Method and device for identifying complex components of road runoff based on single-port wire netting

Abstract

The application discloses a single-port vector network-based road surface runoff complex component identification method and device, and the method comprises the following steps: collecting a road surface runoff sample, and measuring electromagnetic response parameters of the sample by using a single-port vector network analyzer; performing wavelet denoising and fast Fourier transform on the electromagnetic response parameters, extracting multi-frequency band characteristic peak values, and forming a characteristic vector; performing component identification according to the characteristic vector by using a microwave pyrolysis model, and obtaining an identification result; wherein, a pyrolysis process of the microwave pyrolysis model comprises a fast volatilization stage and a slow carbonization stage, the dielectric loss factor is substituted into a volume heating equation in real time in the component identification process, the simulated temperature field of the fast volatilization stage and the slow carbonization stage is dynamically updated, and the microwave pyrolysis model parameters are optimized by using a simulated annealing algorithm; and according to the identification result, an analysis report of the road surface runoff complex component is obtained. The application improves the identification precision and enhances the real-time monitoring capability.

Description

Technical Field

[0001] This application relates to the field of environmental monitoring technology, and in particular to a method and apparatus for identifying complex components of road runoff based on a single-port vector network. Background Technology

[0002] Road runoff, a major source of urban non-point source pollution, contains complex components such as organic pollutants and heavy metal compounds, posing a serious threat to water bodies and the ecological environment. Existing technologies for identifying road runoff components mainly include optical sensors, acoustic vector sensors, and deep neural network analysis. For example, acoustic vector sensors are used to detect road surface water layers, or complex network analysis is used to estimate runoff pollution under the influence of traffic events. However, these methods often rely on image processing or indirect measurement, resulting in poor real-time performance, low accuracy, and insufficient identification of complex components.

[0003] Therefore, existing technologies cannot achieve dynamic and accurate identification of complex components in road runoff. This invention aims to solve the above problems by providing an efficient, non-contact identification method. Summary of the Invention

[0004] This application provides a method and device for identifying complex components of road runoff based on a single-port vector network, which improves the identification accuracy and enhances the real-time monitoring capability.

[0005] This application provides the following solution:

[0006] According to a first aspect, a method for identifying complex components in road runoff based on a single-port vector network analyzer is provided. The method includes: collecting road runoff samples and measuring the electromagnetic response parameters of the samples using a single-port vector network analyzer, the electromagnetic response parameters including reflection coefficient and dielectric loss factor; performing wavelet denoising and fast Fourier transform on the electromagnetic response parameters to extract multi-band characteristic peaks, forming a feature vector containing organic matter and heavy metal complexes; using a microwave pyrolysis model to identify components based on the feature vector, obtaining identification results; wherein the pyrolysis process of the microwave pyrolysis model includes a rapid volatilization stage and a slow carbonization stage, and during the component identification process, the dielectric loss factor is substituted into the volume heating equation in real time to dynamically update the simulated temperature field of the rapid volatilization stage and the slow carbonization stage, and the microwave pyrolysis model parameters are optimized using a simulated annealing algorithm; and obtaining an analysis report of complex components in road runoff based on the identification results.

[0007] According to one achievable method in this application embodiment, performing wavelet denoising and fast Fourier transform on the electromagnetic response parameters to extract multi-band characteristic peaks and form a feature vector containing organic matter and heavy metal complexes includes: applying wavelet transform to perform multi-level decomposition of the electromagnetic response parameters, removing high-frequency noise components and reconstructing a clean signal; performing fast Fourier transform on the denoised signal to convert the time-domain data into a frequency-domain spectrum and identifying the dominant frequency components; dividing the frequency-domain spectrum into low-frequency and mid-frequency ranges, and extracting the peak amplitude, frequency position, and width of each frequency band as feature parameters; based on the extracted feature parameters, constructing a multi-dimensional feature vector, and mapping it to the fingerprint pattern of organic matter and heavy metal complexes through a clustering algorithm.

[0008] According to one achievable method in the embodiments of this application, the implementation method of the microwave pyrolysis model includes: initializing dual-distribution activation energy parameters based on the electromagnetic response parameters of the road runoff sample, including a low activation energy distribution in the rapid volatilization stage and a high activation energy distribution in the slow carbonization stage; constructing a reaction kinetic equation, dividing the pyrolysis process into a rapid volatilization stage and a slow carbonization stage, and incorporating microwave power density as a variable into the reaction kinetic equation to simulate its influence on temperature change.

[0009] According to one achievable method in the embodiments of this application, during the component identification process, the dielectric loss factor measured by the vector network analyzer is substituted into the volume heating equation in real time to dynamically update the temperature field of the rapid volatilization stage and the slow carbonization stage. This includes: collecting dielectric loss factor data of road runoff samples at preset time intervals using a single-port vector network analyzer, and simultaneously recording the current microwave power density and sample density parameters; substituting the dielectric loss factor into the volume heating equation to calculate the instantaneous heating rate and temperature gradient caused by microwave absorption, wherein the volume heating equation considers the product relationship between the dielectric loss factor and the microwave power density, and outputs the local temperature increment and gradient vector; based on the calculation results, dynamically adjusting the temperature field distribution of the rapid volatilization stage, including mapping the instantaneous heating rate to the volatile matter release area according to the low activation energy distribution threshold to achieve gridded iterative updates of the temperature field; and simultaneously adjusting the temperature field distribution of the slow carbonization stage, including applying a high activation energy distribution model to integrate the temperature gradient into the residue stabilization path and optimize the thermal balance of the gradual carbonization process.

[0010] According to one achievable method in the embodiments of this application, the optimization of the microwave pyrolysis model parameters using the simulated annealing algorithm includes: initializing the parameter vector, including the mean activation energy, standard deviation, microwave power density, and residence time, and setting an initial temperature; generating a new state with Gaussian-Cauchy mixed perturbation, calculating the composite objective function value, including identification error and energy consumption, and applying the acceptance criterion: if the new state is better than the current one, it is accepted; otherwise, it is accepted with exponential probability; performing adaptive reheating, and if there is no continuous improvement, temporarily increasing the temperature to enhance exploration; using exponential cooling to reduce the temperature until the termination condition is met, and outputting the optimized parameters.

[0011] According to one of the embodiments of this application, the microwave pyrolysis model further includes: using an electromagnetic-thermal-mass three-field coupled finite element mesh to periodically refresh the hot spot distribution and pollutant diffusion cloud map within the microwave cavity.

