An agricultural non-point source pollution on-line monitoring system and method

Abstract

The present application relates to the technical field of spectrophotometric brightness analysis measurement, and discloses an agricultural non-point source pollution online monitoring system and method, which comprises the following steps: a multi-wavelength collinear transmission light-emitting module emits a complex color parallel light beam to a fluid detection chamber module; a central signal processing module calculates the first-order differential absolute value of the current change rate of a reference detection channel module, controls a charge shunt control module to be in a shunt conduction state to discharge transient overload charge when the first-order differential absolute value is greater than a shunt mutation threshold value; and when the output current difference of a first signal detection module and a second signal detection module is greater than a light path disturbance threshold value, interference data of the current period is removed, the present application uses feedforward shunt topology to balance transient overload charge and eliminate light path deflection error, solves the technical problem that the detection channel is saturated under high turbidity flushing, and improves the long-term robustness of characteristic absorption brightness measurement in a multiphase fluid environment.

Description

Technical Field

[0001] This invention relates to an online monitoring system and method for agricultural non-point source pollution, belonging to the field of spectrophotometric brightness analysis and measurement technology. Background Technology

[0002] Current in-situ spectrophotometry techniques utilize a standard spectral collinear transmission array to acquire the transmitted light intensity after passing through a characteristic measurement channel. The characteristic absorbance values ​​of the target component are interpreted by measuring the relative attenuation of the multispectral current. This inversion is based on Beer-Lambert's law, which requires the excitation beam to pass through a homogeneous fluid, causing the exponential dissipation of the transmitted intensity to monotonically correspond to the intrinsic absorption interception of dissolved molecules. This establishes a definite concentration inversion relationship and is a conventional method for in-situ long-term measurement. However, in irrigation districts where runoff and drainage channels converge, strong shear turbulence and water level fluctuations cause dense surges in heterogeneous suspended sediment particles. When the excitation beam passes through a sand-laden multiphase fluid, the sediment particles induce nonlinear strong volume scattering and step-like beam obstruction during the transient time it takes to truncate the characteristic measurement optical axis. This disturbance, deviating from the homogeneous assumption, causes the spectral channel current to instantly bottom out. The pre-impedance circuit, lacking feedforward response capability, falls into the noise blind zone, or exceeds the linear range due to instantaneous charge accumulation, leading to measurement errors in the system.

[0003] Traditional methods often employ multi-stage filter cartridges to physically trap suspended particles, or introduce post-processing units to apply polynomial functions for mathematical fitting. However, long-term deployment of these filters can lead to biological adhesion and secondary blockage, increasing maintenance costs. Digital smoothing requires that the receiving channel not experience physical overload. Once the spectral current experiences a step-like hard saturation due to sudden blockage by sediment, the post-processing algorithm loses effective reference data, revealing a conflict between the fixed impedance topology and rigid measurement optical path and the multiphase fluid matrix. Some solutions attempt to adapt to different media by adjusting the physical optical path, but hardware structures relying solely on mechanical displacement exhibit dynamic response lag when dealing with sudden heterogeneous disturbances. Furthermore, the accompanying control methods are insufficient in suppressing transient overloads and identifying discrete anomalies. For example, [the following is a separate section, published in CN...] Chinese invention patent application 102042965A discloses an online broadband water quality analyzer. It uses a stepper motor to drive the optical window to move and adjust the absorption optical path, and uses an ultrasonic transducer to clean the window. This architecture is based on the premise that the water matrix is ​​in a relatively stable and slowly changing state. When faced with extreme agricultural runoff conditions containing dense sediment microclusters and strong shear microbubbles, the mechanical optical path adjustment cannot achieve real-time countermeasurement in the face of millisecond-level transient pulsations. Moreover, the control system lacks feedforward discharge logic for charge accumulation in the pre-amplifier stage, which causes the characteristic measurement channel to hard saturate and paralyze under sudden high turbidity flushing. At the same time, due to the lack of a multi-channel differential tuning time-domain elimination algorithm, it is unable to eliminate the pseudo-feature absorption noise caused by mechanical axial deflection, resulting in disordered divergence of the interpreted concentration parameters.

[0004] Therefore, the technical problem to be solved by this invention is how to utilize the intrinsic physical micro-wave characteristics generated by the spectral reference channel, reconstruct the load impedance topology in a feedforward manner to balance overload charge, and coordinate the control of the physical geometry of the measurement channel to eliminate microbubble scattering interference. Summary of the Invention

[0005] To address the problems in the background art, the technical solution of the present invention is as follows: An online monitoring system for agricultural non-point source pollution, comprising:

[0006] A multi-wavelength collinear transmission light emission module and a photoelectric detection module are respectively fixed on both sides of the fluid detection chamber module;

[0007] The photoelectric detection module includes a reference detection channel module, a first signal detection module, a second signal detection module, and a charge shunting control module, with the charge shunting control module connected in parallel to the signal output terminal of the reference detection channel module.

[0008] The central signal processing module is connected to the reference detection channel module, the first signal detection module, the second signal detection module, and the charge shunt control module. The central signal processing module is used to complete the following steps based on the transient electrical signal flow of each module during measurement within the fluid detection chamber module:

[0009] Step S1: Periodically collect and calculate the absolute value of the first derivative of the change in output current between adjacent sampling points of the reference detection channel module divided by the sampling interval time. When the absolute value of the first derivative is greater than the shunt change threshold, control the charge shunt control module to be in the shunt conduction state to discharge the instantaneous overload charge.

[0010] Step S2: Continuously calculate the absolute value of the output current difference between the first signal detection module and the second signal detection module in real time. When the absolute value of the current difference is greater than the optical path disturbance threshold, remove the interference data of the current sampling period in the current absorption brightness interpretation process.

[0011] Preferably, the system also includes a concentration analysis module; the concentration analysis module is connected to the central signal processing module. When the absolute value of the first derivative is less than or equal to the shunt mutation threshold, the concentration analysis module is used to obtain the output voltage ratio sequence of the characteristic wavelength channel and the reference detection channel module in the multi-wavelength collinear transmission emission module, and calculate the second-order variance of the output voltage ratio sequence within a defined time window; when the second-order variance is greater than the bubble interference threshold, the switching frequency of the thin-layer flow channel switching module in the fluid detection chamber module is adjusted according to the increment of the second-order variance.

