediment flux is less than or equal to the minimum effective flux threshold, the interception efficiency will not be calculated, and no level of warning signal (first, second, or third level) will be triggered. At the same time, a status message indicating that the interception efficiency assessment is paused will be sent to the remote monitoring terminal. If the first total sediment flux is greater than the minimum effective flux threshold for three consecutive judgment periods, then the normal interception efficiency calculation and early warning logic will be restored. During the pause, the on-site audible and visual alarm only indicates that the system is in standby monitoring mode by flashing green light, without emitting any sound alarm.

[0070] In the above technical solution, before each calculation of interception efficiency, the system determines the effectiveness of the upstream sediment flux and comprehensively delineates the minimum effective flux limit by combining the basic cross-sectional parameters and reference sediment transport indicators. The reference concentration can be selected as 0.01 kg / m³. 3 The bedload baseline reference value can be selected as 0.001 kg / (m·s). When the flux value does not reach the effective standard, the system suspends efficiency calculation and early warning triggering, only retaining the green light standby indicator. The light flashing frequency can be set to once every 2 seconds, and the equipment does not emit an audible alarm. The judgment period can be set to 10 seconds. After the flux has stably reached the standard for three consecutive periods, the system automatically resumes all monitoring calculations and early warning response functions, switching to normal working mode.

[0071] Minimum effective flux threshold Q m ᵢ n Calculated dynamically using the following formula: Q min = A × C sat × v + B × q b0 in: A represents the real-time water flow area of ​​the first cross-section (m²). 2 The result is calculated from the first water pressure value combined with the first water depth-area relationship curve; B is the average width of the first cross section (m), which is directly read or measured from the water surface width corresponding to the first water depth-area relationship curve. v is the average flow velocity (m / s) of the first cross section, which is collected in real time by the first flow meter; C sat To determine the effective lower limit concentration for suspended matter monitoring, a value of 0.01 kg / m³ was used. 3 This value is determined based on the detection limit of the selected suspension concentration meter; qb0 The reference sediment transport rate per unit width (kg / (m·s)) corresponding to the critical velocity for bedload initiation is determined as follows: First, based on the median particle size d of the riverbed sediment at the first cross section... 50 The dimensionless critical shear force θ was obtained using the Shields curve. c Then from θ c = τ c / [(ρ s - ρ)·g·d 50 Calculate the critical shear force τ c (ρ) s ρ is the density of sediment particles (natural river sand can be taken as 2650 kg / m³), ρ is the density of water, and g is the acceleration due to gravity (9.8 m / s²). 2 ), and then from τ c = ρ·g·h·J (where h is the water depth and J is the water surface gradient) to calculate the critical water depth condition, and then use the Manning formula to calculate the critical velocity u for bedload initiation. c In indoor water tank tests, at the critical flow velocity u c The measured bedload transport rate per unit width under the given conditions is q. b0 If experimental conditions are lacking, the Meyer-Peter & Müller formula can be used to calculate q under critical starting conditions. b0 Approximate value.

[0072] The physical meaning of the above formula is: when the total sediment transport from the upstream flow is lower than the sum of the minimum suspended sediment flux corresponding to the effective detection limit of the suspended sediment concentration meter and the critical flux for bedload initiation, the monitoring system cannot guarantee the validity and reliability of the flux data, and the interception efficiency calculation should be suspended at this time. Q min The minimum effective flux threshold is dynamically and adaptively adjusted in real time as the cross-sectional area A, cross-sectional width B, and flow velocity v change.

[0073] By adopting this technical solution, the present invention can clearly distinguish between low sand transport standby conditions and effective monitoring conditions, stop unnecessary calculations and alarm actions during periods when there is no actual monitoring significance, reduce unnecessary consumption of system resources, avoid false warnings caused in low throughput scenarios, and ensure that the judgment results obtained in the effective monitoring stage have reliable reference value.

[0074] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and embodiments shown and described herein.