[0012] According to one achievable method in the embodiments of this application, obtaining an analysis report of complex components of road runoff based on the identification results includes: summarizing the identification results, which include the types, concentrations, and confidence intervals of organic pollutants and heavy metal compounds; calculating the risk index of each component and assessing the potential environmental pollution impact based on concentration thresholds; generating visualization elements to display the component distribution and spatiotemporal variation trends, and integrating metadata, which includes sample collection time, location, and measurement conditions, to form a structured report document.

[0013] According to a second aspect, a device for identifying complex components of road runoff based on a single-port vector network analyzer is provided. The device includes: a road data acquisition unit configured to acquire road runoff samples and measure the electromagnetic response parameters of the samples using a single-port vector network analyzer, the electromagnetic response parameters including reflection coefficient and dielectric loss factor; a data preprocessing unit configured to perform wavelet denoising and fast Fourier transform on the electromagnetic response parameters, extracting multi-band characteristic peaks to form a feature vector containing organic matter and heavy metal complexes; a component identification unit configured to use a microwave pyrolysis model to identify components based on the feature vector and obtain identification results; wherein the pyrolysis process of the microwave pyrolysis model includes a rapid volatilization stage and a slow carbonization stage, and during the component identification process, the dielectric loss factor is substituted into the volume heating equation in real time to dynamically update the simulated temperature field of the rapid volatilization stage and the slow carbonization stage, and the microwave pyrolysis model parameters are optimized using a simulated annealing algorithm; and an analysis report generation unit configured to obtain an analysis report of complex road runoff components based on the identification results.

[0014] According to a third aspect, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0015] According to a fourth aspect, an electronic device is provided, comprising: one or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any one of the first aspects.

[0016] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0017] This application achieves non-contact data acquisition by collecting road runoff samples and measuring electromagnetic response parameters such as reflection coefficient and dielectric loss factor, avoiding the destructiveness and complexity of traditional methods. Subsequently, wavelet denoising and fast Fourier transform are applied to the electromagnetic response parameters to extract multi-band characteristic peaks, forming feature vectors for organic matter and heavy metal ions, thus improving the accuracy and robustness of component identification. The core of this method lies in using a microwave pyrolysis model for simulated identification, dividing the pyrolysis process into a rapid volatilization stage and a slow carbonization stage. The dielectric loss factor is dynamically updated in real-time by incorporating it into the volume heating equation, and model parameters, such as activation energy distribution and microwave power density, are optimized through a simulated annealing algorithm to ensure the adaptability and efficiency of the identification process. This method significantly improves identification accuracy, enhances real-time monitoring capabilities, is suitable for pollutant analysis in urban drainage systems, reduces energy consumption and environmental interference, and provides an innovative solution for environmental protection.

[0018] Of course, any product implementing this application does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart of a method for identifying complex components of road runoff based on a single-port vector network analysis provided in this application embodiment;

[0021] Figure 2 A structural block diagram of a road runoff complex component identification device based on a single-port vector network provided in this application embodiment;

[0022] Figure 3 A schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0023] 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, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0024] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0025] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0026] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0027] Figure 1 A flowchart illustrating the method for identifying complex components of road runoff based on a single-port vector network analysis (VNA) provided in this application embodiment. Figure 1 As shown, the method may include the following steps:

[0028] Step 101: Collect road runoff samples and use a single-port vector network analyzer to measure the electromagnetic response parameters of the samples, including the reflection coefficient and dielectric loss factor.

[0029] Step 102: Perform wavelet denoising and fast Fourier transform on the electromagnetic response parameters to extract multi-band characteristic peaks and form a feature vector containing organic matter and heavy metal complexes.

[0030] Step 103: Using a microwave pyrolysis model, component identification is performed based on the feature vector to obtain identification results; wherein, the pyrolysis process of the microwave pyrolysis model includes a rapid volatilization stage and a slow carbonization stage. During the component identification process, the dielectric loss factor is substituted into the volume heating equation in real time to dynamically update the simulated temperature field of the rapid volatilization stage and the slow carbonization stage, and the microwave pyrolysis model parameters are optimized using a simulated annealing algorithm.

[0031] Step 104: Based on the identification results, obtain an analysis report of the complex components of road runoff.

[0032] As can be seen from the above process, this application achieves non-contact data acquisition by collecting road runoff samples and measuring electromagnetic response parameters such as reflection coefficient and dielectric loss factor, avoiding the destructiveness and complexity of traditional methods. Subsequently, wavelet denoising and fast Fourier transform are performed on the electromagnetic response parameters to extract multi-band characteristic peaks, forming feature vectors for organic matter and heavy metal ions, thus improving the accuracy and robustness of component characteristics. The core lies in using a microwave pyrolysis model for simulated identification, dividing the pyrolysis process into a rapid volatilization stage and a slow carbonization stage. The dielectric loss factor is substituted into the volume heating equation in real time to dynamically update the simulated temperature field, and the model parameters, such as activation energy distribution and microwave power density, are optimized through a simulated annealing algorithm to ensure the adaptability and efficiency of the identification process. This method significantly improves identification accuracy, enhances real-time monitoring capabilities, is suitable for pollutant analysis in urban drainage systems, reduces energy consumption and environmental interference, and provides an innovative solution for environmental protection.

[0033] First, the above step 101, namely "collecting road runoff samples and measuring the electromagnetic response parameters of the samples using a single-port vector network analyzer, wherein the electromagnetic response parameters include the reflection coefficient and the dielectric loss factor", will be described in detail with reference to the embodiments.

[0034] This step begins with collecting pollutant-laden stormwater runoff samples from typical areas such as urban roads, parking lots, or drainage outlets. Sample collection is typically conducted immediately after a rainfall event to ensure that pollutant concentrations reflect actual road erosion conditions. Standard sampling bottles or automated samplers can be used, with sample volumes generally between 500 ml and 2 liters to meet subsequent measurement requirements. After collection, samples should be transferred to the measuring device as quickly as possible to prevent prolonged storage that could lead to component volatilization or degradation, thus ensuring the accuracy of the electromagnetic response data.

[0035] The core feature of this technology is the use of a single-port vector network analyzer to measure the sample. A single-port vector network analyzer is a high-precision radio frequency instrument that characterizes the electromagnetic properties of materials by emitting microwave signals and receiving reflected signals. In this invention, the instrument's operating frequency range is typically set to 1 GHz to 10 GHz, a frequency band that can effectively excite the differences in dielectric response of different pollutants in road runoff. During measurement, the sample is placed in a dedicated test fixture or coaxial probe, with the probe in full contact with the sample to ensure that the microwave signal penetrates liquid or suspended pollutants. The instrument directly acquires the amplitude and phase information of the reflected signal through a single-port S-parameter measurement mode.

