River turbidity monitoring device and method for river monitoring

Abstract

This invention relates to the field of river monitoring technology, providing a river turbidity monitoring device and method for river monitoring. The river turbidity monitoring device includes: a floating main frame, a water sample collection chamber, multi-source sensor nodes, a self-cleaning mechanism, and a central data processing module. The floating main frame provides support for water surface suspension. The water sample collection chamber is located at the bottom of the floating main frame, and has flow guide holes on both sides to allow river water to flow through. The multi-source sensor nodes and the self-cleaning mechanism are both installed inside the water sample collection chamber. The self-cleaning mechanism is located outside the measuring end face of the multi-source sensor nodes. This invention utilizes a submersible integrated structure design for the floating main frame and the water sample collection chamber, allowing the multi-source sensor nodes to be directly immersed in the river water flowing through the water sample collection chamber, thus achieving in-situ real-time monitoring.

Description

Technical Field

[0001] This invention belongs to the field of river monitoring technology, and in particular relates to a river turbidity monitoring device and method for river monitoring. Background Technology

[0002] River turbidity is a key water quality parameter reflecting the suspended solids content in water bodies, directly affecting drinking water safety, aquatic ecosystem health, and the effectiveness of comprehensive river management. However, existing river turbidity monitoring methods have many technical shortcomings: the traditional manual sampling-laboratory analysis model has poor timeliness and cannot reflect real-time dynamic changes in turbidity; while fixed online monitoring stations can achieve continuous monitoring, they involve large equipment investments and high maintenance costs, and data reliability decreases significantly over time due to blockages caused by sediment in water intake pipes. Although optical buoy-type turbidity monitoring devices, which have emerged in recent years, have achieved in-situ immersion measurement, they still face three core challenges:

[0003] Firstly, optical turbidity probes are submerged in water for a long time, and their measurement windows are easily covered by algal biofilms and silt deposits, which leads to optical path attenuation and changes in scattering characteristics, causing the measured values ​​to gradually drift. The existing method of relying on manual periodic cleaning is difficult to restore the probe accuracy in a timely manner.

[0004] Secondly, the optical measurement of turbidity is significantly affected by changes in water temperature—temperature changes the refractive index of water and the intensity of Brownian motion of particles, which in turn affects the intensity of backscattered light, and existing devices lack systematic correction methods for temperature drift.

[0005] Third, under conditions of heavy rainfall or high flow velocity during the flood season, the sediment at the bottom of the water is washed and tumbled, causing the turbidity of the water to rise sharply in an instant. This "artifact-like" surge in turbidity is not a true reflection of the degree of water pollution. However, existing algorithms cannot effectively distinguish between the interference introduced by the water flow and the real water quality deterioration signal, resulting in frequent false alarms.

[0006] Therefore, there is an urgent need for a river turbidity monitoring device and method with self-cleaning capability, simultaneous sensing of multiple physical quantities, and intelligent data correction function to achieve long-term stable and accurate output of turbidity data. Summary of the Invention

[0007] The purpose of this invention is to provide a river turbidity monitoring device for river monitoring, which aims to solve the above-mentioned problems.

[0008] The present invention is implemented as follows: a river turbidity monitoring device for river monitoring, comprising: a floating main frame, a water sample collection chamber, a multi-source sensor node, a self-cleaning mechanism, and a central data processing module;

[0009] The main floating frame is used to provide support for water surface suspension. The water sample collection chamber is located at the bottom of the main floating frame, and the chamber has flow guide holes on both sides to allow river water to flow through.

[0010] Both the multi-source sensing node and the self-cleaning mechanism are installed inside the water sample collection chamber; the self-cleaning mechanism is located outside the measuring end face of the multi-source sensing node.

[0011] The central data processing module is sealed and installed on the top of the main frame of the floating body, and is electrically connected to the multi-source sensing node and the self-cleaning mechanism respectively.

[0012] A further technical solution is that the self-cleaning mechanism includes a micro drive motor, a lead screw guide rail, a reciprocating mechanical scraper, and an ultrasonic transducer.

[0013] The micro drive motor drives the reciprocating mechanical scraper to slide against the measuring end face of the multi-source sensing node via the lead screw guide rail;

[0014] The ultrasonic transducer is embedded in the inner wall of the water sample collection chamber and is used to peel off the biofilm and sediment scale attached to the sensing node through the high-frequency cavitation effect.

[0015] 3. The device according to claim 2, wherein the multi-source sensing node is embedded with an optical turbidity probe based on the backscattering principle, an ultrasonic Doppler flow velocity probe, and a high-precision thermistor temperature probe, which are used to simultaneously acquire water turbidity, river flow velocity, and water temperature data, and transmit the acquired signals to the central data processing module in real time.

[0016] A method for monitoring river turbidity based on any one of the above-described devices includes the following steps:

[0017] S1: Device deployment, anchoring the device to the river channel to be measured, and synchronously collecting raw turbidity, real-time flow velocity and water temperature data through multi-source sensor nodes;

[0018] S2: Self-cleaning trigger: The central data processing module drives the self-cleaning mechanism to perform mechanical scraping and ultrasonic descaling based on the set time period or abnormal turbidity fluctuation threshold.