Claims

1. A method for real-time monitoring and early warning of sediment curtain interception efficiency, characterized in that, Includes the following steps: A first current meter, a first suspended sediment concentration meter, a first bedload flux measurement module, and a first pressure sensor are installed at a first cross-section upstream of the sediment curtain. A second current meter, a second suspended sediment concentration meter, a second bedload flux measurement module, and a second pressure sensor are installed at a second cross-section downstream of the sediment curtain. A first water depth-area relationship curve for the first cross-section and a second water depth-area relationship curve for the second cross-section are pre-acquired. The first bedload flux measurement module includes a first impact plate fixedly installed on the riverbed surface of the first cross-section and a first accelerometer fixed to the back of the first impact plate. The second bedload flux measurement module includes a second impact plate fixedly installed on the riverbed surface of the second cross-section and a second accelerometer fixed to the back of the second impact plate. The system continuously collects the first cross-sectional average flow velocity output by the first flow meter, the first cross-sectional average suspended sediment concentration output by the first suspended sediment concentration meter, the first vibration signal output by the first accelerometer, and the first water pressure value output by the first pressure sensor, as well as the second cross-sectional average flow velocity output by the second flow meter, the second cross-sectional average suspended sediment concentration output by the second suspended sediment concentration meter, the second vibration signal output by the second accelerometer, and the second water pressure value output by the second pressure sensor. Based on the first vibration signal, the total bedload flux of the first cross section is calculated through a pre-established first calibration relationship, wherein the first calibration relationship is the correspondence between the characteristic values ​​of the vibration signal and the bedload transport rate; based on the second vibration signal, the total bedload flux of the second cross section is calculated through a pre-established second calibration relationship, wherein the second calibration relationship is the correspondence between the characteristic values ​​of the vibration signal and the bedload transport rate; The water depth of the first cross-section is calculated based on the first water pressure value, and the real-time area of ​​the first cross-section is calculated according to the first water depth-area relationship curve; the water depth of the second cross-section is calculated based on the second water pressure value, and the real-time area of ​​the second cross-section is calculated according to the second water depth-area relationship curve. The first suspended sediment flux is obtained by multiplying the real-time first cross-sectional area, the average flow velocity of the first cross-section, and the average suspended sediment concentration of the first cross-section. This first suspended sediment flux is then added to the total bedload flux of the first cross-section to obtain the first total sediment flux. Similarly, the second suspended sediment flux is obtained by multiplying the real-time second cross-sectional area, the average flow velocity of the second cross-section, and the average suspended sediment concentration of the second cross-section. This second suspended sediment flux is then added to the total bedload flux of the second cross-section to obtain the second total sediment flux. The interception efficiency is calculated by dividing the difference between the first and second total sediment flux by the first total sediment flux and then multiplying by 100%. The water pressure difference rate is obtained by dividing the difference between the first and second water pressure values ​​by the first water pressure value and then multiplying by 100%. When the interception efficiency decreases at a rate exceeding 2% per second for 5-60 consecutive seconds, a first-level warning signal is triggered; when the interception efficiency falls below the first warning threshold in the range of 40% to 60%, a second-level warning signal is triggered; when the second total sediment flux exceeds the second warning threshold in the range of 75% to 90% of the first total sediment flux, or when the water pressure difference rate is less than the third warning threshold in the range of 5% to 10%, a third-level warning signal is triggered. The first-level, second-level, and third-level early warning signals are sent to the remote monitoring terminal, and the on-site audible and visual alarms are activated simultaneously.