[0036] The reflection coefficient is a key parameter measuring the degree of microwave signal reflection at a sample interface. Its value is usually represented by S11 and reflects the impedance mismatch between the sample and free space. Different pollutants, due to their different molecular structures and charge distributions, will exhibit specific changes in the spectral characteristics of the reflection coefficient. For example, organic pollutants may produce strong absorption peaks at specific frequencies, while heavy metal complexes will change their reflection amplitude due to increased conductivity. By analyzing the reflection coefficient versus frequency curve, the electromagnetic fingerprint of pollutants can be preliminarily distinguished.

[0037] The dielectric loss factor is an important indicator characterizing a material's ability to absorb microwave energy, and its value is closely related to the sample's loss tangent. This parameter directly reflects the energy dissipation characteristics of pollutants in a microwave field; for example, heavy metal ions dominate the low-frequency response through conductive loss. During the measurement process, the dielectric loss factor and reflection coefficient are acquired simultaneously, together forming a complete picture of the sample's electromagnetic response, providing high-dimensional and rich input data for subsequent feature extraction and model recognition. Through this non-contact, high-sensitivity measurement method, this invention achieves rapid electromagnetic characterization of complex components.

[0038] The following describes in detail step 102, namely, "performing wavelet denoising and fast Fourier transform on the electromagnetic response parameters, extracting multi-band characteristic peaks, and forming a feature vector containing organic matter and heavy metal complexes," with reference to the embodiments.

[0039] In practical measurements, single-port vector network analyzers inevitably introduce environmental noise, instrument thermal noise, or random disturbances caused by sample flow. These noises mainly manifest as high-frequency random fluctuations, severely interfering with the extraction of useful signals. Wavelet denoising, through multi-resolution analysis, decomposes the original time-domain signal into wavelet coefficients of different scales. Low-scale coefficients correspond to high-frequency noise, while high-scale coefficients preserve the main contour of the signal. By setting appropriate thresholds, soft-thresholding or hard-thresholding is applied to the high-frequency coefficients, followed by signal reconstruction, effectively filtering out noise while preserving the original spectral characteristics of the reflection coefficient and dielectric loss factor to the greatest extent. Commonly used wavelet basis functions include the Daubechies series, whose tight support and orthogonality ensure computational efficiency and signal fidelity in the denoising process.

[0040] After denoising, performing a Fast Fourier Transform (FFT) on the signal is the core operation for converting time-domain data into a frequency-domain spectrum. The FFT utilizes a divide-and-conquer strategy to reduce the computational complexity of the Discrete Fourier Transform (DFT) from quadratic to logarithmic levels, significantly improving processing speed. After the transform, the complex fluctuations originally superimposed in the time domain are clearly separated into different frequency components, forming amplitude and phase spectra. The amplitude spectrum directly reflects the energy distribution of the signal at each frequency, while different pollutants in road runoff produce characteristic absorption or reflection peaks in specific frequency bands due to molecular polarization, ionic conductivity, or particle scattering mechanisms. The location, intensity, and width of these peaks together constitute the electromagnetic fingerprint of the pollutants.

[0041] Extracting multi-band characteristic peaks is the core step in feature engineering. The frequency domain spectrum is divided into low-frequency, mid-frequency, and high-frequency bands, each corresponding to a response dominated by a different physical mechanism. For example, the low-frequency band is mainly affected by the conductivity loss of heavy metal composites, while the mid-frequency band reflects the dipole relaxation of organic materials. Within each band, local maxima are automatically detected, and their amplitude, center frequency, and full width at half maximum (FWHM) are recorded as feature parameters. These parameters are then normalized to form a high-dimensional feature vector.

[0042] The final result is a feature vector containing organic matter and heavy metal complexes, achieving a patterned representation of pollutant types. This vector not only preserves the spectral details of the original signal but also compresses the complex electromagnetic response into a computable digital fingerprint through dimensionality reduction and clustering mapping. Subsequent microwave pyrolysis models use this vector as input to simulate pyrolysis behavior and invert component types and concentrations, thus achieving high-precision identification. This feature extraction process ensures a lossless conversion from noisy data to structured fingerprints, laying the foundation for the robustness and accuracy of the entire identification process.

[0043] As an feasible approach, this application applies wavelet transform to perform multi-level decomposition of the electromagnetic response parameters, removes high-frequency noise components, and reconstructs a clean signal; performs fast Fourier transform on the denoised signal to convert the time-domain data into a frequency-domain spectrum and identifies the dominant frequency components; divides the frequency-domain spectrum into low-frequency, mid-frequency, and high-frequency bands, and extracts the peak amplitude, frequency position, and width of each band as feature parameters; based on the extracted feature parameters, constructs a multi-dimensional feature vector, and maps it to the fingerprint pattern of organic matter and heavy metal complexes through a clustering algorithm.

[0044] Specifically, wavelet transform utilizes a multi-resolution analysis framework to decompose the original time-domain signal layer by layer into approximation coefficients and detail coefficients. Approximation coefficients preserve the low-frequency trend of the signal, while detail coefficients capture high-frequency fluctuations. By applying a threshold to the detail coefficients at each decomposition level, such as using a soft thresholding method, noise coefficients below the threshold are zeroed or reduced. The signal is then reconstructed using inverse wavelet transform. This process effectively eliminates high-frequency noise while preserving the true variation characteristics of the reflection coefficient and dielectric loss factor to the greatest extent possible, ensuring that subsequent analysis is based on a clean signal.

[0045] The Fast Fourier Transform (FFT) significantly improves computational efficiency through a divide-and-conquer algorithm, mapping discrete time-domain sampling points to amplitude and phase spectra in the frequency domain. In the frequency domain, the originally superimposed complex fluctuations are clearly separated into harmonic components of different frequencies. The identification of dominant frequency components relies on the peak detection of the amplitude spectrum, which corresponds to the resonant absorption or scattering response of pollutants in a microwave field. For example, polarization relaxation of organic molecules forms broad peaks in specific frequency bands, while heavy metal complexes exhibit low-frequency enhancement due to conductivity losses. Through this transformation, structured information in the frequency domain is extracted from the continuous waveform in the time domain.