[0019] S3: Multidimensional data coupling processing, which corrects the collected raw turbidity using a mathematical model based on temperature compensation and flow rate calibration;

[0020] S4: Status warning. Calculate the comprehensive river health warning index based on the corrected final turbidity value and trigger a remote alarm when the level exceeds the standard.

[0021] A further technical solution, in step S3, addresses the measurement error caused by changes in water temperature leading to alterations in the refractive index and optical scattering characteristics of water vapor. A turbidity compensation model based on temperature dimensionlessness is established to correct the original turbidity measurement value. The specific calculation method is as follows:

[0022] First, calculate the real-time water temperature and subtract the historical lowest water temperature threshold for that river section to obtain the temperature difference. Then, calculate the historical highest water temperature threshold and subtract the lowest water temperature threshold for that river section to obtain the temperature range. Divide the temperature difference by the temperature range to obtain the normalized temperature variable. Multiply the normalized temperature variable by the linear compensation coefficient to obtain the linear correction term. Multiply the normalized temperature variable by the square of the linear thermal drift compensation coefficient to obtain the quadratic correction term. Add the linear correction term to the quadratic correction term to obtain the temperature compensation factor. Finally, multiply the original turbidity measurement value by the temperature compensation factor to obtain the temperature-compensated turbidity value.

[0023] A further technical solution, in step S3, addresses the problem of artifact interference in the baseline turbidity value caused by the instantaneous tumbling of sediment at the bottom of the water due to heavy rainfall or high flow velocity during the flood season. A dynamic turbidity calibration model based on flow velocity normalization is further established. The specific calculation method is as follows:

[0024] First, calculate the real-time river velocity minus the static flow velocity threshold to obtain the velocity difference; then calculate the full-load flood flow velocity upper limit threshold minus the static flow velocity threshold to obtain the velocity range; divide the velocity difference by the velocity range to obtain the normalized velocity variable; then calculate the sediment deposition disturbance damping coefficient multiplied by the normalized velocity variable, and take the negative value of the product to calculate the exponential function value with the natural constant e as the base; finally, multiply the temperature-compensated turbidity value by the exponential function value to obtain the true river calibration turbidity value output after eliminating water flow disturbance interference.

[0025] A further technical solution, in step S4, involves constructing a comprehensive early warning index model for transient evaluation and trend prediction of river water quality using time-series coupled logic. The specific calculation method is as follows:

[0026] First, calculate the final calibrated turbidity value of the current sampling period and divide it by the upper limit of safe turbidity specified in the river water environmental protection standard to obtain the static exceedance factor;

[0027] The absolute value of the turbidity change within two adjacent sampling periods is then calculated and divided by the product of the safe turbidity upper limit and the sampling time interval. This product is then multiplied by a normalized time constant with a fixed value of 1 second to obtain the dynamic deterioration trend factor.

[0028] The comprehensive early warning index is obtained by multiplying the static evaluation state weight coefficient by the static exceedance factor, and adding the dynamic deterioration trend weight coefficient by the dynamic deterioration trend factor; the sum of the two weight coefficients is equal to one.

[0029] When the final calibrated turbidity value is greater than or equal to the upper limit of safe turbidity, or when the comprehensive early warning index is greater than or equal to one, the system determines that the water body is in a state of severe turbidity and automatically triggers high-frequency sampling and alarm.

[0030] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0031] The device of this invention adopts an integrated submersible structure design of the floating main frame and the water sample collection chamber. The multi-source sensing nodes are directly immersed in the river water flowing through the water sample collection chamber, realizing in-situ real-time monitoring. This avoids the problems of sediment blockage and water sample lag caused by traditional water intake pipeline sampling, and ensures the authenticity and timeliness of the monitoring data.

[0032] This invention employs a dual self-cleaning mechanism that combines mechanical scraping and ultrasonic cavitation. The reciprocating mechanical scraper slides against the measuring end face via a lead screw guide, physically removing adhering contaminants. The ultrasonic transducer generates a high-frequency cavitation effect, stripping away biofilms and micron-sized scale. The two work alternately or in synergy, effectively extending the probe maintenance cycle and ensuring the accuracy and stability of long-term monitoring data.

[0033] This invention establishes a turbidity compensation model based on dimensionless temperature, using the first and second terms of the normalized temperature variable to correct the temperature drift error of optical scattering, effectively eliminating the systematic deviation of turbidity measurement caused by seasonal changes in water temperature and diurnal temperature differences. Furthermore, it constructs a dynamic turbidity calibration model based on flow velocity normalization, using an exponential decay function to mathematically remove artifacts caused by high-velocity agitation of sediment, ensuring that the final output turbidity value truly reflects the concentration of suspended solids in the water body, significantly reducing the false alarm rate during heavy rainfall and flood seasons.