2. The method for real-time monitoring and early warning of sediment curtain interception efficiency as described in claim 1, characterized in that, The steps for calculating the total bedload flux of the first cross section based on the first vibration signal include: Wavelet packet decomposition is performed on the first vibration signal output by the first accelerometer to extract the energy component corresponding to the impact characteristic frequency band of the bedload particles and obtain the first impact energy characteristic value; at the same time, the first water pressure value output by the first pressure sensor and the first cross-sectional average flow velocity output by the first flow meter are used to calculate the water flow shear force at the first cross-section. The pre-established first calibration relationship is: the piecewise linear relationship between the first impact energy characteristic value and the dimensionless ratio of the water flow shear force at the first cross-section and the bedload transport rate at the first cross-section; the segmentation point of this piecewise linear relationship is automatically adjusted according to the rate of change of the average suspended sediment concentration at the first cross-section output by the first suspended sediment concentration meter. Calculate the total bedload flux of the first cross section based on the first calibration relationship; The steps for calculating the total bedload flux of the second cross section based on the second vibration signal include: Wavelet packet decomposition is performed on the second vibration signal output by the second accelerometer to extract the energy component corresponding to the impact characteristic frequency band of the bedload particles, and the second impact energy characteristic value is obtained. At the same time, based on the second water pressure value output by the second pressure sensor and the average flow velocity of the second cross section output by the second flow meter, the shear force of the water flow in the second cross section is calculated. The pre-established second calibration relationship is: the piecewise linear relationship between the second impact energy characteristic value and the dimensionless ratio of the water flow shear force at the second cross-section and the bedload transport rate at the second cross-section; the segmentation point of this piecewise linear relationship is automatically adjusted according to the rate of change of the average suspended sediment concentration at the second cross-section output by the second suspended sediment concentration meter. The total bedload flux of the second section is calculated based on the second calibration relationship.

3. The method for real-time monitoring and early warning of sediment curtain interception efficiency as described in claim 2, characterized in that, Before calculating the interception efficiency, a time delay compensation is first applied to the second total sediment flux based on the water flow propagation time between the upstream first cross section and the downstream second cross section. Specifically, this includes: The Froude number of the first cross-section is calculated based on the average flow velocity and water depth of the first cross-section, and the theoretical water flow propagation delay is calculated based on the river length from the first cross-section to the second cross-section and the average wave velocity. At the same time, the peak delay of the suspended sediment concentration time series of the first cross-section and the second cross-section, as well as the peak delay of the vibration signal energy series of the first accelerometer and the second accelerometer, are calculated using cross-correlation analysis. The vibration signal energy series is used to characterize the hysteresis characteristics of bedload motion. The theoretical water flow propagation delay, the cross-correlation delay of suspended sediment concentration, and the cross-correlation delay of vibration signal energy are weighted and fused to obtain the final total sediment flux compensation delay. The time series of the second total sediment flux is shifted forward by the final compensation delay, and then the shifted second total sediment flux is compared with the first total sediment flux to calculate the interception efficiency.

4. The method for real-time monitoring and early warning of sediment curtain interception efficiency as described in claim 3, characterized in that, It also includes self-check and repair steps for sensor data validity: The average suspended mass concentration of the first section output by the first suspended mass concentration meter and the average suspended mass concentration of the second section output by the second suspended mass concentration meter are continuously compared. When the difference between the two exceeds the preset range and the duration exceeds the first set time, and the absolute value of the average suspended mass concentration of the first section does not change abruptly, the ratio of suspended mass flux to bed mass flux of the first section and the ratio of suspended mass flux to bed mass flux of the second section are calculated respectively. If the ratio of a certain section deviates from the historical average of that section by more than 3 times the standard deviation, it is determined that the suspended sediment concentration meter of that section is blocked. At this time, the water pressure difference rate of that section and the upstream sediment flux are used to estimate the suspended sediment concentration replacement value through a pre-trained neural network model and trigger the sensor maintenance reminder. Meanwhile, by comparing the vibration signal characteristics of the first accelerometer and the second accelerometer, if the characteristic energy of one of the accelerometers is continuously lower than the preset lower energy threshold and the average flow velocity of the corresponding section is higher than the bedload initiation velocity, it is determined that the impact plate of the section is buried or damaged. In this case, the suspended sediment concentration of the same section and the empirical bedload transport formula based on the shear force of the water flow are used to estimate the bedload flux of the section.

