River water level detection method

Abstract

The invention discloses a river water level detection method, and relates to the technical field of computer processing, and the method comprises the following steps: S01, presetting a distributed pressure sensor array in each target collection region to obtain underwater pressure data k and real-time water flow data v of a plurality of target collection regions of a river to be detected; s02, establishing a dynamic pressure water level model based on the water flow direction pressure conduction quantity estimation of the plurality of target acquisition areas in combination with real-time water flow velocity and wave period parameters; s03, performing signal attenuation compensation on the underwater pressure data according to the dynamic pressure water level model to obtain initial water level data; s04, multi-source data fusion calibration is carried out on the preliminary water level data, a river water level detection result is generated, and the result comprises a wave height frequency spectrum and a water level change curve after peak elimination correction; and S05, generating a wave energy spectral density map according to the calibrated water level data, performing real-time comparison according to the wave energy spectral density map and a preset threshold value, and performing result return according to a comparison structure. The small-wave-height short-period wave measurement precision and the water level detection reliability are remarkably improved, the intelligent flood prevention early warning and dynamic adjusting capacity is achieved, and it is guaranteed that data collection is accurate and consistent.

Description

Technical Field

[0001] The present invention relates to the field of computer processing technology, in particular to a river water level detection method. Background Art

[0002] River water level detection is a core component of hydrological monitoring and disaster warning, and its accuracy directly impacts flood control, ecological protection, and water resource management. Traditional water level detection methods, which rely on single-point pressure sensors or float-type water level gauges, have significant limitations in complex flow environments. These methods are susceptible to fluctuations in flow velocity, wave period, and turbulence, leading to pressure signal attenuation or distortion.

[0003] To address this issue, existing technologies often use a static compensation model (refer to China's authorized patent, publication number CN114136395A, which discloses a water level detection method, device, computer equipment, and readable storage medium). This static compensation model cannot correct the dynamic pressure conduction effect in real time, especially under the high-frequency components of waves, where the error is significant. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention adopts a method for detecting river water level, which comprises the following steps:

[0005] S01, presetting a distributed pressure sensor array in each target collection area to obtain underwater pressure data k and real-time water flow data v of multiple target collection areas of the river to be detected;

[0006] S02. Establishing a dynamic pressure water level model based on the estimated pressure conduction in the water flow direction of the target acquisition areas and in combination with real-time water flow velocity and wave period parameters;

[0007] S03, performing signal attenuation compensation on the underwater pressure data according to the dynamic pressure water level model to obtain preliminary water level data;

[0008] S04, performing multi-source data fusion calibration on the preliminary water level data to generate a river water level detection result, the result including a wave height spectrum and a water level change curve after peak elimination correction;

[0009] S05. Generate a wave energy spectrum density map based on the calibrated water level data, perform a real-time comparison between the wave energy spectrum density map and a preset threshold, and report the result based on the comparison structure.

[0010] Preferably, the distributed pressure sensor array in step S01 comprises:

[0011] S11, arrange a 3×3 grid-shaped pressure sensor array along the water flow direction, with the central node being the master pressure sensor and the rest being slave pressure sensors;

[0012] S12. Each pressure sensor is installed on the riverbed base with an installation depth deviation of ≤5cm and connected to the data acquisition terminal via a waterproof cable;

[0013] S13. The master pressure sensor collects raw pressure data at a sampling rate of 16 Hz, and the slave pressure sensor collects auxiliary data at a sampling rate of 8 Hz. The multiple slave pressure sensors are phase-aligned according to the propagation delay Δt=d / v, where d is the spacing between adjacent sensors in the distributed pressure sensor array.

[0014] Preferably, the establishment of the dynamic pressure water level model in step S02 includes:

[0015] S21. Calculate the pressure attenuation coefficient based on the real-time water velocity v and wave period T:

[0016]

[0017] Where β is the environmental correction factor, which is 2;

[0018] S22, based on the following formula: p(x,y)=k·H(x,y)·e -α·x +b Establish a pressure conduction compensation model along the water flow direction;

[0019] Where x is the coordinate along the water flow direction, p(x,y) is the compensated pressure value at the coordinate point (x,y) of the target acquisition area, H(x,y) is the highest water level at the coordinate point (x,y) of the target acquisition area, and b is the pressure offset.

