Method and system for detecting concentration of suspended solids in canal water sample

Abstract

The invention relates to the technical field of water quality detection, and discloses a method and system for detecting the concentration of suspended solids in a canal water sample, and the method comprises the following steps: obtaining environmental parameters of a canal water body in real time, and synchronously collecting a light scattering intensity signal and an acoustic echo attenuation signal of the canal water body through a near-infrared optical sensor and an acoustic Doppler flow velocity profiler; calculating an initial suspended matter concentration value based on the light scattering intensity signal and the acoustic echo attenuation signal; correcting the initial suspended matter concentration value through a dynamic compensation algorithm according to the environmental parameters of the canal water body, and generating concentration data before calibration; inputting the concentration data before calibration into a pre-trained machine learning model, and outputting a final suspended matter concentration value; wherein the pre-trained machine learning model comprises a random forest regression model and a gradient boosting decision tree model. According to the scheme, the detection accuracy under the complex hydrological condition and the real-time response capability of the model under the disturbance scene can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of water quality detection, and in particular relates to a method and system for detecting the concentration of suspended matter in a canal water sample. Background Art

[0002] Measuring suspended solids concentration in canal waters is a critical component of environmental monitoring and aquatic ecosystem management. Traditional methods struggle to overcome the combined interference of optical and acoustic signals in dynamic water environments. This is particularly true in scenarios with frequent waterway disturbances and variable hydrological conditions. Sensor measurements are susceptible to temperature fluctuations, sudden changes in ion concentrations, and water turbulence. This leads to insufficient reliability of monitoring data in practical applications, hindering the accuracy of water quality assessments and the effectiveness of pollution prevention and control decisions.

[0003] At present, traditional solutions only use fixed coefficients to compensate for the single influence of temperature or conductivity, and are unable to cope with multi-parameter coupling interference such as water temperature changes, ionic strength fluctuations, and sudden water flow disturbances, resulting in compensation failure under complex working conditions; conventional algorithms lack adaptive adjustment mechanisms under disturbance conditions such as heavy rain and ship navigation, and the model output lags behind the rapid changes in hydrological conditions.

[0004] Therefore, there is an urgent need to develop a method and system for detecting the suspended matter concentration in canal water samples, which can improve the detection accuracy under complex hydrological conditions and the real-time response capability of the model under disturbance scenarios. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention provides a method and system for detecting the suspended matter concentration in canal water samples, which can improve the detection accuracy under complex hydrological conditions and the real-time response capability of the model under disturbance scenarios.

[0006] The present invention provides a method for detecting the concentration of suspended matter in a canal water sample, the method comprising the following steps: S1. Real-time acquisition of environmental parameters of the canal water body by synchronously collecting the light scattering intensity signal and acoustic echo attenuation signal of the canal water body through a near-infrared optical sensor and an acoustic Doppler current profiler; S2. Calculate the initial suspended solids concentration value based on the light scattering intensity signal and the acoustic echo attenuation signal; S3. According to the environmental parameters of the canal water body, the initial suspended solids concentration value is corrected by a dynamic compensation algorithm to generate pre-calibration concentration data; S4. Input the pre-calibration concentration data into a pre-trained machine learning model to output the final suspended matter concentration value; wherein the pre-trained machine learning model includes a random forest regression model and a gradient boosting decision tree model.

[0007] Furthermore, in S2, the calculation formula for the initial suspended solids concentration value is as follows: ; Among them, C init represents the initial suspended solids concentration, k1 represents the optical scattering coefficient, k2 represents the acoustic attenuation coefficient, I scatter represents the light scattering intensity signal, α measured Represents the measured acoustic attenuation value, α water Indicates the background attenuation value of water body.

[0008] Furthermore, the calculation formula of the water background attenuation value is as follows: ; Where b represents the ground state constant, β represents the temperature sensitivity coefficient of sound wave attenuation, T represents the water temperature of the canal water body, and f represents the sound wave frequency of the acoustic Doppler current profiler.

[0009] Furthermore, in S1, the environmental parameters include the water temperature, conductivity and pH value of the canal water body.