5. The method for real-time monitoring and early warning of sediment curtain interception efficiency as described in claim 4, characterized in that, Before using the suspended sediment concentration and empirical bedload transport formula based on water flow shear force to estimate the bedload flux of a cross section, the real-time water flow shear force of the cross section is calculated based on the water pressure and average flow velocity. Then, it is determined whether the real-time water flow shear force of the cross section is greater than the preset critical shear force for bedload initiation. Only when the real-time water flow shear force of the cross section is greater than the critical shear force for bedload initiation is the empirical bedload transport formula allowed for estimation. If the real-time water flow shear force of the cross section is less than or equal to the critical shear force for bedload initiation, the bedload flux of the cross section is set to zero, and an insufficient water flow intensity reminder message is sent to the remote monitoring terminal.

6. The method for real-time monitoring and early warning of sediment curtain interception efficiency as described in claim 4, characterized in that, The first, second, and third warning thresholds are dynamically adjusted based on upstream sediment inflow conditions. The sliding average value and coefficient of variation of the total sediment flux at the first cross section are calculated in real time. When the coefficient of variation is higher than the preset value, the second warning threshold is raised. When the fluctuation range of the water pressure differential exceeds the upper limit of the historical fluctuation range, the third warning threshold is lowered; and the threshold for the rate of decrease in interception efficiency corresponding to the first warning signal is adjusted according to the noise level of the interception efficiency, which is characterized by the root mean square of the high-frequency component of the interception efficiency.

7. The method for real-time monitoring and early warning of sediment curtain interception efficiency as described in claim 1, characterized in that, After sending the first-level, second-level, and third-level warning signals to the remote monitoring terminal and simultaneously activating the on-site audible and visual alarms, the system also performs interlocking and priority management of the warning signals. When both the first-level and second-level warning signals are triggered simultaneously, only the alarm level corresponding to the second-level warning signal is output, along with the feature code of the first-level warning signal. When a Level 3 warning signal exists simultaneously with a Level 1 or Level 2 warning signal, the Level 3 warning signal will be elevated to the highest priority alarm, and the on-site audible and visual alarms will be forcibly triggered to distinguish them by different flashing colors. An emergency shutdown command will be sent to the remote monitoring terminal when any of the following combinations of conditions are met, and it will be recommended to shut down the upstream water supply or activate the emergency sediment flushing facilities: The sediment curtain is damaged or fails when the interception efficiency is lower than the lower limit of the second warning threshold and the water pressure difference rate is lower than the preset low water pressure difference emergency threshold. The lower limit of the second warning threshold is preset in the range of 40% to 60%, and the low water pressure difference emergency threshold is preset in the range of 5% to 15%. Severe blockage by sediment curtain: The interception efficiency is higher than the preset high interception efficiency threshold, and the water pressure difference rate is higher than the preset high water pressure difference emergency threshold; wherein, the high interception efficiency threshold is preset within the range of 70% to 90%, and the high water pressure difference emergency threshold is preset within the range of 20% to 30%. Furthermore, when the conditions of mud curtain damage or failure and severe mud curtain blockage are triggered simultaneously, it is determined that the sensor data is abnormal or the cross-sectional shape has changed drastically. The automatic shutdown command is suspended, the data self-check level is upgraded, and a prompt message requiring manual review is sent to the remote monitoring terminal.

8. The method for real-time monitoring and early warning of sediment curtain interception efficiency as described in claim 1, characterized in that, Before calculating the interception efficiency by dividing the difference between the first and second total sediment flux by the first total sediment flux and then multiplying by 100%, the low flux determination and processing steps are performed first: The first total sediment flux is compared with the preset minimum effective flux threshold, which is dynamically determined based on the first cross-sectional area, the average flow velocity of the first cross-section, and the reference sediment transport rate corresponding to the critical velocity for bedload initiation. If the first total sediment flux is less than or equal to the minimum effective flux threshold, the interception efficiency will not be calculated, and no level of warning signal (first, second, or third level) will be triggered. At the same time, a status message indicating that the interception efficiency assessment is paused will be sent to the remote monitoring terminal. If the first total sediment flux is greater than the minimum effective flux threshold for three consecutive judgment periods, then the normal interception efficiency calculation and early warning logic will be restored. During the pause, the on-site audible and visual alarm only indicates that the system is in standby monitoring mode by flashing green light, without emitting any sound alarm.