[0012] Preferably, step S2 includes the following sub-steps: Step S21, under the initial calibration state, obtain the initial output current of the first signal detection module and the second signal detection module when the fluid detection chamber module is filled with standard deionized water, and divide the initial output current of the first signal detection module by the initial output current of the second signal detection module to determine the balance ratio coefficient; Step S22, during online monitoring, multiply the real-time output current of the second signal detection module by the balance ratio coefficient to generate a compensation current value, calculate the absolute value of the difference between the real-time output current of the first signal detection module and the compensation current value, and use it as the absolute value of the output current difference; when the absolute value of the output current difference is greater than the optical path disturbance threshold, remove the interference data of the current sampling period in the current absorbance interpretation process. Preferably, the system further includes a window status monitoring module; the window status monitoring module is connected to the central signal processing module, and in the monitoring state where the charge shunting control module is activated, the window status monitoring module is used to record the total number of shunting switching by the charge shunting control module within a set time window, and calculate the first derivative of the total number of shunting switching with time; when the first derivative is greater than the drift threshold, the window status monitoring module outputs a light-transmitting window cleaning control command to the central signal processing module.

[0013] Preferably, the steps for determining the shunt mutation threshold include: acquiring a reference photocurrent data sequence of the calibration water body; calculating the standard deviation of the reference photocurrent data sequence; adding three times the standard deviation to the inherent noise baseline value of the system, and determining the result of the addition as the shunt mutation threshold.

[0014] Preferably, the charge shunt control module uses a field-effect transistor with a conduction delay time of less than 10 μs as an internal switching component, and the response conduction time of the charge shunt control module after receiving the conduction control signal is in the range of 15 μs to 30 μs, and the impedance value of its internal low impedance branch is less than 10% of the input impedance of the preamplifier circuit in the central signal processing module.

[0015] Preferably, the multi-wavelength collinear transmission light emission module includes an ultraviolet light source, a visible light source, and a near-infrared light source; the light beams emitted by the ultraviolet light source, the visible light source, and the near-infrared light source are converged to the same collimated measurement optical axis by a beam combiner group; wherein, the light beam emitted by the near-infrared light source is directed toward the reference detection channel module, and the light beams emitted by the ultraviolet light source and the visible light source are directed toward the first signal detection module and the second signal detection module, respectively.

[0016] Preferably, step S1 includes the following sub-steps: Step S11, using the central signal processing module to obtain the output current change and sampling interval time of the reference detection channel module between adjacent sampling points, and calculating the absolute value of the first derivative; Step S12, when the absolute value of the first derivative is greater than the shunt change threshold twice consecutively, controlling the charge shunt control module to enter the shunt conduction state. Preferably, the system also includes a host protective chamber module and a surface cleaning control module; the host protective chamber module has a fluid detection chamber module and an optical light transmission window assembly embedded in it, the multi-wavelength collinear transmission light emission module and the photoelectric detection module are respectively fixedly arranged on both sides of the fluid detection chamber module, and the central signal processing module is encapsulated inside the host protective chamber module; the surface cleaning control module is arranged opposite to the optical light transmission window assembly, and the surface cleaning control module is used to remove the adhering substances on the surface of the optical light transmission window assembly when it receives the light transmission window cleaning control command output by the window status monitoring module.

[0017] A method for online monitoring of agricultural non-point source pollution, which operates in an online monitoring system for agricultural non-point source pollution, includes the following steps:

[0018] Step S101: A single-axis polychromatic parallel beam is emitted into the fluid detection chamber module using a multi-wavelength collinear transmission light emission module, so that the single-axis polychromatic parallel beam passes perpendicularly through the farmland runoff to be measured in the fluid detection chamber module. The single-axis polychromatic parallel beam contains a 220nm characteristic wavelength and a 700nm reference wavelength.

[0019] Step S102: The central signal processing module periodically collects the current output by the reference detection channel module and calculates the absolute value of the first derivative of the change in output current between adjacent sampling points divided by the sampling interval. When the absolute value of the first derivative is greater than the shunt change threshold, the charge shunt control module is controlled to be in the shunt conduction state to discharge the instantaneous overload charge. Step S103: The central signal processing module continuously calculates the absolute value of the difference between the output current of the first signal detection module and the second signal detection module in real time. When the absolute value of the current difference is greater than the optical path disturbance threshold, the interference data of the current sampling period is removed in the current absorption brightness interpretation process.

[0020] Compared with the prior art, the beneficial effects of the present invention are:

[0021] 1. In online monitoring of agricultural non-point source pollution, the reference detection channel module converts the collected transmission spectrum into a reference photoelectric signal. The central signal processing module monitors the reference voltage value corresponding to this signal. When this voltage value reaches the preset low resistance threshold, it drives the solid-state electronic switch to cut off the first path and connect the second path, causing the front load impedance of the receiving channel to switch to a higher-order static resistance value in a stepwise manner. This improves the photoelectric conversion gain to maintain the output voltage signal-to-noise ratio of the characteristic wavelength channel under high shielding conditions. The central signal processing module also controls the charge shunt control module to be in a shunt conduction state to discharge instantaneous overload charge, eliminate the energy dissipation caused by volume scattering of suspended sediment particles, and avoid the problem of signal hard saturation collapse caused by complete blocking of the characteristic analysis spectrum.

[0022] 2. The central signal processing module monitors the continuous photocurrent signal output by the reference detection channel module and calculates the current change between adjacent sampling points. The change is divided by the sampling interval to obtain the absolute value of the first-order differential coefficient (i.e., the absolute value of the first-order differential) and defined as the optical flow micro-cluster mutation factor. When the factor exceeds the preset mutation boundary, it is determined that there is a mud and sand micro-cluster in the fluid that is about to block the measurement optical axis and a high-level command is issued to drive the solid-state switch in the feedforward shunt control topology to complete the closing action. Before the signal hard saturation, the low-impedance shunt branch is connected in parallel. The feedforward shunt balances the instantaneous overload charge caused by the sudden blockage of mud and sand, ensuring that the output level of the characteristic channel completes the dynamic range expansion before the characteristic beam disappears due to the blockage.

[0023] 3. The central signal processing module collects the output currents from the first and second signal detection modules, which are arranged in a symmetrical and discrete manner. When the fluid impacts the measurement cavity and causes the beam collimation to deflect axially, the surface received light flux of the two modules undergoes asymmetric differential deformation. The central signal processing module performs real-time calculation on the absolute value of the difference between the two currents through an operator. When the absolute value of the difference reaches the preset vibration threshold, it directly determines that the spatial optical path has undergone axial deflection disturbance. The time-domain window data of this sampling period is removed from the subsequent absorbance calculation matrix to block the error transmission of transient mechanical disturbance to the spectral measurement process, eliminate the false feature absorption noise interference caused by strong shear turbulence, and reduce the signal distortion caused by the loss of feature light flux. Attached Figure Description

[0024] Figure 1 This is a diagram showing the connection structure of the online monitoring system modules of the present invention;

[0025] Figure 2 This is a schematic diagram of the working principle of the central signal processing module of the present invention.