[0010] Furthermore, in S3, the initial suspended solids concentration value is corrected by a compensation algorithm based on the environmental parameters of the canal water body to generate pre-calibration concentration data, including: S31, performing water temperature linear compensation on the initial suspended solids concentration value according to the water temperature of the canal water body to obtain compensated concentration data; The calculation formula for water temperature linear compensation is as follows: ; Among them, C temp represents the concentration data after compensation, β t represents the water temperature compensation factor, T represents the water temperature of the canal water body, T ref Indicates the reference water temperature; S32. When the conductivity is greater than the preset conductivity and the pH value is less than the preset pH value, performing ion interference correction on the compensated concentration data to obtain pre-calibration concentration data; otherwise, directly outputting the compensated concentration data as pre-calibration concentration data; The calculation formula for ion interference correction is as follows: ; ; Among them, C cal Indicates the concentration data before calibration, K ion represents the ion interference correction factor, α represents the ion interference sensitivity coefficient, EC represents the conductivity of the canal water, and EC ref Indicates reference conductivity.

[0011] Furthermore, in S4, the pre-calibration concentration data is input into the pre-trained machine learning model to output the final suspended solids concentration value, including: S41, obtaining turbidity time series data and three-dimensional flow velocity vector of the canal water body; S42, determining an operating mode based on the turbidity time series data of the canal water body and the three-dimensional flow velocity vector; S43. According to the working mode, the concentration data before calibration is input into the corresponding pre-trained machine learning model.

[0012] Furthermore, in S42, the operating mode is determined based on the turbidity time series data of the canal water body and the three-dimensional flow velocity vector, including: The turbidity change slope is determined based on the turbidity time series data, and the flow velocity pulsation intensity is determined based on the three-dimensional flow velocity vector; If the turbidity change slope is greater than or equal to a first preset value, and / or the flow velocity pulsation intensity is greater than or equal to a second preset value, then the disturbance mode is determined; Otherwise, it is determined to be the normal working mode.

[0013] Furthermore, in S43, according to the working mode, the pre-calibration concentration data is input into the corresponding pre-trained machine learning model, including: If it is a normal working mode, the concentration data before calibration is input into the corresponding pre-trained random forest regression model; If it is a disturbance operating mode, the concentration data before calibration is input into the corresponding pre-trained gradient boosting decision tree model.

[0014] The present invention also provides a system for detecting the concentration of suspended matter in canal water samples, which is used to perform the above-mentioned method for detecting the concentration of suspended matter in canal water samples. The system includes the following modules: The data acquisition module is used to obtain the environmental parameters of the canal water in real time. It uses a near-infrared optical sensor and an acoustic Doppler current profiler to synchronously collect the light scattering intensity signal and acoustic echo attenuation signal of the canal water. An initial value calculation module is used to calculate the initial suspended matter concentration value based on the light scattering intensity signal and the acoustic echo attenuation signal; A correction module is used to correct the initial suspended solids concentration value through a dynamic compensation algorithm according to the environmental parameters of the canal water body to generate pre-calibration concentration data; The output module is used to input the pre-calibration concentration data into a pre-trained machine learning model and output the final suspended matter concentration value; wherein the pre-trained machine learning model includes a random forest regression model and a gradient boosting decision tree model.

[0015] The embodiments of the present invention have the following technical effects: The present invention improves the accuracy of signal analysis through the joint calculation of optical scattering and acoustic attenuation, and conducts layered correction for the coupling effects of multiple parameters such as temperature, conductivity, and pH by constructing a dynamic compensation chain of water temperature linear compensation and ion interference correction, breaking through the limitations of single compensation. Through dual-signal collaboration and graded compensation of environmental parameters, it effectively resists complex interferences such as sudden temperature changes, high ionic strength, and acidic water bodies, and improves data reliability under complex hydrological conditions; based on the working condition pattern recognition of turbidity time series and flow velocity vector, it drives the adaptive switching of random forest regression and gradient boosting decision tree to achieve dynamic model optimization; based on the intelligent working condition recognition of flow velocity pulsation intensity and turbidity slope, it ensures the real-time response capability of the model in disturbance scenarios such as intensive navigation periods and heavy rain impacts. The adaptive switching mechanism of the machine learning model can significantly reduce the risk of misjudgment of extreme working conditions, providing stable technical support for long-term monitoring of canal water quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the specific embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 is a flow chart of a method for detecting suspended matter concentration in a canal water sample provided by an embodiment of the present invention; Figure 2 This is a logic diagram of a method for correcting an initial suspended solids concentration value through a dynamic compensation algorithm provided by an embodiment of the present invention; Figure 3 It is a structural diagram of a system for detecting the concentration of suspended matter in canal water samples provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0018] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention are described clearly and completely below. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. All other embodiments derived by persons of ordinary skill in the art based on the embodiments of the present invention without inventive effort are also within the scope of protection of the present invention.

