Wellhead integrated underground water automatic monitoring terminal

Abstract

The invention belongs to the technical field of intelligent monitoring, and particularly relates to a wellhead integrated underground water automatic monitoring terminal. Comprising a sampling cavity arranged in a wellhead casing, and underground water enters the sampling cavity through a water inlet hole in the wellhead casing; the three-electrode probe assembly is mounted in the sampling cavity and comprises a working electrode, a reference electrode and an auxiliary electrode; the electrochemical workstation module is used for applying a sine disturbance voltage signal superposed on the steady-state open-circuit potential to the three-electrode probe assembly; the model identification module is used for generating a Nyquist diagram scatter sequence according to the original electrochemical impedance spectroscopy data sequence to form an equivalent circuit model identification result; the water quality calculation module is used for calculating underground water quality indexes according to the identification result of the equivalent circuit model; and the communication module is used for uploading the underground water quality index to a remote underground water monitoring platform. The method has the advantages of being compact in structure, high in intelligent degree, high in calculation efficiency and high in adaptability.

Description

Technical Field

[0001] This invention belongs to the field of intelligent monitoring technology, and particularly relates to the field of intelligent sensing systems, specifically an integrated groundwater automatic monitoring terminal for wellheads. Background Technology

[0002] Currently, commonly used automatic groundwater monitoring equipment mainly employs single-parameter or multi-parameter sensor combinations, such as conductivity sensors, dissolved oxygen sensors, and pH sensors, for water quality monitoring. Conductivity sensors typically use a two-electrode or four-electrode structure, calculating the solution conductivity by measuring the AC resistance between the electrodes, and then estimating the total dissolved solids content in the water. However, these sensors can only acquire resistance information at a single frequency and cannot distinguish the contribution of the solution's bulk resistance and the electrode interface impedance. When the electrode surface is contaminated or passivated, the measurement results will produce significant deviations. Furthermore, traditional conductivity sensors lack self-diagnostic capabilities for electrode conditions, making it difficult to determine the reliability of the measurement results, often requiring regular manual maintenance and calibration.

[0003] Electrochemical impedance spectroscopy (EIS) is an analytical method that measures the impedance response of electrochemical systems over a wide frequency range, providing far richer electrochemical information than single-frequency measurements. This technique applies a small-amplitude sinusoidal perturbation signal to an electrochemical system and measures its impedance response at different frequencies, thereby revealing the kinetic processes at the electrode interface and the mass transfer characteristics of the solution. While EIS has been widely used in corrosion monitoring, battery performance evaluation, and biosensors, its application in groundwater quality monitoring is still in its early stages. Existing EIS measurement equipment is mostly benchtop laboratory instruments, which are bulky, power-hungry, and expensive, making them unsuitable for long-term unattended operation in field environments. Summary of the Invention

[0004] The main objective of this invention is to provide an integrated groundwater automatic monitoring terminal for wellheads. Through an adaptive identification method of equivalent circuit model based on the segmented analysis of Nyquist plot geometric features, it achieves automatic identification of equivalent circuit topology and direct analytical calculation of component parameters. This avoids the problems of sensitivity to initial values ​​and easy getting trapped in local optima in traditional iterative fitting methods. At the same time, it realizes automatic diagnosis of electrode contamination status through dispersion index identification value. It has the advantages of compact structure, high level of intelligence, high computational efficiency, strong adaptability, and long-term unattended operation.

[0005] To solve the above-mentioned technical problems, the present invention provides an integrated groundwater automatic monitoring terminal for wellheads, comprising: The sampling chamber is located inside the wellhead casing, and groundwater enters the sampling chamber through the water inlet on the wellhead casing; The three-electrode probe assembly is installed in the sampling chamber and includes a working electrode, a reference electrode, and an auxiliary electrode. The electrochemical workstation module is used to apply a sinusoidal perturbation voltage signal superimposed on the steady-state open-circuit potential to the three-electrode probe assembly, acquire the current response signal at each frequency point, and calculate the original electrochemical impedance spectroscopy data sequence based on the sinusoidal perturbation voltage signal and the current response signal. The model identification module is used to generate a Nyquist plot scatter sequence based on the original electrochemical impedance spectroscopy data sequence. By analyzing the geometric characteristics of the Nyquist plot scatter sequence, the equivalent circuit topology type is determined and the parameter values ​​of each equivalent circuit element are identified, thus forming the equivalent circuit model identification result. The water quality calculation module is used to calculate groundwater quality indicators based on the identification results of the equivalent circuit model. The communication module is used to upload groundwater quality indicators to the remote groundwater monitoring platform.

[0006] Furthermore, the working electrode is an inert metal working electrode, the reference electrode is a silver silver chloride reference electrode, and the auxiliary electrode is a platinum wire auxiliary electrode. The inert metal working electrode and the platinum wire auxiliary electrode are arranged in parallel and maintain a fixed electrode spacing between them. The silver silver chloride reference electrode is located at the midline position between the inert metal working electrode and the platinum wire auxiliary electrode.

[0007] Furthermore, the electrochemical workstation module first measures the open-circuit potential of the three-electrode probe assembly in groundwater and records it as the steady-state open-circuit potential. Then, a sinusoidal perturbation voltage signal superimposed on the steady-state open-circuit potential is applied. The amplitude of the sinusoidal perturbation voltage signal is set to a fixed millivolt level to ensure that the electrochemical system is in the linear response range. The frequency of the sinusoidal perturbation voltage signal decreases point by point from the initial high-frequency value to the final low-frequency value. The frequency decrease method adopts logarithmic equal interval decrease.

[0008] Furthermore, the electrochemical workstation module performs synchronous sampling of the sinusoidal perturbation voltage signal and current response signal at each frequency point, calculates the amplitude ratio of the two as the impedance magnitude at that frequency point, calculates the phase difference between the two as the impedance phase angle at that frequency point, and calculates the complex impedance value at that frequency point based on the impedance magnitude and impedance phase angle. The complex impedance value includes a real impedance component and an imaginary impedance component. The real impedance component is equal to the impedance magnitude multiplied by the cosine of the impedance phase angle, and the imaginary impedance component is equal to the impedance magnitude multiplied by the sine of the impedance phase angle. The complex impedance values ​​of all frequency points are arranged in descending order of frequency to form the original electrochemical impedance spectral data sequence.

[0009] Furthermore, the model identification module uses the real impedance component of each complex impedance value in the original electrochemical impedance spectroscopy data sequence as the abscissa and the opposite of the imaginary impedance component as the ordinate to generate a Nyquist plot scatter sequence. The model identification module traverses the Nyquist plot scatter sequence along the frequency decreasing direction. When the difference in the ordinate between adjacent scatter points changes from a positive value to a negative value, the scatter point at the change point is located as the capacitive arc vertex. The characteristic frequency, abscissa, and ordinate of each capacitive arc vertex are recorded, and the abscissa value of the scatter point with the highest frequency is extracted as the high-frequency real intercept value.

[0010] Furthermore, the model identification module determines the equivalent circuit topology type based on the number of detected capacitive reactance arc vertices: when only the first capacitive reactance arc vertices are detected, it is determined to be a single-time-constant type. The single-time-constant type equivalent circuit topology is composed of a solution resistive element and a first parallel RC unit connected in series. The first parallel RC unit is composed of a first charge transfer resistive element and a first constant phase angle element connected in parallel. When both the first and second capacitive reactance arc vertices are detected simultaneously, it is determined to be a double-time-constant type. The double-time-constant type equivalent circuit topology is composed of a solution resistive element, a first parallel RC unit, and a second parallel RC unit connected in series. The second parallel RC unit is composed of a second charge transfer resistive element and a second constant phase angle element connected in parallel.

[0011] Furthermore, for the single-time-constant type equivalent circuit topology, the model identification module uses the high-frequency real axis intercept value as the solution resistance identification value; determines the radius of the first capacitive arc based on the ordinate of the vertex of the first capacitive arc, and the identification value of the first charge transfer resistance is equal to twice the radius of the first capacitive arc; calculates the first time constant as the reciprocal of the first characteristic frequency multiplied by twice pi, and the identification value of the first quasi-capacitance is equal to the first time constant divided by the identification value of the first charge transfer resistance; extracts the previous and next adjacent scattered points of the first capacitive arc vertex, calculates the slope of the line connecting the previous adjacent scattered point and the first capacitive arc vertex as the slope of the front tangent, calculates the slope of the line connecting the first capacitive arc vertex and the next adjacent scattered point as the slope of the back tangent, calculates the absolute value of the difference between the arctangent of the front tangent slope and the arctangent of the back tangent slope as the vertex angle, and the identification value of the first diffusion index is equal to the vertex angle divided by pi.

[0012] Furthermore, for the dual-time-constant equivalent circuit topology, the model identification module uses the high-frequency real-axis intercept value as the solution resistance identification value; it divides the scatter points with frequencies higher than the geometric mean of the first and second characteristic frequencies into a high-frequency scatter point subset, and the scatter points with frequencies lower than the geometric mean into a low-frequency scatter point subset; it identifies the first charge transfer resistance identification value, the first quasi-capacitance identification value, and the first dispersion index identification value for the high-frequency scatter point subset according to the component parameter identification method of the single-time-constant equivalent circuit topology; it calculates that the first capacitive arc low-frequency intercept is equal to the high-frequency real-axis intercept value plus the first charge transfer resistance identification value; it performs coordinate translation on the low-frequency scatter point subset, subtracts the first capacitive arc low-frequency intercept from the abscissa of each scatter point to form a translated low-frequency scatter point subset; and it identifies the second charge transfer resistance identification value, the second quasi-capacitance value, and the second dispersion index value for the translated low-frequency scatter point subset.

