Method and system for detecting trace organic pollutants in water body

Abstract

The application discloses a water body trace organic pollutant detection method and system, relates to the field of environmental water body detection, and comprises the following steps: collecting environmental parameters of a target water body sample to be detected, and acquiring baseline spectrum data of the target water body sample; determining disturbance operation parameters matched with the target water body sample, and exerting micro-disturbance on the target water body sample according to the disturbance operation parameters to acquire disturbance response spectrum data; performing time sequence difference analysis on the disturbance response spectrum data and the baseline spectrum data to obtain disturbance response residual spectrum, and identifying an optimal response characteristic wavelength from a characteristic response wavelength interval of a target trace organic pollutant based on the disturbance response residual spectrum; and solving a concentration correction model according to the optimal response characteristic wavelength, the disturbance operation parameters and the environmental parameters to obtain the concentration of the target trace organic pollutant. Thus, the detection sensitivity and anti-interference capability of trace organic pollutants in a water body sample are improved.

Description

Technical Field

[0001] This application relates to the field of environmental water body detection technology, and in particular to a method and system for detecting trace organic pollutants in water. Background Technology

[0002] With the rapid advancement of industrialization and urbanization, various trace organic pollutants (such as phenol, polycyclic aromatic hydrocarbons, and pesticide residues) are constantly entering surface water and groundwater bodies. These pollutants are characterized by high biotoxicity, high environmental mobility, and difficulty in natural degradation, posing a serious threat to ecosystem security and human health even at trace levels. Therefore, achieving highly sensitive and accurate detection of trace organic pollutants in water bodies has become a pressing technical challenge for environmental science and public health.

[0003] Currently, the composition of environmental water bodies is extremely complex, containing a large number of coexisting components such as natural organic matter, colloids, and inorganic ions. These components not only highly overlap with target pollutants at the optical and chemical signal levels, but also significantly affect the stability and specificity of the overall detection signal due to dynamic changes in environmental conditions (such as temperature, pH, and turbidity). In addition, the concentration of target pollutants is often at extremely low levels, making their signals easily overwhelmed by complex background noise and interference from coexisting components, leading to decreased sensitivity and increased false positive and false negative rates in actual detection processes.

[0004] In addition, the spatial and temporal variations of environmental water bodies are significant, and the background conditions and interference sources vary greatly at different sampling points and at different sampling times. This leads to a series of technical challenges for traditional detection methods in practical applications, such as difficulty in effectively correcting background interference, insufficient signal specificity, and limitations in detection accuracy and stability. Summary of the Invention

[0005] This application provides a method, system, storage medium, computer program product, and electronic device for detecting trace organic pollutants in water, which at least solves the problem that trace organic pollutant signals are easily submerged by water background, impurities, and environmental noise in current related technologies, resulting in low detection sensitivity and poor anti-interference ability.

[0006] In a first aspect, embodiments of this application provide a method for detecting trace organic pollutants in water. The method includes: collecting environmental parameters of a target water sample to be detected and acquiring baseline spectral data of the target water sample; determining perturbation operation parameters matching the target water sample, and applying micro-perturbation to the target water sample according to the perturbation operation parameters to obtain perturbation response spectral data; the perturbation operation parameters include perturbation operation type, perturbation operation amplitude, and perturbation operation duration; performing time-series difference analysis on the perturbation response spectral data and the baseline spectral data to obtain a perturbation response residual spectrum, and identifying the optimal response characteristic wavelength from the characteristic response wavelength range of the target trace organic pollutant based on the perturbation response residual spectrum; solving a concentration correction model according to the optimal response characteristic wavelength, the perturbation operation parameters, and the environmental parameters to obtain the concentration of the target trace organic pollutant; the concentration correction model is expressed by the following formula:

[0007] C target =f(△S(λ) * ,t),A perturb E env ),

[0008] In the formula, C target Let T be the concentration of the target trace organic pollutant, and ΔS(λ) be the time interval of the perturbation. * ,t) represents the residual spectrum of the perturbation response at the optimal response characteristic wavelength λ. * and the response strength at time t of the disturbance, A perturb For the disturbance operation parameters, E env Here, f(·) represents the environmental parameters, and f(·) represents the pre-trained concentration correction model function.

[0009] Secondly, embodiments of this application provide a water trace organic pollutant detection system, the system comprising: a water sample acquisition unit, used to acquire environmental parameters of a target water sample to be detected and obtain baseline spectral data of the target water sample; a sample micro-perturbation unit, used to determine perturbation operation parameters matching the target water sample, and apply micro-perturbation to the target water sample according to the perturbation operation parameters to obtain perturbation response spectral data; the perturbation operation parameters include perturbation operation type, perturbation operation amplitude, and perturbation operation duration; an optimal wavelength identification unit, used to perform time-series difference analysis on the perturbation response spectral data and baseline spectral data to obtain a perturbation response residual spectrum, and identify the optimal response characteristic wavelength from the characteristic response wavelength range of the target trace organic pollutant based on the perturbation response residual spectrum; and a trace concentration identification unit, used to solve a concentration correction model according to the optimal response characteristic wavelength, the perturbation operation parameters, and the environmental parameters to obtain the concentration of the target trace organic pollutant; the concentration correction model is expressed by the following formula:

[0010] C target =f(△S(λ) * ,t),A perturb E env ),

[0011] In the formula, C target Let T be the concentration of the target trace organic pollutant, and ΔS(λ) be the time interval of the perturbation. * ,t) represents the residual spectrum of the perturbation response at the optimal response characteristic wavelength λ. * and the response strength at time t of the disturbance, A perturb For the disturbance operation parameters, E env Here, f(·) represents the environmental parameters, and f(·) represents the pre-trained concentration correction model function.

[0012] Thirdly, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of the water trace organic pollutant detection method of any embodiment of the present application.

[0013] Fourthly, embodiments of this application provide a storage medium storing a computer program thereon, characterized in that, when the program is executed by a processor, it implements the steps of the water trace organic pollutant detection method of any embodiment of this application.

[0014] Fifthly, embodiments of this application provide a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the water trace organic pollutant detection method of any embodiment of this application.

[0015] The method and system for detecting trace organic pollutants in water provided in this application can achieve at least the following technical effects:

[0016] (1) By applying precise and controllable micro-perturbations to the water sample, the dynamic response of the target pollutant molecules is actively stimulated by setting perturbation operation parameters (including perturbation type, amplitude and duration), and by combining the time difference between the baseline spectrum and the perturbation response spectrum, the efficient removal of stable background signals and the significant amplification of weak target signals are achieved, which effectively improves the response sensitivity of the detection system to low concentration pollutants.

[0017] (2) By performing time-series difference analysis between the baseline spectrum before perturbation and the response spectrum after perturbation, the net response characteristics of pollutants after perturbation can be effectively extracted, eliminating the influence of system background drift and irrelevant noise. Furthermore, by comparing and analyzing the candidate wavelengths of the characteristic response wavelength range of the target trace organic pollutants in the residual spectrum of the perturbation response, the optimal response wavelength that best matches the molecular characteristics of the pollutants is automatically determined, thereby significantly enhancing the ability to distinguish between similar spectral lines, avoiding cross-interference caused by wavelength overlap, and improving detection specificity.

[0018] (3) Considering that the pollutant response is not only affected by its own concentration, but also by the joint regulation of the disturbance mode and environmental conditions, an adaptive multi-factor coupled quantitative model with disturbance response intensity, disturbance operation parameters and environmental parameters as independent variables is constructed. Through the pre-training strategy, the dynamic estimation and error compensation of pollutant concentration under different environmental backgrounds and disturbance operations can be realized, achieving high stability and high accuracy of detection results under various complex water conditions.

[0019] This technical solution, by applying micro-perturbations and analyzing the residual spectrum of the perturbation response, combined with the extraction of the optimal response characteristic wavelength, enables the accurate identification and quantitative analysis of the concentration of trace organic pollutants in water. Simultaneously, through dynamic acquisition of environmental parameters and adjustment of perturbation operation parameters, the influence of environmental condition changes on the detection signal can be corrected in real time, reducing false positive and false negative rates and significantly improving the stability and accuracy of the detection. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A flowchart illustrating an example of a method for detecting trace organic pollutants in water according to an embodiment of this application is shown.