[0019] The embodiment of the present invention provides a method for detecting the concentration of suspended matter in a canal water sample. Figure 1 This is a flow chart of a method for detecting suspended matter concentration in a canal water sample provided by an embodiment of the present invention. Figure 1 , the method comprises the following steps: S1. Real-time acquisition of environmental parameters of the canal water body. The light scattering intensity signal and acoustic echo attenuation signal of the canal water body are synchronously collected through near-infrared optical sensors and acoustic Doppler current profilers.

[0020] In some embodiments, the environmental parameters may include the water temperature, conductivity and pH value of the canal water body, among which the water temperature parameter can be measured in real time by using an immersion temperature sensor that directly contacts the water body. The change in water temperature will simultaneously affect the Brownian motion intensity of the suspended particles and the attenuation characteristics of sound wave propagation; the conductivity parameter can be measured in situ using an electrode sensor, which indirectly reflects the concentration of dissolved ions by analyzing the conductivity of the water body. This parameter is coupled with the surface charge distribution of the suspended particles; the pH value parameter can be obtained through a glass electrode sensor, which reflects the activity level of hydrogen ions in the water body and will change the stability and aggregation morphology of suspended colloids in acidic or alkaline environments.

[0021] A near-infrared optical sensor emits near-infrared light and receives scattered light from suspended particles in the water, generating a light scattering intensity signal. An acoustic Doppler current profiler simultaneously transmits ultrasonic waves into the water and captures the attenuation of the acoustic echo caused by particles, ensuring precise temporal alignment of the two signals to avoid sampling delays. The complementary nature of the optical and acoustic signals enhances data integrity, providing a foundation for subsequent calculations.

[0022] S2. Calculate the initial suspended solids concentration value based on the light scattering intensity signal and the acoustic echo attenuation signal.

[0023] In some embodiments, the calculation formula for the initial suspended solids concentration value is as follows: ; Among them, C init represents the initial suspended solids concentration, k1 represents the optical scattering coefficient, k2 represents the acoustic attenuation coefficient, I scatter represents the light scattering intensity signal, α measured Represents the measured acoustic attenuation value, α water Indicates the background attenuation value of water body; The calculation formula of water background attenuation value is as follows: ; Here, b represents the ground-state constant, which represents the fundamental acoustic energy attenuation level of pure water at a specific frequency under standard temperature conditions and can be measured in the laboratory. β represents the temperature sensitivity coefficient of acoustic attenuation, which describes the retardation effect of the increased thermal motion of water molecules caused by temperature changes on the transfer of acoustic energy. Its physical significance lies in the fact that rising temperature increases the frequency of molecular collisions, thereby increasing the rate of acoustic energy loss. T represents the water temperature of the canal water body, and f represents the acoustic wave frequency of the acoustic Doppler current profiler. High-frequency sound waves are easily affected by the relaxation effect of water molecules, while low-frequency sound waves are more easily scattered by bubbles. Targeted correction of the background attenuation value is achieved through frequency parameters. By dynamically modeling the inherent acoustic properties of water bodies, the additional attenuation caused by suspended matter and the background absorption of water bodies can be accurately distinguished, the systematic interference of temperature and sound frequency on acoustic signals can be eliminated, the coupling effect of environmental variables on acoustic signals can be weakened, and the decoupling accuracy of acoustic signals can be improved.