[0026] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0027] The technical solutions of the embodiments of this application will be clearly described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0028] An online monitoring system for agricultural non-point source pollution includes:

[0029] A multi-wavelength collinear transmission light emission module and a photoelectric detection module are respectively fixed on both sides of the fluid detection chamber module;

[0030] The photoelectric detection module includes a reference detection channel module, a first signal detection module, a second signal detection module, and a charge shunting control module, with the charge shunting control module connected in parallel to the signal output terminal of the reference detection channel module.

[0031] The central signal processing module is connected to the reference detection channel module, the first signal detection module, the second signal detection module, and the charge shunt control module. The central signal processing module is used to complete the following steps based on the transient electrical signal flow of each module during measurement within the fluid detection chamber module:

[0032] Step S1: Periodically collect and calculate the absolute value of the first derivative of the change in output current between adjacent sampling points of the reference detection channel module divided by the sampling interval time. When the absolute value of the first derivative is greater than the shunt change threshold, control the charge shunt control module to be in the shunt conduction state to discharge the instantaneous overload charge.

[0033] Step S2: Continuously calculate the absolute value of the output current difference between the first signal detection module and the second signal detection module in real time. When the absolute value of the current difference is greater than the optical path disturbance threshold, remove the interference data of the current sampling period in the current absorption brightness interpretation process.

[0034] Preferably, the system also includes a concentration analysis module; the concentration analysis module is connected to the central signal processing module. When the absolute value of the first derivative is less than or equal to the shunt mutation threshold, the concentration analysis module is used to obtain the output voltage ratio sequence of the characteristic wavelength channel and the reference detection channel module in the multi-wavelength collinear transmission emission module, and calculate the second-order variance of the output voltage ratio sequence within a defined time window; when the second-order variance is greater than the bubble interference threshold, the switching frequency of the thin-layer flow channel switching module in the fluid detection chamber module is adjusted according to the increment of the second-order variance.

[0035] Preferably, step S2 includes the following sub-steps: Step S21, under the initial calibration state, obtain the initial output current of the first signal detection module and the second signal detection module when the fluid detection chamber module is filled with standard deionized water, and divide the initial output current of the first signal detection module by the initial output current of the second signal detection module to determine the balance ratio coefficient; Step S22, during online monitoring, multiply the real-time output current of the second signal detection module by the balance ratio coefficient to generate a compensation current value, calculate the absolute value of the difference between the real-time output current of the first signal detection module and the compensation current value, and use it as the absolute value of the output current difference; when the absolute value of the output current difference is greater than the optical path disturbance threshold, remove the interference data of the current sampling period in the current absorbance interpretation process. Preferably, the system further includes a window status monitoring module; the window status monitoring module is connected to the central signal processing module, and in the monitoring state where the charge shunting control module is activated, the window status monitoring module is used to record the total number of shunting switching by the charge shunting control module within a set time window, and calculate the first derivative of the total number of shunting switching with time; when the first derivative is greater than the drift threshold, the window status monitoring module outputs a light-transmitting window cleaning control command to the central signal processing module.

[0036] Preferably, the steps for determining the shunt mutation threshold include: acquiring a reference photocurrent data sequence of the calibration water body; calculating the standard deviation of the reference photocurrent data sequence; adding three times the standard deviation to the inherent noise baseline value of the system, and determining the result of the addition as the shunt mutation threshold.

[0037] Preferably, the charge shunt control module uses a field-effect transistor with a conduction delay time of less than 10 μs as an internal switching component, and the response conduction time of the charge shunt control module after receiving the conduction control signal is in the range of 15 μs to 30 μs, and the impedance value of its internal low impedance branch is less than 10% of the input impedance of the preamplifier circuit in the central signal processing module.

[0038] Preferably, the multi-wavelength collinear transmission light emission module includes an ultraviolet light source, a visible light source, and a near-infrared light source; the light beams emitted by the ultraviolet light source, the visible light source, and the near-infrared light source are converged to the same collimated measurement optical axis by a beam combiner group; wherein, the light beam emitted by the near-infrared light source is directed toward the reference detection channel module, and the light beams emitted by the ultraviolet light source and the visible light source are directed toward the first signal detection module and the second signal detection module, respectively.

[0039] Preferably, step S1 includes the following sub-steps: Step S11, using the central signal processing module to obtain the output current change and sampling interval time of the reference detection channel module between adjacent sampling points, and calculating the absolute value of the first derivative; Step S12, when the absolute value of the first derivative is greater than the shunt change threshold twice consecutively, controlling the charge shunt control module to enter the shunt conduction state. Preferably, the system also includes a host protective chamber module and a surface cleaning control module; the host protective chamber module has a fluid detection chamber module and an optical light transmission window assembly embedded in it, the multi-wavelength collinear transmission light emission module and the photoelectric detection module are respectively fixedly arranged on both sides of the fluid detection chamber module, and the central signal processing module is encapsulated inside the host protective chamber module; the surface cleaning control module is arranged opposite to the optical light transmission window assembly, and the surface cleaning control module is used to remove the adhering substances on the surface of the optical light transmission window assembly when it receives the light transmission window cleaning control command output by the window status monitoring module.

[0040] A method for online monitoring of agricultural non-point source pollution, which operates in an online monitoring system for agricultural non-point source pollution, includes the following steps:

[0041] Step S101: A single-axis polychromatic parallel beam is emitted into the fluid detection chamber module using a multi-wavelength collinear transmission light emission module, so that the single-axis polychromatic parallel beam passes perpendicularly through the farmland runoff to be measured in the fluid detection chamber module. The single-axis polychromatic parallel beam contains a 220nm characteristic wavelength and a 700nm reference wavelength.

[0042] Step S102: The central signal processing module periodically collects the current output by the reference detection channel module and calculates the absolute value of the first derivative of the change in output current between adjacent sampling points divided by the sampling interval. When the absolute value of the first derivative is greater than the shunt change threshold, the charge shunt control module is controlled to be in the shunt conduction state to discharge the instantaneous overload charge. Step S103: The central signal processing module continuously calculates the absolute value of the difference between the output current of the first signal detection module and the second signal detection module in real time. When the absolute value of the current difference is greater than the optical path disturbance threshold, the interference data of the current sampling period is removed in the current absorption brightness interpretation process.