[0046] Frequency band division is based on the physical mechanisms of pollutant electromagnetic responses. The low-frequency band typically covers below 1 GHz and mainly reflects the conductivity and ion migration effects of heavy metal complexes; the mid-frequency band is located between 1 and 5 GHz, dominating dipole relaxation and molecular vibrations of organic compounds. Within each frequency band, local maxima are detected, and their amplitude, center frequency, and full width at half maximum (FWHM) are recorded. These parameters are standardized and incorporated as independent feature dimensions into the vector space.

[0047] Multidimensional feature vectors combine amplitude, position, and width parameters of each frequency band into a high-dimensional point cloud. Clustering algorithms, such as K-means or Gaussian mixture models, are used to classify the point cloud into patterns. Using pre-labeled training samples, the algorithm learns the typical fingerprint distributions of two types of pollutants: organic matter corresponds to a broad-peak, low-frequency shift pattern, while heavy metal complexes exhibit a low-frequency, strong-peak pattern. Finally, the feature vector of each measurement sample is mapped to the nearest fingerprint cluster, completing the accurate conversion from the original signal to the pollutant type fingerprint, providing a structured, high-information-density input for subsequent microwave pyrolysis model identification.

[0048] The following describes in detail step 103, namely, "using a microwave pyrolysis model to identify components based on the feature vector and obtain identification results; wherein the pyrolysis process of the microwave pyrolysis model includes a rapid volatilization stage and a slow carbonization stage, and during the component identification process, the dielectric loss factor is substituted into the volume heating equation in real time to dynamically update the simulated temperature field of the rapid volatilization stage and the slow carbonization stage, and the microwave pyrolysis model parameters are optimized using a simulated annealing algorithm."

[0049] The microwave pyrolysis model is a numerical simulation model based on the microwave heating mechanism used to predict the pyrolysis kinetics of materials in a microwave field. This model does not perform actual pyrolysis on road runoff samples; instead, it uses numerical simulation and feature vectors obtained from electromagnetic measurements to invert the pyrolysis behavior of pollutants in a microwave field, thus achieving non-destructive identification. The feature vectors, as model inputs, contain electromagnetic fingerprint information of organic matter and heavy metal complexes at different frequency bands. By simulating the temperature response and mass transformation processes of these components under microwave heating, the model infers their chemical composition and concentration distribution. The identification results are output as the type, content, and confidence interval of each component, providing quantitative evidence for subsequent environmental assessments.

[0050] This invention, for the first time, explicitly divides the microwave pyrolysis process into a rapid volatilization stage and a slow carbonization stage, and employs a dual-distribution activation energy model to characterize their kinetic behavior, differing from the commonly used three-stage method or single-distribution model in existing technologies. This division, combined with real-time dielectric loss factor measurements from a single-port vector network, dynamically updates the simulated temperature fields of the two stages, achieving accurate and non-destructive identification of complex components in road runoff, exhibiting significant non-obviousness. The rapid volatilization stage corresponds to a low activation energy reaction, primarily simulating the rapid release of light organic matter and volatile components; this stage is characterized by a rapid heating rate and significant mass loss. The slow carbonization stage corresponds to a high activation energy reaction, describing the gradual condensation and stabilization process of residual organic matter and heavy metal complexes, characterized by a slow temperature rise and a gradual plateauing of mass loss. This dual-stage division conforms to the theoretical framework of the dual-distribution activation energy model and can effectively capture the dynamic differences in the non-uniform pyrolysis behavior of multiple components in road runoff.

[0051] As an feasible approach, the implementation method of the microwave pyrolysis model in this application includes: initializing dual-distribution activation energy parameters based on the electromagnetic response parameters of the road runoff sample, including a low activation energy distribution in the rapid volatilization stage and a high activation energy distribution in the slow carbonization stage; constructing a reaction kinetic equation, dividing the pyrolysis process into a rapid volatilization stage and a slow carbonization stage, and incorporating microwave power density as a variable into the reaction kinetic equation to simulate its influence on temperature changes.

[0052] Specifically, the electromagnetic response parameters include the reflection coefficient and dielectric loss factor measured by a single-port vector network analyzer. These data directly reflect the absorption and scattering characteristics of microwaves by different pollutants in the sample. The initialization process first estimates the initial activation energy distribution for the rapid volatilization and slow carbonization stages based on the spectral characteristics of the reflection coefficient and the frequency band distribution of the dielectric loss factor. The rapid volatilization stage corresponds to a low activation energy distribution, typically ranging from 30 to 80 kJ / mol, mainly characterizing the rapid release of light organic matter and volatile components. The slow carbonization stage corresponds to a high activation energy distribution, typically ranging from 120 to 220 kJ / mol, describing the gradual condensation and stabilization of residual organic matter and heavy metal complexes. The activation energies of both stages follow independent Gaussian distributions, and initial values ​​are assigned through a weighted mapping of electromagnetic parameters.

[0053] Constructing the reaction kinetic equation is the core step in model implementation. This equation, based on Arrhenius's law, is extended to a bidistribution form, where the overall conversion α is composed of the superposition of two independent subprocesses:

[0054]

[0055] Where A1 and A2 are the pre-exponential factors for the rapid volatilization stage and the slow carbonization stage, respectively; E1 and E2 are the corresponding activation energies; R is the gas constant; T is the instantaneous temperature; and f1(E1) and f2(E2) are Gaussian distribution functions.

[0056]

[0057] To reflect the heating characteristics of microwave bodies, the change of temperature T over time is controlled by the volumetric heating equation:

[0058]

[0059] Where ρ is the sample density, C P Let P be the specific heat capacity, tanδ be the microwave power density, tanδ be the dielectric loss factor, and k be the thermal conductivity. This equation incorporates microwave power density as a variable into the reaction kinetics system, making the pyrolysis rate not only dependent on the chemical energy barrier but also responding in real time to the temperature field changes caused by electromagnetic absorption, thereby achieving dynamic coupling of the two-stage kinetics.

[0060] During component identification, the dielectric loss factor measured by a single-port vector network analyzer is substituted into the volumetric heating equation in real time to achieve dynamic updates of the simulated temperature field. The volumetric heating equation is based on the volumetric heating mechanism of microwaves in a medium, considering the coupling effect of the dielectric loss factor and microwave power density to calculate the instantaneous heat generation rate per unit volume. The real-time updated dielectric loss factor reflects the microwave absorption capacity of the sample in its current state; substituted into the equation, it can accurately predict the local temperature gradient and heating rate. The temperature field during the rapid volatilization stage is iteratively updated using a gridded method based on a low activation energy threshold to ensure that the thermal response of the volatile release region is consistent with reality. The temperature field during the slow carbonization stage is optimized using a high activation energy distribution model to improve the residual char stabilization path and avoid excessive hotspot concentration. This closed-loop update mechanism allows the simulation process to adapt to real-time changes in sample components, significantly improving identification accuracy.