[0024] The optical scattering coefficient, k1, is calibrated in the laboratory and represents the intensity of the light scattering response caused by a unit concentration of suspended solids under specific water quality conditions. The acoustic attenuation coefficient, k2, is also calibrated in the laboratory and describes the absorption characteristics of suspended solids to sound wave energy. The water background attenuation value represents the inherent absorption of sound waves by pure water and must be subtracted from the measured total attenuation to avoid systematic errors. This separation process clearly distinguishes the contribution of suspended solids from the acoustic interference of water molecules themselves.

[0025] The real-time collected light scattering intensity signal is multiplied by the optical scattering coefficient to obtain the light signal contribution value. Simultaneously, the net attenuation value of the acoustic signal is calculated. This is done by subtracting the water background attenuation value from the measured acoustic attenuation value to eliminate water background interference, and then multiplying the result by the acoustic attenuation coefficient to obtain the acoustic signal contribution value. Finally, the two signal contribution values ​​are weighted and fused to generate the initial suspended solids concentration value. Through this dual-modal signal collaborative calculation mechanism, when one type of signal is distorted due to environmental factors (such as high turbidity causing optical signal saturation), the other signal can provide compensatory data support, significantly improving the anti-interference ability of the initial calculation result. The optical scattering intensity signal reflects the particle size distribution, while the acoustic echo attenuation signal indicates the suspended solids density characteristics. By weightedly fusing the two signals to output the initial concentration value, the deviation caused by reliance on a single signal is avoided, thereby improving the robustness of the initial result.

[0026] S3. According to the environmental parameters of the canal water body, the initial suspended matter concentration value is corrected through a dynamic compensation algorithm to generate pre-calibration concentration data.

[0027] In some embodiments, Figure 2 This is a logic diagram of a method for correcting the initial suspended solids concentration value by a dynamic compensation algorithm provided by an embodiment of the present invention, see Figure 2 , S3 includes the following sub-steps: S31, performing water temperature linear compensation on the initial suspended solids concentration value according to the water temperature of the canal water body to obtain compensated concentration data; The calculation formula for water temperature linear compensation is as follows: ; Among them, C temp represents the concentration data after compensation, β t represents the water temperature compensation factor, that is, the response sensitivity of suspended matter concentration to temperature changes, T represents the water temperature of the canal water body, T ref Indicates the reference water temperature.

[0028] For example, in a low-temperature environment, the particle agglomeration effect is intensified, the optical scattering signal is enhanced, and the acoustic attenuation is weakened. At this time, the compensation algorithm automatically adjusts the output weight according to the positive and negative values ​​of the temperature difference, so that the compensated concentration data is closer to the actual distribution state of the particles.

[0029] S32. When the conductivity is greater than the preset conductivity and the pH value is less than the preset pH value, performing ion interference correction on the compensated concentration data to obtain pre-calibration concentration data; otherwise, directly outputting the compensated concentration data as pre-calibration concentration data; The calculation formula for ion interference correction is as follows: ; ; Among them, C cal Indicates the concentration data before calibration, K ion represents the ion interference correction factor, α represents the ion interference sensitivity coefficient, EC represents the conductivity of the canal water, and EC ref Indicates reference conductivity.

[0030] Specifically, if the conductivity detection value continuously exceeds the threshold and the pH is below the lower limit, it indicates that the water body has a complex scenario where high ionic strength and acidic conditions are superimposed. At this time, the ion interference sensitivity coefficient is applied. This coefficient establishes a logarithmic response relationship based on the ion interference sensitivity coefficient, so that the correction factor increases nonlinearly with the change in conductivity. In the calculation formula of ion interference correction, the reference conductivity represents the interference-free baseline state. The deviation of the actual conductivity from the baseline is amplified by the sensitivity coefficient to generate the correction intensity parameter. Finally, the temperature compensation result is multiplied by the correction factor to output the pre-calibration concentration data; if there are no complex interference conditions, the compensated concentration data is directly transmitted.