[0020] Preferably, the signal attenuation compensation in step S03 includes:

[0021] S31, performing frequency domain analysis on the underwater pressure data k to extract the wavelet high-frequency component Tm;

[0022] S32, according to the wave period parameter T≤2s and the wave height threshold H≤0.3m, the high frequency component is gain compensated, and the compensation amplitude is Among them, γ is the peak elimination correction factor, and τ is the time, which is 0.5s.

[0023] Preferably, the dynamic adjustment of the peak elimination correction factor γ includes:

[0024] S100, collecting historical peak elimination rate η, water flow turbulence intensity I and sensor signal-to-noise ratio SNR to build a training data set;

[0025] S101. Use the random forest regression algorithm to train the model, with the input features being [η, I, SNR] and the output being γ. The MAE of the model test set is ≤ 0.05.

[0026] S102. Calculate the current parameters and predict γn every 5 minutes, and perform smooth updates using γ(t) = 0.8·γ(t-1) + 0.2·γn. When the SNR is less than 20 dB, freeze the γ value and enable the backup compensation strategy.

[0027] Among them, γ(t) is the peak elimination correction factor at time t, γ(t-1) is the correction factor at the previous moment, and γn is the new correction factor predicted by the machine learning model.

[0028] Preferably, the multi-source data fusion calibration in step S04 includes:

[0029] S41, obtaining the acoustic ranging water level data and the historical wave height database of the same detection point, wherein the acoustic ranging water level data is sampled at a rate of f s ≥10Hz acquisition, and aligning timestamps through interpolation algorithm;

[0030] S42. Establish a Kalman filter state equation, define the state vector X = [H, dH / dt]T and the observation vector Z = [Hp, Hs]T;

[0031] S43, dynamically setting an observation noise matrix according to a pressure sensor signal-to-noise ratio (SNR) and an acoustic device measurement error, wherein the pressure sensor signal-to-noise ratio (SNR) is ≥ 30 dB and the acoustic device measurement error is ±0.5 cm;

[0032] S44. Output the smoothed calibration water level value Hc every 10 seconds. If the fusion residual ε(t) = |Hp-Hs| exceeds 2 cm for three consecutive times, a sensor abnormality alarm is triggered.

[0033] Where H is the water level value at the current moment, dH / dt is the water level change rate, Hp is the water level value initially calculated by the pressure sensor, and Hs is the water level value measured by the acoustic rangefinder.

[0034] Preferably, the wave energy spectrum density map in step S05 is compared with a preset threshold in real time, and the result is reported according to the comparison structure, including calculating the wave height change rate in real time:

[0035]

[0036] The threshold value θ is set to 0.1 m / min. If ΔH / Δt ≥ θ for 5 minutes, it is determined to be a sudden change in wave height. If ΔH / Δt < θ for 5 minutes, it is determined to be a normal floating level.

[0037] Where ΔH is the change in wave height, H(t) is the calibrated water level at the current moment, H(t-Δt) is the water level at the previous sampling moment, and Δt is the sampling time interval.

[0038] The present invention has at least the following beneficial effects:

[0039] 1. By establishing a dynamic pressure water level model, calculating the pressure attenuation coefficient based on the real-time water velocity and wave period parameters, and establishing a pressure conduction compensation model along the water flow direction, the underwater pressure data is compensated for signal attenuation. This can reduce the impact of pressure conduction loss and peak elimination on the measurement results, thereby improving the measurement accuracy of small wave heights and short-period waves.

[0040] 2. A multi-source data fusion calibration method is used to obtain acoustic ranging water level data and a historical wave height database from the same detection point. By establishing a Kalman filter state equation, the observation noise matrix is ​​dynamically set based on the pressure sensor signal-to-noise ratio and acoustic device measurement error. A smoothed calibration water level value is output every 10 seconds. If the fusion residual exceeds the threshold three consecutive times, a sensor anomaly alarm is triggered. This method leverages the strengths of multiple data sources, complementing and validating each other to effectively improve the reliability of water level detection results.

