A particulate matter trapping method and system based on electrostatic enhancement and surface catalysis

Abstract

The application discloses a kind of particle trapping method and system based on electrostatic enhancement and surface catalysis, it is related to motor vehicle exhaust aftertreatment technical field, the method of the present application is by constructing multivariate breakdown voltage prediction model, calculate pulse voltage safe operation window and generate feedforward control lookup table;Establish the quantitative mapping model of coating microstructure parameters and contact point density, by atomic layer deposition in nano-protrusion tip selectively load active component;Collect full-condition combined performance data, train Gaussian process regression combined performance proxy model, solve the joint optimal configuration matrix of charging parameter and coating contact density, and execute real-time collaborative control based on table lookup interpolation.The application combines engine ECU signal foresight prediction, adjusts target voltage in advance when exhaust temperature rapidly transient, so that electrostatic charging is stably operated in pulse voltage safe operation window in full temperature domain.

Description

Technical Field

[0001] This invention relates to the field of motor vehicle exhaust aftertreatment technology, specifically to a highly efficient particulate matter capture method and system based on electrostatic enhancement and surface catalysis. Background Technology

[0002] Particulate matter emitted by motor vehicles is a significant source of fine particulate matter (PM2.5). Existing particulate matter capture technologies typically optimize the electrostatic discharge (ESD) system and catalytic coating as two independent modules. However, these two subsystems are mutually restrictive. Under high-temperature conditions, the gas breakdown voltage decreases, requiring a reduction in pulse voltage to avoid arcing. This leads to a decrease in particulate charge, weakening the electrostatic attraction between the particulates and the catalytic coating, and reducing the probability of close contact. Furthermore, traditional catalytic coatings are predominantly mesoporous, with pore sizes that do not match the size of the soot particles, preventing them from entering the mesopores and resulting in low utilization of numerous internal active sites. When the independent optimization results of the two subsystems are combined, the overall capture efficiency falls below the theoretical optimal level of joint optimization. This leads to technical problems in existing technologies, such as the inability of ESD parameters to track changes in exhaust conditions across the entire temperature range, low effective utilization of active sites in the catalytic coating, and overall efficiency loss due to independent optimization of the two subsystems.

[0003] Therefore, there is an urgent need for a highly efficient particulate matter capture method and system based on electrostatic enhancement and surface catalysis to solve the technical problems mentioned above. Summary of the Invention

[0004] The summary section of this application is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description section below. This summary section is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0005] To address the problems and shortcomings of existing technologies, the present invention aims to provide a particulate matter capture method and system based on electrostatic enhancement and surface catalysis, thereby solving the technical problems in related technologies such as the lack of joint optimization of charge parameters and catalytic coating contact density, the inability to dynamically adjust the safe operating window of pulse voltage according to operating conditions, resulting in low particulate matter capture efficiency and the risk of breakdown discharge.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] As a first aspect of this application, the present invention discloses a particulate matter capture method based on electrostatic enhancement and surface catalysis, comprising the following steps:

[0008] Step 1: Generate a multivariate breakdown voltage prediction model based on exhaust temperature, gas density, and gas component concentration;

[0009] Step 2: Calculate the pulse voltage safe operating window under each operating condition based on the multivariable breakdown voltage prediction model, and then generate a feedforward control lookup table based on the rate of change of exhaust temperature according to the rate of change of throttle opening and the rate of change of speed.

[0010] Step 3: Obtain programmed temperature oxidation data and electron microscopy observation data of the catalytic coating, generate a quantitative mapping model between coating microstructure parameters and contact point density, and determine the parameters of the nanoprotrusion array.

[0011] Step 4: Using the nano-protrusion array parameters determined by the quantitative mapping model, selectively load catalytically active components on the nano-protrusion tips of the catalytic coating surface through atomic layer deposition;

[0012] Step 5: Conduct a full-condition performance acquisition experiment based on the multi-dimensional combination of operating parameters and system parameters, record the particulate matter capture efficiency under each combination, and generate a joint performance database.

[0013] Step 6: Using the joint performance database as the training set, optimize the kernel function hyperparameters using maximum likelihood estimation and use the cross-validation results as the accuracy acceptance criterion, and obtain the joint performance surrogate model through regression training.

[0014] Step 7: Use the joint performance proxy model to solve for the joint optimal combination of charge parameters and coating contact density at each operating point to generate a joint optimal configuration matrix;

[0015] Step 8: Based on the real-time exhaust temperature, particulate matter concentration, and exhaust flow rate, look up values ​​in the joint optimal configuration matrix and interpolate to synchronously output the charge pulse control signal and alternating electric field control signal to the corresponding execution unit.

[0016] Preferably, the generation of the multivariate breakdown voltage prediction model in step 1 includes:

[0017] The temperature-corrected form of the Peek formula was used to fit the relationship between the initiation voltage and breakdown voltage and temperature and gas density.

[0018] HC concentration in the fitting residuals and The effect of concentration was regressed and calibrated, and the concentration correction coefficient was obtained and added to the fitting result.

[0019] Each input variable was standardized using Z-score before participating in fitting and regression calibration.

[0020] Preferably, the pulse voltage safe operating window in step 2 is determined by taking 85% of the predicted breakdown voltage as the upper limit of the real-time safe voltage, taking 1.2 times the predicted starting voltage as the lower limit of the effective charge, and the interval between the two as the pulse voltage safe operating window. The center value of the pulse voltage safe operating window is taken as the target pulse peak voltage.

[0021] Preferably, step 2, which generates a feedforward control lookup table based on look-ahead prediction of engine ECU signals, includes:

[0022] Step 2.1: Obtain the throttle opening value and engine speed value at the current moment, and calculate the time change rate of throttle opening and the time change rate of engine speed;

[0023] Step 2.2: Based on the pre-calibrated mapping relationship between the throttle opening change rate, the speed change rate, and the exhaust temperature change rate, calculate the estimated value of the current exhaust temperature change rate.

[0024] Step 2.3: Compare the estimated value of the exhaust temperature change rate with a preset threshold. When the estimated value exceeds the preset threshold, calculate the predicted value of the exhaust temperature at future times based on the estimated change rate and the preset look-ahead time window. Substitute the predicted value of the exhaust temperature into the multivariate breakdown voltage prediction model to obtain the corresponding pulse voltage safe operation window.

[0025] Step 2.4: Summarize the target pulse peak voltages under various combinations of exhaust temperature, atmosphere conditions, and operating parameters to generate a combined feedforward control lookup table for temperature, atmosphere, and operating conditions.

[0026] Preferably, the quantitative mapping model between coating microstructure parameters and contact point density in step 3 includes:

[0027] The difference in carbon soot oxidation peak temperature of the catalytic coating under loose contact conditions and tight contact conditions is calculated as an indicator of contact efficiency.

[0028] The number of close contact points with contact angles less than a preset angle threshold was counted based on electron microscopy observation data.

[0029] Regression fitting was performed on the peak temperature difference and the density of close contact points of multiple groups of coating samples with different microstructures to generate quantitative relationship curves;

[0030] Geometric interlocking conditions were analyzed based on the diameter distribution of carbon soot equivalent spheres and the characteristic size of nano-protrusions on the coating surface.

[0031] A quantitative mapping model is generated by determining the characteristic size range and spatial density parameters of the nanoprotrusion array that enable the density of close contact points to reach the target value.