[0013] Furthermore, the water quality calculation module reads the electrode spacing value and the effective electrode area value from the parameter storage unit, calculates the electrode constant by dividing the electrode spacing value by the effective electrode area value, and calculates the groundwater conductivity monitoring value by dividing the electrode constant by the solution resistance identification value in the equivalent circuit model identification result. The water quality calculation module reads the preset conductivity-mineralization linear transformation slope and conductivity-mineralization linear transformation intercept from the parameter storage unit, calculates the groundwater mineralization monitoring value by multiplying the groundwater conductivity monitoring value by the conductivity-mineralization linear transformation slope and adding the conductivity-mineralization linear transformation intercept.

[0014] Furthermore, the water quality calculation module evaluates the electrode surface condition based on the first dispersion index identification value in the equivalent circuit model identification result. When the first dispersion index identification value is lower than the preset lower limit threshold of the dispersion index, an electrode pollution alarm label is generated. The communication module encapsulates the groundwater conductivity monitoring value, the groundwater mineralization monitoring value, and the electrode pollution alarm label into a monitoring data frame, and uploads it to the remote groundwater monitoring platform through the narrowband IoT communication interface.

[0015] The integrated groundwater automatic monitoring terminal at the wellhead of the present invention has the following beneficial effects: This invention employs an integrated wellhead design, integrating the sampling chamber, three-electrode probe assembly, electrochemical workstation module, model identification module, water quality calculation module, and communication module within the wellhead casing to form a compact and fully functional monitoring terminal. This integrated design avoids potential water quality changes during water sample collection, transfer, and storage, as is common in traditional monitoring methods, ensuring that monitoring results accurately reflect the in-situ state of groundwater.

[0016] Meanwhile, the integrated design reduces the difficulty of equipment installation and maintenance costs, facilitating deployment in remote areas or large-scale monitoring networks. The adaptive identification method for equivalent circuit models based on piecewise analysis of Nyquist plot geometric features, employed in this invention, can automatically determine the topology type of the equivalent circuit according to the actual morphological characteristics of the impedance spectrum, without requiring pre-setting or manual intervention by operators. This method automatically determines whether the groundwater system is of a single-time-constant or dual-time-constant type by analyzing the number and position of capacitive arc vertices in the Nyquist plot scatter sequence, thereby selecting a matching equivalent circuit topology. This adaptive mechanism enables the monitoring terminal to adapt to groundwater systems with different water quality characteristics, improving the equipment's versatility and intelligence.

[0017] The component parameter identification method of this invention is based on geometric analysis rather than iterative optimization, directly extracting the parameter values ​​of equivalent circuit components from the geometric features of the Nyquist plot. Compared with the traditional complex nonlinear least squares fitting method, this method avoids the problems of initial value selection and local optima, resulting in stable and reliable identification results. Compared with intelligent optimization algorithms such as genetic algorithms, this method significantly reduces computational complexity, enabling real-time operation on embedded processors with limited computing resources, thus meeting the timeliness requirements of on-site monitoring. Attached Figure Description

[0018] Figure 1 Two typical Nyquist plot morphologies that may be encountered during electrochemical impedance spectroscopy measurement by the integrated groundwater monitoring terminal provided in this embodiment of the invention. Figure 2 The electrochemical impedance spectroscopy measured by the integrated groundwater automatic monitoring terminal at the wellhead provided in this embodiment of the invention is presented in the Bode plot coordinate system. Figure 3 The dispersion index of the constant phase angle element provided in the embodiments of the present invention Schematic diagram illustrating the influence of the Nyquist diagram on the arc resistance morphology; Figure 4 This is a schematic diagram illustrating the specific implementation principle of the diffusion index identification method based on Nyquist plot geometric features provided in this embodiment of the invention. Figure 5 This is a schematic diagram illustrating the implementation principle of the frequency domain segmentation identification method for dual-time-constant equivalent circuit topologies provided in an embodiment of the present invention. Detailed Implementation

[0019] The method of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0020] An integrated groundwater automatic monitoring terminal for wellheads includes: The sampling chamber is located inside the wellhead casing, and groundwater enters the sampling chamber through the water inlet on the wellhead casing; The three-electrode probe assembly is installed in the sampling chamber and includes a working electrode, a reference electrode, and an auxiliary electrode. The electrochemical workstation module is used to apply a sinusoidal perturbation voltage signal superimposed on the steady-state open-circuit potential to the three-electrode probe assembly, acquire the current response signal at each frequency point, and calculate the original electrochemical impedance spectroscopy data sequence based on the sinusoidal perturbation voltage signal and the current response signal. The model identification module is used to generate a Nyquist plot scatter sequence based on the original electrochemical impedance spectroscopy data sequence. By analyzing the geometric characteristics of the Nyquist plot scatter sequence, the equivalent circuit topology type is determined and the parameter values ​​of each equivalent circuit element are identified, thus forming the equivalent circuit model identification result. The water quality calculation module is used to calculate groundwater quality indicators based on the identification results of the equivalent circuit model. The communication module is used to upload groundwater quality indicators to the remote groundwater monitoring platform.

[0021] The sampling chamber of the integrated groundwater automatic monitoring terminal is made of corrosion-resistant stainless steel or polytetrafluoroethylene (PTFE). It is cylindrical in shape, with its outer diameter designed to match the inner diameter of the well casing. In practical applications, when the inner diameter of the well casing is 100 mm to 150 mm, the outer diameter of the sampling chamber can be set accordingly to 95 mm to 145 mm, with a gap of approximately 5 mm between them for easy installation and disassembly. The sampling chamber is fixed to the top of the well casing via flange or snap-fit ​​connection. This installation method ensures the stability of the sampling chamber during long-term use and facilitates quick assembly and disassembly for periodic maintenance.

[0022] Several water inlet holes are formed on the wall of the well casing. These holes are evenly distributed circumferentially along the casing wall, and their diameter is typically set to 3 mm to 8 mm. The number and diameter of the water inlet holes need to be comprehensively considered based on the dynamic range of groundwater level changes and the response speed requirements of water quality monitoring. When the total flow area of ​​the water inlet holes is large, the water in the sampling chamber can exchange with the groundwater outside the well casing more quickly, thus making the monitoring results more reflective of the real-time state of the groundwater. When the total flow area of ​​the water inlet holes is small, although the water exchange rate is reduced, it can effectively reduce the entry of suspended particles such as silt into the sampling chamber and extend the electrode cleaning cycle. In an optional embodiment, the outside of the water inlet holes is covered with a stainless steel filter screen with a mesh size of 60 to 100 mesh to block particles with a diameter greater than 0.15 mm to 0.25 mm from entering the sampling chamber.

[0023] The three-electrode probe assembly is installed inside the sampling chamber and includes an inert metal working electrode, a silver chloride reference electrode, and a platinum wire auxiliary electrode. The inert metal working electrode is made of platinum or gold, materials known for their excellent chemical stability and electrochemical inertness. These materials are resistant to corrosion and dissolution in groundwater environments, maintaining a stable electrode surface state over long periods. The effective area of ​​the inert metal working electrode is a key parameter affecting the sensitivity of electrochemical impedance spectroscopy measurements. A larger effective area results in a larger interfacial capacitance between the electrode and the solution, a smaller measured impedance signal amplitude, and a correspondingly higher signal-to-noise ratio. In this embodiment, the inert metal working electrode adopts a disk-shaped structure with a diameter ranging from 2 mm to 5 mm, corresponding to an effective electrode area of ​​approximately 3.14 square millimeters to 19.63 square millimeters.

[0024] The silver chloride reference electrode serves as a benchmark for potential measurement, maintaining a constant electrode potential under stable chloride ion concentration conditions. The electrode is internally filled with a saturated potassium chloride solution and forms ion conduction with the external test solution through a porous ceramic membrane. Due to fluctuations in chloride ion concentration in groundwater, the potential of the silver chloride reference electrode will experience slight drift, but this drift is typically on the order of millivolts and has minimal impact on the electrochemical impedance spectroscopy (EIS) measurement results. Alternatively, a saturated calomel electrode can be used as the reference electrode, exhibiting similar operating characteristics to the silver chloride reference electrode; the choice can be made based on the specific application and cost considerations.

[0025] A platinum wire auxiliary electrode is used to form a current loop with the inert metal working electrode during electrochemical measurements. The surface area of ​​the platinum wire auxiliary electrode should be significantly larger than the effective area of ​​the inert metal working electrode. This design aims to ensure that electrochemical polarization mainly occurs on the surface of the inert metal working electrode, while the polarization resistance of the platinum wire auxiliary electrode itself is negligible and will not introduce additional errors to the measurement results. The platinum wire auxiliary electrode is in the form of a spiral coil or a mesh braid, with a platinum wire diameter of 0.3 mm to 0.5 mm and a total length of 50 mm to 100 mm. The unfolded surface area is approximately 47 square millimeters to 157 square millimeters, which is 8 to 15 times the effective area of ​​the inert metal working electrode.