[0043] Example 1: In the specific working condition of the confluence of irrigation and drainage canals in the irrigation district, due to the sudden rain, the runoff is generated by the rainstorm. The normal fluid flow is accompanied by the fluctuation of sediment turbidity between 200 NTU and 1200 NTU. The evolution of the spatial configuration of the scattering matrix caused by the fluctuation of fluid shear rate leads to the instantaneous sweep of suspended sediment particles across the detection optical axis in the form of sudden micro-cluster surges. This working condition, which deviates from the homogeneous assumption, causes the characteristic analysis beam to be abruptly blocked, causing the output current of the spectral measurement channel to hit the bottom instantly and fall into the noise blind zone; or due to the instantaneous accumulation of charge, the linear range is exceeded, causing the fixed impedance topology to saturate the signal at the receiving end. The continuous fitting calculation at the post-end causes calculation divergence due to the distortion of the input variable reference.

[0044] To address the transient overload and signal saturation caused by the swell of sediment micro-clusters, the multi-wavelength collinear transmission emission module emits a uniaxial polychromatic parallel beam containing a 220nm characteristic wavelength and a 700nm reference wavelength into the fluid detection chamber module. The uniaxial polychromatic parallel beam passes perpendicularly through the farmland runoff to be measured within the fluid detection chamber module. The reference detection channel module, the first signal detection module, and the second signal detection module in the photoelectric detection module are fixed on both sides of the fluid detection chamber module to receive the transmitted spectral signal. The central signal processing module periodically acquires the current output from the reference detection channel module. And calculate the absolute value of the first derivative of the change in output current between adjacent sampling points divided by the sampling interval time. When the absolute value of the first differential Greater than the preset shunting mutation threshold At this time, the central signal processing module controls the charge shunt control module connected in parallel to the signal output terminal of the reference detection channel module to turn on. The low-impedance branch inside the charge shunt control module discharges instantaneous overload charge, shunting and balancing the instantaneous overload charge caused by the obstruction of sediment particles. This allows the output level of the characteristic channel to complete dynamic range expansion before the characteristic beam obstruction disappears, preventing the output signal of the photodetector from saturating. For reference, the output current of the detection channel module, For time, As a preset shunt change threshold, in the actual photoelectric conversion circuit, when a high concentration of heterogeneous mud and sand micro-clusters generates a step-like optical axis blockage, the photocurrent generated on the photosensitive detection surface experiences an extremely violent and steep transient drop. This is due to the inherent high time constant RC delay between the high resistance feedback resistor in the pre-impedance circuit and the parasitic junction capacitance of the photodiode.

[0045] This extremely high rate of change of negative current step causes a momentary deviation in the virtual ground balance at the input of the preamplifier, which in turn induces extremely high dynamic transient voltage overshoot on the parasitic capacitance of the feedback loop and the input, forming a nonlinear instantaneous charge accumulation. If this overload charge is not physically eliminated in time, the output stage of the entire amplifier circuit will be locked at the saturation gain edge of the linear range of the preamplifier circuit, resulting in a long recovery period blind zone, unable to respond to the transmitted light intensity signal that recovers as the surge passes. By connecting a low-impedance shunt branch in parallel between the capacitor plates at the input of the preamplifier circuit, a low-impedance path can be provided for these overload charges that accumulate instantaneously due to the deterioration of the input virtual ground to quickly discharge to the system reference ground. This forces the transient charge distribution of the amplifier input node to be balanced from an electrical principle perspective, thereby preventing the device hard saturation paralysis caused by steep negative signal jumps at the preamplifier load impedance topology level.

[0046] When strong shear turbulence causes mechanical vibration inside the fluid detection chamber module, resulting in a transient axial deflection of the collimating measurement optical axis, the surface-received light flux of the symmetrically separated first and second signal detection modules in the photoelectric detection module undergoes differential deformation; the central signal processing module, combined with the balance proportional coefficient... Calculate the output current of the first signal detection module With the output current of the second signal detection module The absolute value of the output current difference after balancing correction is expressed mathematically as follows: To establish a high-precision differential compensation benchmark for mechanical vibration, a dynamic calibration process was introduced into the system. Specifically, in the initial calibration state, the fluid detection chamber module is filled with standard deionized water, the multi-wavelength collinear transmission light-emitting module emits a light beam, and the central signal processing module records the initial output currents of the first and second signal detection modules. The initial output current of the first signal detection module is divided by the initial output current of the second signal detection module to determine the balance ratio coefficient. During online monitoring, the real-time output current of the second signal detection module will be... Multiply by the balance ratio factor Generate compensation current value Calculate the real-time output current of the first signal detection module. The absolute value of the difference between the output current and the compensation current value is taken as the absolute value of the output current difference. When the absolute value of the output current difference exceeds the preset optical path disturbance threshold... At this time, the central signal processing module identifies the current sampling period data as a mechanical axial deflection disturbance and removes the output data of the current sampling period from the current absorbance interpretation process to isolate the asymmetric influence of multiphase fluid concentration changes and external mechanical vibration on the signal. The charge shunt control module's shunt conduction control, in conjunction with the removal of the current sampling period's output data, reduces the overload charge of the photoelectric receiving array. The central signal processing module uses unsaturated current data to determine optical path disturbances. Simultaneously, the removal of the current sampling period's output data blocks the error propagation of transient axial deflection to the spectral measurement process. Within a single system architecture, this balances the charge overload caused by high scattering and the mechanical vibration deflection noise caused by strong shear turbulence. This is the output current of the first signal detection module. This is the output current of the second signal detection module. To achieve the preset optical path disturbance threshold, the charge shunting control module activates the shunting conduction, and the central signal processing module removes the output data of the current sampling period, the output charge overload and pseudo-feature absorption noise interference of the photoelectric detection module are reduced when the fluid detection chamber module faces the conditions of silt micro-clump surges and strong shear turbulence. This improves the long-term robustness of feature absorbance measurement in multiphase fluid environments.