[0041] 3. It can calculate the wave height change rate in real time and set thresholds to identify sudden wave height changes. When a sudden wave height change is detected for a certain period of time, a JSON message containing the location of the change and the wave height trend forecast is pushed to the target user account, and the emergency response protocol is initiated. This can promptly identify potential flood risks, provide timely and accurate information support for flood prevention decisions, and help improve the efficiency and safety of flood prevention work.

[0042] 4. A training dataset is constructed using historical peak-cutting rates, water turbulence intensity, and sensor signal-to-noise ratios. A random forest regression algorithm is used to train a model to predict peak-cutting correction factors. The current parameters are calculated every five minutes for smooth updates. When the sensor signal-to-noise ratio falls below a certain value, the peak-cutting correction factor is frozen and a backup compensation strategy is activated. This dynamic adjustment of the peak-cutting correction factor allows for timely adjustment of compensation parameters based on actual environmental changes, enhancing the system's adaptability and stability and further improving the accuracy of measurement results. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0044] Figure 1A flow chart of a river water level detection method provided in Example 1 of the present invention;

[0045] Figure 2 This is a schematic diagram of an implementation of a data acquisition device provided in Example 1 of the present invention. DETAILED DESCRIPTION

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0047] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way are interchangeable where appropriate so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0048] Embodiment one:

[0049] This embodiment provides a method for detecting river water level, which includes the following steps: Figure 1 As shown:

[0050] S01, presetting a distributed pressure sensor array in each target collection area to obtain underwater pressure data k and real-time water flow data v of multiple target collection areas of the river to be detected;

[0051] Specific, combined Figure 2 As shown, the underwater pressure data k is obtained by a distributed pressure sensor array, and the water flow data v is obtained by a flow velocity sensor. The above-mentioned distributed pressure sensor array and flow velocity sensor are both installed on a base, and then the base is distributed along the riverbed base array. The above-mentioned distributed pressure sensor array is composed of:

[0052] S11, arrange a 3×3 grid-shaped pressure sensor array along the water flow direction, with the central node being the master pressure sensor and the rest being slave pressure sensors;

[0053] S12. Each pressure sensor is installed on the riverbed base with an installation depth deviation of ≤5cm and connected to the data acquisition terminal via a waterproof cable;

[0054] S13. The master pressure sensor collects raw pressure data at a sampling rate of 16 Hz, and the slave pressure sensor collects auxiliary data at a sampling rate of 8 Hz. The multiple slave pressure sensors are phase-aligned according to the propagation delay Δt = d / v, where d is the spacing between adjacent sensors in the distributed pressure sensor array.

[0055] In the above embodiment, each pressure sensor in the distributed pressure sensor array is mounted on a riverbed foundation, with installation depth deviation controlled within a certain range. The sensors are connected to a data acquisition terminal via waterproof cables. The master and slave pressure sensors acquire data at different sampling rates, and the slave sensor data is phase-aligned based on propagation delay. This ensures the accuracy and consistency of data acquisition, providing a reliable data foundation for subsequent water level detection and wave height calculation.

[0056] S02. Estimating the pressure conduction in the water flow direction of multiple target acquisition areas and combining it with real-time water flow velocity and wave period parameters to establish a dynamic pressure water level model;

[0057] The establishment of the above-mentioned dynamic pressure water level model includes:

[0058] S21. Calculate the pressure attenuation coefficient based on the real-time water velocity v and wave period T:

[0059]

[0060] Where β is the environmental correction factor, which is 2;

[0061] S22, based on the following formula: p(x,y)=k·H(x,y)·e -α·x +b Establish a pressure conduction compensation model along the water flow direction;

[0062] Where x is the coordinate along the water flow direction, p(x,y) is the compensated pressure value at the coordinate point (x,y) in the target acquisition area, H(x,y) is the highest water level at the coordinate point (x,y) in the target acquisition area, and b is the pressure offset.