[0032] Preferably, the geometric interlocking condition is that the characteristic height of the nanoprotrusions is not less than a preset ratio of the lower limit of the equivalent diameter of the soot sphere, and the spacing between the nanoprotrusions is not greater than the upper limit of the equivalent diameter of the soot sphere, so that a single soot particle can simultaneously form contact with multiple adjacent nanoprotrusions; within the parameter range that satisfies the geometric interlocking condition, the catalytic active component is selectively loaded at the tip of the nanoprotrusion through atomic layer deposition, and the surface energy difference between the tip and the flat area is used to make the active component precursor preferentially nucleate at the tip position; the coating preparation process parameters are systematically varied through response surface methodology, and the predicted relationship between the preparation parameters and the catalytic contact efficiency is generated by regression analysis.

[0033] Preferably, generating the joint performance database in step 5 includes:

[0034] A full-condition combined experiment was conducted under multiple temperature ranges, concentration gradients, and flow rates, considering exhaust temperature, particulate matter concentration, and exhaust flow rate. Under each condition combination, the peak pulse voltage, pulse frequency, and density of tight contact points of the coating were adjusted, and the corresponding quantity removal efficiency and mass removal efficiency were recorded simultaneously. The results were then compiled into a combined performance database.

[0035] Preferably, step 8 further includes arc discharge protection, used to achieve rapid protection and accumulate abnormal data when arc discharge occurs, including:

[0036] The waveform data of the discharge current pulse is continuously monitored at a microsecond-level sampling frequency;

[0037] Real-time discrimination is based on the difference between the continuous high amplitude and oscillation characteristics of arc discharge waveform and the single-peak exponential decay pattern of normal corona discharge.

[0038] When an arc discharge characteristic is detected, the protection is triggered to shut down. The exhaust temperature, atmosphere parameters and pulse voltage value corresponding to the moment the arc is triggered are recorded as abnormal samples and fed back to the calibration dataset of the multivariate breakdown voltage prediction model described in step 1 for local update of correction coefficients.

[0039] Preferably, it also includes a full lifecycle adaptive update step:

[0040] Continuously record operating parameters, charge parameters, and measured acquisition efficiency data at each moment to generate an incremental operation dataset;

[0041] Based on the deviation trend between the measured capture efficiency in the incremental running dataset and the predicted value of the joint performance proxy model, the drift direction and drift amount of the contact density parameter caused by catalytic coating aging are identified.

[0042] The incremental running dataset is incorporated into the training set of the joint performance proxy model using online incremental learning, the posterior distribution parameters are updated, and the joint optimal configuration is resolved using the feedforward control lookup table, and the joint optimal configuration matrix is ​​updated.

[0043] As a second aspect of this application, the present invention discloses a particulate matter capture system based on electrostatic enhancement and surface catalysis, comprising:

[0044] The breakdown voltage prediction module is used to generate a multivariate breakdown voltage prediction model based on exhaust temperature, gas density, and gas component concentration.

[0045] The safety window calculation module is used to calculate the pulse voltage safety operation window under various operating conditions based on the multivariable breakdown voltage prediction model, and then generate a feedforward control lookup table based on the rate of change of exhaust temperature according to the rate of change of throttle opening and the rate of change of speed.

[0046] The coating microstructure mapping module is used to acquire temperature-programmed oxidation data and electron microscopy observation data of the catalytic coating, and generate a quantitative mapping model between coating microstructure parameters and contact point density.

[0047] The coating preparation parameter module is used to selectively load catalytically active components on the tips of nano-protrusions on the surface of the catalytic coating by atomic layer deposition using the nano-protrusion array parameters determined by the quantitative mapping model.

[0048] The joint performance database generation module is used to conduct full-condition performance acquisition experiments based on multi-dimensional combinations of operating parameters and system parameters, record the particulate matter capture efficiency under each combination, and generate a joint performance database.

[0049] The surrogate model training module is used to obtain the joint performance surrogate model by using the joint performance database as the training set, optimizing the kernel function hyperparameters by maximum likelihood estimation and using the cross-validation results as the accuracy acceptance criteria for regression training.

[0050] The configuration matrix generation module is used to solve the joint optimal combination of charge parameters and coating contact density at each working point using the joint performance proxy model to generate a joint optimal configuration matrix.

[0051] The real-time collaborative control module is used to look up and interpolate values ​​in the joint optimal configuration matrix based on the real-time exhaust temperature, particulate matter concentration, and exhaust flow rate, and synchronously output charge pulse control signals and alternating electric field control signals to the corresponding execution units.

[0052] This invention aims to provide a method and system for particulate matter capture based on electrostatic enhancement and surface catalysis. Compared with the prior art, the advantages of this invention are:

[0053] First, the multivariate breakdown voltage prediction model incorporates exhaust temperature, gas density, and gas component concentration into a unified breakdown voltage prediction framework. It calculates the safe operating window for pulse voltage in real time under various temperature and atmospheric conditions, enabling the charge parameters to track continuous changes in exhaust conditions and avoiding the response lag of fixed voltage or step-by-step switching methods. Combined with forward-looking prediction of engine ECU signals, the target voltage is adjusted in advance when exhaust temperature changes rapidly, narrowing the arc risk time window and ensuring that the static charge operates stably within the safe operating window for pulse voltage across the entire temperature range.

[0054] Secondly, the quantitative relationship curve between peak temperature difference and density of close contact points transforms the contact efficiency of the catalytic coating into quantifiable microstructure parameters. Based on the geometric interlocking conditions of the equivalent diameter of soot spheres and the characteristic size of nanoprotrusions, the range of coating surface structure parameters that enable multi-point embedded contact of soot particles was determined, thereby geometrically improving the proportion of close contact between soot and the catalytic coating. Active components are selectively enriched at the tips of nanoprotrusions through atomic layer deposition, concentrating catalytic active sites at the locations with the highest probability of contact with soot, thus improving the effective utilization rate of active sites.

[0055] Furthermore, the Gaussian process regression joint performance surrogate model incorporates charge parameters and coating contact density into a unified nonlinear optimization framework. The joint optimal configuration matrix for charge parameters and coating contact density eliminates the superposition efficiency loss caused by the intertwining of parameter spaces when the two subsystems are optimized independently. Online incremental learning continuously updates the joint performance surrogate model and the joint optimal configuration matrix, ensuring that the synergistic control effect is maintained throughout the catalytic coating aging process. Attached Figure Description

[0056] The accompanying drawings, which form part of this application, are used to provide a further understanding of the application and to make other features, objects, and advantages of the application more apparent. The illustrative embodiments and descriptions of this application are used to explain the application and do not constitute an undue limitation of the application. In the drawings:

[0057] Figure 1 This is a flowchart of the efficient particulate matter capture method based on electrostatic enhancement and surface catalysis provided in the embodiments of the present invention;

[0058] Figure 2 This is a schematic diagram comparing the measured and predicted values ​​of typical temperature-induced voltage breakdown voltage provided in an embodiment of the present invention.

[0059] Figure 3 This is a schematic diagram of a typical operating point pulse voltage safety operation window provided in an embodiment of the present invention;

[0060] Figure 4 This is a schematic diagram illustrating the effect of nanoprotrusion density and active component loading on peak temperature difference provided in an embodiment of the present invention.

[0061] Figure 5 This is a schematic diagram illustrating the effect of nanoprotrusion density and active component loading on continuous regeneration rate provided in the embodiments of the present invention.

[0062] Figure 6 This is a schematic diagram showing the removal efficiency of particulate matter quantity and mass at typical operating points provided in the embodiments of the present invention;

[0063] Figure 7 This is a schematic diagram illustrating the variation of optimal pulse parameters at typical operating points with exhaust temperature, as provided in the embodiments of the present invention.