[0034] This invention constructs a comprehensive early warning index model that weights and integrates the current static turbidity exceedance factor with the dynamic deterioration trend factor of turbidity change rate. It comprehensively evaluates the water health risk from two dimensions: "whether the current state exceeds the limit" and "whether the future trend will deteriorate". It has both transient evaluation and trend prediction capabilities, improves the accuracy and lead time of early warning, and provides reliable decision support for river water environment management. Attached Figure Description

[0035] Figure 1 This is a block diagram of the overall structure of the river turbidity monitoring device;

[0036] Figure 2 This is a structural block diagram of the core components inside the water sample collection chamber;

[0037] Figure 3 This is a content framework diagram of the overall solution;

[0038] Figure 4 This is a logic diagram for turbidity compensation based on temperature dimensionless conversion. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0040] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0041] like Figures 1-2 As shown, a river turbidity monitoring device for river monitoring is provided in one embodiment of the present invention, including: a floating main frame, a water sample collection chamber, a multi-source sensor node, a self-cleaning mechanism, and a central data processing module;

[0042] The main floating frame is used to provide support for water surface suspension. The water sample collection chamber is located at the bottom of the main floating frame, and the chamber has flow guide holes on both sides to allow river water to flow through.

[0043] Both the multi-source sensing node and the self-cleaning mechanism are installed inside the water sample collection chamber; the self-cleaning mechanism is located outside the measuring end face of the multi-source sensing node.

[0044] The central data processing module is sealed and installed on the top of the main frame of the floating body, and is electrically connected to the multi-source sensing node and the self-cleaning mechanism respectively.

[0045] In this embodiment, a complete structure of a river turbidity monitoring device for river monitoring is provided. The device uses a floating main frame as a floating support platform on the water surface. The floating main frame is made of hollow, sealed, corrosion-resistant polymer material (such as high-density polyethylene or polyurethane float), which has sufficient buoyancy reserve to support all internal modules. At the same time, it is connected to the riverbed through bottom anchor chains or positioning piles, so that the device maintains a relatively fixed position within the monitoring section and avoids drifting with the water flow.

[0046] The water sampling chamber is located at the bottom of the main floating structure, completely submerged at a certain depth below the water surface (usually 0.5~1.5m). This depth effectively avoids surface floating debris and wave disturbance, while also preventing contact with the riverbed sediment layer. The water sampling chamber is a cylindrical or rectangular cavity open at both ends, with flow guide holes on its side walls. The design of the flow guide holes allows river water to flow continuously and naturally through the cavity, creating a "flow-through" water sampling environment. This eliminates the need for additional water pumps or water intake pipelines, reducing energy consumption and avoiding secondary pollution caused by pipeline sedimentation and biological growth.

[0047] The multi-source sensor node is installed in the central area inside the water sample collection chamber. Its measuring end face is exposed and faces the direction of water flow to ensure full contact with the fresh river water flowing through it, so as to obtain real-time water quality and flow rate data.

[0048] The self-cleaning mechanism is also installed inside the water sample collection chamber and located on the outside of the measurement end face of the multi-source sensor node. Its function is to actively remove contaminants attached to the surface of the sensor end face under preset triggering conditions, so as to ensure the measurement accuracy of the sensor node during long-term operation.

[0049] The central data processing module is sealed and installed on the top of the main frame of the floating body. It is encapsulated in an IP68-rated waterproof sealed shell and integrates a low-power embedded processor, a data storage unit, and a wireless communication module. It is electrically connected to the multi-source sensor nodes and the self-cleaning mechanism through waterproof connectors. On the one hand, it receives raw signals such as turbidity, flow rate, and temperature uploaded by the sensor nodes. On the other hand, it determines the start time and working mode of the self-cleaning mechanism based on the built-in algorithm.

[0050] This embodiment utilizes an integrated submersible structure design for the floating main frame and the water sample collection chamber. Multi-source sensing nodes are directly immersed in the river water flowing through the chamber, enabling in-situ real-time monitoring. This avoids the problems of sediment blockage and water sample lag caused by traditional water intake pipeline sampling, ensuring the authenticity and timeliness of the monitoring data.

[0051] In a preferred embodiment of the present invention, the self-cleaning mechanism includes a micro drive motor, a lead screw guide rail, a reciprocating mechanical scraper, and an ultrasonic transducer.

[0052] The micro drive motor drives the reciprocating mechanical scraper to slide against the measuring end face of the multi-source sensing node via the lead screw guide rail;

[0053] The ultrasonic transducer is embedded in the inner wall of the water sample collection chamber and is used to peel off the biofilm and sediment scale attached to the sensing node through the high-frequency cavitation effect.

[0054] In this embodiment, the specific structure and working principle of the self-cleaning mechanism are described in detail. The self-cleaning mechanism consists of four parts: a miniature drive motor, a lead screw guide rail, a reciprocating mechanical scraper, and an ultrasonic transducer.