[0063] As mentioned above, by considering the influence of water flow velocity and wave period on pressure attenuation, a more accurate pressure conduction compensation model is established, which can more accurately reflect the relationship between underwater pressure and water level and improve the accuracy of water level calculation. The introduction of the environmental correction factor β enables the model to adapt to different water environments, enhancing the versatility and reliability of the model.

[0064] S03, performing signal attenuation compensation on the underwater pressure data according to the dynamic pressure water level model to obtain preliminary water level data;

[0065] Specifically, signal attenuation compensation includes:

[0066] S31, performing frequency domain analysis on the underwater pressure data k and extracting the wavelet high-frequency component Tm;

[0067] S32, according to the wave period parameter T≤2s and the wave height threshold H≤0.3m, the high frequency component is gain compensated, and the compensation amplitude is Among them, γ is the peak elimination correction factor, and τ is the time, which is 0.5s.

[0068] The dynamic adjustment of the peak elimination correction factor γ mentioned above includes:

[0069] S100, collecting historical peak elimination rate η, water flow turbulence intensity I and sensor signal-to-noise ratio SNR to build a training data set;

[0070] S101. Use the random forest regression algorithm to train the model, with the input features being [η, I, SNR] and the output being γ. The MAE of the model test set is ≤ 0.05.

[0071] S102. Calculate the current parameters and predict γn every 5 minutes, and perform smooth updates using γ(t) = 0.8·γ(t-1) + 0.2·γn. When the SNR is less than 20 dB, freeze the γ value and enable the backup compensation strategy.

[0072] Among them, γ(t) is the peak elimination correction factor at time t, γ(t-1) is the correction factor at the previous moment, and γn is the new correction factor predicted by the machine learning model.

[0073] The above embodiment extracts and compensates for the high-frequency components of the wavelet, enhances the high-frequency signal attenuated by pressure conduction loss, reduces the impact of peak clipping on the measurement results, and dynamically adjusts the compensation amplitude according to the wave period and wave height threshold, making the compensation more precise and improving the accuracy and stability of water level measurement.

[0074] Furthermore, the peak-shaving correction factor is dynamically adjusted based on historical data and real-time parameters, making the compensation more precise and adaptable to varying current and wave conditions. Furthermore, a random forest regression algorithm is used to train the model, improving its accuracy and generalization capabilities and reducing peak-shaving correction errors. A smooth update mechanism and backup compensation strategy ensure the stability and reliability of the peak-shaving correction factor, ensuring the system functions properly under various conditions.

[0075] It should be noted that the above-mentioned collected historical peak elimination rate η is the average value of the peak elimination rates collected for the river channel in the past three years, the water flow turbulence intensity I is the maximum value of the water flow turbulence collected in the past three years, and the sensor signal-to-noise ratio refers to the signal-to-noise ratio of each pressure sensor in the distributed pressure sensor array.

[0076] S04. Perform multi-source data fusion calibration on the preliminary water level data to generate river water level detection results, including wave height spectrum and water level change curve after peak elimination correction;

[0077] Specifically, the multi-source data fusion calibration mentioned above includes:

[0078] S41, obtain the acoustic ranging water level data and historical wave height database of the same detection point, wherein the acoustic ranging water level data is sampled at a rate of f s ≥10Hz acquisition, and aligning timestamps through interpolation algorithm;

[0079] S42. Establish a Kalman filter state equation, define the state vector X = [H, dH / dt]T and the observation vector Z = [Hp, Hs]T;

[0080] S43. Dynamically set an observation noise matrix according to the pressure sensor signal-to-noise ratio and the acoustic device measurement error, wherein the pressure sensor signal-to-noise ratio SNR is ≥ 30 dB and the acoustic device measurement error is ±0.5 cm.

[0081] S44. Output the smoothed calibration water level value Hc every 10 seconds. If the fusion residual ε(t) = |Hp-Hs| exceeds 2 cm for three consecutive times, a sensor abnormality alarm is triggered.

[0082] Where H is the water level value at the current moment, dH / dt is the water level change rate, Hp is the water level value initially calculated by the pressure sensor, and Hs is the water level value measured by the acoustic rangefinder.