[0064] Figure 8 This is a schematic diagram illustrating the relationship between the predicted quantity removal efficiency and the exhaust flow rate under the joint optimal configuration provided in this embodiment of the invention.

[0065] Figure 9 This is a schematic diagram comparing the density parameters of the tight contact points before and after coating aging, provided in an embodiment of the present invention. Detailed Implementation

[0066] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0067] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0068] Example

[0069] This invention discloses a particulate matter capture method based on electrostatic enhancement and surface catalysis. The following will describe this disclosure in detail with reference to the accompanying drawings and embodiments.

[0070] In the capture and treatment of particulate matter in motor vehicle exhaust, electrostatic enhancement applies charge to the particles through high-voltage corona discharge, using Coulomb force to drive the particles toward the collection surface; surface catalysis loads active components onto the filter coating, achieving continuous passive regeneration of soot through catalytic oxidation. The entire motor vehicle journey covers multiple typical stages, including cold start, low-speed urban driving, medium-speed suburban driving, and high-speed driving, with exhaust temperatures ranging from room temperature to over 700°C, and particulate matter concentration, gas composition, and flow rate all fluctuating continuously. Existing technologies typically optimize the electrostatic charging system and the catalytic coating as two independent modules. However, these two subsystems are mutually restrictive: under high-temperature conditions, the gas breakdown voltage decreases, requiring a reduction in pulse voltage to avoid arc discharge, leading to a decrease in particulate charge and weakened electrostatic attraction between the particles and the catalytic coating, reducing the probability of close contact; traditional coatings are predominantly mesoporous, with pore sizes mismatched with the bulk soot particle size, preventing soot from entering the mesopores and resulting in low utilization of numerous internal active sites; the overall efficiency, when the independent optimization results of the two subsystems are superimposed, is lower than the theoretical optimal level of joint optimization. Figure 1 As shown, the specific steps include:

[0071] Step 1: Generate a multivariate breakdown voltage prediction model based on exhaust temperature, gas density, and gas component concentration;

[0072] Step 2: Calculate the safe operating window of pulse voltage under various operating conditions based on the multivariable breakdown voltage prediction model, and then generate a feedforward control lookup table based on the rate of change of exhaust temperature according to the rate of change of throttle opening and the rate of change of speed.

[0073] Step 3: Obtain programmed temperature oxidation data and electron microscopy observation data of the catalytic coating, generate a quantitative mapping model between coating microstructure parameters and contact point density, and determine the parameters of the nanoprotrusion array.

[0074] Step 4: Using the nano-protrusion array parameters determined by the quantitative mapping model, selectively load catalytic active components onto the nano-protrusion tips on the catalytic coating surface through atomic layer deposition.

[0075] Step 5: Conduct a full-condition performance acquisition experiment based on the multi-dimensional combination of operating parameters and system parameters, record the particulate matter capture efficiency under each combination, and generate a joint performance database.

[0076] Step 6: Using the joint performance database as the training set, optimize the kernel function hyperparameters using maximum likelihood estimation and use the cross-validation results as the accuracy acceptance criterion. Obtain the joint performance surrogate model through regression training.

[0077] Step 7: Use the joint performance proxy model to solve for the joint optimal combination of charge parameters and coating contact density at each operating point to generate the joint optimal configuration matrix;

[0078] Step 8: Based on the real-time exhaust temperature, particulate matter concentration, and exhaust flow rate, look up the values ​​in the joint optimal configuration matrix and interpolate to synchronously output the charge pulse control signal and alternating electric field control signal to the corresponding execution unit.

[0079] For step 1, a multivariate breakdown voltage prediction model is generated based on exhaust temperature, gas density, and gas component concentration. The exhaust temperature range is 100℃ to 800℃, and the exhaust component concentration is... concentration, Concentration and Measured data of corona initiation voltage and breakdown voltage under multiple level conditions were collected for each concentration. The temperature-corrected form of the Peek formula was used to fit the relationship between initiation voltage, breakdown voltage, temperature, and gas density. The inputs were exhaust temperature and gas density, and the outputs were the fitted values ​​of initiation voltage and breakdown voltage at the corresponding temperatures. The HC concentration in the fitting residuals was then analyzed. The effect of concentration was regressed and calibrated to obtain a concentration correction coefficient. This coefficient was then superimposed onto the fitting result to generate a result incorporating exhaust temperature, gas density, HC concentration, and... A multivariate breakdown voltage prediction model that takes concentration as input and outputs predicted initiation voltage and predicted breakdown voltage. The Peek formula is used to calculate the initiation field strength of the conductor corona; its temperature correction is not directly based on temperature. Instead of using it as a variable, it is determined by the relative density of air. This indirectly reflects the effect of temperature. The Peek formula with temperature correction is expressed as:

[0080] ;

[0081] in, Indicates the starting voltage. This represents the temperature-corrected relative density of air. Indicates the reference gas breakdown voltage. It is expressed as the radius of the line electrode.

[0082] It should be noted that the above regression calibration refers to using the residual between the measured breakdown voltage and the fitted value at each operating point as the dependent variable, based on the temperature correction fitting of the Peek formula, and using the HC concentration at the corresponding operating point and Concentration was used as the independent variable in a multiple linear regression to obtain the independent correction coefficients for each gas component on the breakdown voltage. The concentration correction coefficients reflect the influence of electronegative gas molecules in the exhaust gas on the electron adhesion rate, exhibiting a consistent correction direction across different temperature points. Before participating in fitting and regression calibration, the input variables of the multivariate breakdown voltage prediction model, including exhaust temperature, gas density, HC concentration, and... Concentrations were standardized using Z-scores to eliminate the influence of differences in the dimensions of the input variables on the estimation of regression coefficients.

[0083] For step 2, the safe operating window for pulse voltage under various operating conditions is calculated based on a multivariate breakdown voltage prediction model. Then, a feedforward control lookup table is generated by predicting the rate of change of exhaust temperature based on the throttle opening and engine speed changes. 85% of the predicted breakdown voltage is used as the upper limit of the real-time safe voltage, and 1.2 times the predicted initial voltage is used as the lower limit of the effective charge. The interval between these two values ​​constitutes the safe operating window for pulse voltage under that condition. The center value of the safe operating window is taken as the target pulse peak voltage. A safety upper limit lower than the predicted breakdown voltage is used to reserve a safety margin, while an effective charge lower limit higher than the predicted initial voltage is used to ensure that the pulse voltage maintains effective corona discharge.

[0084] To address the issue of target voltage response lag during rapid transient changes in exhaust temperature, step 2 further includes acquiring the throttle opening signal and engine speed signal output by the engine ECU, and performing forward prediction of the exhaust temperature change rate based on the throttle opening change rate and engine speed change rate. Specifically, this includes the following steps:

[0085] Step 2.1: Obtain the throttle opening value and engine speed value at the current moment, and calculate the time change rate of throttle opening and the time change rate of engine speed.

[0086] Step 2.2: Based on the pre-calibrated mapping relationship between the throttle opening change rate, the speed change rate, and the exhaust temperature change rate, calculate the estimated value of the current exhaust temperature change rate.

[0087] Step 2.3 compares the estimated rate of change of exhaust temperature with a preset threshold. When the estimated rate of change exceeds the preset threshold, the predicted exhaust temperature for future moments is calculated based on the estimated rate of change and a preset look-ahead time window. The look-ahead time window is a preset fixed value, determined based on engine dynamic response characteristics, and ranges from 1 to 3 seconds. The predicted exhaust temperature is then substituted into the multivariate breakdown voltage prediction model to obtain the corresponding pulse voltage safety operating window, allowing for advance adjustment of the target pulse peak voltage.