[0022] Figure 2 A flowchart illustrating an example of determining perturbation operation parameters that match a target water sample, according to an embodiment of this application, is shown.

[0023] Figure 3 A schematic diagram of the simulated ultraviolet absorption spectrum response of an example of phenolic pollutants under physical micro-oscillations is shown.

[0024] Figure 4A schematic diagram of the simulation results of an example of the Raman spectral response of trace organic amine pollutants under different pH conditions is shown.

[0025] Figure 5 A simulation diagram illustrating an example of the fluorescence spectral response of trace polycyclic aromatic hydrocarbons under different ionic intensities is shown.

[0026] Figure 6 A flowchart illustrating an example of identifying the optimal response characteristic wavelength based on the perturbation response residual spectrum according to an embodiment of this application is shown.

[0027] Figure 7 A simulation diagram showing the comparison of prediction error distributions for an example of a trace organic pollutant concentration prediction experiment is presented.

[0028] Figure 8 A structural block diagram of an example of a water trace organic pollutant detection system according to an embodiment of this application is shown. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] It should be noted that with the continuous development of biomaterials and intelligent technologies, some relatively effective methods for detecting trace organic pollutants in water have emerged in the current related technologies, but all of them have some shortcomings.

[0031] Specifically, laboratory analytical techniques such as gas chromatography-mass spectrometry (GC-MS) and high-performance liquid chromatography-mass spectrometry (HPLC-MS) are widely used for the qualitative and quantitative analysis of trace organic pollutants due to their extremely high sensitivity and selectivity. These techniques can achieve detection limits at ppb (parts per billion) or even lower levels. However, in practical applications, they suffer from drawbacks such as cumbersome sample pretreatment processes, long analysis cycles, high equipment costs, and demanding operational skills, making it difficult to meet the real-time, wide-range, and high-throughput on-site detection requirements in water environment monitoring.

[0032] Spectroscopy-based non-destructive testing methods, such as absorption spectroscopy, fluorescence spectroscopy, and Raman spectroscopy, have been increasingly applied to rapid on-site screening and dynamic monitoring of water pollutants in recent years due to their advantages, including fast detection speed, ability to achieve online or in-situ automated monitoring, and high degree of information integration. However, in complex aquatic environments, these methods are often affected by factors such as multi-component background interference, high overlap of spectral features, instrument baseline drift, and fluctuations in environmental parameters. This leads to reduced detection sensitivity and increased false positive or false negative rates, making it difficult to achieve highly reliable identification of pollutants at extremely low concentrations.

[0033] With the development of big data and artificial intelligence technologies, some detection systems have begun to incorporate algorithms such as machine learning and pattern recognition to improve signal analysis capabilities and the accuracy of detection results. These methods can achieve efficient deconstruction of complex spectral signals and quantitative analysis of pollutants under specific sample conditions. However, their algorithm performance is highly dependent on the representativeness and quality of the training data, and in real-world environments, they are easily affected by factors such as fluctuations in the quality of acquired signals, insufficient sample labeling, and limited model generalization ability. Therefore, their practical application effectiveness and stability are still insufficient to meet the high demands of on-site environmental monitoring.

[0034] It should be understood that the above description of the relevant technologies is intended only to help the public better understand the inventive spirit and motivation of this application, and is not intended to limit this application. Furthermore, the technical solutions described in the above-mentioned relevant technologies are not prior art, and may also be undisclosed technical solutions, such as those under research or in the laboratory stage.

[0035] Figure 1 A flowchart illustrating an example of a method for detecting trace organic pollutants in water according to an embodiment of this application is shown.

[0036] Regarding the execution subject of the method in the embodiments of this application, it can be any controller or processor with computing or processing capabilities. By introducing a combination of perturbation operation and time-series differential analysis, the signal processing flow is optimized and an accurate calibration model is established, which can effectively improve the detection sensitivity and accuracy of trace organic pollutants in water.

[0037] In some examples, it can be integrated into an electronic device or terminal through software, hardware, or a combination of both, and the type of terminal or electronic device can be diverse, such as mobile phones, tablets, or desktop computers, etc.

[0038] like Figure 1 As shown, in step S110, environmental parameters of the target water sample to be detected are collected, and baseline spectral data of the target water sample are obtained.

[0039] First, environmental parameters of the target water sample must be collected. These parameters include at least one of the following: water temperature, pH value, conductivity, turbidity, or dissolved oxygen concentration. Environmental parameters have varying effects on the stability and characteristics of spectral signals. For example, changes in temperature affect the concentration of dissolved oxygen in the water, thus altering the spectral characteristics of pollutants; changes in pH value affect the molecular structure and chemical reactivity of organic pollutants. By obtaining these environmental parameters, the influence of different environmental factors can be effectively compensated for.

[0040] Furthermore, baseline spectral data of water samples were acquired using a high-resolution spectrometer. Baseline spectra reflect the spectral response of the water body when undisturbed, representing its natural spectral response without external disturbance. They primarily include the optical absorption characteristics of all components in the water body (such as water and dissolved organic matter). Analysis of these baseline data allows for an understanding of the spectral characteristics of the water body under undisturbed conditions and provides a precise reference standard for subsequent disturbance response analysis and correction.

[0041] In step S120, perturbation operation parameters that match the target water sample are determined, and micro-perturbations are applied to the target water sample according to the perturbation operation parameters to obtain perturbation response spectral data.

[0042] Here, appropriate perturbation operation parameters are selected based on the characteristics of the target pollutant. These parameters include the perturbation operation type, perturbation operation amplitude, and perturbation operation duration. The perturbation operation type can be, for example, temperature change, pH adjustment, or dissolved oxygen concentration change. It can vary depending on the type of pollutant to be detected, aiming to selectively amplify the spectral response of the target pollutant.

[0043] In some examples of embodiments of this application, the perturbation operation type includes at least one of the following: physical micro-oscillation, pH value fine-tuning, ion strength micro-change, or light intensity periodic fluctuation.

[0044] Specifically, physical micro-oscillations can employ techniques such as micro-magnetic stirring, ultrasonic micro-disturbance, and low-frequency vibrators to stimulate the internal fluid dynamics of water samples, disrupt the static stability of the microenvironment, and enhance the physical separation of pollutants from the background components of the water. pH fine-tuning can be achieved through automatic titration or micro-acid / base injection, allowing dynamic fluctuations within a very small range to stimulate characteristic responses of pH-sensitive organic pollutants and highlight their intrinsic chemical differences. Micro-changes in ionic strength can be achieved through the automatic addition of trace amounts of salts (such as NaCl and KCl) to adjust the ionic strength of the water, inducing changes in the dissolved, complexed, or adsorbed states of target organic pollutants and amplifying their spectral characteristics. Periodic fluctuations in light intensity can be achieved by controlling the periodic intensity changes of the detection light source or auxiliary excitation light source, suitable for excitation-response scenarios such as Raman and fluorescence, highlighting the excited-state dynamics of the target analyte.

[0045] It should be noted that different types of trace pollutants (such as phenols, polycyclic aromatic hydrocarbons, and organic amines) have fundamentally different molecular structures, energy level distributions, polarities, and sensitivities to external physicochemical environments. Furthermore, different types of molecules often respond more strongly to different types of perturbations that trigger excitation. For example, phenolic pollutants are highly sensitive to physical micro-oscillations because perturbations easily break intermolecular hydrogen bonds and distribution patterns; the fluorescence signals of polycyclic aromatic hydrocarbon pollutants (such as naphthalene and phenanthrene) are easily modulated by ionic intensity, requiring the use of trace salt perturbations; and organic amine pollutants are extremely sensitive to pH conditions, allowing pH fine-tuning to be used as the primary perturbation method. Therefore, the type of perturbation operation can be customized based on the molecular recognition characteristics of the pollutant. By selecting the correct excitation channel, the signal selective amplification of the target pollutant can be maximized while the main background signal remains essentially unchanged, thereby improving detection sensitivity.