[0061] As an implementable approach, this application dynamically updates the temperature field of the rapid volatilization stage and the slow carbonization stage by substituting the dielectric loss factor measured by vector network analyzer into the volume heating equation in real time during component identification. This includes: collecting dielectric loss factor data of road runoff samples at preset time intervals using a single-port vector network analyzer, and simultaneously recording the current microwave power density and sample density parameters; substituting the dielectric loss factor into the volume heating equation to calculate the instantaneous heating rate and temperature gradient caused by microwave absorption, wherein the volume heating equation considers the product relationship between the dielectric loss factor and the microwave power density, and outputs the local temperature increment and gradient vector; dynamically adjusting the temperature field distribution of the rapid volatilization stage based on the calculation results, including mapping the instantaneous heating rate to the volatile matter release area according to the low activation energy distribution threshold to achieve gridded iterative updates of the temperature field; and simultaneously adjusting the temperature field distribution of the slow carbonization stage, including applying a high activation energy distribution model to integrate the temperature gradient into the residue stabilization path and optimize the thermal balance of the gradual carbonization process.

[0062] Specifically, the volume heating equation describes the volume heat generation mechanism of microwaves within the medium. The first term on the right-hand side of the equation, P·tanδ, represents the instantaneous volume heat source caused by microwave absorption, and the second term represents heat conduction and diffusion. Substituting the latest tanδ at each time step, the partial differential equation is solved numerically, outputting the local temperature increment ΔT and the temperature gradient vector ▽T, providing accurate heat input for subsequent staged temperature field updates.

[0063] Based on the calculation results, the temperature field distribution during the rapid volatilization stage is dynamically adjusted to achieve accurate simulation of the low activation energy reaction. The rapid volatilization stage corresponds to a low activation energy distribution, and its reaction rate is extremely sensitive to temperature. The adjustment process first identifies the volatile matter-dominant region based on the low activation energy threshold (e.g., E1 < 80 kJ / mol), and then maps the instantaneous heating rate P·tanδ to the finite element mesh nodes of this region. The temperature field update is solved iteratively using either explicit or implicit finite difference methods.

[0064]

[0065] Where i represents a grid node and n represents a time step. Grid-based iteration ensures that the temperature in the volatile release region responds rapidly to changes in microwave absorption, achieving dynamic matching between the heating curve and the actual pyrolysis behavior at this stage.

[0066] Synchronously adjusting the temperature field distribution during the slow carbonization stage is crucial for optimizing the stabilization process of high-energy-barrier residues. The slow carbonization stage relies on a high activation energy distribution (E2 > 120 kJ / mol), requiring a high temperature threshold for reaction initiation. The system applies a high activation energy distribution model, using the aforementioned temperature gradient ∇T as input, to optimize the thermal balance along the char formation path. The update strategy employs a weighted conduction mechanism.

[0067]

[0068] Among them, T char Let be the temperature of the carbonization region, and α and β be weighting coefficients, (P·tanδ). delayed This represents the microwave heat input after thermal conduction delay. This mechanism avoids excessive hotspot concentration and ensures a gradual and stable carbonization process. Through synchronous iterative updates of the two-stage temperature field, the model achieves closed-loop simulation of the pyrolysis behavior of complex components, significantly improving the real-time performance and accuracy of identification.

[0069] Optimizing the parameters of the microwave pyrolysis model using the simulated annealing algorithm is a key step in improving the identification accuracy of this invention. The simulated annealing algorithm is a global optimization method that simulates the energy minimization principle in the metal annealing process. It can effectively escape local optima and is suitable for complex nonlinear searches in the parameter space of microwave pyrolysis models. The optimization objective is to simultaneously minimize component identification error and simulated energy consumption, constructing a composite objective function. The algorithm uses activation energy distribution parameters, microwave power density, and residence time as optimization variables, and gradually approximates the globally optimal parameter combination through iterative perturbation and acceptance criteria, thereby ensuring that the simulated pyrolysis behavior of the model is highly consistent with the actual electromagnetic response.

[0070] The algorithm first initializes the parameter vector and temperature state. The parameter vector includes the mean and standard deviation of the activation energy during the rapid volatilization phase, the mean and standard deviation of the activation energy during the slow carbonization phase, the microwave power density, and the pyrolysis residence time. The initial values ​​are set based on an empirical mapping of electromagnetic measurement data. The initial temperature is set to a relatively high value, such as 1000 Kelvin, to allow the algorithm to explore the parameter space extensively in the early stages and avoid premature convergence. The composite objective function is defined as a weighted sum of identification error and energy consumption, where the identification error is calculated by the root mean square error of the simulated conversion rate and the predicted values ​​of the eigenvector, and the energy consumption is estimated by the integral of the power density and residence time.

[0071] In each iteration, the algorithm generates new parameter states and evaluates their quality. New states are generated through a Gaussian-Cauchy hybrid perturbation, applying a random perturbation to the current parameter vector, with the perturbation amplitude decreasing with temperature. The Gaussian component ensures fine-grained local search, while the Cauchy component enhances global jump capability. The new states are then substituted into the microwave pyrolysis model to quickly solve for the two-stage conversion rate and calculate the objective function value. If the objective function value of the new state is lower than that of the current state, it is unconditionally accepted; otherwise, it is accepted with an exponential probability exp(-ΔE / T), where ΔE is the increment of the objective function and T is the current temperature. This probabilistic acceptance mechanism allows the algorithm to tolerate inferior solutions at high temperatures, preserving exploration diversity.

[0072] Furthermore, to prevent the algorithm from getting trapped in local optima, an adaptive reheating mechanism is introduced. When the objective function fails to improve after multiple iterations, a reheating operation is triggered, temporarily increasing the temperature by 20% to 50% to reactivate the exploration capability. Subsequently, an exponential cooling strategy is adopted, with the temperature adjusted according to T... k+1 =αT k The temperature is gradually reduced, with a cooling coefficient α typically set to 0.95. This cooling process simulates the slow cooling of physical annealing, ensuring the algorithm focuses on fine-tuning near the optimal solution in later stages. The termination condition is set when the temperature falls below a preset threshold or the maximum number of iterations is reached. Finally, the optimized parameter vector is output and fed back to the microwave pyrolysis model, achieving closed-loop adaptive optimization of the recognition process.