[0031] A dynamic compensation algorithm is implemented to correct initial concentration values ​​based on environmental parameters. The impact of water temperature on suspended solids concentration is reflected in the fact that changes in water temperature can cause particle aggregation or dispersion. The compensation algorithm uses a linear model to adjust the initial value based on the measured water temperature, with the reference water temperature set as a fixed benchmark to ensure smooth temperature drift suppression. The algorithm also determines whether to perform ion interference correction through a collaborative analysis of conductivity and pH parameters. When an abnormally high conductivity or low pH is detected, the algorithm automatically applies a correction factor to offset the masking effect of sudden changes in ionic strength on the signal, generating pre-calibration concentration data. This layered compensation mechanism improves data accuracy, addresses environmental interference coupling, and ensures the generation of reliable intermediate data under both normal and extreme conditions.

[0032] Specifically, water temperature fluctuations directly affect the dynamic characteristics of particles, and the combination of high conductivity and low pH will change the surface potential of suspended matter, resulting in artificially high optical scattering signals and phase shifts in acoustic signals. The dynamic compensation algorithm separates the independent action paths of temperature and ion interference, avoiding both insufficient compensation of a single parameter and overcompensation. This embodiment can significantly reduce the risk of measurement distortion in scenarios with sudden seasonal temperature changes or industrial pollution discharges. For example, during the spring snowmelt period, when low temperatures and snow-melting salt input occur simultaneously, dual-stage compensation can simultaneously suppress temperature drift and chloride ion interference, ensuring data output stability and providing high-quality input for machine learning models.

[0033] S4. Input the pre-calibration concentration data into the pre-trained machine learning model and output the final suspended matter concentration value.

[0034] Among them, the pre-trained machine learning models include random forest regression models and gradient boosting decision tree models. Model training is based on historical water sample datasets covering different hydrological conditions to ensure generalization ability. The historical water sample datasets used for pre-training must cover at least the following multi-dimensional features: hydrological condition range: water temperature gradient 5-35℃ (step size 5℃), conductivity range 100-1500μS / cm (including salinity mutation scenarios), pH fluctuation range 6.0-8.5, turbidity dynamic range 10-500NTU, disturbance event samples: ship navigation disturbance (flow velocity pulsation intensity ≥0.3m 2 / s 2 ), rainstorm runoff impact (turbidity change slope ≥ 15NTU / min), and simultaneously obtain optical scattering intensity, acoustic attenuation value, three-dimensional flow velocity vector and laboratory standard weight method concentration value, with no less than 50 groups of valid samples for each type of hydrological condition; the random forest regression model is good at handling stable states, while the gradient boosting decision tree model is adapted to high-variability scenarios. The corresponding machine learning model is selected according to different working conditions to enhance adaptability to complex water environments, reduce the impact of environmental factors, and improve the stability and repeatability of canal water quality monitoring.

[0035] In some embodiments, S4 includes the following sub-steps: S41. Obtain turbidity time series data and three-dimensional flow velocity vector of the canal water body.

[0036] Specifically, sudden changes in flow velocity can stir up sediment and increase turbidity, and abnormal changes in turbidity are often accompanied by changes in flow velocity patterns in specific directions. Turbidity time series data is acquired through continuous sampling at fixed intervals using an optical turbidity sensor. This data sequence reflects the dynamic evolution of water turbidity, and its changing trends imply information about events such as velocity shocks or external pollution inputs. Three-dimensional velocity vectors are measured in layers on vertical sections using an acoustic Doppler current profiler. The horizontal, longitudinal, and vertical velocity components are obtained through beam array analysis to form a spatial velocity distribution model.

[0037] S42. Determine the operating mode based on the turbidity time series data of the canal water body and the three-dimensional flow velocity vector.

[0038] Specifically include: The turbidity change slope is determined based on the turbidity time series data, and the flow velocity pulsation intensity is determined based on the three-dimensional flow velocity vector; Specifically, the turbidity slope is calculated by differencing time series data, representing the rate of turbidity change per unit time. High slope values ​​indicate sudden disturbances such as heavy rain or ship agitation. Velocity pulsation intensity is calculated by calculating the variance of the three-dimensional velocity vector, reflecting the severity of water turbulence.