[0083] In the above embodiment, by fusing acoustic ranging water level data with pressure sensor data and using the Kalman filter algorithm to perform data fusion and calibration, the measurement error of a single sensor is effectively eliminated, thereby improving the accuracy of water level measurement. Secondly, by setting a fusion residual threshold, the pressure sensor status is monitored in real time, and sensor anomalies are promptly detected, thus enabling abnormal quantity monitoring and ensuring the reliability and stability of the measurement system.

[0084] S05. Generate a wave energy spectrum density map based on the calibrated water level data, perform a real-time comparison between the wave energy spectrum density map and a preset threshold, and report the results based on the comparison structure.

[0085] Specifically, the wave energy spectrum density map is compared with the preset threshold in real time, and the results are reported based on the comparison structure, while the wave height change rate is calculated in real time:

[0086]

[0087] The threshold value θ is set to 0.1 m / min. If ΔH / Δt ≥ θ for 5 minutes, it is determined to be a sudden change in wave height. If ΔH / Δt < θ for 5 minutes, it is determined to be a normal floating level.

[0088] Where ΔH is the change in wave height, H(t) is the calibrated water level at the current moment, H(t-Δt) is the water level at the previous sampling moment, and Δt is the sampling time interval.

[0089] The above-mentioned embodiment, by calculating the wave height change rate in real time, can promptly detect sudden wave height changes, providing timely information support for flood prevention decision-making. This application pushes JSON messages containing information such as the location of the sudden change and wave height trend forecasts, enabling users to fully understand sudden wave height changes and take effective countermeasures. This allows relevant departments and personnel to quickly organize flood prevention efforts and reduce losses caused by flood disasters.

[0090] It should be noted that when the wave height is at a normal floating level, no JSON message will be generated for push.

[0091] In summary, this application addresses the problem of severe pressure conduction losses and peak-cutting-induced measurement distortion in pressure-type tide gauges deployed year-round in river waters with low wave heights and short periods. By using a dynamic pressure-water-level model and signal attenuation compensation, the impact of peak-cutting is effectively reduced, improving water-level measurement accuracy. Furthermore, by generating real-time water-level measurement results and wave energy spectral density maps, sudden changes in wave height can be detected and flood warnings triggered, providing strong support for flood prevention decisions.

[0092] In the first embodiment described above, a dynamic pressure water level model is established, the pressure attenuation coefficient is calculated based on the real-time water flow velocity and wave period parameters, and a pressure conduction compensation model along the water flow direction is established to compensate for signal attenuation of the underwater pressure data. This can reduce the impact of pressure conduction loss and peak elimination on the measurement results, thereby improving the measurement accuracy of small wave heights and short-period waves. Secondly, a multi-source data fusion calibration method is adopted to obtain acoustic ranging water level data and a historical wave height database at the same detection point. By establishing a Kalman filter state equation, the observation noise matrix is ​​dynamically set based on the pressure sensor signal-to-noise ratio and acoustic equipment measurement error, and a smoothed calibration water level value is output every 10 seconds. When the fusion residual exceeds the threshold three times in a row, a sensor abnormality alarm is triggered. This method fully utilizes the advantages of multiple data sources, complements and verifies each other, and can effectively improve the reliability of water level detection results. Furthermore, the wave height change rate can be calculated in real time, and a threshold can be set to determine sudden wave height changes. When a sudden change in wave height is detected for a certain period of time, a JSON message containing the location of the change and a predicted wave height trend is pushed to the target user account, and the emergency response protocol is initiated. This can promptly identify potential flood risks, provide timely and accurate information support for flood prevention decisions, and help improve the efficiency and safety of flood prevention work. Finally, a training dataset is constructed using historical peak-shaving rates, water turbulence intensity, and sensor signal-to-noise ratio. A random forest regression algorithm is used to train a model to predict the peak-shaving correction factor, and the current parameters are calculated and updated smoothly every 5 minutes. When the sensor signal-to-noise ratio falls below a certain value, the peak-shaving correction factor value is frozen and a backup compensation strategy is activated. This method, through dynamic adjustment of the peak-shaving correction factor, allows the compensation parameters to be adjusted in a timely manner according to actual environmental changes, enhancing the adaptability and stability of the system and further improving the accuracy of measurement results.