[0088] Step 2.4: Summarize the target pulse peak voltages under various combinations of exhaust temperature, atmosphere conditions, and operating parameters to generate a feedforward control lookup table that combines temperature, atmosphere, and operating conditions.

[0089] For step 3, the programmed temperature-rise oxidation data and electron microscopy observation data of the catalytic coating are obtained, a quantitative mapping model of coating microstructure parameters and contact point density is generated, and the parameters of the nano-protrusion array are determined. The difference in carbon soot oxidation peak temperature under two conditions is calculated, and the difference in carbon soot oxidation peak temperature is used as a contact efficiency index. High-resolution transmission electron microscopy observation data of the same catalytic coating sample are obtained, the contact angle distribution between carbon soot particles and the coating surface is statistically analyzed, and the number of close contact points is obtained based on the number of contact points with contact angles less than a preset angle threshold. Peak temperature difference and close contact point number data are obtained for multiple groups of coating samples with different microstructures, and regression fitting is performed on the peak temperature difference and close contact point density to generate a quantitative relationship curve between peak temperature difference and close contact point density. Based on the distribution of carbon soot equivalent sphere diameter and the characteristic size of protrusions on the coating surface, the geometric interlocking conditions are analyzed. Specifically, the following steps are also included:

[0090] Step 3.1: Obtain the equivalent sphere diameter distribution data of carbon soot particles to determine the main particle size range.

[0091] Step 3.2: Based on the main particle size range, calculate the characteristic size constraint of the nano-protrusion array on the coating surface required to achieve multi-point embedded contact of the soot particles. The constraint is that the characteristic height of the nano-protrusion is not less than a preset ratio of the lower limit of the equivalent diameter of the soot sphere, and the spacing between the nano-protrusions is not greater than the upper limit of the equivalent diameter of the soot sphere, so that a single soot particle can simultaneously form contact with multiple adjacent nano-protrusions.

[0092] Step 3.3: Within the parameter range that satisfies the geometric interlocking conditions, based on the quantitative relationship curve between the peak temperature difference and the density of the close contact points, determine the optimal feature size range and spatial density parameters of the nano-protrusion array required to make the density of the close contact points reach the target value, and generate a quantitative mapping model between the coating microstructure parameters and the contact point density.

[0093] For step 4, the parameters of the nanoprotrusion array are determined based on a quantitative mapping model of coating microstructure parameters and contact point density. Catalytic active components are selectively loaded onto the tips of the nanoprotrusions on the catalytic coating surface via atomic layer deposition, allowing active sites to preferentially accumulate at locations with the highest probability of contact with soot. Peak temperature difference and continuous regeneration rate data are obtained for different combinations of nanoprotrusion density and active component loading. Response surface regression analysis is performed using nanoprotrusion density and active component loading as input variables, and peak temperature difference and continuous regeneration rate as response variables, to generate a response surface prediction relationship between coating preparation parameters and catalytic contact efficiency. It should be noted that the aforementioned selective loading refers to utilizing the surface energy difference between the nanoprotrusion tips and flat areas during atomic layer deposition, allowing catalytic active component precursors to preferentially nucleate at the tip locations. The nanoprotrusion tips have a small radius of curvature and high surface free energy, resulting in higher adsorption energy for precursor molecules at these locations compared to flat areas, thus achieving preferential enrichment of active components at the tip locations within the same number of deposition cycles.

[0094] For step 5, a full-condition performance acquisition experiment was conducted based on multi-dimensional combinations of operating parameters and system parameters, recording the particulate matter capture efficiency under each combination to generate a joint performance database. For each operating point, the overall particulate matter number removal efficiency and mass removal efficiency were recorded under different combinations of pulse charging parameters and different combinations of catalytic coating contact densities. The operating parameters, charging parameters, coating contact density parameters, and corresponding removal efficiency data for each operating point were summarized to generate a joint performance database of electrostatic enhancement and surface catalysis. It should be noted that the typical temperature ranges covered by the above operating points include the room temperature range corresponding to the cold start stage, the 150℃ to 300℃ range corresponding to urban low speed, the 300℃ to 450℃ range corresponding to suburban medium speed, and the 450℃ to 700℃ range corresponding to high-speed driving. Multiple particulate matter concentration levels and exhaust flow rate levels were combined within each temperature range. The pulse charging parameters include the pulse peak voltage and pulse frequency. The catalytic coating contact density parameter is the close contact point density value output by the quantitative mapping model of coating microstructure parameters and contact point density in step 3.

[0095] Step 6: Using the joint performance database as the training set, maximum likelihood estimation is employed to optimize the kernel function hyperparameters, and cross-validation results are used as the accuracy acceptance criterion. A joint performance surrogate model is obtained through regression training. Specifically, the combined vector of operating parameters and system parameters for each operating point in the electrostatic enhancement and surface catalysis joint performance database is used as input, with the overall particulate matter capture efficiency as the output. Gaussian process regression is used to train the electrostatic enhancement and surface catalysis joint performance database. The operating parameters include exhaust temperature, particulate matter concentration, and exhaust flow rate; the system parameters include peak pulse voltage, pulse frequency, and coating tight contact point density. Before combining the operating parameters and system parameters into the input vector, mean normalization based on the range is performed on exhaust temperature, particulate matter concentration, exhaust flow rate, peak pulse voltage, pulse frequency, and coating tight contact point density to scale each variable to a uniform numerical range, thereby eliminating the influence of dimensional differences on the kernel function distance metric.

[0096] Gaussian process regression optimizes the kernel function hyperparameters using maximum likelihood estimation and uses the negative log-marginal likelihood as the loss function. For the input vector... The joint performance proxy model outputs the predicted mean. and prediction variance ,in The input vector is composed of a combination of operating condition parameters and system parameters. As a point estimate of overall particulate matter capture efficiency, It reflects the uncertainty of the prediction. Given by a weighted linear combination of the observations in the training set:

[0097] ;

[0098] in, For test points The kernel function value vector between all training points has a dimension equal to the number of training samples. ; for The kernel matrix of the training set, its dimensional training set, The element is the first The and the first Kernel function values ​​between training points; To observe the noise variance; for 3D identity matrix; for dimensional training set capture efficiency observation vector; Indicates transpose; This represents the number of training samples; and The training sample index is used. The loss function is the negative log-marginal likelihood.

[0099] ;

[0100] in, for dimensional training set input matrix, The number of training samples. For the input vector dimension, For determinant, Pi is the mathematical constant of a circle. Through... The gradient of the kernel function hyperparameters is calculated and optimized using maximum likelihood estimation to determine the optimal kernel function hyperparameters. The prediction error at independent test points is evaluated through cross-validation. When the prediction error meets the preset accuracy requirements, a joint performance surrogate model is generated.

[0101] Step 7: Utilize the joint performance surrogate model to solve for the joint optimal combination of charge parameters and coating contact density at each operating point, generating a joint optimal configuration matrix. Specifically, based on the joint performance surrogate model, for each operating point, with the objective of maximizing overall particulate matter removal efficiency and using the pulse voltage safe operating window from Step 2 as the charge parameter constraint, solve for the joint optimal combination of charge parameters and coating contact density for each operating point. Arrange the joint optimal combinations of all operating points according to three dimensions: exhaust temperature, particulate matter concentration, and exhaust flow rate, generating a joint optimal configuration matrix of charge parameters and coating contact density covering all operating conditions.