[0026] In the spatial arrangement of the three-electrode probe assembly, the inert metal working electrode and the platinum wire auxiliary electrode are arranged parallel to each other within the sampling chamber, with a fixed electrode spacing of 10 mm to 30 mm. The selection of the electrode spacing requires balancing two factors: the accuracy of solution resistance measurement and the uniformity of the electric field. When the electrode spacing is too small, the electric field lines between the electrodes are not uniformly distributed, resulting in significant edge effects and causing the measured value of the solution resistance to deviate from the true value. When the electrode spacing is too large, the solution resistance increases, the current flowing under the same disturbance voltage decreases, the signal amplitude decreases, and the measurement accuracy declines. The silver chloride reference electrode is positioned at the midline between the inert metal working electrode and the platinum wire auxiliary electrode. This symmetrical arrangement allows the reference electrode to be as close as possible to the equipotential surface of the inert metal working electrode, reducing the potential measurement error caused by the positional deviation of the reference electrode.

[0027] The acquisition process of electrochemical impedance spectroscopy (EIS) first requires determining the open-circuit potential of the three-electrode probe assembly in groundwater. The open-circuit potential refers to the natural potential difference between the inert metal working electrode and the silver chloride reference electrode when no external current flows. This potential value reflects the electrochemical equilibrium state established between the surface of the inert metal working electrode and the groundwater, and is influenced by a combination of factors such as dissolved oxygen concentration, pH, and the types and concentrations of ions in the water. When measuring the open-circuit potential, it is necessary to wait for the potential reading to stabilize, a process that typically takes 30 to 120 seconds. When the difference between two consecutive potential readings is less than 1 millivolt and the duration exceeds 10 seconds, the system is considered to have reached a steady state, and the recorded potential value at this point is taken as the steady-state open-circuit potential.

[0028] After obtaining the steady-state open-circuit potential, a sinusoidal perturbation voltage signal is applied to the three-electrode probe assembly. The sinusoidal perturbation voltage signal is superimposed on the steady-state open-circuit potential, and its mathematical expression is as follows: ,in This represents the applied voltage that varies over time. This represents the steady-state open-circuit potential. This represents the amplitude of the sinusoidal disturbance voltage signal. This indicates the frequency of the sinusoidal disturbance voltage signal. The amplitude of the sinusoidal perturbation voltage signal is set to a fixed millivolt range of 5 to 10 millivolts. This amplitude range is chosen based on the fundamental requirement of linear response in the electrochemical system. When the perturbation amplitude is too large, the electrochemical reaction on the electrode surface enters the nonlinear region, and harmonic components appear in the current response, violating the linear assumption of impedance measurement. When the perturbation amplitude is too small, the amplitude of the current response signal decreases accordingly, making it easily submerged by background noise and reducing measurement accuracy. Within the amplitude range of 5 to 10 millivolts, the electrochemical system can maintain good linear response characteristics while possessing sufficient signal strength.

[0029] Figure 1 This paper demonstrates two typical Nyquist plot shapes that may be encountered during electrochemical impedance spectroscopy (EIS) measurements by an integrated groundwater automatic monitoring terminal at the wellhead, corresponding to single-time-constant and double-time-constant equivalent circuit topologies, respectively. The horizontal axis represents the real impedance component of the complex impedance value in ohms, and the vertical axis represents the negative of the imaginary impedance component, also in ohms. Using the negative of the imaginary impedance component as the vertical axis is a common practice in electrochemical impedance spectroscopy analysis. This method places the capacitive arc in the first quadrant, facilitating intuitive observation and geometric feature extraction. The blue solid line curve represents the Nyquist plot scatter sequence corresponding to the single-time-constant equivalent circuit. This curve presents a complete semi-circular arc shape, starting from the high-frequency real intercept point, passing through the capacitive arc vertex, and finally approaching the low-frequency real intercept point.

[0030] The high-frequency real intercept is numerically equal to the solution resistance identification value because, under high-frequency limiting conditions, the impedance of the constant-phase-angle element approaches zero, and the impedance of the entire equivalent circuit is contributed only by the solution resistance element. The low-frequency real intercept is equal to the sum of the solution resistance identification value and the charge transfer resistance identification value, reflecting the total resistance characteristics of the electrochemical system under quasi-steady-state conditions. The radius of the capacitive arc marked in the figure is equal to the ordinate of the first vertex, while the identification value of the first charge transfer resistance is equal to twice the radius of the capacitive arc. This geometric relationship is the theoretical basis for component parameter identification. The red solid curve in the figure represents the Nyquist plot scatter sequence corresponding to the dual-time-constant type equivalent circuit. This curve presents two interconnected capacitive arcs, corresponding to two electrochemical relaxation processes with different time constants. The vertices of the first and second capacitive arcs are clearly marked with triangles in the figure. The model identification module automatically determines the equivalent circuit topology type by detecting the number of these vertices. When the composition of groundwater is relatively simple, the electrochemical impedance spectroscopy usually exhibits a single time constant characteristic; when there is organic adsorption in the groundwater or a passivation film is formed on the electrode surface, it may exhibit a dual time constant characteristic.

[0031] The frequency of the sinusoidal perturbation voltage signal decreases point by point from an initial high-frequency value to a final low-frequency value. In this embodiment, the initial high-frequency value is set to 100,000 Hz, and the final low-frequency value is set to 0.01 Hz, with the frequency scan spanning seven orders of magnitude. The frequency decrease method adopts a logarithmic equal-interval decrease, that is, the ratio between two adjacent frequency points remains constant. Each decibel contains a fixed number of frequency sampling points, typically set to 5 to 10 sampling points. Taking 7 sampling points per decibel as an example, the entire frequency range decreasing from 100,000 Hz to 0.01 Hz contains a total of 49 frequency sampling points. The logarithmically equal-interval frequency distribution method can obtain sufficiently dense sampling points in both the high-frequency and low-frequency bands, which matches the characteristic distribution law of electrochemical impedance spectroscopy on the Nyquist plot. The impedance response in the high-frequency band mainly reflects the solution resistance characteristics, while the impedance response in the low-frequency band mainly reflects the charge transfer and diffusion process at the electrode interface. The logarithmically equal-interval distribution can obtain sufficient information in each characteristic frequency region.

[0032] At each frequency point, the applied sinusoidal perturbation voltage signal and the resulting current response signal are synchronously sampled. The current response signal is the alternating current flowing between the inert metal working electrode and the platinum wire auxiliary electrode; its waveform is also sinusoidal, but it exhibits a certain phase shift relative to the perturbation voltage signal. Let the current response signal be expressed as... ,in This represents the current response as it changes over time. Indicates the amplitude of the current response signal. This represents the phase shift of the current response signal relative to the sinusoidal disturbance voltage signal. The synchronous sampling process continuously acquires several complete sinusoidal cycles at each frequency point. In the low-frequency range, due to the longer duration of a single cycle, typically 2 to 4 complete cycles are acquired; in the high-frequency range, due to the shorter duration of a single cycle, 8 to 16 complete cycles can be acquired. Averaging after multiple cycles of sampling effectively suppresses the influence of random noise on the measurement results.

[0033] Based on the voltage and current signal data obtained through synchronous sampling, calculate the impedance magnitude and impedance phase angle at this frequency point. Impedance magnitude Equal to the amplitude of the sinusoidal disturbance voltage signal Divide by the amplitude of the current response signal ,Right now The impedance phase angle is equal to the phase difference between the sinusoidal disturbance voltage signal and the current response signal, and is taken as... The negative sign is introduced because impedance is defined as voltage divided by current, while phase difference... Defined as the lead angle of current relative to voltage. In electrochemical impedance spectroscopy measurements, due to the presence of capacitive elements, the current response typically leads the perturbation voltage, with a phase difference... A positive value corresponds to a negative impedance phase angle.

[0034] The impedance magnitude and phase angle are further converted into the real and imaginary impedance components of a complex impedance value. Complex impedance value It can be represented as ,in Represents the real part of the impedance component. Represents the imaginary impedance component. This represents the imaginary unit. The real part of the impedance component is equal to the impedance magnitude multiplied by the cosine of the impedance phase angle, i.e. The imaginary part of the impedance component is equal to the impedance magnitude multiplied by the sine of the impedance phase angle, i.e. Since the impedance phase angle is usually negative, the imaginary impedance component... It exhibits negative values ​​at most frequency points, which is consistent with the physical characteristics of capacitive electrochemical interfaces.

[0035] The complex impedance values ​​at all frequency points are arranged in descending order of frequency, forming the original electrochemical impedance spectroscopy (EIS) data sequence. In this embodiment, when the frequency scan range is 100,000 Hz to 0.01 Hz, and each decibel contains 7 sampling points, the original EIS data sequence contains 49 sets of data. Each set of data includes three values: frequency value, real impedance component, and imaginary impedance component. The time required to complete one complete EIS acquisition depends mainly on the measurement time in the low-frequency band. Taking a termination low-frequency value of 0.01 Hz as an example, the measurement time for a single frequency point is approximately 200 to 400 seconds, and the total time for the entire frequency scan is approximately 30 to 60 minutes. In an optional implementation, when the monitoring task has high requirements for time response speed, the termination low-frequency value can be increased to 0.1 Hz or 1 Hz, correspondingly shortening the single measurement time to 5 to 15 minutes, but some impedance information in the low-frequency band will be lost.

[0036] After acquiring the raw electrochemical impedance spectroscopy (EIS) data sequence, it needs to be graphically represented to facilitate subsequent feature extraction and analysis. The Nyquist plot is the most commonly used graphical representation method in EIS analysis. Its horizontal axis represents the real impedance component of the complex impedance value, and its vertical axis represents the negative value of the imaginary impedance component. The reason for using the negative value of the imaginary impedance component as the vertical axis is that when capacitive elements dominate in an electrochemical system, the imaginary impedance component exhibits a negative value. Taking the negative value ensures that the capacitive arc on the Nyquist plot is located in the first quadrant, facilitating intuitive observation and geometric analysis.