[0047] The concentration of inorganic components in the farmland runoff to be measured within the fluid detection chamber module is interpreted by the central signal processing module. The photoelectric detection module returns to the absorbance interpretation state under homogeneous flow conditions after the sediment micro-particle surge ends. The charge shunt control module and the central signal processing module trigger pre-impedance adjustment and data time-domain window correction based on the external physical boundary state, forming an absorbance measurement architecture that does not rely on external mechanical filtering components. In order to establish a complete and objective interpretation closed loop from the underlying photoelectric conversion voltage to the final macroscopic target component concentration index, the central signal processing module embeds a standard linear calibration inversion model based on Lambert-Beer's law. In the actual concentration analysis process, after the central signal processing module completes the removal of interference data and feedforward shunt balancing within the sampling period, it reads in real time the digital output voltage of the 220 nm characteristic wavelength channel and the digital output of the 700 nm reference wavelength channel under the current bubble-free and unsaturated state. The central signal processing module divides the voltage value of the aforementioned characteristic wavelength channel by the voltage value of the reference wavelength channel to obtain the current real-time ratio. Then, it calls the built-in mathematical operator to perform a natural logarithmic operation on the ratio sequence, thereby extracting the characteristic absorbance scalar value after excluding interference from light source thermal attenuation and scattering background. The central signal processing module inputs the absorbance scalar value into a pre-stored first-order analytical mapping equation. The slope calibration coefficient and intercept constant of this equation are pre-fixed in the memory by measuring the absorbance of five groups of standard nitrate nitrogen pollutant samples with concentrations ranging from 10 mg / L to 50 mg / L under the power-on calibration state and performing linear fitting regression using the least squares method. After multiplying the characteristic absorbance scalar value by the slope calibration coefficient and adding the intercept constant, the central signal processing module can directly demodulate and output the measured absolute concentration value of the target inorganic component pollutant in the farmland runoff, in milligrams per liter.

[0048] The thin-layer flow channel switching module includes an electromagnetic reversing valve and parallel long-path and short-path thin-layer flow channels. The physical measured optical path of the long-path flow channel is 10 mm, and the physical measured optical path of the short-path thin-layer flow channel is 1 mm. During online monitoring, when the second-order variance of the output voltage ratio sequence within a fixed time window of 50 milliseconds is less than or equal to the bubble interference threshold, the central signal processing module outputs a low-level control signal to drive the electromagnetic reversing valve to remain open. Only the long-path flow channel is conducted in the fluid detection chamber module. When the second-order variance exceeds the bubble interference threshold, the variance increment is obtained by subtracting the bubble interference threshold from the second-order variance. The switching frequency of the electromagnetic reversing valve is determined according to a preset linear mapping relationship. The switching frequency is equal to the reference frequency of 2 Hz plus the product of the frequency gain coefficient and the variance increment. The frequency gain coefficient is 5 Hz per unit variance. A pulse width modulation signal is sent to the electromagnetic reversing valve to alternately switch the conduction state of the long-path flow channel and the short-path thin-layer flow channel, shortening the scattering distance of characteristic photons through the gas-liquid two-phase medium.

[0049] Example 2: In the confluence of irrigation and drainage canals in farmland, the charge overload of the multispectral absorbance channels under the scouring of suspended solids and the optical path deflection interference in the turbulent mixing zone were addressed. A fluid property measurement platform equipped with a collinear transmission analysis beam and a multi-channel differential tuning receiver circuit was used for comparative verification. The high-frequency sampling component, with a current resolution of 0.1 μA and a conversion sampling rate of 50 kHz, was used to capture the polarization current change waveform when heterogeneous solid microparticles transiently pass over the probe optical axis. The sampling period of the high-frequency sampling component... To balance the timeliness of time-domain capture with the computational load of the central signal processing module, the quantization process involves setting a lower limit for the sampling frequency based on twice the highest spectral frequency of the signal variation to meet anti-aliasing constraints. Under the condition that the time constant for a heterogeneous solid micro-particle gliding across the optical axis is 10 ms, the sampling period is calculated through anti-aliasing constraint resolution. The sampling period was set to a fixed 100 μs. In the test fluid loop, the shear flow driven by the water pump was used to apply a continuous mechanical-hydraulic impact to the outer shell of the fluid detection chamber module. This actively injected optical axis deflection disturbance with first-order attenuation characteristics into the measurement system to simulate the scouring interference of multiphase flow in the external field. In the actual circuit execution, from the time the high-frequency sampling component captured the abnormal over-limit of the first-order differential absolute value at the current sampling point, to the time the central signal processing module completed the digital judgment, issued the control command, and the field-effect transistor switch was completely closed and turned on, the entire hardware control chain inevitably had a physical delay dead zone formed by the superposition of the fixed sampling period of 100 microseconds and the inherent response delay of the device of 15 microseconds to 30 microseconds. The total timing lag was between 115 microseconds and 130 microseconds. In order to eliminate the staged physical hard saturation blind zone caused by the steep scouring of mud and sand in this hundred-microsecond transition dead zone, the system was designed to eliminate the physical hard saturation blind zone caused by the steep scouring of mud and sand in the system.

[0050] This invention adds a low-pass RC hysteresis absorption network, consisting of a 20 picofarad miniature chip capacitor and a 10 kΩ resistor, to the input pin of the preamplifier circuit. This RC hysteresis absorption network provides a physical signal smoothing hysteresis of approximately 0.2 microseconds at the input preamplifier stage. This effectively delays the absolute rate at which transient polarization charge caused by sudden blocking accumulates on the amplifier core feedback plate, thus providing sufficient time control for the algorithm determination of the central signal processing module and the turn-on of the MOSFET. Within the transition dead zone of this mode transition, the preamplifier circuit relies on the transient charge storage effect of the RC hysteresis absorption network to maintain itself at the edge of the linear operating range. Once the MOSFET is fully turned on within 130 microseconds, the overload charge accumulated at the preamplifier input is immediately discharged by the parallel low-impedance branch, thereby ensuring that the system does not trigger signal saturation failure during the system state abrupt transition period.

[0051] Test samples were introduced under multiple suspended solids concentration gradient conditions. Structures lacking impedance recombination shunt circuits were designated as the control group, while structures with complete charge discharge and disturbance interception circuits were designated as the present invention's sample group. When the fluid was at the lower limit turbidity of 200 NTU, the reference detection channel module output current generated as the inorganic particles swept across the measurement optical axis in the control group was measured. The value is 2.1 μA, derived from the absolute value of the first-order differential between adjacent time-domain sampling points obtained from demodulation. The current is 0.05 A / s, which does not meet the preset shunt surge threshold. Furthermore, the fluid transmittance is abundant, and the characteristic absorbance output of both the control group and the sample group of this invention remains within a stable range of 0.155. However, when the medium is changed to a median turbidity of 700 NTU, a sudden dense formation of solid micro-clusters causes a momentary accumulation of photogenerated charges in the control group, resulting in a decrease in the output current of the reference detection channel module. A step drop occurs, and the absolute value of its first differential is... Increasing it to 0.45 A / s causes the absolute value of the first derivative to exceed the preset shunt abrupt change threshold of 0.20 A / s. In this invention, the parallel charge shunt control module within the sample group connects to the low-impedance discharge branch to balance transient overload charges, ensuring that the measured absorbance output by the spectroscopic interpretation unit remains at 0.482 and its characteristic voltage does not exceed the linear range. Conversely, in the control group, the receiver signal is locked at the full-scale saturation upper limit of 5.00V due to overload of the photogenerating circuit. When the medium further undergoes regular variation to the upper limit of turbidity of 1200 NTU, resulting in a multiple scattering overload state, the output current of the reference detection channel module measured by the sample group of this invention... The slope of the fall intensifies, and the absolute value of the first derivative increases. Reaching 0.85 A / s, continuously exceeding the preset shunt surge threshold. Under the constraints, the transistor shunt branch of the present invention maintains a constant conduction state to expand the preconditioning range, thereby outputting an absorbance resolution value of 0.812 under a large-scale obscuring state of the feature analysis beam, while the current of the spectrophotometer channel in the control group reaches its limit, causing the measurement output to be interrupted; wherein For reference, the output current of the detection channel module, For time, This is the preset shunt mutation threshold.