[0055] The miniature drive motor is a waterproof stepper motor or a miniature DC geared motor, and its output shaft is connected to the end of the lead screw of the lead screw guide rail via a coupling. The lead screw guide rail consists of a precision lead screw and a matching slider. The lead screw is mounted on both sides of the measuring end face of the multi-source sensing node, and the slider is fitted on the lead screw and can move back and forth along the lead screw axis. The reciprocating mechanical scraper (i.e., scraper blade) is fixed on the slider. The cutting edge of the scraper blade is made of wear-resistant polyurethane or silicone rubber that will not damage the optical window. After installation, the scraper surface maintains slight contact with the measuring end face of the sensing node.

[0056] When the central data processing module issues a cleaning command, the micro drive motor starts and drives the lead screw to rotate in the forward direction. The slider drives the scraper to slide at a constant speed from one end of the measuring end face to the other end, scraping away the biofilm, algae and loose mud and sand attached to the end face. After reaching the end point, the motor rotates in the reverse direction, and the scraper returns to the starting position, completing a complete reciprocating scraping stroke.

[0057] The ultrasonic transducer is embedded in the inner wall of the water sample collection chamber, with its radiating surface facing the measurement end face of the multi-source sensing node. The transducer is encapsulated with a piezoelectric ceramic wafer and is driven by a high-frequency oscillation circuit to generate ultrasonic waves of 20~80kHz. When the ultrasonic waves propagate in the liquid medium, they generate tens of thousands of tiny bubbles through high-frequency alternating compression and stretching. These bubbles grow rapidly and implode violently under the alternating positive and negative pressure of the ultrasonic field (i.e., cavitation effect), generating extremely high instantaneous temperatures and shock waves locally. This effectively removes biofilms formed by extracellular polymers secreted by microorganisms and dense silt and sand deposits at the nanometer to micrometer scale from the measurement end face. The level of cleaning precision is unattainable by mechanical scraping.

[0058] In actual operation, the system can be set to start a complete "mechanical scraping + ultrasonic cavitation" combined cleaning sequence once a week or daily, or to actively trigger cleaning when the central data processing module detects abnormal slow drift in turbidity data (such as a monotonous increase in the turbidity reference value exceeding the set threshold for 24 consecutive hours).

[0059] This embodiment achieves deep cleaning of the measurement end face of the sensing node through the synergistic effect of mechanical scraping and ultrasonic cavitation. Mechanical scraping removes soft deposits, while ultrasonic cavitation peels off stubborn biofilms and micron-sized scale. The two work alternately or in synergy, effectively extending the maintenance-free cycle of the probe and ensuring the accuracy and stability of long-term monitoring data.

[0060] In a preferred embodiment of the present invention, the multi-source sensing node is embedded with an optical turbidity probe based on the backscattering principle, an ultrasonic Doppler flow velocity probe, and a high-precision thermistor temperature probe, which are used to simultaneously acquire water turbidity, river flow velocity, and water temperature data, and transmit the acquired signals to the central data processing module in real time.

[0061] In this embodiment, the specific probe configuration and synchronous measurement function of the multi-source sensing node are described. The multi-source sensing node is a miniaturized sensor group integrating three probes with different principles, embedded in the mounting base in the center of the water sample collection chamber, with the measuring end face of each probe exposed in the flowing water sample.

[0062] The first type is an optical turbidity probe based on the backscattering principle. It integrates a near-infrared LED light source (wavelength about 850nm) and a photodetector. The light beam emitted by the light source passes through the measurement window and enters the water body. The suspended particles in the water body scatter the incident light. The detector receives the intensity of the backscattered light at an angle of 90° or 140°. Based on the positive correlation between the intensity of the scattered light and turbidity, the optical signal is converted into a turbidity value (unit: NTU) in real time through a built-in calibration curve.

[0063] The second type is the ultrasonic Doppler flow velocity probe, which uses the acoustic Doppler effect to measure flow velocity. The probe emits ultrasonic pulses of a fixed frequency into the water. When the ultrasonic waves encounter suspended particles or tiny bubbles in the water flow, they produce reflected echoes. Due to the movement of the water flow, the echo frequency undergoes a Doppler frequency shift. By detecting the frequency shift, the velocity of the water flow relative to the probe can be calculated, that is, the real-time flow velocity of the river water (in m / s).

[0064] The third type is a high-precision thermistor temperature probe, which uses a negative temperature coefficient thermistor as the sensing element. Its resistance decreases non-linearly with increasing temperature. Through a precision voltage divider circuit and analog-to-digital conversion, the change in resistance is converted into a temperature value (in °C), with a measurement accuracy of ±0.1 °C. Under the unified clock synchronization trigger of the central data processing module, the three types of probes synchronously collect turbidity, flow rate, and temperature data at the same sampling frequency (such as once every 10 seconds or once per minute), and transmit the three timestamped signals to the central data processing module in real time.

[0065] This synchronous acquisition mechanism ensures that turbidity, flow rate, and temperature values ​​correspond one-to-one on the time axis during subsequent data fusion processing, providing time-matched data input for the temperature and flow rate-based turbidity correction model. The synchronous configuration of the three probes enables the device to acquire the core turbidity index while simultaneously acquiring two key environmental physical quantities that affect the accuracy of turbidity measurement, providing a complete raw data foundation for subsequent multi-dimensional data coupling correction.