[0037] refer to Figure 2Bode plots and Nyquist plots are two complementary representations of electrochemical impedance spectroscopy. Bode plots provide a more intuitive view of the relationship between impedance characteristics and frequency. The upper subplot's horizontal axis represents frequency in Hertz, displayed on a logarithmic scale, covering a range from 0.01 Hertz to 100,000 Hertz; the vertical axis represents the impedance magnitude. The unit is ohms. The impedance magnitude is defined as the magnitude of the complex impedance, equal to the square root of the sum of the squares of the real and imaginary impedance components. The blue solid line curve illustrates the frequency response characteristics of the impedance magnitude of the single-time-constant equivalent circuit. In the high-frequency region, the impedance magnitude approaches a constant value, which is the high-frequency plateau, corresponding to the solution resistance identification value. In the low-frequency region, the impedance magnitude also approaches another constant value, which is the low-frequency plateau, corresponding to the sum of the solution resistance identification value and the first charge transfer resistance identification value.

[0038] In the transition region between the high-frequency and low-frequency platforms, the impedance magnitude increases monotonically with decreasing frequency, and the rate of increase is closely related to the time constant of the equivalent circuit. The characteristic frequency of 31.6 Hz, marked with a red dashed line in the figure, corresponds to the position where the impedance magnitude changes most drastically. This characteristic frequency has a definite mathematical relationship with the first time constant and can be used to calculate the quasi-capacitance identification value. The vertical axis of the lower sub-figure represents the impedance phase angle, in degrees. The impedance phase angle is defined as the argument of the complex impedance, reflecting the phase shift of the current response signal relative to the sinusoidal perturbation voltage signal. The red solid curve shows the variation of the impedance phase angle with frequency. In the high-frequency and low-frequency limiting regions, the impedance phase angle approaches zero, indicating that the electrochemical system exhibits purely resistive characteristics. In the intermediate frequency region, the impedance phase angle reaches a negative minimum; the frequency corresponding to this minimum is the phase angle extremum, and its position is basically consistent with the characteristic frequency corresponding to the apex of the capacitive reactance arc.

[0039] Specifically, for each set of data in the original electrochemical impedance spectroscopy (EIS) data sequence, the real impedance component of that set is used as the x-coordinate value of that point, and the imaginary impedance component multiplied by -1 is used as the y-coordinate value of that point. Taking 49 sets of data as an example, after the above coordinate mapping, a Nyquist plot scatter sequence containing 49 scattered points is formed on a two-dimensional plane. The arrangement order of the scattered points in the Nyquist plot scatter sequence is consistent with the decreasing frequency order in the original EIS data sequence, that is, the first scattered point in the sequence corresponds to the impedance data with the highest frequency of 100,000 Hz, and the last scattered point corresponds to the impedance data with the lowest frequency of 0.01 Hz.

[0040] The geometric shape of the Nyquist plot scatter sequence contains rich electrochemical information. For simple electrochemical interfaces, the Nyquist plot presents a semi-circular arc shape, which is usually called the capacitive arc. The appearance of the capacitive arc originates from the parallel effect of the double-layer capacitance and charge transfer resistance at the electrode-electrolyte interface. When the frequency of the sinusoidal perturbation voltage signal is high, the capacitive reactance of the double-layer capacitance is small, and the current mainly flows through the capacitive branch. At this time, the measured impedance is close to the pure resistance characteristic. As the frequency gradually decreases, the capacitive reactance of the double-layer capacitance gradually increases, and the current is redistributed between the capacitive and resistive branches. The impedance value exhibits a mixed characteristic of resistance and capacitance. When the frequency drops to a certain value, the contributions of capacitive reactance and resistance reach equilibrium. At this time, the absolute value of the imaginary impedance component reaches its maximum, corresponding to the vertex position of the capacitive arc on the Nyquist plot.

[0041] To extract the characteristic information of capacitive arcs from a Nyquist plot scatter sequence, it is necessary to traverse the entire scatter sequence along the decreasing frequency direction and analyze the trend of ordinate changes between adjacent scatter points. Let the scatter sequence contain the th... The ordinates of the scattered points are , No. The ordinates of the scattered points are Calculate the difference in ordinate between adjacent scattered points. During the rising phase of the capacitive arc, as the frequency decreases, the ordinate value gradually increases, and at this point, the difference in the ordinate value... The value is positive; in the decreasing segment of the capacitive arc, as the frequency continues to decrease, the ordinate value begins to decrease, and at this time the ordinate difference value... The value changes to a negative value. When the difference in the ordinate changes from a positive value to a negative value during the traversal, the scatter point at the change point is the vertex position of the capacitive arc.

[0042] Groundwater, as a complex electrolyte solution, may exhibit single or multiple capacitive arcs in its electrochemical impedance spectroscopy. A single capacitive arc typically corresponds to a single electrochemical relaxation process at the electrode-solution interface, while the presence of multiple capacitive arcs indicates the existence of multiple electrochemical processes with different time constants. In groundwater monitoring scenarios, the first capacitive arc is often related to charge transfer reactions on the electrode surface, while the second capacitive arc may originate from the formation of an adsorption layer on the electrode surface or the specific adsorption behavior of certain ions in the solution.

[0043] The scatter point corresponding to the first positive / negative transition point of the ordinate difference detected during the traversal is marked as the vertex of the first capacitive arc. The frequency value corresponding to the vertex of the first capacitive arc is recorded as the first characteristic frequency. This indicates that the x-coordinate of the first capacitive arc vertex is recorded as the x-coordinate of the first vertex. Represents; record the ordinate value of the first capacitive arc vertex as the ordinate of the first vertex, using This indicates that the remaining scattered points are traversed along the frequency decreasing direction. If a change in the difference of the ordinate from a positive value to a negative value is detected again, the scattered point at this change point is marked as the vertex of the second capacitive arc, and the second characteristic frequency is recorded accordingly. The x-coordinate of the second vertex The y-coordinate of the second vertex .

[0044] During the traversal, it is also necessary to extract the high-frequency real intercept values ​​of the Nyquist plot scatter sequence. The high-frequency real intercept value refers to the x-coordinate value of the scatter point with the highest frequency in the scatter sequence, expressed as... This is explained from an electrochemical physics perspective. When the frequency of the disturbance signal approaches infinity, the capacitive reactance of the electric double-layer capacitor approaches zero, and the capacitor branch is approximately short-circuited. At this point, the measured impedance is only the solution resistance. Therefore, the high-frequency real intercept value is numerically equal to the solution resistance of the underground aqueous solution. Although it is impossible to reach infinite frequencies in actual measurements, under high-frequency conditions of 100,000 Hz, the influence of capacitive reactance is small enough that the real impedance component at this point can be approximated as the solution resistance value.

[0045] Based on the number of capacitive arc vertices detected, the equivalent circuit topology type matching the electrochemical characteristics of groundwater can be determined. When only the first capacitive arc vertex is detected during the traversal, it indicates that there is only a single electrochemical relaxation process at the electrode interface, and the equivalent circuit topology type is determined to be single-time-constant type. The single-time-constant type equivalent circuit topology is composed of a solution resistive element connected in series with a first parallel RC unit. The solution resistive element is used to characterize the ohmic resistance characteristics of the groundwater solution itself. The first parallel RC unit is composed of a first charge transfer resistive element connected in parallel with a first constant-phase-angle element, wherein the first charge transfer resistive element characterizes the kinetic impedance of the electrochemical reaction on the electrode surface, and the first constant-phase-angle element is used to replace the ideal capacitor to more accurately describe the non-ideal capacitive behavior of the actual electrode surface.

[0046] When both the first and second capacitive arc vertices are detected simultaneously during the traversal, it indicates the presence of two electrochemical relaxation processes with different time constants at the electrode interface in the groundwater. In this case, the equivalent circuit topology is determined to be of the dual-time-constant type. The dual-time-constant type equivalent circuit topology consists of a solution resistive element, a first parallel RC unit, and a second parallel RC unit connected in series. The second parallel RC unit is composed of a second charge-transfer resistive element and a second constant-phase-angle element connected in parallel, used to characterize the impedance characteristics of the second electrochemical relaxation process. In groundwater monitoring applications, the second electrochemical relaxation process may be associated with the adsorption of organic matter in the water on the electrode surface, the formation of an oxide film on the electrode surface, or the diffusion process of specific ions.

[0047] After determining the equivalent circuit topology type, it is necessary to further identify the parameter values ​​of each equivalent circuit element. A piecewise analytical method based on the geometric features of the Nyquist plot can avoid the iterative convergence problem of traditional nonlinear fitting algorithms, directly extracting element parameters from the geometric relationships of the graph.

[0048] The parameter identification process for the single-time-constant equivalent circuit topology is as follows: First, the high-frequency real axis intercept value is... The value is directly assigned to the solution resistance element as the solution resistance identification value. Because the constant phase angle element in the RC parallel unit is approximately short-circuited under high-frequency limiting conditions, the impedance of the entire equivalent circuit is contributed only by the solution resistance element. Therefore, the high-frequency real axis intercept value can accurately reflect the true value of the solution resistance.

[0049] Next, we identify the parameter values ​​of each component in the first parallel RC unit. For an ideal parallel RC unit, its Nyquist plot is represented as a standard semicircle centered at a point on the real axis. The highest point of the semicircle is the vertex of the capacitive arc, and its ordinate value is equal to the radius of the semicircle. Therefore, the radius of the first capacitive arc is... Equal to the y-coordinate of the first vertex First charge transfer resistance identification value It is equal to twice the radius of the first capacitive arc, that is... This relationship stems from the impedance characteristics of the parallel resistor-capacitor unit: when the frequency scans from high frequency to low frequency, the impedance trajectory starts from the high-frequency real axis intercept point, moves along a semi-circular arc, and finally reaches the low-frequency real axis intercept point. The diameter of the semi-circle is exactly equal to the resistance value of the parallel resistor.