[0052] During the heterogeneous pulsating interference section subjected to transient hydrodynamic impact, the angular displacement vibration of the fluid detection chamber module housing causes the collimation measurement optical axis to deviate from the geometric center line, resulting in an imbalance in the received light intensity distribution on the surfaces of the symmetrically arranged first and second signal detection modules. At this time, because the control group has not undergone differential gain balancing, the output current of its first signal detection module will be affected. Output current of the second signal detection module When an asymmetric deviation occurs, the absolute value of the uncorrected difference is measured. The optical path disturbance reaches 1.2 μA and continuously exceeds the preset 0.5 μA threshold. Because the control group directly included the differential deformation error, which included the intrinsic hardware asymmetry error, in the continuous photoelectric calculation process, the final interpreted inorganic component concentration data exhibited a 45% random fluctuation. In contrast, the sample group of this invention balanced the proportionality coefficient... To balance the output current, the absolute value of the calculated output current difference is used. Exceeding the optical path perturbation threshold The sampling time-domain window closes the data output bus, eliminating disturbed discrete anomaly parameters to block the propagation of geometric deflection error to the concentration inversion topology, keeping the multispectral measurement error under strong shear vibration interference within 1.8%. Simultaneously, to determine the adaptive convergence limit of the controllable measurement physical boundary, the concentration of suspended inorganic solids is further increased to 1500 NTU, exceeding the protection range. At this point, due to the volume scattering caused by the extremely high solid density, multiple nonlinear photoresistance saturation occurs in the characteristic absorbance, resulting in the energy transmittance of the characteristic analysis beam falling below the lower edge of the photoelectric detection array's detection limit. The transmitted photocurrent output by the sample group of this invention undergoes nonlinear decay and converges to zero. After crossing the working inflection point, the absorbance gain curve enters a flat plateau region with a slope approaching zero, indicating that 1200 NTU is the upper boundary of the physical operating condition where the feedforward reconstruction circuit and the time-domain elimination program work together to maintain stable interpretation. Furthermore, when the parameters are modified to reduce the shunt abrupt change threshold... When adjusted to an extremely low value below the window threshold, the preamplifier branch frequently experiences malfunctioning current conduction under normal random noise disturbances, leading to periodic disordered impedance steps in the differential output voltage and reducing the system's characteristic absorbance resolution. The above test results indicate that the defined threshold window and control response logic constitute a stable operating window that balances resistance to high-turbidity charge impacts and flow field mechanical vibrations. This is the output current of the first signal detection module. This is the output current of the second signal detection module. Based on the preset optical path disturbance threshold and the physical property test data under multiple concentration gradients and mechanical vibration interference, the technique of using the absolute value of the first derivative of the output current of the reference detection channel module to determine the heterogeneous particle blocking event and feedforward to trigger the conduction of the low impedance branch, combined with the time-domain elimination of the asymmetric axial deflection noise of the optical path by the absolute value of the output current difference of the dual signal detection module, allows the farmland runoff under test inside the fluid detection chamber module to stably interpret and output the corresponding inorganic component concentration measurement value after experiencing transient high-scattering sediment solid overload and heterogeneous flow field impact interference. Furthermore, the photoelectric detection module automatically recovers to the absorbance inversion state under the homogeneous conventional flow condition after the sediment micro-particle surge passes through the measurement optical axis. Thus, the complex spatiotemporal physical and chemical interference in the multiphase fluid environment is converted into a trigger signal source for adjusting the circuit impedance topology and the data time-domain window, forming an in-situ online chemical component monitoring architecture for absorbance that does not require an external cascaded mechanical filter and offsets multiple environmental disturbances through internal photoelectric topology logic.

[0053] Example 3: This example combines Figures 1 to 2 This section describes the online monitoring system and methods for agricultural non-point source pollution, such as... Figure 1 As shown, a multi-wavelength collinear transmission light-emitting module emits a single-axis polychromatic parallel beam. This beam passes perpendicularly through and is delivered to the fluid detection chamber module. The fluid detection chamber module contains the farmland runoff fluid to be measured and performs transmitted light dissipation distribution. A reference detection channel module periodically collects photoelectric outputs between adjacent sampling points and transmits a reference output current to the central signal processing module. A charge shunting control module is connected in parallel to the signal output terminal to shunt and discharge instantaneous overload charges. A first signal detection module collects a first-channel transient transmitted light signal and transmits a first-channel output current to the central signal processing module. A second signal detection module collects a second-channel transient transmitted light signal and transmits it to the central signal processing module. The first processing module transmits the second output current. The central signal processing module integrates the feedforward balancing strategy in step S1 and the data elimination strategy in step S2. The feedforward balancing strategy in step S1 is used to calculate the absolute value of the first derivative of the change in the output current of the reference detection channel. When it is determined that the absolute value of the first derivative is greater than the shunt change threshold, a shunt conduction state command is issued to control the charge shunt control module to conduct. The data elimination strategy in step S2 is used to continuously calculate the absolute value of the difference between the output current of the first signal detection module and the second signal detection module in real time. When it is determined that the absolute value of the current difference is greater than the optical path disturbance threshold, the interference data of the current sampling period is eliminated.

[0054] like Figure 2As shown, the farmland runoff to be measured is input into the system. The system periodically collects the output current of the reference detection channel module. The central signal processing module is connected to the behavior signal of the periodically collected reference detection channel module output current. The central signal processing module is also connected to control the charge shunt control module to be in the shunt conduction state. The behavior of the periodically collected reference detection channel module output current and the behavior of the charge shunt control module being in the shunt conduction state jointly guide the discharge of instantaneous overload charge. At the same time, when strong shear turbulence is generated externally, the central signal processing module continuously calculates the absolute value of the difference between the output current of the first signal detection module and the second signal detection module in real time. This value is then directly used to eliminate interference data in the current sampling period. The central signal processing module maintains signal connection with both the behavior of continuously calculating the absolute value of the difference between the output current of the first signal detection module and the second signal detection module and the behavior of eliminating interference data in the current sampling period.