[0066] like Figure 3 , Figure 4 As shown, a method for monitoring river turbidity based on any of the above-described devices includes the following steps:

[0067] S1: Device deployment, anchoring the device to the river channel to be measured, and synchronously collecting raw turbidity, real-time flow velocity and water temperature data through multi-source sensor nodes;

[0068] S2: Self-cleaning trigger: The central data processing module drives the self-cleaning mechanism to perform mechanical scraping and ultrasonic descaling based on the set time period or abnormal turbidity fluctuation threshold.

[0069] S3: Multidimensional data coupling processing, which corrects the collected raw turbidity using a mathematical model based on temperature compensation and flow rate calibration;

[0070] S4: Status warning. Calculate the comprehensive river health warning index based on the corrected final turbidity value and trigger a remote alarm when the level exceeds the standard.

[0071] In this embodiment, a complete method for monitoring river turbidity based on the above-mentioned device is provided, comprising four steps.

[0072] Step S1 involves device deployment and data acquisition: The device is anchored at a representative monitoring section of the river to be measured. The draft of the main floating frame is adjusted so that the water sample collection chamber is completely submerged below the water surface. The anchoring system ensures that the device is not washed away from its original position by the water flow. After the device is ready, the central data processing module activates the multi-source sensor nodes to enter continuous acquisition mode. The optical turbidity probe, ultrasonic Doppler flow velocity probe, and thermistor temperature probe simultaneously acquire raw turbidity, real-time flow velocity, and water temperature data. All raw signals are amplified, filtered, and converted from analog to digital before being stored in the local cache.

[0073] Step S2 is the self-cleaning trigger: The central data processing module has built-in cleaning scheduling logic, which independently triggers self-cleaning based on two conditions: one is a set time period (such as forcibly executing a mechanical scraping and ultrasonic descaling combination sequence every 24 hours or every 72 hours); the other is an abnormal turbidity fluctuation threshold (when the turbidity value collected for several consecutive times shows a monotonous and slow upward trend compared to the historical benchmark value and the cumulative change exceeds a preset percentage, it is determined that the probe end face is dirty, and cleaning is triggered). During the cleaning action, the system pauses data acquisition and records cleaning markers. Data acquisition resumes after cleaning is completed and the water body stabilizes.

[0074] S3 is a multi-dimensional data coupling processing: The central data processing module calls the embedded mathematical model to perform two-stage correction on the original turbidity based on temperature compensation and flow velocity calibration. First, the synchronously acquired water temperature data is substituted into the temperature compensation model to eliminate optical measurement drift caused by water temperature changes. Then, the temperature-compensated turbidity value and the synchronous flow velocity data are substituted into the flow velocity calibration model to remove the artifact interference introduced by sediment rolling under high flow velocity and output the true river calibration turbidity value.

[0075] Step S4 is the status warning: Based on the corrected final turbidity value, the central data processing module uses time-coupled logic to calculate the river health comprehensive warning index. This index simultaneously considers the static degree of turbidity exceeding the standard and the dynamic deterioration trend of the turbidity change rate. When the final turbidity value directly exceeds the safety upper limit or the comprehensive warning index exceeds the threshold, the system determines that the water body is in a state of severe turbidity, automatically triggers high-frequency sampling and encrypted monitoring, and sends alarm information to the remote monitoring platform through the built-in 4G / satellite communication module.

[0076] This embodiment achieves an upgrade of turbidity monitoring from "passive manual maintenance" to "proactive intelligent operation and maintenance" through a fully automated design of "deployment and collection - self-cleaning maintenance - data correction - intelligent early warning". This significantly reduces the frequency of on-site maintenance and improves data quality and early warning reliability.

[0077] In a preferred embodiment of the present invention, in step S3, to address the measurement errors caused by changes in water temperature leading to alterations in the refractive index and optical scattering characteristics of water vapor, a turbidity compensation model based on temperature dimensionlessness is established to correct the original turbidity measurement values. The following corrections are made:

[0078]

[0079] In the formula, The turbidity value is after temperature compensation and is in NTU. θ is the raw turbidity measurement value output by the optical turbidity probe, with the dimension NTU; θ is the real-time water temperature measured by the temperature probe, with the dimension °C. and These are the highest and lowest water temperature thresholds in the historical statistics for the target river section, both in °C. The normalized dimensionless temperature variable; α is the linear compensation coefficient, and β is the quadratic thermal drift compensation coefficient, both of which are dimensionless empirical constants fitted experimentally.

[0080] In this embodiment, the meaning and acquisition method of each letter parameter in the formula are explained one by one as follows:

[0081] The turbidity value after temperature compensation, in NTU (Nephelometric Turbidity Units), is the corrected turbidity value output by the model.

[0082] The raw turbidity measurement value output by the optical turbidity probe, in NTU, is obtained directly by the optical turbidity probe based on the backscattering principle.

[0083] θ: Real-time water temperature measured synchronously by a high-precision thermistor temperature probe, in °C.