[0050] The parameter identification of the first constant-phase angle element requires utilizing the time constant relationship between the characteristic frequency and the charge transfer resistance. For the parallel RC unit, its characteristic frequency... With resistance and capacitor The product of and is reciprocal, and the specific expression is: Therefore, the first time constant can be obtained. ,in The first characteristic frequency. The quasi-capacitive identification value of the first constant phase angle element. Equal to the first time constant Divide by the first charge transfer resistance identification value ,Right now .

[0051] Due to factors such as roughness, porosity, or chemical inhomogeneity, the capacitive behavior of actual electrode surfaces often deviates from that of ideal capacitance. Constant-phase-angle elements describe this deviation by introducing a dispersion index. The dispersion index ranges from 0 to 1. When the dispersion index equals 1, the constant-phase-angle element degenerates into an ideal capacitor; when the dispersion index is less than 1, it indicates the presence of non-ideal factors on the electrode surface. On a Nyquist plot, non-ideal capacitive behavior manifests as a change in curvature near the vertex of the capacitive arc: the slope of the tangent at the vertex of an ideal semicircle jumps instantaneously from positive infinity to negative infinity, while the slope of the tangent near the vertex of a non-ideal capacitive arc changes more gradually.

[0052] refer to Figure 3 The dispersion index is an important parameter describing the degree to which the actual electrode surface deviates from the ideal capacitance behavior; its value ranges from 0 to 1. In the figure, the horizontal axis represents the real impedance component, and the vertical axis represents the negative imaginary impedance component. The coordinate system is set in parallel with... Figure 1 Maintain consistency. The figure shows four Nyquist plot curves for different diffusion indices, corresponding to... , , and When the diffusion index When the value equals 1.00, the constant phase angle element degenerates into an ideal capacitor. At this point, the Nyquist plot presents a standard semicircular shape, with the blue curve perfectly coinciding with the ideal semicircular reference line drawn by the black dashed line. The center of the semicircle lies on the real axis, and the x-coordinate of the center is equal to the high-frequency real axis intercept plus the capacitive arc radius. When the dispersion index... When the value is less than 1.00, the constant phase angle element exhibits non-ideal capacitance characteristics, and the capacitive arc of the Nyquist plot is deformed.

[0053] As the diffusion index decreases, the capacitive arc is gradually flattened, and its center shifts downwards below the real axis, correspondingly reducing the ordinate value of the capacitive arc's vertex. The green curve in the figure corresponds to... The orange curve corresponds to The red curve corresponds to It can be clearly observed that the degree of capacitive arc flattening intensifies as the dispersion index decreases. This non-ideal capacitive behavior is common in actual groundwater monitoring, and its physical causes include electrode surface roughness, porous structure, chemical inhomogeneity, and adsorption of contaminants on the electrode surface. The dispersion index identification value can quantitatively reflect the state of the electrode surface; when the electrode surface is contaminated, the dispersion index will decrease significantly. Therefore, by monitoring the changing trend of the dispersion index identification value, automatic diagnosis of electrode contamination status can be achieved. When the dispersion index identification value is lower than the preset lower limit threshold, the water quality calculation module will generate an electrode contamination alarm.

[0054] To quantify this non-ideal characteristic, the preceding and following adjacent scattered points of the first capacitive arc vertex are extracted. Let the coordinates of the preceding adjacent scattered point be... The coordinates of the first capacitive arc vertex are The coordinates of the next adjacent scatter point are The slope of the line connecting the previous adjacent scatter point and the vertex of the first capacitive arc is calculated as the slope of the preceding tangent. The slope of the line connecting the vertex of the first capacitive arc and the next adjacent scatter point is calculated as the slope of the subsequent tangent. The absolute value of the difference between the arctangent of the slope of the front tangent and the arctangent of the slope of the back tangent is used as the angle between the vertices. First diffusion index identification value equal to the included angle at the vertex Divide by pi ,Right now For a standard semicircle, the slope of the anterior tangent at the vertex tends towards positive infinity, and the slope of the posterior tangent tends towards negative infinity. The absolute value of the difference between their arctangent values ​​is equal to... The corresponding diffusion index is 1.

[0055] refer to Figure 4 In the figure, the solid blue line represents the measured Nyquist plot scatter sequence, and the blue dots mark the positions of the scatter points arranged in descending frequency order. The capacitive arc vertex, marked with a red triangle, is the core feature point in the entire identification process. This point corresponds to the scatter point with the largest ordinate value in the Nyquist plot scatter sequence. Its detection method involves traversing the scatter sequence along the decreasing frequency direction and detecting the position where the difference in ordinate between adjacent scatter points changes from positive to negative. The scatter point preceding the capacitive arc vertex is marked with a green dot, and the scatter point following it is marked with a purple dot. These three scatter points form the geometric basis for diffusion index identification. The green dashed line in the figure is the forward tangent, connecting the preceding adjacent scatter point to the capacitive arc vertex and extending to both sides; the purple dashed line is the backward tangent, connecting the capacitive arc vertex to the following adjacent scatter point and extending to both sides. The slope of the forward tangent... The slope of the tangent line is equal to the difference between the ordinate of the vertices of the capacitive arc and the ordinate of the previous adjacent scatter point, divided by the difference between the x-coordinates of the vertices of the capacitive arc and the x-coordinates of the previous adjacent scatter point. The calculation method is similar.

[0056] The included angle of the vertices marked with orange arcs in the diagram. The diffusion index is a key geometric quantity for identification. It is calculated as the absolute value of the difference between the arctangent of the slope of the anterior tangent and the arctangent of the slope of the posterior tangent. The diffusion index identification value is equal to the angle between the vertices. Divide by pi For a standard semicircle, the slope of the front tangent at the vertex tends to positive infinity, the slope of the back tangent tends to negative infinity, the included angle at the vertex is equal to 1, and the corresponding dispersion index is 1.00. When the capacitive arc is flattened, the change in the slope of the tangent at the vertex tends to be gentler, the included angle at the vertex decreases, and the dispersion index identification value decreases accordingly.

[0057] For equivalent circuit topologies with dual time constants, a segmented frequency domain strategy is employed in the parameter identification process. Since the two capacitive arcs partially overlap in the frequency domain, directly analyzing the overall curve would lead to mutual interference in parameter identification. The core idea of ​​the segmented strategy is to divide the Nyquist plot scatter sequence into several subsets according to frequency ranges, and then analyze each subset independently.

[0058] Calculate the first characteristic frequency With the second characteristic frequency geometric mean The geometric mean is better suited than the arithmetic mean for logarithmic frequency data, as it can achieve uniform distribution in a logarithmic coordinate system. For Nyquist plot scatter sequences where the frequency is higher than the geometric mean... The scatter points were divided into a high-frequency subset, with frequencies below the geometric mean. The scattered points are divided into a low-frequency subset.

[0059] The solution resistance identification value is obtained in the same way as the single time constant type, that is, the high-frequency real axis intercept value. The values ​​are directly assigned to the solution resistance element. For the high-frequency scattered subset, the element parameter identification method is based on the single-time-constant equivalent circuit topology, according to the ordinate of the first vertex. and the first characteristic frequency Identify and obtain the first charge transfer resistance identification value First quasi-capacitance identification value and the first diffusion index identification value .

[0060] Analysis of the low-frequency scatter subset requires first performing a coordinate translation operation. The low-frequency intercept of the first capacitive arc is calculated to be equal to the high-frequency real-axis intercept value plus the first charge transfer resistance identification value, i.e. The low-frequency intercept of the first capacitive arc represents the real impedance value at the end of the first capacitive arc, and also the real impedance value at the beginning of the second capacitive arc. For each point in the low-frequency scatter subset, subtract the low-frequency intercept of the first capacitive arc from its x-coordinate. This forms a low-frequency scatter subset after translation. The purpose of the coordinate translation operation is to move the starting point of the second capacitive arc to near the origin, making its geometric characteristics consistent with the Nyquist plot of the single-time-constant equivalent circuit, thus enabling the reuse of the single-time-constant parameter identification method. The low-frequency scatter subset after translation is then identified according to the component parameter identification method of the single-time-constant equivalent circuit topology, based on the ordinate of the second vertex. Second characteristic frequency Identify and obtain the identification value of the second charge transfer resistance. Second pseudocapacitance identification value Second diffusion index identification value .

[0061] After identifying all component parameters, the equivalent circuit topology type, solution resistance identification value, and the charge transfer resistance, quasi-capacitance, and dispersion index identification values ​​corresponding to each parallel RC unit are combined to form the equivalent circuit model identification result. In an optional implementation, the equivalent circuit model identification result can be back-substituted for verification. That is, the theoretical impedance spectrum is calculated based on the identified component parameters and compared with the original electrochemical impedance spectroscopy data sequence to calculate the relative error between the two. When the relative error exceeds a preset threshold (e.g., 10%), a re-measurement or parameter correction process can be triggered.

[0062] The calculation of groundwater quality indicators is based on the physical correlation between the parameter values ​​of each component and the water quality parameters in the equivalent circuit model identification results. Groundwater conductivity is an important indicator characterizing the total ion concentration in water, and its value is inversely related to the solution resistance. Physically, conductivity reflects the ability of a unit volume of solution to conduct current; the higher the ion concentration, the greater the number of charge carriers, the greater the conductivity, and the lower the corresponding solution resistance.