[0102] Step 8: Based on the real-time exhaust temperature, particulate matter concentration, and exhaust flow rate, perform table lookup and interpolation in the joint optimal configuration matrix, and simultaneously output the charging pulse control signal and alternating electric field control signal to the corresponding execution unit. Specifically, using the real-time collected exhaust temperature, particulate matter concentration, and exhaust flow rate as input, perform table lookup and interpolation in the joint optimal configuration matrix of charging parameters and coating contact density, and output the corresponding target values ​​for charging pulse peak voltage, pulse frequency, and catalytic contact auxiliary alternating electric field power. The charging pulse control signal and alternating electric field control signal are then synchronously sent to the corresponding execution unit.

[0103] In this embodiment of the application, in order to achieve rapid protection when arc discharge occurs and accumulate abnormal data for model correction, arc discharge protection is also included in step 8. Specifically, this includes the following steps:

[0104] Step 8.1: Continuously monitor the waveform data of the discharge current pulse at a microsecond-level sampling frequency in the discharge circuit.

[0105] Step 8.2: Real-time discrimination is performed based on the characteristic differences between the arc discharge waveform and the normal corona discharge waveform. The current pulse of a normal corona discharge exhibits a single-peak shape with a steep rise edge and exponentially decaying fall edge; the current waveform of an arc discharge exhibits a sustained high amplitude and oscillating characteristics. When arc discharge characteristics are detected, protection shutdown is triggered, cutting off the pulse voltage output. The rise time of the current pulse in a normal corona discharge does not exceed 100 ns, the pulse half-width does not exceed 500 ns, the peak current amplitude is in the range of milliamperes to tens of milliamperes, and the waveform exhibits a single peak followed by exponential decay; the current amplitude of an arc discharge exceeds a preset multiple threshold (taken as 10 times) of the peak current of a normal corona discharge, and the duration exceeds a preset time threshold (taken as 1 μs), or the waveform contains an oscillating component with a frequency lower than a preset frequency threshold. When any of the above conditions are met, it is determined to be an arc discharge.

[0106] Step 8.3: Record the exhaust temperature, atmosphere parameters and pulse voltage value corresponding to the moment the electric arc is triggered as abnormal samples, and feed them back to the calibration dataset of the multivariate breakdown voltage prediction model to locally update the correction coefficients of the multivariate breakdown voltage prediction model.

[0107] In this embodiment, to adapt the joint performance proxy model and the joint feedforward control lookup table for temperature, atmosphere, and operating conditions to the aging changes of the catalytic coating over time, a full lifecycle adaptive update step is added in addition to step 8. Specifically, this includes the following steps:

[0108] Step 9.1: During operation, continuously record the operating parameters, charge parameters, and corresponding measured acquisition efficiency data at each moment to generate an incremental operation dataset.

[0109] Step 9.2: Based on the deviation trend between the measured capture efficiency and the predicted value of the joint performance surrogate model in the incremental running dataset, identify the drift direction and drift amount of the contact density parameter caused by the aging of the catalytic coating.

[0110] Step 9.3: Integrate the incremental running dataset into the training set of the joint performance proxy model using online incremental learning, and update the posterior distribution parameters of the joint performance proxy model. Based on the updated joint performance proxy model, resolve the joint optimal combination for each operating point, update the joint optimal configuration matrix of charge parameters and coating contact density and the joint feedforward control lookup table of temperature, atmosphere and operating conditions, and generate a life-cycle adaptive joint optimal configuration.

[0111] Furthermore, online incremental learning is used to incorporate the incremental running dataset into the training set of the joint performance surrogate model. A recursive Bayesian update method is employed, appending the kernel function value vector corresponding to newly added sample points to the kernel matrix. The posterior mean and posterior covariance parameters of the joint performance surrogate model are then updated using a Bayesian formula, eliminating the need for retraining the entire dataset. When identifying the drift direction and amount of contact density parameters caused by catalytic coating aging, the residual sequence between the measured trapping efficiency and the predicted value of the joint performance surrogate model is used as input. A linear trend regression is used to fit the slope of the residual change with cumulative running time, and the slope direction is taken as the drift direction. The residual mean offset is then back-calculated into the contact density parameter space to obtain an estimate of the drift amount.

[0112] This invention incorporates exhaust temperature, gas density, and gas component concentration into a unified breakdown voltage prediction framework through a multivariate breakdown voltage prediction model. It calculates the safe operating window of pulse voltage in real time under various temperature and atmospheric conditions, enabling the charge parameters to track continuous changes in exhaust conditions and avoiding response lag under fixed voltage or step-by-step switching modes. Combined with forward prediction of ECU signals, it adjusts the target voltage in advance during rapid temperature changes, further reducing the arc risk time window and ensuring that the electrostatic charge operates stably within the safe operating window of pulse voltage across the entire temperature range.

[0113] By using a quantitative relationship curve between peak temperature difference and density of close contact points, the contact efficiency of the catalytic coating is transformed from a qualitative description into a quantifiable microstructure parameter. Based on the geometric interlocking conditions of the equivalent diameter of soot spheres and the characteristic size of nanoprotrusions, the range of coating surface structure parameters that enable multi-point embedded contact of soot particles is determined, thereby geometrically improving the proportion of close contact between soot and the catalytic coating. Active components are selectively enriched at the tips of nanoprotrusions through atomic layer deposition, concentrating catalytic active sites at the positions with the highest probability of contact with soot, thus improving the effective utilization rate of active sites.

[0114] By incorporating the charge parameters and coating contact density into a unified nonlinear optimization framework through a Gaussian process regression joint performance surrogate model, the joint optimal configuration matrix of charge parameters and coating contact density eliminates the superposition efficiency loss caused by the intertwining of parameter spaces when the two subsystems are optimized independently. Real-time collaborative control synchronously outputs charge pulse control signals and alternating electric field control signals based on lookup table interpolation, ensuring operation near the joint optimum under all operating conditions. Online incremental learning continuously updates the joint performance surrogate model and the joint optimal configuration matrix of charge parameters and coating contact density, ensuring that the collaborative control effect is maintained continuously during the catalytic coating aging process.

[0115] See Figures 2 to 9This invention discloses the following application example: A heavy-duty diesel truck's after-treatment system needs to continuously capture and treat exhaust particulate matter throughout the vehicle's entire journey during actual road operation. The truck is equipped with an electrostatically enhanced particulate matter trap, and system calibration and deployment were completed in B month of 20XX. The vehicle departs from an urban distribution center and experiences four typical stages: cold start, low-speed urban driving, medium-speed suburban driving, and high-speed driving. The exhaust temperature gradually rises from room temperature to over 650℃, and the particulate matter concentration and exhaust flow rate continuously change with the operating conditions. The system needs to coordinate and regulate the charging pulse parameters and the auxiliary electric field of the catalytic coating in real time throughout the entire journey to ensure that the capture efficiency remains at the optimal combined level.

[0116] Step 1 corresponds to the generation process of the multivariate breakdown voltage prediction model. Within the exhaust temperature range of 100℃ to 800℃, different... , and Concentration levels were monitored, and measured values ​​of corona initiation voltage and breakdown voltage were collected at each operating point. Using exhaust temperature and gas density as inputs, a preliminary fitting was performed using the Peek formula with temperature correction. The fitting residuals at each operating point were then analyzed using HC concentration and... Concentration was used as the independent variable in a multiple linear regression to obtain the concentration correction coefficient. All input variables were Z-score standardized before fitting. Table 1 below shows the measured data and fitting correction results for four typical temperature points.