[0084] and These are the highest and lowest historical water temperature thresholds for the target river section, respectively, both measured in degrees Celsius (°C). They are obtained through statistical analysis of historical water temperature monitoring data for at least one complete year (covering all four seasons) at the monitoring section. Typically, the highest and lowest temperatures from the historical data are used, but they can also be set based on local hydrological and meteorological data, for example, using... =0℃、 =35℃ to cover the annual variation range of most natural river channels.

[0085] : For dimensionless temperature variables that have undergone range normalization, the real-time temperature is mapped to the [0,1] interval, eliminating the temperature dimension and making the model universal across river sections.

[0086] The first-order compensation coefficient is a dimensionless empirical constant obtained through indoor calibration experiments. Specifically, in a constant-temperature turbidity standard solution bath, different temperature gradients (e.g., 5℃, 10℃, 15℃, 20℃, 25℃, 30℃, 35℃) are set, and the turbidity readings of the same turbidity standard solution at each temperature point are measured. A linear regression is performed with the normalized temperature as the independent variable and the relative turbidity deviation as the dependent variable; the resulting slope is the coefficient of performance. .

[0087] The quadratic thermal drift compensation coefficient is a dimensionless empirical constant, and... The quadratic term was obtained by simultaneously performing a quadratic regression fitting on the calibration experimental data mentioned above. The purpose of introducing the quadratic term is to correct the nonlinear relationship between temperature and turbidity deviation—the temperature drift curves of some optical devices exhibit obvious bending characteristics, and a single linear correction is insufficient to completely eliminate the error. Introducing the quadratic term can significantly improve the correction accuracy.

[0088] This model effectively eliminates the systematic biases caused by seasonal changes in water temperature and diurnal temperature differences in optical turbidity measurements through a combination of linear and quadratic terms, making the temperature-compensated turbidity value x_c more accurately reflect the concentration of suspended solids in the water. and Once calibrated, it is written as a system constant into the storage area of ​​the central data processing module, and the temperature compensation calculation is automatically executed after each data acquisition.

[0089] As a preferred embodiment of the present invention, in step S3, to address the problem of artifact interference in the baseline turbidity value caused by the instantaneous tumbling of sediment at the bottom of the water due to heavy rainfall or high flow velocity during the flood season, a dynamic turbidity calibration model based on flow velocity normalization is further established:

[0090]

[0091] In the formula, The actual river calibration turbidity value output after eliminating water flow disturbance interference is in NTU dimension; v is the real-time river velocity measured by the Doppler velocity probe, in m / s dimension. The static water flow velocity threshold. Both are the upper limit threshold values ​​for the full-load flood flow velocity, and their dimensions are m / s; is a dimensionless parameter after normalizing the velocity boundary; k is the sediment settling disturbance damping coefficient, which is a dimensionless constant.

[0092] In this embodiment, the meaning and acquisition method of each letter parameter in the formula are explained one by one as follows:

[0093] The true river calibration turbidity value output after eliminating water flow disturbance interference, with the dimension NTU, is the final output value of the model and represents the true estimate of the suspended solids concentration in the water body at that moment.

[0094] The turbidity value after temperature compensation, in NTU, is calculated by the temperature compensation model described in claim 5.

[0095] v: Real-time river flow velocity measured by an ultrasonic Doppler velocity probe, in m / s.

[0096] The static flow velocity threshold, measured in m / s, is a preset constant. Its physical meaning is that when the flow velocity is below this value, the shear stress on the bottom sediments is insufficient to initiate their suspension, and the interference of the flow velocity on turbidity measurement is negligible. v0 is usually calculated based on the sediment particle size distribution of the monitored river section using a sediment initiation velocity formula (such as the Shields formula), or determined through analysis of velocity-turbidity scatter plots from historical low-flow periods (such as the dry season)—typically taken as 0.1~0.3 m / s.

[0097] The upper limit threshold for full-load flood flow velocity, measured in m / s, is a preset constant representing the maximum design flow velocity at the monitoring section under extreme flood conditions. It can be obtained from the hydrological design data of the river channel or from historical maximum flood flow velocity statistics.

[0098] : Regarding the flow velocity at [ , The dimensionless parameter after normalization mapping within the interval. When ≤ When this parameter is set to 0 (or zeroed out), the exponent term... , That is, no flow rate correction is performed; when ≥ When this parameter is 1, the exponent term is... That is, applying the maximum intensity of attenuation correction; When the value is between the two, a correction of the corresponding strength is applied in a linear proportion.

[0099] k: The sediment deposition disturbance damping coefficient, a dimensionless constant, controls the intensity of velocity correction. A larger k value results in stronger attenuation at the same velocity, suitable for river sections with fine-grained, easily suspended sediment; a smaller k value results in gentler attenuation, suitable for coarse-grained riverbeds. k is obtained by nonlinear least-squares fitting calibration of "velocity-turbidity increment" data from multiple rainfall events in the monitored river section.