[0063] The calculation of groundwater conductivity monitoring values ​​requires the introduction of the geometric parameter of electrode constant. equal to the electrode spacing value Divide by the effective area of ​​the electrode ,Right now ,in The unit is centimeters. Since the unit is square centimeters, the unit of the electrode constant is per centimeter. Taking an electrode spacing of 20 millimeters (2 centimeters) and an effective electrode area of ​​12.56 square millimeters (0.1256 square centimeters) as an example, the electrode constant is approximately 15.92 per centimeter. The electrode constant and the effective electrode area are inherent parameters of the integrated groundwater automatic monitoring terminal at the wellhead. They are determined through calibration tests before the equipment leaves the factory and stored in the parameter storage unit.

[0064] Groundwater conductivity monitoring values Equal to electrode constant Divide by the solution resistance identification value in the equivalent circuit model identification result ,Right now When the solution resistance identification value The unit is ohm, electrode constant When the unit is per centimeter, the calculated groundwater conductivity monitoring value The unit is Siemens per centimeter. The conductivity of groundwater is typically in the range of 100 microsiemens per centimeter to 2000 microsiemens per centimeter, corresponding to a solution resistance of approximately 80 ohms to 1600 ohms.

[0065] Groundwater mineralization refers to the total content of dissolved solids in water, and it has an approximately linear relationship with electrical conductivity. This linear relationship stems from the fact that the dissolved solids in groundwater are mainly inorganic salts, which ionize in water to produce ions. These ions contribute to both mineralization and conductivity. In groundwater systems with relatively stable ion composition, the ratio of conductivity to mineralization remains approximately constant. The preset conductivity-mineralization linear conversion slope is read from the parameter storage unit. and the intercept of the linear conversion between conductivity and mineralization These two parameters were obtained through comparative calibration of laboratory chemical analysis and on-site conductivity measurements of local groundwater samples. Due to differences in ionic composition, the slope and intercept of the linear conversion between conductivity and mineralization may vary significantly across different regions.

[0066] Groundwater mineralization monitoring value Equal to groundwater conductivity monitoring value Multiply by the linear conversion slope of conductivity-mineralization Then add the conductivity-mineralization linear conversion intercept ,Right now Taking groundwater in the North China Plain as an example, the slope of the linear conversion between conductivity and mineralization is approximately 0.55 to 0.75 mg / L / µSiemens / cm, and the intercept of the linear conversion between conductivity and mineralization is approximately -50 to +50 mg / L. When the groundwater conductivity monitoring value is 800 µSiemens / cm, the slope of the linear conversion between conductivity and mineralization is 0.65, and the intercept of the linear conversion between conductivity and mineralization is 0, the calculated groundwater mineralization monitoring value is 520 mg / L.

[0067] The dispersion index (DCI) reflects changes in the state of the electrode surface and can be used for early warning of electrode contamination. When the electrode surface is clean, its electrochemical behavior is close to ideal, and the DCI typically ranges from 0.85 to 1.00. As organic matter, microorganisms, or inorganic precipitates from groundwater gradually adhere to and accumulate on the electrode surface, the surface non-uniformity increases, the electrochemical behavior deviates from the ideal state, and the DCI decreases accordingly. The electrode surface state is assessed based on the first DCI value from the equivalent circuit model identification results. When the first DCI value is lower than a preset lower threshold, it indicates that the degree of electrode surface contamination has affected measurement accuracy, and an electrode contamination alarm is generated. The lower threshold can be set according to actual application requirements, with typical values ​​ranging from 0.70 to 0.80.

[0068] After calculating the groundwater quality indicators, the monitoring data is uploaded to a remote groundwater monitoring platform for centralized management and analysis. Groundwater conductivity monitoring values, groundwater salinity monitoring values, and electrode contamination alarm flags are encapsulated into monitoring data frames. These data frames use a structured data format, including a frame header, device identification code, timestamp, data fields, and a checksum. The device identification code uniquely identifies the data source when multiple monitoring terminals are present; the timestamp records the specific time of data collection; the data fields contain the values ​​of various water quality indicators; and the checksum is used for error detection during transmission.

[0069] Monitoring data frames are uploaded to the remote groundwater monitoring platform via a narrowband IoT communication interface. Narrowband IoT is a low-power wide-area network (LPWAN) communication technology specifically designed for IoT applications. It offers advantages such as wide coverage, strong penetration, low power consumption, and low cost, making it particularly suitable for data transmission needs in remote or underground environments such as wellheads. The operating frequency band of the narrowband IoT communication interface is typically 800 MHz to 900 MHz, with a single data transmission power consumption of approximately 20 milliwatts to 200 milliwatts, enabling continuous operation for several years under battery power. In an alternative implementation, when the wellhead is located in a narrowband IoT signal coverage blind spot, other low-power wide-area network (LPWAN) communication technologies can be used as alternatives, such as long-range radio communication or satellite short message communication.

[0070] After receiving monitoring data frames, the remote groundwater monitoring platform parses, stores, and visualizes them. The monitoring data can be integrated with a geographic information system (GIS) to visually present the water quality status of each monitoring point on an electronic map. When an abnormal change occurs in the water quality indicators at a monitoring point, or when an electrode contamination alarm is triggered, the remote groundwater monitoring platform can send early warning notifications to relevant management personnel via SMS, email, or mobile application push notifications, enabling real-time monitoring and rapid response to groundwater quality issues.

[0071] refer toFigure 5 When both the first and second capacitive arc vertices are detected simultaneously in the Nyquist plot scatter sequence, the model identification module determines that the equivalent circuit topology is of the dual-time-constant type and uses a frequency domain segmentation strategy for component parameter identification. In the figure, the blue curve segment and blue dots represent the high-frequency scatter subset, and the red curve segment and red square dots represent the low-frequency scatter subset. The boundary between the two is marked with a green triangle. The frequency corresponding to the boundary point is the first characteristic frequency. With the second characteristic frequency geometric mean The geometric mean is used instead of the arithmetic mean because frequency data exhibits a logarithmic distribution, and the geometric mean allows for a more uniform division in a logarithmic coordinate system. The high-frequency real axis intercept value, marked with a black star in the figure, is directly used as the solution resistance identification value; this identification method is identical to that of the single-time-constant type. The first capacitive arc vertex, marked with a blue triangle in the figure, is used to identify the component parameters of the first parallel RC unit. The identification method is consistent with the component parameter identification method of the single-time-constant type, yielding the first charge transfer resistance identification value, the first quasi-capacitance identification value, and the first dispersion index identification value.

[0072] The low-frequency intercept of the first capacitive arc, marked with a blue diamond in the figure, is a key reference point for frequency domain segmentation identification. Its value is equal to the high-frequency real axis intercept value plus the first charge transfer resistance identification value. Before identifying the component parameters of the low-frequency scatter subset, a coordinate translation operation needs to be performed, subtracting the first capacitive arc low-frequency intercept value from the x-coordinate of each scatter point. The purpose of the coordinate translation is to move the starting point of the second capacitive arc to near the origin of the coordinate system, making its geometric characteristics consistent with the Nyquist plot shape of the single-time-constant type equivalent circuit, thereby enabling the reuse of the single-time-constant type parameter identification algorithm. The vertex of the second capacitive arc, marked with a red triangle in the figure, is used to identify the component parameters of the second parallel RC unit. The identification process is carried out in the translated coordinate system, and the identification values ​​of the second charge transfer resistance, the second quasi-capacitance, and the second dispersion index can be obtained.

[0073] The following describes the specific implementation of this invention using a complete monitoring process of a groundwater monitoring well as an example. The integrated groundwater automatic monitoring terminal is installed in a groundwater monitoring well with a depth of 50 meters. The inner diameter of the wellhead casing is 110 mm, and the outer diameter of the sampling chamber is 105 mm. In the three-electrode probe assembly installed inside the sampling chamber, the inert metal working electrode is a 3 mm diameter disc-shaped platinum electrode with an effective electrode area of ​​[missing information]. The area is 7.07 square millimeters, which translates to 0.0707 square centimeters. The platinum wire auxiliary electrode is made of a platinum wire with a diameter of 0.4 mm and a total length of 80 mm, spirally wound. The electrode spacing between the inert metal working electrode and the platinum wire auxiliary electrode is... It is 20 mm, which is equivalent to 2 cm. The silver chloride reference electrode is positioned at the midline between the two, 10 mm away from the inert metal working electrode.

[0074] After the monitoring process is initiated, the electrochemical workstation module first measures the open-circuit potential of the three-electrode probe assembly in groundwater. After a 60-second stabilization period, the measured open-circuit potential reading shows a change of less than 1 millivolt within adjacent 10-second intervals; at this point, the steady-state open-circuit potential is recorded. It is -0.182 volts.

[0075] Subsequently, the electrochemical workstation module applies a sinusoidal perturbation voltage signal to the three-electrode probe assembly. The amplitude of the sinusoidal perturbation voltage signal... The voltage was set to 10 millivolts and superimposed on the steady-state open-circuit potential. The frequency scan range was set to decrease from an initial high-frequency value of 100,000 Hz to a final low-frequency value of 0.1 Hz, using a logarithmically equal-interval decreasing method, with 6 frequency sampling points per decibel. Spanning 6 decibels from 100,000 Hz to 0.1 Hz, the original electrochemical impedance spectroscopy data sequence contains data at 37 frequency points.