[0073] Furthermore, the microwave pyrolysis model in this application also includes: employing a coupled electromagnetic, thermal, and mass field finite element mesh to periodically refresh the hotspot distribution and pollutant diffusion cloud map within the microwave cavity. This mesh discretizes the microwave cavity into thousands to tens of thousands of volume elements, each of which simultaneously carries the solution variables for the electromagnetic field, temperature field, and pollutant concentration field. The electromagnetic field describes microwave propagation and absorption, the temperature field characterizes the volumetric heating effect, and the pollutant concentration field tracks the spatial diffusion of components. The three fields drive each other through a set of coupled equations, forming a closed-loop numerical system to ensure that the simulation process truly reflects the interaction behavior between microwaves and the complex components of road runoff.

[0074] The mathematical foundation for electromagnetic-thermal-mass coupling is built upon Maxwell's equations, the heat conduction equation, and the convection-diffusion equation. The electromagnetic field is governed by time-harmonic Maxwell's equations, calculating the electric field intensity distribution and deriving local power dissipation as a heat source. The temperature field is based on the volumetric heating equation, receiving electromagnetic power dissipation and outputting the temperature gradient. The pollutant concentration field follows Fick's diffusion law and the convection term, with the temperature gradient-driven thermophoretic effect further influencing component migration. The three fields are solved sequentially or in parallel within each time step, with data transmitted in real time between grid nodes, achieving full-field coupling.

[0075] Periodically refreshing the hotspot distribution and contaminant diffusion cloud map within the microwave cavity is a key mechanism for dynamic visualization and model calibration. Every preset time step, typically 10 to 100 milliseconds, the electric field focusing region is recalculated, generating a hotspot cloud map that displays the location and intensity of the high-temperature region where microwave energy is deposited. Simultaneously, the contaminant diffusion cloud map is rendered based on the concentration field gradient, presenting the spatial migration paths of organic matter and heavy metal complexes. The refreshing process employs adaptive mesh refinement technology, automatically increasing node density in hotspot and interface regions to improve local solution accuracy.

[0076] Through three-field coupling and periodic refresh, the model can not only predict macroscopic pyrolysis conversion rates but also reveal non-uniform heating and component separation phenomena at the microscopic scale. Hotspot distribution guides the localized acceleration of temperature rise during the rapid volatilization stage, while pollutant diffusion cloud maps provide spatial evidence for residual char formation during the slow carbonization stage. The final output dynamic cloud map sequence provides intuitive verification of the identification results and supports parameter optimization and anomaly diagnosis, significantly enhancing the model's adaptability and interpretability to complex systems.

[0077] The following describes step 104, namely "obtaining an analysis report of complex road runoff components based on the identification results," in detail with reference to an embodiment.

[0078] This step transforms the optimized identification results from the microwave pyrolysis model into a structured and visualized environmental assessment document, summarizing the types, concentrations, and confidence intervals of organic pollutants and heavy metal compounds, providing a scientific basis for urban drainage management and pollution source tracing. The report generation process encompasses data integration, risk quantification, visualization rendering, and document encapsulation, ensuring comprehensive, readable, and actionable information.

[0079] First, the identification results are summarized, and quantitative information for each component is extracted. The identification results are output by the microwave pyrolysis model, including the chemical category, mass concentration, and statistical confidence interval for each pollutant. The confidence interval is obtained through multiple iterative convergence assessments using a simulated annealing algorithm, reflecting the uncertainty of the identification. Organic pollutants are classified by functional groups, such as hydrocarbons and esters; heavy metal complexes are listed with their main ionic forms, such as lead and nickel. This information is organized in tabular form, forming the core data layer of the report.

[0080] Subsequently, risk indices for each component are calculated to achieve a quantitative assessment of pollution impact. The risk index is calculated based on a weighted average of concentrations relative to environmental thresholds, referencing national surface water quality standards or World Health Organization guidelines. For example, when heavy metal concentrations exceed the threshold, the index increases exponentially. The calculation results are presented as heatmaps or radar charts, visually displaying the components exceeding standards and potential ecological threats, assisting decision-makers in quickly identifying high-risk pollutants.

[0081] Generating visualizations is a key step in improving report readability. This involves creating spatiotemporal distribution maps of pollutants, showing the correlation between concentration gradients at sampling points and rainfall; generating radar charts to compare the relative intensities of different components; and creating bar charts to display confidence intervals and historical data trends. Visualizations use a unified color scheme and legend to ensure consistency across reports. Simultaneously, metadata, including sampling time, geographic coordinates, rainfall intensity, and measurement conditions, is integrated to form a complete traceability chain. First, during the sampling phase, metadata, including geographic coordinates, rainfall intensity, sampling time, ambient temperature, and road surface type, is collected via mobile terminals or IoT sensors. This data is uploaded to a central database in real-time in JSON format, generating unique identifiers. During the measurement phase, a single-port vector network analyzer automatically appends instrument parameters, such as frequency range, power output, and calibration time, binding them to the sample identifier. During the identification phase, the microwave pyrolysis model dynamically records simulation parameters, optimization iterations, and confidence intervals, also stored using the identifier as an index. When the report is generated, the database is retrieved using the identifier, the metadata table is automatically populated, and the data is embedded in the report header and appendix. Meanwhile, the metadata supports hash verification to ensure data integrity; the visualization module can filter metadata by time or location to generate dynamic pollution maps. This approach enables end-to-end traceability from source to report, enhancing the credibility and decision-making value of the analysis report.

[0082] Finally, a structured report document is packaged, supporting multiple output formats. The report is primarily in PDF format, embedding interactive charts and hyperlinks for easy online viewing; an Excel attachment is also provided, containing raw data and calculation formulas for secondary analysis. The document automatically adds a version number, generation time, and technical statement to ensure traceability. Through this complete process, the analysis report not only outputs identification results but also transforms them into actionable evidence for environmental management, significantly improving the efficiency and scientific rigor of road runoff pollution control.

[0083] The methods provided in this application can be applied to various scenarios, including but not limited to: in post-rainstorm runoff monitoring, a portable single-port vector network analyzer can be deployed at drainage outlets to identify the concentrations of organic pollutants and heavy metal compounds in real time, and generate pollution heat maps by combining metadata, assisting municipal departments in quickly locating high-risk areas and initiating emergency interception. In industrial park stormwater discharge monitoring, fixed monitoring stations can be connected to continuously analyze runoff component change trends, automatically generate compliance reports, and support environmental law enforcement and enterprise pollution source tracing. Furthermore, in coastal city flood control and drainage projects, this method can be linked with tide level sensors to predict the diffusion path of runoff pollutants into the sea, generate dynamic risk assessment reports, and provide decision-making basis for marine ecological protection.