[0100] This model employs an exponential decay form, based on the physical principle that at high flow velocities, the resuspension of sediment is roughly proportional to a high power of the flow velocity, while its contribution to turbidity measurement is linearly related to the suspended solids concentration. The exponential decay function effectively describes this nonlinear relationship of "high-velocity interference—turbidity response." Using this model, at lower flow velocities ( ≈ ), The model does not attenuate turbidity because there is no sediment tumbling interference at this point; at higher flow velocities (v significantly greater than 100%), the turbidity is reduced. ), Significantly smaller than The model mathematically removes artifacts caused by sediment tumbling, ensuring that the final turbidity value accurately reflects the concentration of suspended solids in the water. This correction mechanism significantly reduces the false alarm rate during heavy rainfall and flood season.

[0101] In a preferred embodiment of the present invention, in step S4, a comprehensive early warning index model for transient evaluation and trend prediction of river water quality is constructed using time-coupled logic:

[0102]

[0103] In the formula, The comprehensive early warning index is a dimensionless scalar. This is the final calibrated turbidity value for the current sampling period; The safe upper limit of turbidity specified in the water environment protection standard for this river channel, both of which are NTU in dimension, and their ratio constitutes a dimensionless static exceedance factor; The change in turbidity between two adjacent sampling periods is expressed in NTU. The sampling time interval is expressed in seconds (s); τ is the normalized time constant, which is fixed at 1 second. and These are the dimensionless weight coefficients for the static evaluation state and the dynamic deterioration trend, respectively, and they satisfy... ;

[0104] when or When the system determines that the water body is in a state of severe turbidity, it will automatically trigger high-frequency sampling and alarm.

[0105] In this embodiment, the meaning and acquisition method of each letter parameter in the formula are explained one by one as follows:

[0106] The comprehensive early warning index is a dimensionless scalar quantity and is the core decision indicator for the system to determine the turbidity status of water bodies.

[0107] The final calibrated turbidity value for the current sampling period (after temperature compensation and flow rate calibration), in NTU, is output in real time by the model described in claim 6.

[0108] The safe upper limit of turbidity specified in the water environment protection standard for this river channel is measured in NTU. This is based on the "Surface Water Environmental Quality Standard" (GB 3838) or the water quality management targets set by the local environmental protection department for this river section—for example, 5 NTU for primary protection zones of drinking water sources, 20 NTU for general landscape water use, and 30 NTU for industrial water use.

[0109] The static exceedance factor is dimensionless. When this ratio is greater than 1, it indicates that the current turbidity has exceeded the standard limit; when it is less than 1, it indicates that the current turbidity is within the safe range. This factor reflects the "instantaneous exceedance level" at the current moment; the larger the value, the worse the current water quality.

[0110] The change in turbidity between two adjacent sampling periods (i.e., the change in turbidity in this sampling period) Subtract the last time The dimension is NTU. Positive values ​​indicate an increase in turbidity, while negative values ​​indicate a decrease.

[0111] The time interval between two consecutive samples, measured in seconds (s), is set by the sampling timer of the central data processing module (e.g., ...). =60s, 300s, etc.).

[0112] τ: Normalized time constant, fixed at 1 second. This constant is used only to keep the third term of the formula dimensionless—because The dimension of is 1 / s, which is multiplied by τ (in s) to make it a dimensionless quantity.

[0113] The dynamic deterioration trend factor is dimensionless and reflects the relative rate of change of turbidity per unit time. The larger the value, the more rapidly the turbidity of the water body is increasing, and the greater the risk of a sharp deterioration in water quality.

[0114] and : These are the dimensionless weight coefficients for the static evaluation state and the dynamic deterioration trend, respectively, satisfying... The two coefficients can be configured according to the focus of river management—if more attention is paid to whether the current water quality meets the standards, then set... >0.5; If more attention is paid to sudden changes in water quality trends to achieve early warning, then set >0.5. The default recommended value is... =0.6、 =0.4.

[0115] The warning judgment logic is: when ≥ At any time, regardless At what value will the system directly determine that the water body is in a state of severe turbidity and trigger an alarm, because the absolute value of turbidity exceeding the standard is sufficient to constitute a water quality incident; when but When I_w is ≥1, the system also determines it as a state of severe turbidity and triggers an alarm. This situation corresponds to an urgent trend of "currently not exceeding the standard but rapidly approaching it." For example, the current turbidity is 80% of the safe limit, but the rate of change is extremely fast, predicting that it will exceed the standard in the next sampling period. At this time, I_w may exceed 1, triggering an early alarm. After the alarm is triggered, the system automatically switches the sampling frequency from the normal mode to the high-frequency encryption mode (e.g., increasing from once every 5 minutes to once every 30 seconds), and packages and sends the current turbidity value, temperature, flow rate, and location information to the remote monitoring platform through the communication module. This model, through the weighted fusion of static and dynamic factors, has both transient evaluation and trend prediction capabilities. Compared with the single threshold method, it is more scientific and sensitive, providing reliable decision support for river water environment management.