[0039] If the turbidity change slope is greater than or equal to a first preset value, and / or the flow velocity pulsation intensity is greater than or equal to a second preset value, it is determined to be a disturbed operating mode; otherwise, it is determined to be a normal operating mode.

[0040] Among them, the first preset value is the turbidity change slope threshold, which is used to determine the critical value of sudden water disturbance events, and the second preset value is the flow velocity pulsation intensity threshold, which is used to quantify the critical value of water turbulence intensity; illustratively, based on the turbidity time series data of historical disturbance events in the canal (such as ship navigation, rainstorm runoff), the distribution range of the turbidity change slope when the event occurs is calculated, and the turbidity change slope corresponding to the minimum detectable event intensity of the sudden change in suspended matter concentration in the canal water body is selected as the first preset value; the three-dimensional flow velocity vector data during ship navigation and rainstorm runoff is obtained by using an acoustic Doppler current profiler, and the flow velocity variance is calculated. The pulsation intensity corresponding to the critical energy density of the sudden change in water body kinetic energy can be determined through turbulence energy spectrum analysis as the second preset value.

[0041] That is, when the slope of turbidity change remains high or the flow velocity pulsation intensity shows a violent fluctuation characteristic, the disturbance condition determination flag is triggered; otherwise, the normal operating mode is maintained. Traditional methods that rely solely on the absolute value of turbidity are prone to overlooking slowly changing pollution events, while isolated flow velocity monitoring cannot distinguish between natural turbulence and human disturbances. For example, this embodiment uses a dual-channel collaborative verification mechanism to enable accurate judgments to be triggered by a combination of slight turbidity changes (such as propellers stirring up bottom mud) and sudden changes in vertical flow velocity in low-speed ship navigation scenarios; when the canal water body is in a stable hydrological state during the dry season, the normal operating condition flag is maintained when the turbidity curve is flat and the low turbulence intensity is superimposed.

[0042] S43. According to the working mode, the concentration data before calibration is input into the corresponding pre-trained machine learning model.

[0043] Specifically include: If it is a normal working mode, the concentration data before calibration is input into the corresponding pre-trained random forest regression model; If it is a disturbance operating mode, the concentration data before calibration is input into the corresponding pre-trained gradient boosting decision tree model.

[0044] In some embodiments, a random forest regression model call instruction is triggered in the normal operating mode. The model is trained and generated by a historical normal hydrological data set, and its internal structure includes a parallel operation mechanism of multiple decision trees; each decision tree constructs branching rules based on the characteristics of the optical acoustic signal, such as basic nodes such as the division of the light scattering intensity signal interval and the judgment of the acoustic attenuation value threshold. The model averages the output results of multiple trees through an integrated learning strategy to reduce the sensitivity of a single tree to noise data. When the pre-calibration concentration data is input, the model first parses the optical signal feature branch to determine the interval range of the light scattering intensity signal; then enters the acoustic attenuation value branch and performs secondary classification based on the acoustic attenuation after temperature compensation; finally, the multi-path decision results are converged to the regression output layer to generate a concentration prediction value. This model has significant advantages under stable hydrological conditions. For example, when the turbidity fluctuations in the canal are smooth during the flat water period, the model can accurately capture the linear correlation characteristics of the optical acoustic signal.

[0045] In some embodiments, the model switches to a gradient boosting decision tree model in the disturbance mode. This model utilizes a serialized decision tree construction approach, with subsequent trees continuously correcting the prediction residuals of preceding trees, forming an iterative optimization chain. During the model training phase, data from disturbance scenarios such as ship navigation and rainstorm runoff impact can be introduced, enabling the decision tree to learn signal distortion compensation rules for high-turbulence conditions. In addition to receiving pre-calibration concentration data, the model inputs also include real-time turbidity slope and three-dimensional velocity vector directional characteristic parameters. The first decision tree generates a baseline prediction based on the turbidity slope; the second tree corrects the optical signal distortion component based on the velocity pulsation intensity characteristics; and subsequent trees optimize the acoustic signal phase offset error layer by layer. For example, when a sudden change in the vertical velocity component triggers bubble interference, the model automatically reduces the weight coefficient of the acoustic attenuation value based on historically learned bubble shielding effect rules. This chain correction mechanism is particularly critical in flood-prone scenarios, effectively counteracting the blockage of the near-infrared optical path by entrained debris.