[0117] 150 0.74 180 6.2 18.4 17.9 18.3 280 0.51 310 9.5 14.7 14.2 14.6 450 0.38 420 12.1 11.2 10.6 11.1 620 0.29 550 14.8 8.6 8.0 8.5

[0118] Table 1

[0119] In Table 1, the deviations between the predicted and measured values ​​after concentration correction are all within 0.2 kV, indicating that the concentration correction coefficient effectively compensates for HC and The multivariate breakdown voltage prediction model has the ability to predict the impact of electron adhesion rate across temperature ranges.

[0120] Step 2 calculates the safe operating window for pulse voltage at each operating point based on the predicted initial voltage and breakdown voltage output from Step 1. The upper limit of safety is set to 85% of the predicted breakdown voltage, the lower limit of effective charge is set to 1.2 times the predicted initial voltage, and the target pulse peak voltage is set to the midpoint of the two values. The throttle opening and engine speed signals from the engine ECU are read synchronously, and the rates of change of opening and engine speed are calculated. The exhaust temperature change rate is estimated through calibration mapping. When the rate of change exceeds a preset threshold, the target voltage is updated in advance based on the predicted temperature at future times using a look-ahead time window. Table 2 below shows the safe operating window and target value for pulse voltage at four temperature points.

[0121] 150 10.8 18.3 12.96 15.56 14.26 280 8.5 14.6 10.20 12.41 11.31 450 6.4 11.1 7.68 9.44 8.56 620 4.9 8.5 5.88 7.23 6.56

[0122] Table 2

[0123] The target pulse peak voltages at the four temperature points in Table 2 above decrease monotonically with increasing temperature, reflecting the constraint of reduced gas breakdown voltage on charge parameters under high-temperature conditions. The target values ​​at each operating point are all between the lower limit of effective charge and the upper limit of safety, preserving sufficient safety margin. The target pulse peak voltages under all combinations of temperature, atmosphere, and operating parameters are summarized to generate a joint feedforward control lookup table for temperature, atmosphere, and operating conditions.

[0124] Step 3 completes the quantitative characterization of the catalytic coating contact efficiency and the analysis of geometric interlocking conditions. Temperature-programmed oxidation experiments were conducted on multiple groups of coating samples with different microstructures. The difference in peak carbon soot oxidation temperature under loose and tight contact conditions was calculated as an indicator of contact efficiency. Simultaneously, the number of tight contact points with contact angles less than a preset threshold was counted using high-resolution transmission electron microscopy, and a quantitative relationship curve between peak temperature difference and tight contact point density was fitted. Based on the equivalent sphere diameter distribution of carbon soot, the geometric interlocking constraints of the nano-protrusion array were determined, as shown in Table 3.

[0125] Lower limit of soot equivalent sphere diameter 85 nm Upper limit of soot equivalent sphere diameter 310 nm Main particle size concentration range 120 to 220 nm Lower limit of nano-protrusion feature height 51 nm (60% of lower limit) Upper limit of nano-protrusion pitch 310 nm Target close contact point density 18 / μm2

[0126] Table 3

[0127] Table 3 illustrates the carbon soot particle size distribution and the geometric interlocking constraints of the nanoprotrusions. Based on a main particle size range of 120 to 220 nm, a nanoprotrusion feature height of no less than 51 nm and a spacing of no more than 310 nm allows a single carbon soot particle to simultaneously form embedded contact with multiple adjacent nanoprotrusions. Based on the quantitative relationship curve between peak temperature difference and the density of close contact points, the optimal feature height range of the nanoprotrusion array required to achieve a close contact point density of 18 points / μm² was determined to be 65 to 90 nm, and the spacing range to be 150 to 250 nm. This generated a quantitative mapping model between the coating microstructure parameters and the contact point density.

[0128] Step 4: Based on the nanoprotrusion array parameters determined in Step 3, selectively load catalytic active components onto the tips of the nanoprotrusions via atomic layer deposition. Peak temperature difference and continuous regeneration rate data are obtained for different combinations of nanoprotrusion density and active component loading. Response surface regression analysis is performed using both as response variables, as shown in Table 4.

[0129] 12 1.2 38 0.41 16 1.8 52 0.63 20 2.4 67 0.82 24 3.0 71 0.88

[0130] Table 4

[0131] Table 4 shows the response surface regression analysis data for the combined nanoprotrusion density and loading. The results indicate that increasing the nanoprotrusion density from 12 nanoprotrusions / μm² to 20 nanoprotrusions / μm² significantly improves both the peak temperature difference and the continuous regeneration rate. Beyond 20 nanoprotrusions / μm², the increase in the response variable tends to level off, indicating a saturation effect. Combining the peak temperature difference and continuous regeneration rate responses, the optimal nanoprotrusion density was determined to be 20 nanoprotrusions / μm², and the active component loading was determined to be 2.4 g / L. The response surface regression analysis then describes the relationship between the coating preparation parameters and the catalytic contact efficiency.

[0132] Step 5 covers four temperature stages across the entire operating range: cold start, urban low-speed (150 to 300°C), suburban medium-speed (300 to 450°C), and high-speed driving (450 to 700°C). Within each stage, multiple particulate matter concentration levels and exhaust flow rates are recorded. The overall particulate matter number removal efficiency and mass removal efficiency under different combinations of pulse charge parameters and coating contact density are recorded, and the results are compiled to generate a combined performance database of electrostatic enhancement and surface catalysis. Table 5 shows typical operating condition samples for the combined performance database.

[0133] 200 85 4.2 13.50 180 18 88 91 350 62 6.8 10.10 160 18 84 87 500 48 9.5 8.00 140 18 79 83 630 35 12.3 6.20 120 18 74 78

[0134] Table 5

[0135] Step 6 uses the joint performance database shown in Table 5 as the training set, and performs mean normalization based on the range on the six variables: exhaust temperature, particulate matter concentration, exhaust flow rate, pulse peak voltage, pulse frequency, and coating tight contact point density. The results are then combined into a 6-dimensional input vector. The overall particulate matter capture efficiency is used as the output, and Gaussian process regression is employed for training. Negative logarithmic marginal likelihood is used as the output. The loss function is defined, and the kernel function hyperparameters are optimized using maximum likelihood estimation. After training, cross-validation is performed on independent test points. When the prediction error meets the preset accuracy requirement, a joint performance surrogate model is generated, outputting the prediction mean for each test point. and prediction variance Taking an exhaust temperature of 420℃, particulate matter concentration of 55 mg / m³, exhaust velocity of 8.1 m / s, pulse peak voltage of 9.20 kV, pulse frequency of 150 Hz, and coating tight contact point density of 18 / μm² as an example, after normalization, the predicted mean is calculated by inputting it into a trained Gaussian process regression model. (i.e., quantity removal efficiency of 82%), prediction variance The low prediction uncertainty indicates that the operating condition falls within the coverage area of ​​the training set.

[0136] Step 7, based on the joint performance proxy model, aims to maximize the overall particulate matter removal efficiency. Using the pulse voltage safety operating window at each operating point in Step 2 as a constraint, it solves for the joint optimal combination of charge parameters and coating contact density at each operating point. The results are then arranged according to three dimensions: exhaust temperature, particulate matter concentration, and exhaust flow rate, generating a joint optimal configuration matrix. Table 6 shows the joint optimal configuration matrix for typical operating points.