[0046] Furthermore, the amplitude (i.e., the intensity of the perturbation) and duration (i.e., the application time of the perturbation) of the perturbation operation are designed to elicit changes in the spectral response of pollutant molecules, enabling them to exhibit characteristics that are undetectable under undisturbed conditions. For the same class of trace organic pollutants, their response mechanisms to physical or chemical perturbations are similar, and the same type of perturbation operation can be used. However, the optimal perturbation amplitude and duration (i.e., perturbation intensity and application time window) need to be optimized for different water samples, different backgrounds, and different instrument conditions to effectively enhance the spectral response of trace pollutants in the currently detected water samples, thereby effectively improving the accuracy and reliability of detection.

[0047] In step S130, time-series difference analysis is performed on the perturbation response spectral data and the baseline spectral data to obtain the perturbation response residual spectrum, and the optimal response characteristic wavelength is identified from the characteristic response wavelength range of the target trace organic pollutant based on the perturbation response residual spectrum.

[0048] Here, the difference between the perturbation response spectrum and the baseline spectrum reflects the response characteristics of pollutants in the water body to perturbation operations. Through time-series difference analysis, the interference of environmental background noise and natural components in the water body (such as natural organic matter and inorganic ions) can be effectively removed, effectively eliminating background interference and highlighting the spectral characteristic responses of pollutants affected by perturbation operations.

[0049] Furthermore, based on the perturbation response residual spectrum, the optimal response characteristic wavelength is identified from the characteristic response wavelength range of the target trace organic pollutant. For example, based on the response peak in the residual spectrum, the most representative and sensitive response wavelength can be screened from multiple candidate wavelengths. This provides the maximum pollutant signal, ensuring efficient differentiation of target pollutants and accurate identification of their characteristics in complex water conditions, thus enhancing the specificity and sensitivity of the signal.

[0050] It should be noted that each trace organic pollutant possesses one or more highly specific and sensitive "characteristic response wavelength ranges," which often correspond to the molecular absorption peaks, emission peaks, or scattering characteristics of the pollutant in spectral domains such as ultraviolet, visible, fluorescence, or Raman. For example, the main ultraviolet absorption peak of phenol is usually around 270 nm, polycyclic aromatic hydrocarbons have multiple absorption peaks in the 220–300 nm range, and some pesticides will produce fluorescence emission at specific wavelengths under specific excitation, and so on.

[0051] While the aforementioned characteristic response range pinpoints the "molecular fingerprint" range of a specific pollutant, the actual spectral response within this range does not always exhibit the largest, most stable, or most representative signal. Furthermore, the strongest response signal, lowest noise, and best sensitivity are often found at a specific wavelength, making it the optimal location for "quantitative detection." In addition, due to the influence of water background, instrument system, and other impurities, sporadic noise, background interference, or overlapping false peaks can easily occur. Moreover, other wavelengths within the characteristic response range often contain significant background noise or non-specific interference, which can dilute the target signal, lower the detection limit, and even lead to false positives / false negatives when used collectively in the calculation. Therefore, selecting the optimal response characteristic wavelength can greatly improve the accuracy and robustness of concentration quantification.

[0052] In step S140, the concentration correction model is solved based on the optimal response characteristic wavelength, perturbation operation parameters, and environmental parameters to obtain the concentration of the target trace organic pollutant.

[0053] Here, the core task of the concentration correction model is to establish a mathematical model that can quantitatively map the perturbation response signal to the target pollutant concentration by solving for the identified optimal response characteristic wavelength, perturbation operating parameters, and environmental parameters. This is achieved by mapping the input relevant signal data to the corresponding pollutant concentration value.

[0054] For example, the concentration correction model is expressed by the following equation:

[0055] C target =f(△S(λ) * ,t),A perturb E env Equation (1)

[0056] In the formula, C target Let T be the concentration of the target trace organic pollutant, and ΔS(λ) be the time interval of the perturbation. * ,t) represents the residual spectrum of the perturbation response at the optimal response characteristic wavelength λ. * and the response intensity at time t of the disturbance, A perturb For the disturbance operation parameters, E env Here, f(·) represents the environmental parameters, and f(·) represents the pre-trained concentration correction model function.

[0057] Specifically, the concentration correction model can be trained based on a large amount of experimental data or historical sample data. It can automatically correct signal fluctuations according to disturbance response and environmental changes, adapt to different water environments, and accurately estimate the concentration of pollutants.

[0058] It should be understood that the type of concentration correction model function can be diverse and will not be limited here. For example, the concentration correction model can employ multiple regression models (such as linear regression, support vector regression, Lasso regression, etc.) or machine learning methods (such as neural networks, random forests, etc.) to map known perturbation responses, characteristic wavelengths, and environmental parameters to the concentration of the target pollutant. Furthermore, there is generally a nonlinear relationship between pollutant concentration and response signal (especially when pollutant concentration is low), and various non-restricted nonlinear models (such as support vector machine regression, random forests, neural networks, etc.) can be used to set the concentration correction model function.

[0059] Through the embodiments of this application, by comprehensively utilizing perturbation operation, time-series difference analysis, optimal response wavelength identification, and multi-factor concentration correction model, and by flexibly adjusting perturbation operation parameters and dynamically selecting response characteristic wavelengths, the sensitivity, specificity, and stability issues of trace organic pollutant detection in environmental water bodies are effectively solved. It can also provide accurate and reliable concentration data for detection scenarios with different pollutants and environmental conditions.

[0060] Figure 2 A flowchart illustrating an example of determining perturbation operation parameters that match a target water sample, according to an embodiment of this application, is shown.

[0061] like Figure 2 As shown, in step S210, the target pollutant type of the target trace organic pollutant is obtained.

[0062] It should be noted that different types of pollutants have different molecular structures, physicochemical properties, and response mechanisms to disturbances. Only by first identifying the type of pollutant can we ensure that the excitation of disturbances is specific and efficient.

[0063] Here, the methods for obtaining the target pollutant type can be diverse, such as specifying it based on user input or automatic identification. For example, the system automatically parses the type of pollutant to be detected by analyzing the detection task sheet, such as "phenols," "polycyclic aromatic hydrocarbons," or "organic amines."

[0064] In step S220, a set of calibration parameters matching the target pollutant type is determined from a preset pollutant-calibration parameter set. The pollutant-calibration parameter set pre-stores multiple pollutant types and corresponding calibration parameter sets.

[0065] Here, a preset pollutant-calibration parameter set establishes a mapping relationship between pollutant types and optimal perturbation parameters. By directly calling the pre-stored parameters, the system can quickly obtain matching perturbation conditions, improving detection efficiency and consistency. An example of the mapping relationship is as follows:

[0066] Phenolic compounds: physical micro-oscillation, 20–35 Hz, 60–80 seconds;

[0067] Polycyclic aromatic hydrocarbons: slight change in ionic strength, NaCl 0.5-0.8 mM, stirring for 25-35 seconds;

[0068] Organic amines: fine-tune the pH to 8.0–9.0 and titrate slowly for 2–3 minutes.

[0069] In this way, the system can automatically match the corresponding calibration parameter set according to the test task sheet. The calibration parameter set can define multiple combinations of parameters for disturbance operation amplitude and duration. As a result, the debugging and pre-calibration time can be significantly shortened, the test throughput and engineering efficiency can be improved, and the high standardization and reproducibility of disturbance operations under different test batches and different sites can be guaranteed.

[0070] In step S230, a leading perturbation operation is applied to the target water sample based on multiple calibration perturbation operation parameters in the calibration parameter group.

[0071] It should be noted that different types of pollutants require different specific disturbance types; for the same type of pollutant, the disturbance type is consistent, and the parameters can be adaptively optimized. Here, each calibration disturbance operation parameter is predefined with a corresponding disturbance operation type, disturbance operation amplitude, and disturbance operation duration. Specifically, to adapt to differences in background water conditions and pollutant composition in different water bodies, multiple sets of predefined disturbance operation parameters are used for pilot experiments, forming a standardized and engineered parameter screening process.

[0072] For example, calibration parameter group Combination of multiple disturbance parameters (T) i A i D i Composed of, where T i Indicates the type of disturbance operation (e.g., physical micro-oscillation, pH fine-tuning, etc.), A i D represents the amplitude of the disturbance operation (such as oscillation amplitude, pH variation, etc.). i This indicates the duration of the perturbation operation. For each set of parameters, the system automatically controls the corresponding perturbation execution unit to sequentially apply each set of leading perturbation operations to multiple water samples. For example, for the detection of phenols, differentiated perturbation operations can be achieved by implementing differentiated oscillation frequencies and oscillation durations for each water sample.