[0055] Example 4: When the system faces changes in background temperature of farmland runoff caused by seasonal changes, and the light source luminous efficiency decay accompanying the continuous operation of the multispectral collinear transmission light emission module in the field for more than 720 hours, the current reference at the receiving end of the photoelectric detection module will deviate from the zero point in one direction. This physical variation causes the circuit switch of the pre-stage to generate control word triggering deviation, which in turn causes the absorbance data of inorganic components inside the flow channel to output abnormal concentration indicators due to the characteristic signal deviating from the calibration working point.

[0056] Under the calibration state of the fluid detection chamber module being powered on or injected with pollution-free standard demineralized water, the multi-wavelength collinear transmission light-emitting module emits a uniaxial polychromatic parallel beam towards the fluid detection chamber module, and the central signal processing module continuously acquires the output current of the reference detection channel module at a fixed sampling period of 100μs. The 50 continuously collected data points are temporarily stored in a circular sliding buffer to calculate the output current. random noise variance within this time window Simultaneously, the output current of the first signal detection module is collected. Output current of the second signal detection module And calculate the combined balance ratio coefficient. Absolute value of the output current difference after balancing The variance of the difference in random noise within this time window The central signal processing module will calculate the random noise variance value. Multiply the square root of the arithmetic number by 3 and then divide by the sampling period. To determine the shunt mutation threshold At the same time, the variance of the difference random noise value Multiply the square root of the arithmetic root by the number 3 to determine the optical path perturbation threshold. .

[0057] During online measurement in the fluid detection chamber module, the central signal processing module receives the digital electrical signal from the reference probe channel module via the universal asynchronous transceiver interface and converts it into the current output current. The difference between the current output current and the temporary output current stored in the previous cycle is calculated and divided by the sampling period. Obtain the real-time absolute value of the first-order derivative. When the absolute value of the first differential Greater than the shunt mutation threshold At the same time, the central signal processing module writes a shunt conduction signal to the control register of the charge shunt control module, causing the low-impedance pipeline connected in parallel to the signal output terminal of the reference detection channel module to conduct, thereby drawing in and discharging the instantaneous overload charge generated by the high turbidity flushing, and simultaneously reading the output current of the first signal detection module. Output current of the second signal detection module And calculate the absolute value of the difference between the output currents of the two. When the absolute value of the output current difference Greater than the optical path disturbance threshold At this time, the central signal processing module shuts off the output update of the data bus during the absorption luminance interpretation process, skips the data points of the current sampling period and directly calls the historical temporary current data value that was not disturbed in the previous sampling period as the replacement data value, thus blocking the transmission of the mechanical axial deflection error caused by strong shear turbulence to the subsequent inversion network.

[0058] The central signal processing module triggers benchmark calibration with a 24-hour clock cycle, assuming that the farmland runoff is in a homogeneous, conventional flow state and the absolute value of the first derivative is less than the shunt abrupt change threshold for 1000 consecutive sampling points. When it reaches 10%, calculate the output current of the reference detection channel module during that time period. The discrete arithmetic mean is used as the zero-point calibration source. The average current is multiplied by a correction factor of 0.2 and weighted and summed with 0.8 times the historical reference current value in the original memory address. The summation result is used to overwrite the reference in the memory address to correct the unidirectional drift of the photocurrent baseline caused by the long-term thermal decay of the light-emitting device. This allows the spectral detection module to maintain the current linear conditioning range under environmental changes and operating conditions exceeding 720 hours, and outputs scalar data values ​​of desensitized farmland runoff absorbance measurement without signal saturation and data divergence.

[0059] Example 5: When the system faces a multi-point distributed cluster deployment of newly built irrigation and drainage canal nodes, due to differences in the channel geometry and bottom reflectivity at different geographical locations, the initial photocurrent reference of the receiving component in the absence of suspended solids deviates from the theoretical preset. Directly calling fixed parameters will cause the feedforward shunt control program to malfunction. The fluid detection chamber module performs on-site pre-deployment calibration before water intake; the central signal processing module reads the first static dark current component of the reference detection channel module when the detection channel is emptied. The electric drainage valve is opened to inject clarified water at a flow rate of 0.5 m / s into the fluid detection chamber module. The high-frequency sampling component continuously collects 60,000 sampling data points within 60 seconds to obtain the second static apparent current component. The central signal processing module follows the formula. Calculate the absolute background conductance gain scalar of the current physical field. Using this as a physical background conversion factor, the shunting mutation threshold is recalculated. The multiplier base; where, The absolute background conductance gain scalar, This is the second static bright current component. This is the first static dark current component.

[0060] Establishing the absolute background conductivity gain scalar Then, the central signal processing module will scalarize the absolute background conductance gain. The optical path perturbation threshold is corrected by multiplying the initial perturbation coefficient stored in the memory. When transient blocking occurs in suspended sediment particles within the water flow shear layer, causing a sharp change in absorbance, the updated optical path perturbation threshold is used. Match the scattering attenuation envelope of the current monitoring section and output current in the first signal detection module. Output current of the second signal detection module Balance ratio coefficient Absolute value of the output difference after balancing Greater than the optical path disturbance threshold At that time, the time-domain filtering loop performs discrete window data locking to eliminate the method judgment benchmark distortion caused by the cross-regional migration of hardware nodes. This allows the monitoring system to achieve stable inversion output of the measured values ​​of farmland runoff absorbance and target pollutant concentration, unaffected by background light field distortion, after local convergence and conditioning of field parameters. The initial perturbation coefficient, which is stored in the memory, is determined by statistically analyzing continuous static baseline current sampling of a symmetrical detection quadrant array under ideal static conditions without any external mechanical vibration or flow field deflection in a standard static desalination anechoic chamber environment in the laboratory for up to 48 hours. Specifically, this initial perturbation coefficient is defined as the first... Between the first and second signal detection modules, the intrinsic dark current noise difference standard deviation caused by microscopic differences in semiconductor manufacturing processes is three times that of the second module. In this embodiment, the fixed initial perturbation coefficient is quantified as a dimensionless constant of 0.15. During the field calibration process, by multiplying the absolute background conductivity gain scalar with the initial perturbation coefficient, the static hardware asymmetric intrinsic physical defects can be seamlessly transformed into a dynamic current noise tolerance baseline that matches the background light field intensity of the current deployment section. This establishes a clear engineering statistical and physical theoretical boundary for the corrected optical path perturbation threshold, effectively avoiding misjudgments caused by hardware migration across regions.