[0084] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0085] According to another embodiment, a device for identifying complex components of road runoff based on a single-port vector network is provided. Figure 2 A schematic block diagram of a road runoff complex component identification device based on a single-port vector network is shown according to one embodiment. Figure 2 As shown, the device 200 includes:

[0086] The road surface data acquisition unit 201 is configured to acquire road surface runoff samples and use a single-port vector network analyzer to measure the electromagnetic response parameters of the samples, including the reflection coefficient and dielectric loss factor.

[0087] The data preprocessing unit 202 is configured to perform wavelet denoising and fast Fourier transform on the electromagnetic response parameters, extract multi-band feature peaks, and form a feature vector containing organic matter and heavy metal complexes.

[0088] The component identification unit 203 is configured to use a microwave pyrolysis model to identify components based on the feature vector and obtain identification results. The pyrolysis process of the microwave pyrolysis model includes a rapid volatilization stage and a slow carbonization stage. During the component identification process, the dielectric loss factor is substituted into the volume heating equation in real time to dynamically update the simulated temperature field of the rapid volatilization stage and the slow carbonization stage, and the microwave pyrolysis model parameters are optimized using a simulated annealing algorithm.

[0089] The analysis report generation unit 204 is configured to obtain an analysis report of complex components of road runoff based on the identification results.

[0090] As an implementable approach, the data preprocessing unit 202, when performing wavelet denoising and fast Fourier transform on the electromagnetic response parameters to extract multi-band characteristic peaks and form a feature vector containing organic matter and heavy metal complexes, can be configured as follows: applying wavelet transform to perform multi-level decomposition on the electromagnetic response parameters, removing high-frequency noise components and reconstructing a clean signal; performing fast Fourier transform on the denoised signal to convert the time-domain data into a frequency-domain spectrum and identify the dominant frequency components; dividing the frequency-domain spectrum into low-frequency and mid-frequency regions, and extracting the peak amplitude, frequency position, and width of each frequency band as feature parameters; based on the extracted feature parameters, constructing a multi-dimensional feature vector, and mapping it to the fingerprint pattern of organic matter and heavy metal complexes through a clustering algorithm.

[0091] As an feasible approach, the implementation method of the microwave pyrolysis model includes: initializing dual-distribution activation energy parameters based on the electromagnetic response parameters of the road runoff sample, including a low activation energy distribution in the rapid volatilization stage and a high activation energy distribution in the slow carbonization stage; constructing a reaction kinetic equation, dividing the pyrolysis process into a rapid volatilization stage and a slow carbonization stage, and incorporating microwave power density as a variable into the reaction kinetic equation to simulate its influence on temperature changes.

[0092] As an implementable approach, the component identification unit 203 can be configured to dynamically update the temperature fields of the rapid volatileization stage and the slow carbonization stage by substituting the dielectric loss factor measured by the vector network analyzer into the volume heating equation in real time during the component identification process. This can be achieved by: collecting dielectric loss factor data of the road runoff sample at preset time intervals using a single-port vector network analyzer, and simultaneously recording the current microwave power density and sample density parameters; substituting the dielectric loss factor into the volume heating equation to calculate the instantaneous heating rate and temperature gradient caused by microwave absorption, wherein the volume heating equation considers the product relationship between the dielectric loss factor and the microwave power density, and outputs the local temperature increment and gradient vector; dynamically adjusting the temperature field distribution of the rapid volatileization stage based on the calculation results, including mapping the instantaneous heating rate to the volatile matter release area according to the low activation energy distribution threshold to achieve gridded iterative updates of the temperature field; and simultaneously adjusting the temperature field distribution of the slow carbonization stage, including applying a high activation energy distribution model to integrate the temperature gradient into the residue stabilization path and optimize the thermal balance of the gradual carbonization process.

[0093] As an implementable approach, the component identification unit 203, when optimizing the microwave pyrolysis model parameters using the simulated annealing algorithm, can be configured to: initialize the parameter vector, including the mean activation energy, standard deviation, microwave power density, and residence time, and set an initial temperature; generate a new state of Gaussian-Cauchy mixed perturbation, calculate the composite objective function value, including identification error and energy consumption, and apply an acceptance criterion: if the new state is better than the current one, it is accepted; otherwise, it is accepted with exponential probability; perform adaptive reheating, and if there is no continuous improvement, temporarily increase the temperature to enhance exploration; use exponential cooling to reduce the temperature until the termination condition is met, and output the optimized parameters.

[0094] As an feasible approach, the microwave pyrolysis model also includes: using an electromagnetic-thermal-mass three-field coupled finite element mesh to periodically refresh the hotspot distribution and contaminant diffusion cloud map within the microwave cavity.

[0095] As an implementable approach, the analysis report generation unit 204, when obtaining an analysis report on complex components of road runoff based on the identification results, can be configured to: summarize the identification results, including the types, concentrations, and confidence intervals of organic pollutants and heavy metal compounds; calculate the risk index of each component and assess the potential environmental pollution impact based on concentration thresholds; generate visualization elements to display the component distribution and spatiotemporal variation trends, and integrate metadata, including sample collection time, location, and measurement conditions, to form a structured report document.

[0096] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. 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. Those skilled in the art can understand and implement this without creative effort.

[0097] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0098] In addition, embodiments of this application also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.

[0099] And an electronic device, comprising:

[0100] One or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any of the foregoing method embodiments.

[0101] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.

[0102] in, Figure 3 The architecture of an electronic device is illustrated, which may include a processor 310, a video display adapter 311, a disk drive 312, an input / output interface 313, a network interface 314, and a memory 320. The processor 310, video display adapter 311, disk drive 312, input / output interface 313, network interface 314, and memory 320 can communicate with each other via a communication bus 330.

[0103] The processor X10 can be implemented using a general-purpose CPU, microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits to execute relevant programs and implement the technical solution provided in this application.

[0104] The memory X20 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 320 can store the operating system 321 for controlling the operation of the electronic device 300, and the basic input / output system (BIOS) 322 for controlling the low-level operations of the electronic device 300. Additionally, it can store a web browser 323, a data storage management system 324, and a road runoff complex component identification device 325 based on a single-port vector network, etc. The aforementioned road runoff complex component identification device 325 based on a single-port vector network can be the application program that specifically implements the aforementioned steps in this embodiment. In summary, when implementing the technical solution provided in this application through software or firmware, the relevant program code is stored in the memory 320 and executed by the processor 310.