[0073] In step S240, for each calibration perturbation operation parameter, the perturbation response residual spectrum corresponding to the calibration perturbation operation parameter is obtained, and the maximum target response of the perturbation response residual spectrum is determined in the characteristic wavelength range and perturbation time range of the target trace organic pollutant, and the maximum background response is determined in the background wavelength range and perturbation time range unrelated to the target trace organic pollutant.

[0074] Here, after each set of perturbation parameters is applied, the perturbation response spectrum is acquired in real time, and the difference between the spectrum and the baseline spectrum is calculated to extract the maximum target response and the maximum background response, so as to reflect the sensitivity of the target pollutant and the system noise characteristics.

[0075] Specifically, for the i-th set of perturbation parameters, the perturbation response spectrum S is collected. pert,i (λ,t), and obtain the baseline spectrum S base (λ,t0), thus calculating the residual spectrum of the perturbation response:

[0076] △S i (λ,t)=S pert,i (λ,t)-S base (λ,t0), Equation (2)

[0077] Within the characteristic wavelength range Λ of the target pollutant target Within the time interval T of the disturbance, determine the maximum target response:

[0078]

[0079] In the background-independent wavelength range Λ bg Within the time interval T of the disturbance, determine the maximum background response:

[0080]

[0081] Here, the maximum value extraction method is used to fully avoid the influence of occasional disturbances and noise, highlight the real physical signal under the disturbance-response mechanism, and achieve accurate quantification of the response sensitivity of each disturbance parameter and the system noise intensity.

[0082] In step S250, the ratio of the maximum target response to the maximum background response for each calibrated perturbation operation parameter is calculated to obtain the target correlation score.

[0083] Here, the target response and background response are normalized as a ratio to highlight the parameter set with the optimal signal-to-noise ratio, ensuring that the selected perturbation parameters can both greatly excite the target pollutant response and suppress background noise to the greatest extent.

[0084] Specifically, for each set of perturbation parameters, the target correlation score is calculated:

[0085]

[0086] In the formula, ε is a very small positive number to prevent the denominator from being zero; all G i Record it as a target relevance score vector.

[0087] Therefore, by optimizing the signal-to-noise ratio to calculate the target correlation score, the system can automatically avoid high noise interference areas, thereby improving the detection capability of target pollutants and the method's resistance to disturbances.

[0088] In step S260, the perturbation operation parameter with the highest correlation score to the target is selected as the perturbation operation parameter that matches the target water sample.

[0089] Here, the perturbation parameter group with the highest target relevance score is selected globally as the optimal perturbation strategy for the current sample.

[0090] Specifically, the optimal set of perturbation parameters is selected using the maximum value criterion:

[0091] (T * A * D * ) = argmax i G i Equation (6)

[0092] Then, the optimal set of perturbation parameters is automatically sent to the system control unit to perform perturbation operations on the target water sample and corresponding trace organic pollutant detection operations.

[0093] Through the embodiments of this application, a leading perturbation screening mechanism based on multiple sets of calibrated perturbation operation parameters is employed. This mechanism can systematically apply a series of perturbation parameter combinations to different target water samples, taking into account their physicochemical backgrounds and differences in potential pollutant components. The ratio of the maximum target response to the background response is calculated for the residual spectrum of the perturbation response for each set of parameters. Therefore, the optimization of all parameters is based on the objective spectral response (i.e., the quantitative ratio of the maximum signal response in the characteristic wavelength range to the background noise), ensuring accurate screening of the optimal perturbation operation strategy. This achieves maximum response enhancement to target trace organic pollutants and minimizes background interference, significantly improving the signal-to-noise ratio and accuracy of detection.

[0094] The technical implementation principles of the embodiments of this application will be explained below:

[0095] The existence state of trace organic pollutants in water is often affected by solubility, complexation, adsorption, and microenvironment. Their intrinsic spectral signals are often submerged by the main components of the water, impurities, and background noise, making them difficult to distinguish directly. Applying micro-perturbations, such as physical micro-oscillations, pH fine-tuning, slight changes in ionic strength, or periodic illumination, essentially alters the interaction between pollutant molecules and their environment, causing systematic changes in the energy level structure, electron cloud distribution, or dissociation state of pollutant molecules in solution. This change leads to repeatable and controllable dynamic responses in the absorption / fluorescence / Raman spectral signals of pollutants at specific wavelengths, while background components, due to the stability of their physicochemical properties, respond weakly or almost unresponsive to the same perturbations.

[0096] Therefore, perturbation residual spectroscopy can effectively reveal the specific kinetic signals of pollutants. For the same sample, the difference (i.e., residual) between the spectra collected before and after perturbation / during the perturbation period can effectively subtract the water baseline, background, and external interference, achieving differential amplification of pollutant signals and background masking. Furthermore, different perturbation operating parameters (type, amplitude, duration) form unique "perturbation-response" kinetic matching relationships with different pollutant molecules; such responses contain the chemical fingerprint information of the pollutants themselves.

[0097] Experimental Example 1: Physical micro-oscillations enhance the ultraviolet absorption of phenolic pollutants.

[0098] Specifically, ultraviolet absorption spectra were collected for low-concentration phenol water samples under undisturbed conditions and under conditions of low-amplitude mechanical oscillation (e.g., 20 Hz, micro-oscillation for 1 minute).

[0099] Figure 3 A schematic diagram of the simulated ultraviolet absorption spectrum response of an example of phenolic pollutants under physical micro-oscillations is shown.

[0100] Specifically, micro-oscillations can promote a more uniform distribution of phenol molecules and their brief exposure to the water surface, resulting in a short-term increase in absorption intensity at characteristic absorption wavelengths (e.g., 270 nm), while baseline and irrelevant components show minimal changes. For example... Figure 3 As shown, the yellow solid line is the UV absorption spectrum of the undisturbed phenol sample, where the characteristic absorption peak of phenol (approximately 270 nm) is relatively weak; while the orange solid line is the UV absorption spectrum acquired after applying micro-oscillation (e.g., 20 Hz, 1 minute), where the characteristic absorption peak is significantly enhanced; the dashed line represents the residual spectrum after "micro-oscillation - undisturbed", clearly showing the enhanced dynamic response at the characteristic wavelength of phenol.

[0101] Therefore, through Figure 3 It is known that physical micro-oscillations can effectively amplify the spectral signal of phenol molecules, while residual analysis significantly improves the identifiability and quantitative sensitivity of trace pollutants in complex water backgrounds.

[0102] Experimental Example 2: pH fine-tuning induces changes in the Raman signal of organic amines.

[0103] Specifically, Raman spectra were collected from water samples with low concentrations of organic amines at pH 7.0 and pH 8.5, respectively.

[0104] Figure 4 A schematic diagram of the simulation results of an example of the Raman spectral response of trace organic amine pollutants under different pH conditions is shown.

[0105] like Figure 4 As shown, the yellow curve represents the Raman spectrum of the water sample at pH 7.0, displaying characteristic Raman peaks (such as 1350 cm⁻¹). -1 and 1550cm -1 The signal is weak; the orange curve is the Raman spectrum of the same sample after pH adjustment to 8.5, showing a significant increase in the intensity of the main characteristic peak. The dashed line represents the difference between the two (i.e., the "residual spectrum"), producing a distinct positive peak at the characteristic peak position, highlighting the amplification effect of pH fine-tuning on the Raman response of the target pollutant. The gray vertical lines mark the shift positions of the characteristic peaks.

[0106] These results indicate that pH perturbation can induce changes in the Raman activity of target organic amine molecules, while residual analysis can significantly enhance the spectral identifiability and quantitative sensitivity of pollutants in complex backgrounds.

[0107] Figure 5 A simulation diagram illustrating an example of the fluorescence spectral response of trace polycyclic aromatic hydrocarbons under different ionic intensities is shown.

[0108] Specifically, the fluorescence intensity of polycyclic aromatic hydrocarbons (such as naphthalene) in water is easily affected by the ionic strength of the solution.