[0061] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0062] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An online monitoring system for agricultural non-point source pollution, characterized in that, include: A multi-wavelength collinear transmission light emission module and a photoelectric detection module are respectively fixed on both sides of the fluid detection chamber module; The photoelectric detection module includes a reference detection channel module, a first signal detection module, a second signal detection module, and a charge shunting control module, with the charge shunting control module connected in parallel to the signal output terminal of the reference detection channel module. The central signal processing module is connected to the reference detection channel module, the first signal detection module, the second signal detection module, and the charge shunt control module. The central signal processing module is used to complete the following steps based on the transient electrical signal flow of each module during measurement within the fluid detection chamber module: Step S1: Periodically collect and calculate the absolute value of the first derivative of the change in output current between adjacent sampling points of the reference detection channel module divided by the sampling interval time. When the absolute value of the first derivative is greater than the shunt change threshold, control the charge shunt control module to be in the shunt conduction state to discharge the instantaneous overload charge. Step S2: Continuously calculate the absolute value of the output current difference between the first signal detection module and the second signal detection module in real time. When the absolute value of the current difference is greater than the optical path disturbance threshold, remove the interference data of the current sampling period in the current absorption brightness interpretation process.

2. The online monitoring system for agricultural non-point source pollution according to claim 1, characterized in that, The system also includes a concentration analysis module; the concentration analysis module is connected to the central signal processing module. When the absolute value of the first derivative is less than or equal to the shunt mutation threshold, the concentration analysis module is used to obtain the output voltage ratio sequence of the characteristic wavelength channel and the reference detection channel module in the multi-wavelength collinear transmission emission module, and calculate the second-order variance of the output voltage ratio sequence within a certain time window; when the second-order variance is greater than the bubble interference threshold, the switching frequency of the thin-layer flow channel switching module in the fluid detection chamber module is adjusted according to the increment of the second-order variance.

3. The online monitoring system for agricultural non-point source pollution according to claim 1, characterized in that, Step S2 includes the following sub-steps: Step S21, under the initial calibration state, acquire the initial output current of the first signal detection module and the second signal detection module when the fluid detection chamber module is filled with standard deionized water, and divide the initial output current of the first signal detection module by the initial output current of the second signal detection module to determine the balance ratio coefficient; Step S22, during online monitoring, multiply the real-time output current of the second signal detection module by the balance ratio coefficient to generate a compensation current value, calculate the absolute value of the difference between the real-time output current of the first signal detection module and the compensation current value, and use it as the absolute value of the output current difference; when the absolute value of the output current difference is greater than the optical path disturbance threshold, remove the interference data of the current sampling period in the current absorbance interpretation process.

4. The online monitoring system for agricultural non-point source pollution according to claim 1, characterized in that, The system also includes a window status monitoring module; the window status monitoring module is connected to the central signal processing module. In the monitoring state where the charge shunting control module is activated, the window status monitoring module records the total number of shunting switches by the charge shunting control module within a set time window and calculates the first derivative of the total number of shunting switches with time; when the first derivative is greater than the drift threshold, the window status monitoring module outputs a light-transmitting window cleaning control command to the central signal processing module.

5. The online monitoring system for agricultural non-point source pollution according to claim 1, characterized in that, The steps for determining the shunt mutation threshold include: acquiring a reference photocurrent data sequence of the calibration water body; calculating the standard deviation of the reference photocurrent data sequence; adding three times the standard deviation to the inherent noise baseline value of the system, and determining the result of the addition as the shunt mutation threshold.

6. The online monitoring system for agricultural non-point source pollution according to claim 1, characterized in that, The charge shunt control module uses a field-effect transistor with a conduction delay time of less than 10μs as an internal switching component. The response conduction time of the charge shunt control module after receiving the conduction control signal ranges from 15μs to 30μs. The impedance value of its internal low-impedance branch is less than 10% of the input impedance of the preamplifier circuit in the central signal processing module.

7. The online monitoring system for agricultural non-point source pollution according to claim 1, characterized in that, The multi-wavelength collinear transmission light emission module includes an ultraviolet light source, a visible light source, and a near-infrared light source; the light beams emitted by the ultraviolet light source, the visible light source, and the near-infrared light source are converged to the same collimated measurement optical axis by a beam combiner group; among them, the light beam emitted by the near-infrared light source is directed toward the reference detection channel module, and the light beams emitted by the ultraviolet light source and the visible light source are directed toward the first signal detection module and the second signal detection module, respectively.

8. The online monitoring system for agricultural non-point source pollution according to claim 1, characterized in that, Step S1 includes the following sub-steps: Step S11, using the central signal processing module to obtain the output current change and sampling interval time of the reference detection channel module between adjacent sampling points, and calculating the absolute value of the first derivative; Step S12, when the absolute value of the first derivative is greater than the shunting change threshold twice in a row, controlling the charge shunting control module to enter the shunting conduction state.

9. The online monitoring system for agricultural non-point source pollution according to claim 4, characterized in that, The system also includes a host protective chamber module and a surface cleaning control module. The host protective chamber module has a fluid detection chamber module and an optical light transmission window assembly embedded in it. The multi-wavelength collinear transmission light emission module and the photoelectric detection module are respectively fixed on both sides of the fluid detection chamber module, and the central signal processing module is encapsulated inside the host protective chamber module. The surface cleaning control module is arranged opposite to the optical light transmission window assembly. The surface cleaning control module is used to remove the adhering substances on the surface of the optical light transmission window assembly when it receives the light transmission window cleaning control command output by the window status monitoring module.

10. A method for online monitoring of agricultural non-point source pollution, applied to an online monitoring system for agricultural non-point source pollution according to any one of claims 1 to 9, characterized in that, Includes the following steps: Step S101: A single-axis polychromatic parallel beam is emitted into the fluid detection chamber module using a multi-wavelength collinear transmission light emission module, so that the single-axis polychromatic parallel beam passes perpendicularly through the farmland runoff to be measured in the fluid detection chamber module. The single-axis polychromatic parallel beam contains a 220nm characteristic wavelength and a 700nm reference wavelength. Step S102: The central signal processing module periodically collects the current output by the reference detection channel module and calculates the absolute value of the first derivative of the change in output current between adjacent sampling points divided by the sampling interval time. When the absolute value of the first derivative is greater than the shunt change threshold, the charge shunt control module is controlled to be in the shunt conduction state to discharge the instantaneous overload charge. Step S103: The central signal processing module continuously calculates the absolute value of the output current difference between the first signal detection module and the second signal detection module in real time. When the absolute value of the current difference is greater than the optical path disturbance threshold, the interference data of the current sampling period is removed in the current absorption brightness interpretation process.

Images