[0105] Input / output interface 313 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0106] Network interface 314 is used to connect a communication module (not shown in the figure) to enable communication between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0107] Bus 330 includes a pathway for transmitting information between various components of the device, such as processor 310, video display adapter 311, disk drive 312, input / output interface 313, network interface 314, and memory 320.

[0108] It should be noted that although the above-described device only shows the processor 310, video display adapter 311, disk drive 312, input / output interface 313, network interface 314, memory 320, bus 330, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the solution of this application, and does not necessarily include all the components shown in the figures.

[0109] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer program product. This computer program product can be stored in a 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 various embodiments or some parts of the embodiments of this application.

[0110] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for identifying complex components of road runoff based on a single-port vector network, characterized in that, The method includes: Road surface runoff samples were collected, and the electromagnetic response parameters of the samples were measured using a single-port vector network analyzer. The electromagnetic response parameters included the reflection coefficient and the dielectric loss factor. Wavelet denoising and fast Fourier transform are performed on the electromagnetic response parameters to extract multi-band characteristic peaks and form a feature vector containing organic matter and heavy metal complexes. Using a microwave pyrolysis model, component identification is performed based on the feature vector to obtain identification results; wherein, the pyrolysis process of the microwave pyrolysis model includes a rapid volatilization stage and a slow carbonization stage, and during the component identification process, the dielectric loss factor is substituted into the volume heating equation in real time to dynamically update the simulated temperature field of the rapid volatilization stage and the slow carbonization stage, and the microwave pyrolysis model parameters are optimized using a simulated annealing algorithm. Based on the identification results, an analysis report on the complex components of road runoff is obtained.

2. The method according to claim 1, characterized in that, Wavelet denoising and fast Fourier transform are performed on the electromagnetic response parameters to extract multi-band characteristic peaks, forming a feature vector containing organic matter and heavy metal complexes, including: The electromagnetic response parameters are decomposed into multiple levels using wavelet transform to remove high-frequency noise components and reconstruct a clean signal. Perform a Fast Fourier Transform on the denoised signal to convert the time-domain data into a frequency-domain spectrum and identify the dominant frequency components; The frequency domain spectrum is divided into low frequency and mid frequency, and the peak amplitude, frequency position and bandwidth of each frequency band are extracted as feature parameters. Based on the extracted feature parameters, a multidimensional feature vector is constructed and mapped to the fingerprint pattern of organic matter and heavy metal complexes through a clustering algorithm.

3. The method according to claim 1, characterized in that, The implementation method of the microwave pyrolysis model includes: Based on the electromagnetic response parameters of the road runoff samples, the dual-distribution activation energy parameters are initialized, including the low activation energy distribution in the rapid volatilization stage and the high activation energy distribution in the slow carbonization stage. A reaction kinetic equation was constructed, dividing the pyrolysis process into a rapid volatilization stage and a slow carbonization stage. Microwave power density was incorporated as a variable into the reaction kinetic equation to simulate its effect on temperature changes.

4. The method for identifying complex components of road runoff according to claim 1, characterized in that, During component identification, the dielectric loss factor measured by vector calorimetry is substituted into the bulk heating equation in real time to dynamically update the temperature fields of the rapid volatilization stage and the slow carbonization stage, including: The dielectric loss factor data of road runoff samples are collected at preset time intervals using a single-port vector network analyzer, and the current microwave power density and sample density parameters are recorded simultaneously. Substituting the dielectric loss factor into the volume heating equation, the instantaneous heating rate and temperature gradient caused by microwave absorption are calculated. The volume heating equation considers the product relationship between the dielectric loss factor and the microwave power density, and outputs the local temperature increment and gradient vector. Based on the calculation results, the temperature field distribution in the rapid evaporation stage is dynamically adjusted, including mapping the instantaneous heating rate to the volatile matter release area according to the low activation energy distribution threshold, so as to realize the gridded iterative update of the temperature field. The temperature field distribution during the slow carbonization stage is adjusted synchronously, including the application of a high activation energy distribution model to incorporate the temperature gradient into the residue stabilization path and optimize the thermal balance of the gradual carbonization process.

5. The method for identifying complex components of road runoff according to claim 1, characterized in that, The optimization of the microwave pyrolysis model parameters using the simulated annealing algorithm includes: Initialize the parameter vector, including the mean activation energy, standard deviation, microwave power density, and dwell time, and set the initial temperature; Generate a new state with Gaussian-Cauchy mixed perturbation, calculate the composite objective function value, including identification error and energy consumption, and apply the acceptance criterion: if the new state is better than the current one, accept it; otherwise, accept it with exponential probability. Perform adaptive reheating; if there is no continuous improvement, temporarily increase the temperature to enhance the exploration. The temperature is reduced using exponential cooling until the termination condition is met, and the optimized parameters are output.

6. The method according to claim 1, characterized in that, The microwave pyrolysis model also includes: An electromagnetic-thermal-mass three-field coupled finite element mesh was used to periodically refresh the hotspot distribution and pollutant diffusion cloud map within the microwave cavity.

7. The method according to claim 1, characterized in that, Based on the identification results, an analysis report on the complex components of road runoff is obtained, including the following steps: The identification results are summarized, including the types, concentrations, and confidence intervals of organic pollutants and heavy metal complexes; Calculate the risk index of each component and assess the potential environmental pollution impact based on concentration thresholds; The system generates visualization elements to display the component distribution and spatiotemporal variation trends, and integrates metadata, including sample collection time, location, and measurement conditions, to form a structured report document.

8. A device for identifying complex components of road runoff based on a single-port vector network, characterized in that, The device includes: The road surface data acquisition unit is configured to acquire road surface runoff samples and use a single-port vector network analyzer to measure the electromagnetic response parameters of the samples, including the reflection coefficient and dielectric loss factor. The data preprocessing unit is configured to perform wavelet denoising and fast Fourier transform on the electromagnetic response parameters, extract multi-band feature peaks, and form a feature vector containing organic matter and heavy metal complexes. The component identification unit is configured to use a microwave pyrolysis model to identify components based on the feature vector and obtain identification results. The pyrolysis process of the microwave pyrolysis model includes a rapid volatilization stage and a slow carbonization stage. During the component identification process, the dielectric loss factor is substituted into the volume heating equation in real time to dynamically update the simulated temperature field of the rapid volatilization stage and the slow carbonization stage, and the microwave pyrolysis model parameters are optimized using a simulated annealing algorithm. The analysis report generation unit is configured to obtain an analysis report of complex components of road runoff based on the identification results.

9. An electronic device, characterized in that, include: One or more processors; And a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 7.

Images