[0109] By slowly adding low concentrations of inorganic salts to water samples and collecting time-series changes in fluorescence spectra, it was found that the characteristic peaks of pollutants exhibited repeatable kinetic responses, while the fluorescence changes in the main background water (such as the solvent itself) were very weak. Figure 5 As shown, the yellow curve represents the fluorescence spectrum of the water sample under initial low ionic intensity, where 375 nm and 410 nm are typical characteristic peaks of polycyclic aromatic hydrocarbons. The orange curve represents the fluorescence spectrum of the same sample after the addition of a small amount of inorganic salt and the increase in ionic intensity; the characteristic peak intensities increase significantly, indicating that the molecular dynamics of the pollutant are excited by the perturbation. The dashed line represents the difference between the two (i.e., the residual spectrum), which forms obvious positive peaks at both characteristic wavelengths, clearly highlighting the fluorescence enhancement signal of the target pollutant under the perturbation response and greatly weakening the influence of background and other interfering components. The gray vertical lines mark the positions of the characteristic peaks.

[0110] Therefore, through Figure 5The results verified that ion intensity perturbation can effectively amplify the spectral response of polycyclic aromatic hydrocarbon pollutants, while residual analysis greatly improves its detection sensitivity and identification ability in complex aquatic environments.

[0111] It should be noted that in real-world water bodies, the signals of trace organic pollutants are often obscured by the main background, impurities, and instrument background noise. The characteristic peaks of the original spectrum are often very weak or even "covered" by the background. Directly analyzing the original spectrum not only has low sensitivity but is also prone to false positives or false negatives.

[0112] In the aforementioned simulation diagrams of perturbation experiments, an idealized clean background was chosen to visualize the "amplification" of the residual response. However, in actual aquatic environments and complex multi-component samples, the original spectral peaks are often chaotic, with peak shape shifts and characteristic peaks submerged, making direct detection often ineffective. In contrast, perturbation can actively excite the molecular dynamics of trace pollutants without damaging the sample structure, causing their response to be "amplified" in a specific wavelength range, while having minimal impact on the main background components. Furthermore, residual analysis (i.e., post-perturbation - pre-perturbation / control group) can effectively "subtract" most of the background, systematic errors, and environmental noise unrelated to the perturbation, retaining only the perturbation response signal itself. Thus, the dynamic response of pollutant molecules can be presented with a high signal-to-noise ratio even in high-noise and complex backgrounds, significantly reducing the false positive rate.

[0113] Figure 6 A flowchart illustrating an example of identifying the optimal response characteristic wavelength based on the perturbation response residual spectrum according to an embodiment of this application is shown.

[0114] like Figure 6 As shown, in step S610, within the characteristic response wavelength range Λ of the target trace organic pollutant... tgt Within this process, for each candidate response wavelength, the main peak response is calculated based on the perturbation response residual spectrum.

[0115] As described above, each type of pollutant has a corresponding characteristic response wavelength range Λ. tgt This can be obtained through literature data or historical measurements.

[0116] For example, Λ tgt ={λ1,λ2,…,λ n}, where λ1,λ2,…,λ n For each typical response wavelength of the target pollutant, the spectral characteristic regions of the target pollutant are precisely located.

[0117] Furthermore, within the characteristic response wavelength range, the response signal of the target pollutant exhibits a dominant peak signal. This is achieved by analyzing each candidate wavelength point λ. jThe maximum response is extracted to ensure that the most representative signal is selected as the basis for concentration quantification.

[0118] Specifically, the residual spectrum of the perturbation response ΔS(λ) acquired after the perturbation operation is applied. j ,t), for each wavelength λ j Perform main peak response extraction, for each wavelength λ j The response strength is determined by the maximum value, that is, the maximum response value within the disturbance time series is taken.

[0119] R(λ j ) = max t∈T △S(λ j ,t), Equation (7)

[0120] In the formula, △S(λ) j ,t) represents the candidate response wavelength λ in the residual spectrum of the perturbation response. j and the response intensity at time t of the disturbance, R(λ) j ) indicates that in λ j The main peak response at the location.

[0121] Therefore, by accurately extracting the main peak response, the response signal of the target pollutant is significantly enhanced, while eliminating interference from irrelevant signals and background noise.

[0122] In step S620, for each candidate response wavelength, the signal-to-noise ratio of the main peak response corresponding to the candidate response wavelength is calculated.

[0123] Signal-to-noise ratio (SNR) is a key metric for evaluating the quality of spectral response. It is calculated by measuring λ at each wavelength. j The signal-to-noise ratio (SNR) can measure the stability and noise level of a target signal.

[0124] Specifically, by applying each wavelength λ j The main peak response R(λ) j The standard deviation of the background noise σ bg (λ j The ratio is calculated to obtain the signal-to-noise ratio (SNR).

[0125] Background noise σ bg (λ j By calculating λ j Estimation of the spectral standard deviation for the non-response range of wavelength.

[0126]

[0127] In the formula, σ bg (λ j ) is in λ j The standard deviation of background noise at that location, SNR(λ) jR(λ) represents the peak response. j The signal-to-noise ratio.

[0128] The higher the signal-to-noise ratio, the stronger the stability of the response signal. By maximizing the signal-to-noise ratio, the response quality of pollutants can be optimized, thereby reducing the possibility of misjudgment.

[0129] In step S630, for each candidate response wavelength, the fingerprint correlation between the main peak response corresponding to the candidate response wavelength and the standard fingerprint template response is calculated.

[0130] It should be noted that the spectral response of each target pollutant has a specific "fingerprint" characteristic. By performing correlation analysis with known standard fingerprint templates, the presence and concentration of the target pollutant can be further confirmed.

[0131] Specifically, the Pearson correlation coefficient is used to analyze each wavelength λ. j The response strength and standard fingerprint template R ref (λ j Matching is performed to confirm the consistency of each wavelength response with the standard template, thereby improving the accuracy of the response signal.

[0132]

[0133] In the formula, R ref (λ j ) indicates that the standard fingerprint template is in λ j The response intensity at point C, where Cov represents the covariance, C(λ) j ) represents fingerprint correlation, σ R (λ j )and They are respectively at λ j The standard deviation of the main peak response and the standard deviation of the standard fingerprint template.

[0134] It should be noted that a standard fingerprint template refers to the baseline reference data on the distribution of the spectral main peak response with wavelength, collected and normalized under a set (or series) of specific experimental / standard conditions for a specific target pollutant. Through the standard fingerprint template, the "characteristic spectral response curve" or "fingerprint curve" of the pollutant at a specific response wavelength under "ideal pure environment" is expressed.

[0135] Therefore, by using correlation matching, we can ensure that the response signal matches the standard template in the database, and ensure that the selected wavelength can meet the standard response pattern of the target pollutant to the greatest extent, thereby further reducing the risk of false identification or missed detection.

[0136] In step S640, the comprehensive discrimination score is calculated.

[0137] Specifically, by comprehensively considering the signal-to-noise ratio and correlation, a comprehensive discrimination score is calculated for each wavelength, thereby comprehensively evaluating the representativeness of each wavelength's response to the target pollutant.

[0138] Specifically, corresponding weighting coefficients α and β are set for different indicators, and the sum of the weights should satisfy α + β = 1. Then, the signal-to-noise ratio and correlation are combined through weighted calculation to obtain a comprehensive discrimination score.

[0139] Q(λ j )=α·SNR(λ j )+β·C(λ j Equation (10)

[0140] In the formula, Q(λ) j ) indicates that in λ j The comprehensive discrimination score is given at the given point, where α and β represent the discrimination weighting weights.

[0141] In this way, by using a weighted scoring method, the signal-to-noise ratio and correlation can be comprehensively considered to ensure that the selected optimal response wavelength can maximize signal quality and detection accuracy.

[0142] In step S650, the wavelength point with the highest comprehensive discrimination score is selected as the optimal response feature wavelength λ. * .

[0143] Here, the most representative wavelength is selected from the candidate response wavelengths by comprehensively judging the scores.

[0144] Specifically, the comprehensive discrimination score Q(λ) for all candidate response wavelengths. j Compare the values ​​and select the wavelength λ corresponding to the maximum value. * As the optimal response characteristic wavelength.

[0145]

[0146] This application's embodiments employ an optimal response characteristic wavelength determination method based on the correlation weighting of the main peak response and the standard fingerprint template. This method enables multi-point matching between the perturbation response residual spectrum of the current sample and the calibrated target pollutant standard fingerprint template. By comprehensively considering the weights of signal-to-noise ratio and correlation, it ensures both the significance of the response signal and the physical specificity of the wavelength selection for the target pollutant. This significantly enhances the anti-interference capability and specificity of trace organic pollutant detection in complex water environments, reducing the probability of false positives and component overlap interference.