[0137] 200 85 4.2 13.50 185 18 89 350 62 6.8 10.10 165 18 85 500 48 9.5 8.00 142 18 80 630 35 12.3 6.20 122 18 75

[0138] Table 6

[0139] Step 8: During the actual operation of the truck, the system continuously monitors the discharge current waveform at a microsecond-level sampling frequency, and collects exhaust temperature, particulate matter concentration, and exhaust flow rate in real time. It then performs lookup interpolation in the joint optimal configuration matrix, simultaneously outputting the target values ​​for the peak voltage of the charged pulse, the pulse frequency, and the catalytic contact auxiliary alternating electric field power. Two control signals are then sent to the corresponding execution units. When arc discharge characteristics (a continuous high-amplitude oscillating waveform) are detected, protection shutdown is triggered, and the exhaust temperature, atmosphere parameters, and pulse voltage value at the trigger moment are recorded as abnormal samples. These are fed back to the calibration dataset of the multivariate breakdown voltage prediction model, and the correction coefficients are locally updated.

[0140] After approximately 3000 hours of cumulative vehicle operation, step 9 continuously records the operating parameters, charge parameters, and measured capture efficiency at each time point, forming an incremental operation dataset. By comparing the deviation trend between the measured capture efficiency and the predicted value of the joint performance surrogate model, it is identified that the density parameter of close contact points caused by coating aging has decreased by approximately 1.8 points / μm². Incremental data is incorporated into the training set using online incremental learning, updating the posterior distribution parameters of the Gaussian process regression model, resolving the joint optimal combination for each operating point, and simultaneously updating the joint optimal configuration matrix and feedforward control lookup table, so that the cooperative control continues to maintain near the optimal level under coating aging conditions.

[0141] Throughout the implementation process, the data starts from the corona discharge measurement data in step 1, and is used to generate the safe operating window in step 2 through a multivariate breakdown voltage prediction model. The coating microstructure characterization data in steps 3 and 4 determine the quantifiable index of catalytic contact efficiency. Step 5 summarizes the joint measurement results of charge parameters and coating contact density into a joint performance database. Step 6 trains a joint performance surrogate model based on this database. Step 7 solves the joint optimal configuration matrix under the constraints of the surrogate model output and the safe operating window. Step 8 achieves real-time collaborative control based on the configuration matrix and continuously corrects the breakdown voltage model through abnormal sample feedback. The surrogate model and configuration matrix are updated through online incremental learning closed-loop. The output data of each step is passed step by step, forming a complete data flow link from offline calibration to real-time control and then to full life cycle adaptation.

[0142] To achieve the above embodiments, this application may further include a particulate matter capture system based on electrostatic enhancement and surface catalysis, comprising: a breakdown voltage prediction module for generating a multivariate breakdown voltage prediction model based on exhaust temperature, gas density, and gas component concentration; a safety window calculation module for calculating the pulse voltage safety operating window under various operating conditions based on the multivariate breakdown voltage prediction model, and then generating a feedforward control lookup table based on the rate of change of exhaust temperature according to the rate of change of throttle opening and the rate of change of engine speed; a coating microstructure mapping module for acquiring programmed temperature-ramp oxidation data and electron microscopy observation data of the catalytic coating, and generating a quantitative mapping model of coating microstructure parameters and contact point density; and a coating preparation parameter module for using the nano-protrusion array parameters determined by the quantitative mapping model to deposit atomic layers on the nano-protrusion tips on the surface of the catalytic coating. The system includes: a selective loading module for catalytically active components; a joint performance database generation module for conducting full-condition performance acquisition experiments based on multi-dimensional combinations of operating parameters and system parameters, recording particulate matter capture efficiency under each combination to generate a joint performance database; a surrogate model training module for using the joint performance database as a training set, optimizing kernel function hyperparameters using maximum likelihood estimation, and using cross-validation results as accuracy acceptance criteria to perform regression training to obtain a joint performance surrogate model; a configuration matrix generation module for using the joint performance surrogate model to solve for the optimal combination of charge parameters and coating contact density at each operating point to generate a joint optimal configuration matrix; and a real-time collaborative control module for looking up and interpolating values ​​in the joint optimal configuration matrix based on real-time exhaust temperature, particulate matter concentration, and exhaust flow rate, and synchronously outputting charge pulse control signals and alternating electric field control signals to the corresponding execution units.

[0143] To implement the above embodiments, this application also discloses an electronic device. The electronic device may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) or a program loaded from a storage device into a random access memory (RAM). Various programs and data required for the operation of the electronic device are also stored in the RAM. The processing unit, ROM, and RAM are interconnected via a bus. An input / output (I / O) interface is also connected to the bus. Typically, the following devices can be connected to the I / O interface: input devices including, for example, a touchscreen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices including, for example, a liquid crystal display (LCD), speaker, vibrator, etc.; storage devices including, for example, magnetic tape, hard disk, etc.; and communication devices. The communication device allows the electronic device to communicate wirelessly or wiredly with other devices to exchange data. Although electronic devices with various devices are shown, it should be understood that it is not required to implement or possess all of the shown devices. More or fewer devices may be implemented or possessed alternatively.

[0144] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer storage medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processing device, it performs the functions defined above in the methods of some embodiments of this disclosure.

[0145] It should be noted that the computer storage medium in some embodiments of this disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof.

[0146] In some embodiments of this disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer storage medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer storage medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0147] In other implementations, clients and servers may communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol), and may interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and end-to-end networks (e.g., ad hoc end-to-end networks), as well as any currently known or future-developed networks.

[0148] The aforementioned computer storage medium may be included in the aforementioned electronic device, or it may exist independently without being assembled into the electronic device. The aforementioned computer storage medium carries one or more programs, which, when executed by the electronic device, enable the electronic device to implement a particulate matter capture method based on electrostatic enhancement and surface catalysis.

[0149] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0150] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. Each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions. Units described in some embodiments of the present disclosure may be implemented in software or hardware. The described units may also be located in a processor, and the names of these units do not necessarily constitute a limitation on the unit itself.

[0151] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0152] All technologies not described in detail in this invention are existing technologies. The above descriptions are merely some preferred embodiments of this disclosure and explanations of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalent features without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A particulate matter capture method based on electrostatic enhancement and surface catalysis, characterized in that, Includes the following steps: Step 1: Generate a multivariate breakdown voltage prediction model based on exhaust temperature, gas density, and gas component concentration; Step 2: Calculate the pulse voltage safe operating window under each operating condition based on the multivariable breakdown voltage prediction model, and then generate a feedforward control lookup table based on the rate of change of exhaust temperature according to the rate of change of throttle opening and the rate of change of speed. Step 3: Obtain programmed temperature oxidation data and electron microscopy observation data of the catalytic coating, generate a quantitative mapping model between coating microstructure parameters and contact point density, and determine the parameters of the nanoprotrusion array. Step 4: Using the nano-protrusion array parameters determined by the quantitative mapping model, selectively load catalytically active components on the nano-protrusion tips of the catalytic coating surface through atomic layer deposition; Step 5: Conduct a full-condition performance acquisition experiment based on the multi-dimensional combination of operating parameters and system parameters, record the particulate matter capture efficiency under each combination, and generate a joint performance database. Step 6: Using the joint performance database as the training set, optimize the kernel function hyperparameters using maximum likelihood estimation and use the cross-validation results as the accuracy acceptance criterion, and obtain the joint performance surrogate model through regression training. Step 7: Use the joint performance proxy model to solve for the joint optimal combination of charge parameters and coating contact density at each operating point to generate a joint optimal configuration matrix; Step 8: Based on the real-time exhaust temperature, particulate matter concentration, and exhaust flow rate, look up values ​​in the joint optimal configuration matrix and interpolate to synchronously output the charge pulse control signal and alternating electric field control signal to the corresponding execution unit.