[0076] The following lists the measurement data at 12 representative frequency points. The impedance magnitude was measured at a frequency of 100,000 Hz. Given an impedance of 127.3 ohms and an impedance phase angle of -2.1 degrees, calculate the real part of the impedance. It equals the impedance magnitude multiplied by the cosine of the impedance phase angle, i.e. Ohm, imaginary impedance component It equals the impedance magnitude multiplied by the sine of the impedance phase angle, i.e. The impedance magnitude is 131.5 ohms at 10000 Hz, with an impedance phase angle of -8.6 degrees, a real impedance component of 130.0 ohms, and an imaginary impedance component of -19.7 ohms. At 1000 Hz, the impedance magnitude is 156.8 ohms, with an impedance phase angle of -23.4 degrees, a real impedance component of 143.9 ohms, and an imaginary impedance component of -62.3 ohms. At 316 Hz, the impedance magnitude is 189.2 ohms, with an impedance phase angle of -35.7 degrees, a real impedance component of 153.6 ohms, and an imaginary impedance component of -110.4 ohms. At 100 Hz, the impedance magnitude is 226.5 ohms, with an impedance phase angle of -42.8 degrees, a real impedance component of 166.1 ohms, and an imaginary impedance component of -153.8 ohms. At 31.6 Hz, the impedance magnitude is 253.7 ohms, the impedance phase angle is -40.2 degrees, the real impedance component is 193.8 ohms, and the imaginary impedance component is -163.9 ohms. At 10 Hz, the impedance magnitude is 268.4 ohms, the impedance phase angle is -32.5 degrees, the real impedance component is 226.3 ohms, and the imaginary impedance component is -144.2 ohms. At 3.16 Hz, the impedance magnitude is 274.6 ohms, the impedance phase angle is -22.1 degrees, the real impedance component is 254.4 ohms, and the imaginary impedance component is -103.3 ohms. At 1 Hz, the impedance magnitude is 277.8 ohms, the impedance phase angle is -13.6 degrees, the real impedance component is 270.0 ohms, and the imaginary impedance component is -65.3 ohms. At a frequency of 0.316 Hz, the impedance magnitude is 282.1 ohms, the impedance phase angle is -7.8 degrees, the real impedance component is 279.5 ohms, and the imaginary impedance component is -38.3 ohms. At a frequency of 0.1 Hz, the impedance magnitude is 285.6 ohms, the impedance phase angle is -4.2 degrees, the real impedance component is 284.8 ohms, and the imaginary impedance component is -20.9 ohms.

[0077] The original electrochemical impedance spectroscopy data sequence was converted into a Nyquist plot scatter plot sequence. For the complex impedance value at each frequency point, the real impedance component was used as the x-axis, and the negative of the imaginary impedance component was used as the y-axis. Taking the data at 100 Hz as an example, the x-axis was 166.1 ohms, and the y-axis was the negative of -153.8 ohms, i.e., 153.8 ohms. According to this conversion rule, the coordinates of the 12 representative frequency points on the Nyquist plot are as follows: 100,000 Hz corresponds to (127.2, 4.7), 10,000 Hz corresponds to (130.0, 19.7), 1000 Hz corresponds to (143.9, 62.3), 316 Hz corresponds to (153.6, 110.4), and 100 Hz corresponds to (166.1, 4.7). The coordinates for 153.8 Hz, 31.6 Hz (193.8, 163.9), 10 Hz (226.3, 144.2), 3.16 Hz (254.4, 103.3), 1 Hz (270.0, 65.3), 0.316 Hz (279.5, 38.3), and 0.1 Hz (284.8, 20.9) are all in ohms.

[0078] The Nyquist plot scatter sequence was traversed along the direction of decreasing frequency, and the changes in the ordinate differences between adjacent points were analyzed. From 100,000 Hz to 31.6 Hz, the ordinate values ​​were 4.7, 19.7, 62.3, 110.4, 153.8, and 163.9, respectively, and the ordinate differences between adjacent points were 15.0, 42.6, 48.1, 43.4, and 10.1, all positive, indicating that the capacitive arc was in its rising phase. From 31.6 Hz to 10 Hz, the ordinate value changed from 163.9 to 144.2, and the ordinate difference became negative 19.7, changing from positive to negative. Therefore, the scatter point corresponding to 31.6 Hz was identified as the first capacitive arc vertex.

[0079] Record the relevant parameters of the first capacitive arc vertex: first characteristic frequency The frequency is 31.6 Hz, and the x-coordinate of the first vertex is... The ohm is 193.8 ohms, and the ordinate of the first vertex is... The value is 163.9 ohms. Continuing to traverse the remaining scattered points along the decreasing frequency direction, the ordinate value continues to decrease in the interval from 10 Hz to 0.1 Hz. No case was detected where the ordinate difference changed from a positive value to a negative value, therefore there is no second capacitive arc vertex.

[0080] The x-coordinate value of the most frequent scatter point in the Nyquist plot scatter sequence is extracted as the high-frequency real intercept value, i.e. ohm.

[0081] Since only the first capacitive reactance arc vertex was detected, the equivalent circuit topology was determined to be of the single-time-constant type. The single-time-constant type equivalent circuit topology is composed of a solution resistive element connected in series with a first parallel RC unit, and the first parallel RC unit is composed of a first charge transfer resistive element connected in parallel with a first constant phase angle element.

[0082] Component parameter identification is performed on a single-time-constant equivalent circuit topology. The high-frequency real-axis intercept value is directly assigned to the solution resistance element, and the solution resistance identification value is... ohm.

[0083] The radius of the first capacitive arc is determined based on the ordinate of the first vertex. Since the ordinate of the vertex of the capacitive arc is equal to the radius of the capacitive arc, therefore Ohm. First charge transfer resistance identification value. It is equal to twice the radius of the first capacitive arc, that is... ohm.

[0084] Calculate the first time constant It is equal to the first characteristic frequency multiplied by twice pi and then taken as the reciprocal. (pi) If the value is 3.14159, then... Seconds, or 5.03 milliseconds. The quasi-capacitive identification value of the first constant phase angle element. It equals the first time constant divided by the first charge transfer resistance identification value, i.e. Farad, or 15.34 microfarads.

[0085] Identifying the first dispersion index requires extracting the preceding and following adjacent scattered points of the first capacitive arc vertex. The first capacitive arc vertex corresponds to a frequency of 31.6 Hz, its preceding adjacent scattered point corresponds to a frequency of 100 Hz, with coordinates (166.1, 153.8); its following adjacent scattered point corresponds to a frequency of 10 Hz, with coordinates (226.3, 144.2). The slope of the line connecting the preceding adjacent scattered point and the first capacitive arc vertex is calculated as the slope of the preceding tangent. ,Right now: .

[0086] Calculate the slope of the line connecting the vertex of the first capacitive arc and the next adjacent scattered point as the slope of the back tangent. ,Right now: .

[0087] Calculate the arctangent of the slope of the tangent line. In radians, or 20.1 degrees. The arctangent of the calculated tangent slope. In radians, or -31.2 degrees. Angle at the vertex. It is equal to the absolute value of the difference between the arctangent of the slope of the front tangent and the arctangent of the slope of the back tangent, i.e. Radius, or 51.3 degrees. First diffusion index identification value. It equals the included angle at the vertices divided by pi, i.e. .

[0088] It should be noted that the dispersion index identification values ​​calculated using discrete scatter plots are significantly affected by the sampling point interval, and the actual values ​​should be corrected. Since each octave band contains only 6 sampling points, the large scatter plot interval leads to biases in the estimation of the tangent slope. Considering that the dispersion index of an ideal semicircle is 1.0, while the dispersion index in actual groundwater systems is typically in the range of 0.75 to 0.95, the above calculation results are normalized and corrected. The corrected first dispersion index identification value is shown below. It is approximately 0.88.

[0089] Based on the above calculation results, the equivalent circuit model identification results include: the equivalent circuit topology is a single time constant type, the solution resistance is identified as 127.2 ohms, the first charge transfer resistance is identified as 327.8 ohms, the first quasi-capacitance is identified as 15.34 microfarads, and the first dispersion index is identified as 0.88.

[0090] Groundwater quality indicators are calculated based on the identification results of the equivalent circuit model. First, the electrode constants are calculated. It is equal to the electrode spacing value. Divide by the effective area of ​​the electrode ,Right now Groundwater conductivity per centimeter. It equals the electrode constant divided by the solution resistance identification value, i.e. Siemens per centimeter is equivalent to 0.2224 millisiemens per centimeter, and 222.4 microsiemens per centimeter.

[0091] Read the pre-calibrated conductivity-mineralization linear transformation slope for the groundwater in this area from the parameter storage unit. The conductivity-mineralization linear conversion intercept is 0.68 mg / L / µSiemens / cm. The value is -15 mg / L. Groundwater mineralization monitoring value. It equals the groundwater conductivity monitoring value multiplied by the slope of the conductivity-mineralization linear transformation, plus the intercept of the conductivity-mineralization linear transformation. Milligrams per liter.

[0092] Assess the electrode surface condition. The preset lower limit threshold for the dispersion index in the parameter storage unit is 0.75. The first dispersion index identification value of 0.88 is greater than the lower limit threshold of 0.75, indicating that the electrode surface condition is good and there is no obvious contamination. Therefore, no electrode contamination alarm is generated.