[0147] Regarding the selection of the concentration correction model function f(·) in the embodiments of this application, it should be noted that traditional linear or multiple linear regression models can only fit simple ideal scenarios where "the response changes linearly with concentration." In real aquatic environments, they are easily affected by systematic errors, interaction effects, and nonlinear background disturbances, resulting in large errors and poor adaptability. When the relationship between disturbances and environmental variables and concentration exhibits exponential, saturation, threshold, or other complex nonlinear changes, linear models are unable to fit the data and may even provide completely incorrect concentration predictions.

[0148] In contrast, kernel mapping functions (such as Gaussian RBF kernels, Sigmoid kernels, neural networks, etc.) can effectively map the high-dimensional, complex, nonlinear relationships of input variables (normalized peak response, perturbation parameters, environmental parameters, etc.) to a "space that is easier to linearly partition," automatically capturing the implicit coupling relationships and dynamic characteristics between variables. Furthermore, the model can not only characterize complex responses in one go, but also continuously adapt to changes in the field through incremental learning (fine-tuning parameters), automatically correcting the effects of long-term system drift, background changes, and new perturbation modes, thus improving the reliability and applicability of long-term operation.

[0149] In some examples of embodiments of this application, the concentration correction model function adopts a nonlinear kernel mapping model based on Gaussian radial basis functions. More specific advantages compared to other kernel functions will be elaborated in other parts below in conjunction with the experimental section.

[0150] Specifically, the spectral intensity values ​​of the residual spectrum of the perturbation response within the optimal response characteristic wavelength and the perturbation time interval are normalized.

[0151] Here, the residual spectrum of the disturbance response directly reflects the dynamic spectral changes of the water sample under micro-disturbance, but the absolute spectral intensity is easily affected by multiple sources such as sample background, instrument sensitivity, and system noise. By normalizing the residual spectrum at each disturbance time t, sample background, instrument drift, and batch effects are eliminated.

[0152] For each sample, at the optimal response characteristic wavelength λ * Within the time interval of the disturbance, the residual spectrum of the response to the disturbance ΔS(λ) * ,t) is normalized according to the following formula:

[0153]

[0154] In the formula, △S norm (λ * ,t) represents the optimal response characteristic wavelength λ * The normalized spectral intensity of the residual spectrum of the perturbation response at perturbation time t; μ bg (λ * ) and σ bg (λ *) respectively represent at λ * The mean and standard deviation of all spectral intensity values ​​before the perturbation is applied.

[0155] The disturbance operation parameters and environmental parameters are normalized separately.

[0156] Similarly, perturbation operation parameters (such as amplitude, pH adjustment range, etc.) and environmental parameters (such as temperature, conductivity, etc.) have different physical dimensions and statistical distributions. Normalizing them can avoid the imbalance of feature importance and the instability of optimization values ​​during subsequent modeling.

[0157] Specifically, for the disturbance parameter A perturb and environmental parameter E env Perform mean-standard deviation normalization separately:

[0158]

[0159] In the formula, μ A and σ A Let A represent the mean and standard deviation of the disturbance parameter, respectively. norm E represents the normalized perturbation operation parameter vector; norm μ represents the normalized environment parameter vector. E and σ E These represent the mean and standard deviation of the environmental parameters, respectively.

[0160] The above normalization operation ensures that all input features are of the same dimension and numerical scale, providing a balanced feature space input for Gaussian kernel mapping.

[0161] The normalized features are concatenated to construct the input vector x:

[0162] x=[△S norm (λ * ,t),A norm E norm Equation (14)

[0163] A nonlinear kernel mapping model based on Gaussian radial basis kernel function is used to process the input vector in order to predict the concentration of the target trace organic pollutant.

[0164]

[0165] In the formula, K represents the number of hidden nodes in the Gaussian radial basis function kernel, and w k Let b represent the weight parameters of the k-th hidden node obtained through pre-training, and c represent the bias term obtained through pre-training. k Let σ represent the center vector of the k-th hidden node. k The width parameter of the k-th Gaussian radial basis kernel is ‖xc k‖ represents the Euclidean distance between the input feature and the center of the k-th hidden node.

[0166] Here, by automatically learning the high-order nonlinear coupling mapping between input features through the Gaussian Radial Basis Function (RBF), the complex influence mechanism of perturbation signals, perturbation parameters, and environmental parameters on the response of trace organic pollutants can be adaptively characterized, thereby improving the quantitative accuracy and system robustness under low concentration and dynamic environments.

[0167] The Gaussian kernel forms a "local peak" for each sample point in the input variable space, meaning the model's influence on each sample decreases exponentially with increasing distance from the input variables. It is suitable for modeling the influence of perturbation responses and changes in environmental parameters on the main peak response (i.e., there are local differences in the perturbation responses among water samples rather than a globally linear pattern). Furthermore, the Gaussian kernel can map the original finite-dimensional space to an infinite-dimensional high-dimensional feature space, effectively capturing the nonlinear relationship between target concentration and multiple factors such as perturbation parameters and environmental parameters, involving higher-order coupling.

[0168] Through the embodiments of this application, the system integrates the normalization processing of the residual spectrum of the disturbance response, the normalization of multi-source parameters and feature stitching, and uses a nonlinear mapping model with Gaussian radial basis kernel to predict the concentration of the normalized feature vector. This realizes the automatic modeling and engineering application of the complex nonlinear coupling relationship between the disturbance response signal, the disturbance operation parameters and the environmental background parameters, which significantly improves the model's adaptability to dynamic disturbances and the quantitative accuracy of complex environmental changes.

[0169] To verify the effectiveness of the nonlinear concentration correction model based on Gaussian RBF kernel proposed in this application for the detection of trace organic pollutants in water, several water samples of different types were selected. For each sample, spectral intensity time-series data (ΔS(λ)) at its optimal response characteristic wavelength were collected. * ,t)), disturbance operating parameters (A perturb The data includes environmental parameters (such as water temperature, pH, conductivity, etc., which are uniformly normalized) and the corresponding known real pollutant concentrations. The data sample covers the range of low concentration (0.01 mg / L) to high concentration (1.0 mg / L), and has high representativeness and engineering authenticity.

[0170] This experiment used Gaussian RBF kernel support vector regression (SVR), multinomial kernel SVR, Sigmoid kernel SVR, and linear kernel SVR to model the data. All models used ΔS(λ) as the basis for the model. * ,t)A perturb The system takes environmental parameters as joint input and pollutant concentration as output. Each model automatically fine-tunes its parameters through grid search and cross-validation to ensure optimal model performance.

[0171] To comprehensively evaluate model performance, the following three metrics are primarily used:

[0172] Goodness of fit R 2 : Measures the explanatory power of the predicted results on the actual concentration;

[0173] Root mean square error (RMSE): The average error between the predicted value and the true value of a measurement;

[0174] Anomaly detection rate: The percentage of samples with an absolute prediction error greater than 0.1 mg / L.

[0175] Table 1. Comparative Experimental Results

[0176]

[0177]

[0178] As shown in Table 1, the Gaussian RBF kernel model achieved the highest goodness of fit and the lowest root mean square error on the test set, with a significantly lower outlier detection rate compared to other kernel functions. While the polynomial kernel possesses some high-order fitting capability, it exhibits slight overfitting as the variable dimension increases, and its generalization ability is inferior to the Gaussian RBF kernel. The Sigmoid and linear kernels, however, fail to effectively capture the nonlinear relationships under complex perturbations and environmental variables, resulting in higher prediction errors and outlier detection rates.

[0179] Figure 7 A simulation diagram showing the comparison of prediction error distributions for an example of a trace organic pollutant concentration prediction experiment using different kernel function models is presented.

[0180] like Figure 7 The results show that the Gaussian RBF kernel model exhibits the most concentrated residual distribution and the smallest fluctuation amplitude across the entire concentration range, with errors consistently close to zero, significantly outperforming the polynomial, sigmoid, and linear kernel models. Other kernel models show greater error fluctuations and outliers across different concentration ranges, demonstrating lower fitting accuracy and engineering stability. Therefore, this verifies the high accuracy and robustness of the Gaussian RBF kernel nonlinear correction model in predicting trace organic pollutant concentrations in complex aquatic environments.