[0046] The machine learning cluster in this embodiment maintains low-power and high-precision operation of the random forest model under stable water flow conditions in the dry season, and the gradient boosting decision tree model takes over the processing tasks in a timely manner when sudden disturbances occur in the flood season. The dual-model collaborative mechanism can provide adaptive technical support for the full-cycle monitoring of the canal hydrology.

[0047] The present invention improves the accuracy of signal analysis through the joint calculation of optical scattering and acoustic attenuation, and conducts layered correction for the coupling effects of multiple parameters such as temperature, conductivity, and pH by constructing a dynamic compensation chain of water temperature linear compensation and ion interference correction, breaking through the limitations of single compensation. Through dual-signal collaboration and graded compensation of environmental parameters, it effectively resists complex interferences such as sudden temperature changes, high ionic strength, and acidic water bodies, and improves data reliability under complex hydrological conditions; based on the working condition pattern recognition of turbidity time series and flow velocity vector, it drives the adaptive switching of random forest regression and gradient boosting decision tree to achieve dynamic model optimization; based on the intelligent working condition recognition of flow velocity pulsation intensity and turbidity slope, it ensures the real-time response capability of the model in disturbance scenarios such as intensive navigation periods and heavy rain impacts. The adaptive switching mechanism of the machine learning model can significantly reduce the risk of misjudgment of extreme working conditions, providing stable technical support for long-term monitoring of canal water quality.

[0048] The embodiment of the present invention further provides a system for detecting the concentration of suspended matter in a canal water sample, which is used to execute the above-mentioned method for detecting the concentration of suspended matter in a canal water sample. Figure 3 This is a structural diagram of a detection system for suspended solids concentration in canal water samples provided by an embodiment of the present invention, see Figure 3 , the system includes the following modules: The data acquisition module is used to obtain the environmental parameters of the canal water in real time. It uses a near-infrared optical sensor and an acoustic Doppler current profiler to synchronously collect the light scattering intensity signal and acoustic echo attenuation signal of the canal water. An initial value calculation module is used to calculate the initial suspended matter concentration value based on the light scattering intensity signal and the acoustic echo attenuation signal; A correction module is used to correct the initial suspended solids concentration value through a dynamic compensation algorithm according to the environmental parameters of the canal water body to generate pre-calibration concentration data; The output module is used to input the pre-calibration concentration data into a pre-trained machine learning model and output the final suspended matter concentration value; wherein the pre-trained machine learning model includes a random forest regression model and a gradient boosting decision tree model.

[0049] The system embodiment corresponds one-to-one to the above method embodiment, and will not be repeated here.

[0050] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting the concentration of suspended matter in a canal water sample, characterized in that: The method comprises the following steps: S1. Real-time acquisition of environmental parameters of the canal water body by synchronously collecting the light scattering intensity signal and acoustic echo attenuation signal of the canal water body through a near-infrared optical sensor and an acoustic Doppler current profiler; S2. Calculating an initial suspended matter concentration value based on the light scattering intensity signal and the acoustic echo attenuation signal; S3. Correcting the initial suspended solids concentration value using a dynamic compensation algorithm based on the environmental parameters of the canal water body to generate pre-calibration concentration data; S4. Inputting the pre-calibration concentration data into a pre-trained machine learning model to output a final suspended solids concentration value; wherein the pre-trained machine learning model includes a random forest regression model and a gradient boosting decision tree model.

2. The method for detecting the concentration of suspended matter in a canal water sample according to claim 1, wherein: In S2, the calculation formula for the initial suspended solids concentration value is as follows: ； Among them, C init represents the initial suspended solids concentration, k1 represents the optical scattering coefficient, k2 represents the acoustic attenuation coefficient, I scatter represents the light scattering intensity signal, α measured Represents the measured acoustic attenuation value, α water Indicates the background attenuation value of water body.