[0093] Embodiment 2:

[0094] An embodiment of the present invention provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by a processor to implement the steps:

[0095] A distributed pressure sensor array is preset in each target collection area to obtain underwater pressure data k and real-time water flow data v of multiple target collection areas of the river to be detected;

[0096] Based on the estimation of the pressure conduction in the water flow direction of the target acquisition areas, and in combination with the real-time water flow velocity and wave period parameters, a dynamic pressure water level model is established;

[0097] Performing signal attenuation compensation on the underwater pressure data according to the dynamic pressure water level model to obtain preliminary water level data;

[0098] Performing multi-source data fusion calibration on the preliminary water level data to generate a river water level detection result, the result including a wave height spectrum and a water level change curve after peak elimination correction;

[0099] A wave energy spectrum density map is generated based on the calibrated water level data, and a real-time comparison is performed between the wave energy spectrum density map and a preset threshold value, and the result is reported based on the comparison structure.

[0100] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0101] Those skilled in the art will clearly understand that for the sake of convenience and brevity in description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0102] Embodiment three:

[0103] An embodiment of the present invention provides an electronic device, including a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the following steps:

[0104] A distributed pressure sensor array is preset in each target collection area to obtain underwater pressure data k and real-time water flow data v of multiple target collection areas of the river to be detected;

[0105] Based on the estimation of the pressure conduction in the water flow direction of the target acquisition areas, and in combination with the real-time water flow velocity and wave period parameters, a dynamic pressure water level model is established;

[0106] Performing signal attenuation compensation on the underwater pressure data according to the dynamic pressure water level model to obtain preliminary water level data;

[0107] Performing multi-source data fusion calibration on the preliminary water level data to generate a river water level detection result, the result including a wave height spectrum and a water level change curve after peak elimination correction;

[0108] A wave energy spectrum density map is generated based on the calibrated water level data, and a real-time comparison is performed between the wave energy spectrum density map and a preset threshold value, and the result is reported based on the comparison structure.

[0109] The above description is merely a preferred embodiment of the present invention and does not constitute any form of limitation to the present invention. Although the present invention has been disclosed as a preferred embodiment as above, it is not intended to limit the present invention. Any technician familiar with the present profession can make some changes or modifications to equivalent embodiments of the technical contents disclosed above without departing from the scope of the technical solution of the present invention. However, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention are still within the scope of the technical solution of the present invention.

Claims

1. A method for detecting river water level, characterized in that: The method comprises the following steps: S01, presetting a distributed pressure sensor array in each target collection area to obtain underwater pressure data k and real-time water flow data v of multiple target collection areas of the river to be detected; S02. Establishing a dynamic pressure water level model based on the estimated pressure conduction in the water flow direction of the target acquisition areas and in combination with real-time water flow velocity and wave period parameters; S03, performing signal attenuation compensation on the underwater pressure data according to the dynamic pressure water level model to obtain preliminary water level data; S04, performing multi-source data fusion calibration on the preliminary water level data to generate a river water level detection result, the result including a wave height spectrum and a water level change curve after peak elimination correction; S05. Generate a wave energy spectrum density map based on the calibrated water level data, perform a real-time comparison between the wave energy spectrum density map and a preset threshold, and report the result based on the comparison structure.

2. A river water level detection method according to claim 1, characterized in that: The distributed pressure sensor array in step S01 comprises: S11, arrange a 3×3 grid-shaped pressure sensor array along the water flow direction, with the central node being the master pressure sensor and the rest being slave pressure sensors; S12. Each pressure sensor is installed on the riverbed base with an installation depth deviation of ≤5cm and connected to the data acquisition terminal via a waterproof cable; S13. The master pressure sensor collects raw pressure data at a sampling rate of 16 Hz, and the slave pressure sensor collects auxiliary data at a sampling rate of 8 Hz. The multiple slave pressure sensors are phase-aligned according to the propagation delay Δt=d / v, where d is the spacing between adjacent sensors in the distributed pressure sensor array.