2. The particulate matter capture method based on electrostatic enhancement and surface catalysis according to claim 1, characterized in that, The generation of the multivariate breakdown voltage prediction model in step 1 includes: The temperature-corrected form of the Peek formula was used to fit the relationship between the initiation voltage and breakdown voltage and temperature and gas density. HC concentration in the fitting residuals and The effect of concentration was regressed and calibrated, and the concentration correction coefficient was obtained and added to the fitting result. Each input variable was standardized using Z-score before participating in fitting and regression calibration.

3. The particulate matter capture method based on electrostatic enhancement and surface catalysis according to claim 1, characterized in that: In step 2, the pulse voltage safe operating window is determined by taking 85% of the predicted breakdown voltage as the upper limit of the real-time safe voltage and 1.2 times the predicted starting voltage as the lower limit of the effective charge. The interval between the two is the pulse voltage safe operating window, and the center value of the pulse voltage safe operating window is taken as the target pulse peak voltage.

4. The particulate matter capture method based on electrostatic enhancement and surface catalysis according to claim 3, characterized in that, Step 2, which generates a feedforward control lookup table based on look-forward prediction of engine ECU signals, includes: Step 2.1: Obtain the throttle opening value and engine speed value at the current moment, and calculate the time change rate of throttle opening and the time change rate of engine speed; Step 2.2: Based on the pre-calibrated mapping relationship between the throttle opening change rate, the speed change rate, and the exhaust temperature change rate, calculate the estimated value of the current exhaust temperature change rate. Step 2.3: Compare the estimated value of the exhaust temperature change rate with a preset threshold. When the estimated value exceeds the preset threshold, calculate the predicted value of the exhaust temperature at future times based on the estimated change rate and the preset look-ahead time window. Substitute the predicted value of the exhaust temperature into the multivariate breakdown voltage prediction model to obtain the corresponding pulse voltage safe operation window. Step 2.4: Summarize the target pulse peak voltages under various combinations of exhaust temperature, atmosphere conditions, and operating parameters to generate a combined feedforward control lookup table for temperature, atmosphere, and operating conditions.

5. The particulate matter capture method based on electrostatic enhancement and surface catalysis according to claim 1, characterized in that, The quantitative mapping model between coating microstructure parameters and contact point density in step 3 includes: The difference in carbon soot oxidation peak temperature of the catalytic coating under loose contact conditions and tight contact conditions is calculated as an indicator of contact efficiency. The number of close contact points with contact angles less than a preset angle threshold was counted based on electron microscopy observation data. Regression fitting was performed on the peak temperature difference and the density of close contact points of multiple groups of coating samples with different microstructures to generate quantitative relationship curves; Geometric interlocking conditions were analyzed based on the diameter distribution of carbon soot equivalent spheres and the characteristic size of nano-protrusions on the coating surface. A quantitative mapping model is generated by determining the characteristic size range and spatial density parameters of the nanoprotrusion array that enable the density of close contact points to reach the target value.

6. The particulate matter capture method based on electrostatic enhancement and surface catalysis according to claim 5, characterized in that: The geometric interlocking condition is that the characteristic height of the nanoprotrusion is not less than a preset ratio of the lower limit of the equivalent diameter of the soot sphere, and the spacing between the nanoprotrusions is not greater than the upper limit of the equivalent diameter of the soot sphere, so that a single soot particle can simultaneously form contact with multiple adjacent nanoprotrusions; within the parameter range that satisfies the geometric interlocking condition, the catalytic active component is selectively loaded at the tip of the nanoprotrusion through atomic layer deposition, and the surface energy difference between the tip and the flat area is used to make the active component precursor preferentially nucleate at the tip position; By systematically varying the coating preparation process parameters using response surface methodology, the predicted relationship between the preparation parameters and the catalytic contact efficiency was generated through regression analysis.

7. The particulate matter capture method based on electrostatic enhancement and surface catalysis according to claim 1, characterized in that, Step 5, which involves generating the joint performance database, includes: A full-condition combined experiment was conducted under multiple temperature ranges, concentration gradients, and flow rates, considering exhaust temperature, particulate matter concentration, and exhaust flow rate. Under each condition combination, the peak pulse voltage, pulse frequency, and density of tight contact points of the coating were adjusted, and the corresponding quantity removal efficiency and mass removal efficiency were recorded simultaneously. The results were then compiled into a combined performance database.

8. The particulate matter capture method based on electrostatic enhancement and surface catalysis according to claim 1, characterized in that, Step 8 further includes arc discharge protection, used to provide rapid protection and accumulate abnormal data when arc discharge occurs, including: The waveform data of the discharge current pulse is continuously monitored at a microsecond-level sampling frequency; Real-time discrimination is based on the difference between the continuous high amplitude and oscillation characteristics of arc discharge waveform and the single-peak exponential decay pattern of normal corona discharge. When an arc discharge characteristic is detected, the protection is triggered to shut down. The exhaust temperature, atmosphere parameters and pulse voltage value corresponding to the moment the arc is triggered are recorded as abnormal samples and fed back to the calibration dataset of the multivariate breakdown voltage prediction model described in step 1 for local update of correction coefficients.

9. The particulate matter capture method based on electrostatic enhancement and surface catalysis according to claim 8, characterized in that, It also includes a full lifecycle adaptive update step: Continuously record operating parameters, charge parameters, and measured acquisition efficiency data at each moment to generate an incremental operation dataset; Based on the deviation trend between the measured capture efficiency in the incremental running dataset and the predicted value of the joint performance proxy model, the drift direction and drift amount of the contact density parameter caused by catalytic coating aging are identified. The incremental running dataset is incorporated into the training set of the joint performance proxy model using online incremental learning, the posterior distribution parameters are updated, and the joint optimal configuration is resolved using the feedforward control lookup table, and the joint optimal configuration matrix is ​​updated.

10. A particulate matter capture system based on electrostatic enhancement and surface catalysis, characterized in that, include: The breakdown voltage prediction module is used to generate a multivariate breakdown voltage prediction model based on exhaust temperature, gas density, and gas component concentration. The safety window calculation module is used to calculate the pulse voltage safety operation window under various operating conditions based on the multivariable breakdown voltage prediction model, and then generate a feedforward control lookup table based on the rate of change of exhaust temperature according to the rate of change of throttle opening and the rate of change of speed. The coating microstructure mapping module is used to acquire temperature-programmed oxidation data and electron microscopy observation data of the catalytic coating, and generate a quantitative mapping model between coating microstructure parameters and contact point density. The coating preparation parameter module is used to selectively load catalytically active components on the tips of nano-protrusions on the surface of the catalytic coating by atomic layer deposition using the nano-protrusion array parameters determined by the quantitative mapping model. The joint performance database generation module is used to conduct full-condition performance acquisition experiments based on multi-dimensional combinations of operating parameters and system parameters, record the particulate matter capture efficiency under each combination, and generate a joint performance database. The surrogate model training module is used to obtain the joint performance surrogate model by using the joint performance database as the training set, optimizing the kernel function hyperparameters by maximum likelihood estimation and using the cross-validation results as the accuracy acceptance criteria for regression training. The configuration matrix generation module is used to solve the joint optimal combination of charge parameters and coating contact density at each working point using the joint performance proxy model to generate a joint optimal configuration matrix. The real-time collaborative control module is used to look up and interpolate values ​​in the joint optimal configuration matrix based on the real-time exhaust temperature, particulate matter concentration, and exhaust flow rate, and synchronously output charge pulse control signals and alternating electric field control signals to the corresponding execution units.

Images