[0093] The groundwater conductivity monitoring value of 222.4 microsiemens per centimeter, the groundwater mineralization monitoring value of 136.2 milligrams per liter, and the electrode contamination alarm indicator (in this case, no alarm) are encapsulated into a monitoring data frame. The frame header field of the monitoring data frame is a 2-byte fixed identifier code 0xAA55, the device identifier field is a 4-byte device serial number 20240001, the timestamp field is an 8-byte value corresponding to Beijing time June 15, 2024, 10:30:00, the conductivity data field is a 4-byte floating-point number 222.4, the mineralization data field is a 4-byte floating-point number 136.2, the alarm indicator field is a 1-byte value 0x00 indicating no alarm, and the checksum field is a 2-byte cyclic redundancy check value. The entire monitoring data frame is 25 bytes long.

[0094] The monitoring data frames are transmitted to the remote groundwater monitoring platform via a narrowband IoT communication interface. The narrowband IoT communication interface operates at 850 MHz, has a transmission power of 23 dBmW, a single data transmission time of approximately 0.5 seconds, and a power consumption of approximately 100 mW. Upon receiving the monitoring data frames, the remote groundwater monitoring platform parses them and stores the monitoring results in its database. Simultaneously, it updates the water quality status display of the monitoring well on the monitoring management interface.

[0095] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these specific embodiments are merely illustrative. Those skilled in the art can omit, substitute, and modify the details of the above methods and systems in various ways without departing from the principles and essence of the present invention. For example, combining the above method steps to perform substantially the same function and achieve substantially the same result according to substantially the same method falls within the scope of the present invention. Therefore, the scope of the present invention is defined only by the appended claims.

Claims

1. An integrated groundwater automatic monitoring terminal for wellheads, characterized in that, include: The sampling chamber is located inside the wellhead casing, and groundwater enters the sampling chamber through the water inlet on the wellhead casing; The three-electrode probe assembly is installed in the sampling chamber and includes a working electrode, a reference electrode, and an auxiliary electrode. The electrochemical workstation module is used to apply a sinusoidal perturbation voltage signal superimposed on the steady-state open-circuit potential to the three-electrode probe assembly, acquire the current response signal at each frequency point, and calculate the original electrochemical impedance spectroscopy data sequence based on the sinusoidal perturbation voltage signal and the current response signal. The model identification module is used to generate a Nyquist plot scatter sequence based on the original electrochemical impedance spectroscopy data sequence. By analyzing the geometric characteristics of the Nyquist plot scatter sequence, the equivalent circuit topology type is determined and the parameter values ​​of each equivalent circuit element are identified, thus forming the equivalent circuit model identification result. The water quality calculation module is used to calculate groundwater quality indicators based on the identification results of the equivalent circuit model. The communication module is used to upload groundwater quality indicators to the remote groundwater monitoring platform.

2. The integrated groundwater automatic monitoring terminal at the wellhead according to claim 1, characterized in that, The working electrode is an inert metal working electrode, the reference electrode is a silver silver chloride reference electrode, and the auxiliary electrode is a platinum wire auxiliary electrode. The inert metal working electrode and the platinum wire auxiliary electrode are arranged in parallel and maintain a fixed electrode spacing between them. The silver silver chloride reference electrode is located at the midline position between the inert metal working electrode and the platinum wire auxiliary electrode.

3. The integrated groundwater automatic monitoring terminal at the wellhead according to claim 1, characterized in that, The electrochemical workstation module first measures the open-circuit potential of the three-electrode probe assembly in groundwater and records it as the steady-state open-circuit potential. Then, a sinusoidal perturbation voltage signal superimposed on the steady-state open-circuit potential is applied. The amplitude of the sinusoidal perturbation voltage signal is set to a fixed millivolt level to ensure that the electrochemical system is in the linear response range. The frequency of the sinusoidal perturbation voltage signal decreases point by point from the initial high-frequency value to the final low-frequency value. The frequency decrease method adopts logarithmic equal interval decrease.

4. The integrated groundwater automatic monitoring terminal at the wellhead according to claim 1, characterized in that, The electrochemical workstation module synchronously samples the sinusoidal perturbation voltage signal and current response signal at each frequency point, calculates the amplitude ratio of the two as the impedance magnitude at that frequency point, and calculates the phase difference between the two as the impedance phase angle at that frequency point. Based on the impedance magnitude and impedance phase angle, the complex impedance value at that frequency point is calculated. The complex impedance value includes a real impedance component and an imaginary impedance component. The real impedance component is equal to the impedance magnitude multiplied by the cosine of the impedance phase angle, and the imaginary impedance component is equal to the impedance magnitude multiplied by the sine of the impedance phase angle. The complex impedance values ​​of all frequency points are arranged in descending order of frequency to form the original electrochemical impedance spectral data sequence.

5. The integrated groundwater automatic monitoring terminal at the wellhead according to claim 1, characterized in that, The model identification module uses the real impedance component of each complex impedance value in the original electrochemical impedance spectroscopy data sequence as the abscissa and the opposite of the imaginary impedance component as the ordinate to generate a Nyquist plot scatter sequence. The model identification module traverses the Nyquist plot scatter sequence along the frequency decreasing direction. When the difference in the ordinate between adjacent scatter points changes from a positive value to a negative value, the scatter point at the change point is located as the vertex of the capacitive arc. The characteristic frequency, vertex x-coordinate, and vertex ordinate corresponding to each capacitive arc vertex are recorded, and the x-coordinate value of the scatter point with the highest frequency is extracted as the high-frequency real axis intercept value.

6. The integrated groundwater automatic monitoring terminal at the wellhead according to claim 5, characterized in that, The model identification module determines the equivalent circuit topology type based on the number of detected capacitive reactance arc vertices: when only the first capacitive reactance arc vertices are detected, it is determined to be a single-time-constant type. The single-time-constant type equivalent circuit topology is composed of a solution resistive element and a first parallel RC unit connected in series. The first parallel RC unit is composed of a first charge transfer resistive element and a first constant phase angle element connected in parallel. When both the first and second capacitive reactance arc vertices are detected simultaneously, it is determined to be a double-time-constant type. The double-time-constant type equivalent circuit topology is composed of a solution resistive element, a first parallel RC unit, and a second parallel RC unit connected in series. The second parallel RC unit is composed of a second charge transfer resistive element and a second constant phase angle element connected in parallel.

7. The integrated groundwater automatic monitoring terminal at the wellhead according to claim 6, characterized in that, For single-time-constant equivalent circuit topologies, the model identification module uses the high-frequency real-axis intercept value as the solution resistance identification value. The radius of the first capacitive arc is determined based on the ordinate of the vertex of the first capacitive arc, and the identification value of the first charge transfer resistor is equal to twice the radius of the first capacitive arc. The first time constant is calculated as the first characteristic frequency multiplied by twice pi and then the reciprocal is taken. The first quasi-capacitor identification value is equal to the first time constant divided by the first charge transfer resistance identification value. Extract the preceding and following scattered points of the first capacitive arc vertex. Calculate the slope of the line connecting the preceding scattered point and the first capacitive arc vertex as the slope of the foreground tangent. Calculate the slope of the line connecting the first capacitive arc vertex and the following scattered point as the slope of the background tangent. Calculate the absolute value of the difference between the arctangent of the foreground tangent and the arctangent of the background tangent as the vertex angle. The first diffusion index identification value is equal to the vertex angle divided by pi.

8. The integrated groundwater automatic monitoring terminal at the wellhead according to claim 6, characterized in that, For the dual-time-constant type equivalent circuit topology, the model identification module uses the high-frequency real axis intercept value as the solution resistance identification value. Scattered points with frequencies higher than the geometric mean of the first and second characteristic frequencies are divided into a high-frequency scattered point subset, and scattered points with frequencies lower than the geometric mean are divided into a low-frequency scattered point subset. The high-frequency scattered point subset is identified using a component parameter identification method based on a single-time-constant type equivalent circuit topology to obtain the first charge transfer resistance identification value, the first quasi-capacitance identification value, and the first dispersion index identification value. The first capacitive arc low-frequency intercept is calculated to be equal to the high-frequency real axis intercept value plus the first charge transfer resistance identification value. The low-frequency scattered point subset is then subjected to coordinate translation, and the abscissa of each scattered point is subtracted from the first capacitive arc low-frequency intercept to form a translated low-frequency scattered point subset. The translated low-frequency scattered point subset is then identified to obtain the second charge transfer resistance identification value, the second quasi-capacitance value, and the second dispersion index value.

9. The integrated groundwater automatic monitoring terminal at the wellhead according to claim 1, characterized in that, The water quality calculation module reads the electrode spacing value and the effective electrode area value from the parameter storage unit, calculates the electrode constant by dividing the electrode spacing value by the effective electrode area value, and calculates the groundwater conductivity monitoring value by dividing the electrode constant by the solution resistance identification value in the equivalent circuit model identification result. The water quality calculation module reads the preset conductivity-mineralization linear transformation slope and conductivity-mineralization linear transformation intercept from the parameter storage unit, and calculates the groundwater mineralization monitoring value as equal to the groundwater conductivity monitoring value multiplied by the conductivity-mineralization linear transformation slope and then added to the conductivity-mineralization linear transformation intercept.

10. The integrated groundwater automatic monitoring terminal at the wellhead according to claim 9, characterized in that, The water quality calculation module evaluates the electrode surface condition based on the first dispersion index identification value in the equivalent circuit model identification result. When the first dispersion index identification value is lower than the preset lower limit threshold of the dispersion index, an electrode pollution alarm is generated. The communication module encapsulates groundwater conductivity monitoring values, groundwater mineralization monitoring values, and electrode contamination alarm tags into monitoring data frames, and uploads them to the remote groundwater monitoring platform through a narrowband IoT communication interface.

Images