[0181] The following section will continue to detail exemplary implementations of training the concentration correction model.

[0182] Specifically, multiple target trace organic pollutant standard samples with known concentrations are prepared, and the residual spectral data of the perturbation response at the optimal response characteristic wavelength of each standard sample under perturbation operation, as well as the corresponding perturbation operation parameters and environmental parameters, are collected to form a training sample set.

[0183] For example, under laboratory conditions, prepare 10-20 sets of standard water samples with known concentrations, covering the full range of the detection system (e.g., 0.01-1.0 mg / L).

[0184] For each group of water samples, data were collected under uniform perturbation operating parameters (such as oscillation amplitude, pH adjustment range, and light cycle) and environmental parameters (such as temperature and conductivity).

[0185] The collected data includes: the optimal response characteristic wavelength λ for each group of standard samples under perturbation operation. * Perturbation response residual spectrum at Disturbance parameters Environmental parameters and the actual concentration of this group

[0186] The perturbation response residual spectral data are normalized based on the optimal response characteristic wavelength and perturbation time series interval in the training samples, and the principal component analysis method is used to reduce the dimensionality of the normalized time series spectral intensity vector to extract the principal component features.

[0187] Here, the temporal residual spectrum strength between different samples is standardized and principal components are extracted. This not only eliminates non-target factors such as instrument and sample background, but also effectively compresses high-dimensional features, thereby improving model generalization and computational efficiency.

[0188] Specifically, first analyze the residual spectra of the perturbation response for each group of training samples. Normalization:

[0189]

[0190] In the formula, This represents the i-th training sample, with its optimal response characteristic wavelength λ. * and the normalized perturbation response residual spectral sequence over the perturbation time interval t, and These represent all training samples in the background period (or without perturbation) at λ. * The mean and standard deviation are given.

[0191] Normalized spectral intensity sequence for all training samples Construct the matrix using principal component analysis (PCA):

[0192]

[0193] In the formula, Z (train) Let N represent the principal component feature matrix of all training samples, and let N represent the total number of training samples.

[0194] Retain the first d principal components (e.g., 2 or 3), and each group of samples corresponds to the principal component features.

[0195] This effectively removes non-target variables such as noise, background, and instrument drift, and extracts the most core dynamic features that characterize changes in the disturbance response.

[0196] The perturbation operation parameters and environmental parameters in the training samples are normalized.

[0197] Specifically, mean-standard deviation normalization is performed on the perturbation operation parameters and each dimension of the environmental parameters:

[0198]

[0199] in, This represents the normalized value of the perturbation operation parameter for the i-th training sample. and Let represent the mean and standard deviation of the perturbation operation parameters for all training samples, respectively. This represents the normalized value of the environmental parameters (such as temperature, conductivity, pH, etc.) of the i-th training sample. and These represent the mean or standard deviation of all training samples for that environmental parameter component. This ensures that the weights of each input feature are balanced during subsequent kernel function mapping, improving the model's convergence speed and generalization ability.

[0200] For each training sample, the principal component features, normalized perturbation operation parameters, and normalized environment parameters corresponding to the training sample are concatenated into a training input feature vector.

[0201] The K-means clustering algorithm is used to cluster all training input feature vectors to determine the hidden node centers and kernel width parameters of the Gaussian radial basis kernel function.

[0202] Here, features from different sources are fused into a unified vector, and K-means clustering is used to initialize the Gaussian kernel center and kernel width to enhance the model's adaptability to data distribution. Specifically, for each training sample, principal component features, normalized perturbation parameters, and normalized environmental parameters are concatenated to obtain the feature vector:

[0203]

[0204] Use the K-means clustering algorithm to cluster all Clustering, selecting K cluster centers c k As the center of the RBF hidden node, the kernel width σ k Let be the average Euclidean distance between all cluster centers. Therefore, by adaptively selecting the kernel function center and bandwidth using the spatial distribution of training samples, the expressive power and stability of the model under complex data structures are improved.

[0205] Based on the training input feature vector and the corresponding known concentration, the weights and bias parameters of the Gaussian radial basis kernel function are jointly updated by the least squares method and the Adam optimization algorithm, with the goal of minimizing the mean square error loss of the training samples.

[0206] More specifically, normalized features, hidden node centers, and bandwidth are used to achieve a nonlinear mapping from input to concentration output through joint optimization of weights and biases.

[0207] Based on all training samples, the following loss function is constructed to train and optimize the nonlinear kernel mapping model of the Gaussian radial basis kernel function constructed as shown in equation (15):

[0208]

[0209] The least squares method combined with the Adam optimization algorithm is used to jointly train the weights {w}. k The model is adjusted by setting a bias b until the loss function converges, enabling it to automatically adapt to the nonlinear coupling between high-dimensional principal component features and perturbations and environmental variables, thus achieving high-precision concentration inversion.

[0210] Through the embodiments of this application, K-means clustering is used to adaptively determine the center and bandwidth of the Gaussian kernel function, so that the subsequent optimization of model parameters is based on the true characterization of the essential structure of the input feature space. This achieves adaptive initialization of the Gaussian radial basis kernel structure, enabling the concentration correction model to automatically capture the high-dimensional nonlinear coupling relationship between complex water disturbance signals, environmental variables and pollutant concentrations, thereby significantly improving the modeling accuracy and generalization ability of the model.

[0211] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of combined actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Secondly, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application. In the above embodiments, the descriptions of each embodiment have their own emphasis; for parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0212] Figure 8 A structural block diagram of an example of a water trace organic pollutant detection system according to an embodiment of this application is shown.

[0213] like Figure 8 As shown, the water trace organic pollutant detection system 800 includes a water sample collection unit 810, a sample micro-disturbance unit 820, an optimal wavelength identification unit 830, and a trace concentration identification unit 840.

[0214] The water sample acquisition unit 810 is used to collect environmental parameters of the target water sample to be tested and to obtain the baseline spectral data of the target water sample.

[0215] The sample micro-perturbation unit 820 is used to determine the perturbation operation parameters that match the target water sample, and to apply micro-perturbation to the target water sample according to the perturbation operation parameters to obtain perturbation response spectral data; the perturbation operation parameters include perturbation operation type, perturbation operation amplitude and perturbation operation duration.

[0216] The optimal wavelength identification unit 830 is used to perform time-series difference analysis on the perturbation response spectral data and the baseline spectral data to obtain the perturbation response residual spectrum, and to identify the optimal response characteristic wavelength from the characteristic response wavelength range of the target trace organic pollutant based on the perturbation response residual spectrum.

[0217] The trace concentration identification unit 840 is used to solve the concentration correction model based on the optimal response characteristic wavelength, the perturbation operation parameters and the environmental parameters to obtain the concentration of the target trace organic pollutant.

[0218] The concentration correction model is expressed by the following equation:

[0219] C target =f(△S(λ) * ,t),A perturb E env ),

[0220] In the formula, C target Let T be the concentration of the target trace organic pollutant, and ΔS(λ) be the time interval of the perturbation. * ,t) represents the residual spectrum of the perturbation response at the optimal response characteristic wavelength λ. * and the response intensity at time t of the disturbance, A perturb For the disturbance operation parameters, E env Here, f(·) represents the environmental parameters, and f(·) represents the pre-trained concentration correction model function.

[0221] In some embodiments, this application provides a non-volatile computer-readable storage medium storing one or more programs including execution instructions, which can be read and executed by electronic devices (including but not limited to computers, servers, or network devices) to perform the steps of any of the above-described methods for detecting trace organic pollutants in water.

[0222] In some embodiments, this application also provides a computer program product, the computer program product including a computer program stored on a non-volatile computer-readable storage medium, the computer program including program instructions, which, when executed by a computer, cause the computer to perform the steps of any of the above-described methods for detecting trace organic pollutants in water.

[0223] In some embodiments, this application also provides an electronic device comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the steps of a method for detecting trace organic pollutants in water.

[0224] The above-described product can perform the methods provided in the embodiments of this application, and has the corresponding functional modules and beneficial effects for performing the methods. Technical details not described in detail in this embodiment can be found in the methods provided in the embodiments of this application.