3. A river water level detection method according to claim 1, characterized in that: The establishment of the dynamic pressure water level model in step S02 includes: S21. Calculate the pressure attenuation coefficient based on the real-time water velocity v and wave period T: Where β is the environmental correction factor, which is 2; S22, based on the following formula: p(x,y)=k·H(x,y)·e -α·x +b Establish a pressure conduction compensation model along the water flow direction; Where x is the coordinate along the water flow direction, p(x,y) is the compensated pressure value at the coordinate point (x,y) of the target acquisition area, H(x,y) is the highest water level at the coordinate point (x,y) of the target acquisition area, and b is the pressure offset.

4. A river water level detection method according to claim 3, characterized in that: The signal attenuation compensation in step S03 includes: S31, performing frequency domain analysis on the underwater pressure data k to extract the wavelet high-frequency component Tm; S32, according to the wave period parameter T≤2s and the wave height threshold H≤0.3m, the high frequency component is gain compensated, and the compensation amplitude is Among them, γ is the peak elimination correction factor, and τ is the time, which is 0.5s.

5. A river water level detection method according to claim 1, characterized in that: The dynamic adjustment of the peak elimination correction factor γ includes: S100, collecting historical peak elimination rate η, water flow turbulence intensity I and sensor signal-to-noise ratio SNR to build a training data set; S101. Use the random forest regression algorithm to train the model, with the input features being [η, I, SNR] and the output being γ. The MAE of the model test set is ≤ 0.

05. S102. Calculate the current parameters and predict γn every 5 minutes, and perform smooth updates using γ(t) = 0.8·γ(t-1) + 0.2·γn. When the SNR is less than 20 dB, freeze the γ value and enable the backup compensation strategy. Among them, γ(t) is the peak elimination correction factor at time t, γ(t-1) is the correction factor at the previous moment, and γn is the new correction factor predicted by the machine learning model.

6. A river water level detection method according to claim 1, characterized in that: The multi-source data fusion calibration in step S04 includes: S41, obtaining the acoustic ranging water level data and the historical wave height database of the same detection point, wherein the acoustic ranging water level data is sampled at a rate of f s ≥10Hz acquisition, and aligning timestamps through interpolation algorithm; S42. Establish a Kalman filter state equation, define the state vector X = [H, dH / dt]T and the observation vector Z = [Hp, Hs]T; S43. Dynamically set an observation noise matrix according to a pressure sensor signal-to-noise ratio (SNR) and an acoustic device measurement error, wherein the pressure sensor signal-to-noise ratio (SNR) is ≥ 30 dB and the acoustic device measurement error is ±0.5 cm. S44. Output the smoothed calibration water level value Hc every 10 seconds. If the fusion residual ε(t) = |Hp-Hs| exceeds 2 cm for three consecutive times, a sensor abnormality alarm is triggered. Where H is the water level value at the current moment, dH / dt is the water level change rate, Hp is the water level value initially calculated by the pressure sensor, and Hs is the water level value measured by the acoustic rangefinder.

7. A river water level detection method according to claim 1, characterized in that: In step S05, the wave energy spectrum density map is compared with a preset threshold in real time, and the result is reported according to the comparison structure, including calculating the real-time wave height change rate: The threshold value θ is set to 0.1 m / min. If ΔH / Δt ≥ θ for 5 minutes, it is determined to be a sudden change in wave height. If ΔH / Δt < θ for 5 minutes, it is determined to be a normal floating level. Where ΔH is the change in wave height, H(t) is the calibrated water level at the current moment, H(t-Δt) is the water level at the previous sampling moment, and Δt is the sampling time interval.

8. A non-transitory computer-readable storage medium, wherein at least one instruction or at least one program is stored in the non-transitory computer-readable storage medium, characterized in that: The at least one instruction or the at least one program segment is loaded and executed by the processor to implement the steps of the river water level detection method as described in any one of claims 1 to 7.

9. An electronic device, characterized in that: It includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by the processor to implement the steps of the river water level detection method as described in any one of claims 1 to 7.

Images