[0116] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A river turbidity monitoring device for river monitoring, characterized in that, include: The floating main frame, water sample collection chamber, multi-source sensor nodes, self-cleaning mechanism, and central data processing module; The main floating frame is used to provide support for water surface suspension. The water sample collection chamber is located at the bottom of the main floating frame, and the chamber has flow guide holes on both sides to allow river water to flow through. Both the multi-source sensing node and the self-cleaning mechanism are installed inside the water sample collection chamber; the self-cleaning mechanism is located outside the measuring end face of the multi-source sensing node. The central data processing module is sealed and installed on the top of the main frame of the floating body, and is electrically connected to the multi-source sensing node and the self-cleaning mechanism respectively.

2. The apparatus according to claim 1, characterized in that, The self-cleaning mechanism includes a miniature drive motor, a lead screw guide rail, a reciprocating mechanical scraper, and an ultrasonic transducer. The micro drive motor drives the reciprocating mechanical scraper to slide against the measuring end face of the multi-source sensing node via the lead screw guide rail; The ultrasonic transducer is embedded in the inner wall of the water sample collection chamber and is used to peel off the biofilm and sediment scale attached to the sensing node through the high-frequency cavitation effect.

3. The apparatus according to claim 2, characterized in that, The multi-source sensing node is embedded with an optical turbidity probe based on the backscattering principle, an ultrasonic Doppler flow velocity probe, and a high-precision thermistor temperature probe, which are used to simultaneously acquire water turbidity, river flow velocity, and water temperature data, and transmit the acquired signals to the central data processing module in real time.

4. A method for monitoring river turbidity based on the device according to any one of claims 1 to 3, characterized in that, Includes the following steps: S1: Device deployment, anchoring the device to the river channel to be measured, and synchronously collecting raw turbidity, real-time flow velocity and water temperature data through multi-source sensor nodes; S2: Self-cleaning trigger: The central data processing module drives the self-cleaning mechanism to perform mechanical scraping and ultrasonic descaling based on the set time period or abnormal turbidity fluctuation threshold. S3: Multidimensional data coupling processing, which corrects the collected raw turbidity using a mathematical model based on temperature compensation and flow rate calibration; S4: Status warning. Calculate the comprehensive river health warning index based on the corrected final turbidity value and trigger a remote alarm when the level exceeds the standard.

5. The apparatus according to claim 3, characterized in that, In step S3, to address the measurement error caused by changes in water temperature leading to alterations in the refractive index and optical scattering characteristics of water vapor, a turbidity compensation model based on temperature dimensionlessness is established to correct the original turbidity measurement value. The specific calculation method is as follows: First, calculate the real-time water temperature and subtract the historical lowest water temperature threshold for that river section to obtain the temperature difference; then calculate the historical highest water temperature threshold for that river section and subtract the lowest water temperature threshold to obtain the temperature range; divide the temperature difference by the temperature range to obtain the normalized temperature variable; then multiply the normalized temperature variable by the first-order compensation coefficient to obtain the first-order correction term. The quadratic correction term is obtained by multiplying the quadratic thermal drift compensation coefficient by the square of the normalized temperature variable. Add the first correction term and the second correction term to obtain the temperature compensation factor; finally, multiply the original turbidity measurement value by the temperature compensation factor to obtain the temperature-compensated turbidity value.

6. The apparatus according to claim 5, characterized in that, In step S3, to address the issue of artifact interference in the baseline turbidity value caused by the instantaneous turbidity tumbling of sediment due to heavy rainfall or high flow velocities during the flood season, a dynamic turbidity calibration model based on flow velocity normalization is further established. The specific calculation method is as follows: First, calculate the real-time river velocity minus the static flow velocity threshold to obtain the velocity difference; then calculate the full-load flood flow velocity upper limit threshold minus the static flow velocity threshold to obtain the velocity range; divide the velocity difference by the velocity range to obtain the normalized velocity variable; then calculate the sediment deposition disturbance damping coefficient multiplied by the normalized velocity variable, and take the negative value of the product to calculate the exponential function value with the natural constant e as the base; finally, multiply the temperature-compensated turbidity value by the exponential function value to obtain the true river calibration turbidity value output after eliminating water flow disturbance interference.

7. The apparatus according to claim 5, characterized in that, In step S4, a comprehensive early warning index model for transient evaluation and trend prediction of river water quality is constructed using time-series coupled logic. The specific calculation method is as follows: First, calculate the final calibrated turbidity value of the current sampling period and divide it by the upper limit of safe turbidity specified in the river water environmental protection standard to obtain the static exceedance factor; The absolute value of the turbidity change within two adjacent sampling periods is then calculated and divided by the product of the safe turbidity upper limit and the sampling time interval. This product is then multiplied by a normalized time constant with a fixed value of 1 second to obtain the dynamic deterioration trend factor. The comprehensive early warning index is obtained by multiplying the static evaluation state weight coefficient by the static exceedance factor, and adding the dynamic deterioration trend weight coefficient by the dynamic deterioration trend factor; the sum of the two weight coefficients is equal to one. When the final calibrated turbidity value is greater than or equal to the upper limit of safe turbidity, or when the comprehensive early warning index is greater than or equal to one, the system determines that the water body is in a state of severe turbidity and automatically triggers high-frequency sampling and alarm.

Images