[0225] The electronic devices in this application can exist in various forms, including but not limited to: mobile communication devices, ultra-mobile personal computer devices, portable entertainment devices, or other airborne electronic devices with data interaction functions.

[0226] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0227] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented using software plus a general-purpose hardware platform, or of course, using hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0228] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for detecting trace organic pollutants in water, characterized in that, The method includes: Collect environmental parameters of the target water sample to be tested, and obtain the baseline spectral data of the target water sample; Determine perturbation operation parameters that match the target water sample, and apply micro-perturbation to the target water sample according to the perturbation operation parameters to obtain perturbation response spectral data; the perturbation operation parameters include perturbation operation type, perturbation operation amplitude, and perturbation operation duration; the perturbation operation type includes at least one of the following: physical micro-oscillation, pH fine-tuning, ion strength micro-change, or light intensity periodic fluctuation; The perturbation response residual spectrum is obtained by performing time-series difference analysis on the perturbation response spectral data and the baseline spectral data, and the optimal response characteristic wavelength is identified from the characteristic response wavelength range of the target trace organic pollutant based on the perturbation response residual spectrum. The step of identifying the optimal response characteristic wavelength from the characteristic response wavelength range of the target trace organic pollutant based on the perturbation response residual spectrum includes: The characteristic response wavelength range of the target trace organic pollutant Inside, ,in For each typical response wavelength of the target pollutant, for each candidate response wavelength... , The main peak response is calculated based on the residual spectrum of the perturbation response: ， In the formula, To find the candidate response wavelength in the residual spectrum of the perturbation response and disturbance time The response intensity at that point , For the time interval of the disturbance; Indicates in The main peak response at the location; For each candidate response wavelength, calculate the signal-to-noise ratio of the main peak response corresponding to that candidate response wavelength: ， In the formula, In order to be in The standard deviation of background noise at that location Indicates the main peak response Signal-to-noise ratio; For each candidate response wavelength, calculate the fingerprint correlation between the main peak response corresponding to that candidate response wavelength and the standard fingerprint template response: ， In the formula, Indicates standard fingerprint template in The response intensity at that point Describing covariance, Indicates fingerprint correlation. and They are respectively in The standard deviation of the main peak response and the standard deviation of the standard fingerprint template; Calculate the overall discrimination score: ， In the formula, Indicates in The overall judgment score at the point, and This indicates the determination of weighted weights, and ; The wavelength point with the highest comprehensive discrimination score is selected as the optimal response feature wavelength. : ； The concentration correction model is solved based on the optimal response characteristic wavelength, the perturbation operation parameters, and the environmental parameters to obtain the concentration of the target trace organic pollutant; The concentration correction model is expressed by the following equation: ， In the formula, The target is the concentration of trace organic pollutants. To obtain the residual spectrum of the perturbation response at the optimal response characteristic wavelength and the timing of the disturbance The response intensity at that point These are the parameters for disturbance operation. For environmental parameters, This is the function for the pre-trained concentration correction model.

2. The method according to claim 1, characterized in that, The determination of the perturbation operation parameters matching the target water sample includes: Based on multiple calibration perturbation operation parameters in the calibration parameter group, a leading perturbation operation is applied to the target water sample respectively; each calibration perturbation operation parameter is predefined with a corresponding perturbation operation type, perturbation operation amplitude and perturbation operation duration; For each calibration perturbation operation parameter, the perturbation response residual spectrum corresponding to the calibration perturbation operation parameter is obtained, and the maximum target response of the perturbation response residual spectrum is determined in the characteristic wavelength range and perturbation time range of the target trace organic pollutant, and the maximum background response is determined in the background wavelength range and perturbation time range unrelated to the target trace organic pollutant. Calculate the ratio of the maximum target response to the maximum background response for each calibrated perturbation operation parameter to obtain the target correlation score; The perturbation operation parameters with the highest correlation scores to the target are selected as the perturbation operation parameters that match the target water sample.

3. The method according to claim 2, characterized in that, The determination of the calibration parameter set includes: Obtain the target pollutant type of the target trace organic pollutant; A set of calibration parameters matching the target pollutant type is determined from a preset set of pollutant-calibration parameters; the set of pollutant-calibration parameters pre-stores multiple pollutant types and corresponding sets of calibration parameters.

4. The method according to claim 3, characterized in that, The concentration correction model function adopts a nonlinear kernel mapping model based on Gaussian radial basis functions; The step of solving the concentration correction model based on the optimal response characteristic wavelength, the perturbation operating parameters, and the environmental parameters to obtain the concentration of the target trace organic pollutant includes: The spectral intensity values ​​of the perturbation response residual spectrum within the optimal response characteristic wavelength and perturbation time interval are normalized. ， In the formula, Indicates the optimal response characteristic wavelength and disturbance time The normalized spectral intensity value of the residual spectrum of the perturbation response under the given conditions; and They represent in The mean and standard deviation of all spectral intensity values ​​before the perturbation was applied; The disturbance operation parameters and environmental parameters are normalized separately: ， In the formula, and Let represent the mean and standard deviation of the disturbance parameter, respectively. This represents the normalized perturbation operation parameter vector; This represents the normalized environment parameter vector. and These represent the mean and standard deviation of the environmental parameter, respectively. The normalized features are concatenated to construct the input vector. : ， The input vector is processed using a nonlinear kernel mapping model based on Gaussian radial basis kernel function to predict the concentration of the target trace organic pollutant; ， In the formula, This represents the number of hidden nodes in the Gaussian radial basis function kernel. Indicates the first obtained from pre-training The weight parameters of each hidden node. This represents the bias term obtained during pre-training. Indicates the first The center vector of each hidden node Indicates the first The width parameter of a Gaussian radial basis kernel. Indicates the input features and the first The Euclidean distance between the centers of the hidden nodes.

5. The method according to claim 4, characterized in that, Training the concentration correction model includes: Multiple target trace organic pollutant standard samples with known concentrations were prepared. The residual spectral data of the perturbation response at the optimal response characteristic wavelength of each standard sample under perturbation operation, as well as the corresponding perturbation operation parameters and environmental parameters, were collected to form a training sample set. Based on the optimal response feature wavelength and perturbation time series interval in the training samples, the perturbation response residual spectral data is normalized, and the principal component analysis method is used to reduce the dimensionality of the normalized time series spectral intensity vector to extract the principal component features. The perturbation operation parameters and environmental parameters in the training samples are normalized. For each training sample, the principal component features, normalized perturbation operation parameters, and normalized environment parameters corresponding to the training sample are concatenated into a training input feature vector. The K-means clustering algorithm is used to cluster all training input feature vectors to determine the hidden node centers and kernel width parameters of the Gaussian radial basis kernel function. Based on the training input feature vector and the corresponding known concentration, the weights and bias parameters of the Gaussian radial basis kernel function are jointly updated by the least squares method and the Adam optimization algorithm, with the goal of minimizing the mean square error loss of the training samples.

6. A water trace organic pollutant detection system, used to implement the method as described in any one of claims 1-5; characterized in that, The system includes: The water sample collection unit is used to collect environmental parameters of the target water sample to be tested and to obtain the baseline spectral data of the target water sample. The sample micro-perturbation unit is used to determine perturbation operation parameters that match the target water sample, and to apply micro-perturbation to the target water sample according to the perturbation operation parameters to obtain perturbation response spectral data; the perturbation operation parameters include perturbation operation type, perturbation operation amplitude, and perturbation operation duration; The optimal wavelength identification unit is used to perform time-series difference analysis on the perturbation response spectral data and the baseline spectral data to obtain the perturbation response residual spectrum, and to identify the optimal response characteristic wavelength from the characteristic response wavelength range of the target trace organic pollutant based on the perturbation response residual spectrum. The trace concentration identification unit is used to solve the concentration correction model based on the optimal response characteristic wavelength, the perturbation operation parameters and the environmental parameters to obtain the concentration of the target trace organic pollutant; The concentration correction model is expressed by the following equation: ， In the formula, The target is the concentration of trace organic pollutants. To obtain the residual spectrum of the perturbation response at the optimal response characteristic wavelength and the timing of the disturbance The response intensity at that point These are the parameters for disturbance operation. For environmental parameters, This is the function for the pre-trained concentration correction model.

Images