3. The method for detecting the concentration of suspended matter in a canal water sample according to claim 2, wherein: The calculation formula of the water body background attenuation value is as follows: ； Where b represents the ground state constant, β represents the temperature sensitivity coefficient of sound wave attenuation, T represents the water temperature of the canal water body, and f represents the sound wave frequency of the acoustic Doppler current profiler.

4. The method for detecting the concentration of suspended matter in a canal water sample according to claim 1, wherein: In S1, the environmental parameters include the water temperature, conductivity and pH value of the canal water body.

5. The method for detecting the concentration of suspended matter in a canal water sample according to claim 4, characterized in that: In S3, the initial suspended solids concentration value is corrected by a compensation algorithm according to the environmental parameters of the canal water body to generate pre-calibration concentration data, including: S31, performing water temperature linear compensation on the initial suspended solids concentration value according to the water temperature of the canal water body to obtain compensated concentration data; The calculation formula for water temperature linear compensation is as follows: ； Among them, C init Indicates the initial suspended solids concentration value, C temp represents the concentration data after compensation, β t represents the water temperature compensation factor, T represents the water temperature of the canal water body, T ref Indicates the reference water temperature; S32. When the conductivity is greater than a preset conductivity and the pH value is less than a preset pH value, performing ion interference correction on the compensated concentration data to obtain pre-calibration concentration data; otherwise, directly outputting the compensated concentration data as pre-calibration concentration data; The calculation formula for ion interference correction is as follows: ； ； Among them, C cal Indicates the concentration data before calibration, K ion represents the ion interference correction factor, α represents the ion interference sensitivity coefficient, EC represents the conductivity of the canal water, and EC ref Indicates reference conductivity.

6. The method for detecting the concentration of suspended matter in a canal water sample according to claim 1, characterized in that: In S4, the pre-calibration concentration data is input into a pre-trained machine learning model to output a final suspended solids concentration value, including: S41, obtaining turbidity time series data and three-dimensional flow velocity vector of the canal water body; S42, determining an operating mode based on the turbidity time series data and the three-dimensional flow velocity vector of the canal water body; S43. According to the operating mode, the pre-calibration concentration data is input into a corresponding pre-trained machine learning model.

7. The method for detecting the concentration of suspended matter in a canal water sample according to claim 6, characterized in that: In S42, determining the operating mode based on the turbidity time series data of the canal water body and the three-dimensional flow velocity vector includes: determining a turbidity change slope based on the turbidity time series data, and determining a flow velocity pulsation intensity based on the three-dimensional flow velocity vector; If the turbidity change slope is greater than or equal to a first preset value, and / or the flow velocity pulsation intensity is greater than or equal to a second preset value, then the disturbance mode is determined; Otherwise, it is determined to be the normal working mode.

8. The method for detecting the concentration of suspended matter in a canal water sample according to claim 7, characterized in that: In S43, according to the working mode, the pre-calibration concentration data is input into a corresponding pre-trained machine learning model, including: If it is a normal working mode, the concentration data before calibration is input into the corresponding pre-trained random forest regression model; If it is a disturbance operating mode, the pre-calibration concentration data is input into the corresponding pre-trained gradient boosting decision tree model.

9. A system for detecting the concentration of suspended matter in a canal water sample, used for executing the method for detecting the concentration of suspended matter in a canal water sample according to any one of claims 1 to 8, characterized in that: The system includes the following modules: The data acquisition module is used to obtain the environmental parameters of the canal water in real time. It uses a near-infrared optical sensor and an acoustic Doppler current profiler to synchronously collect the light scattering intensity signal and acoustic echo attenuation signal of the canal water. an initial value calculation module, configured to calculate an initial suspended matter concentration value based on the light scattering intensity signal and the acoustic echo attenuation signal; a correction module, configured to correct the initial suspended solids concentration value by a dynamic compensation algorithm according to environmental parameters of the canal water body to generate pre-calibration concentration data; An output module is used to input the pre-calibration concentration data into a pre-trained machine learning model and output a final suspended solids concentration value; wherein the pre-trained machine learning model includes a random forest regression model and a gradient boosting decision tree model.

Images