Diesel vehicle exhaust particle separation device based on dual-frequency sound field and dynamic optimization method

Abstract

This application discloses a diesel vehicle exhaust particulate separation device and dynamic optimization method based on a dual-frequency sound field. The device has a dual-frequency broadband sound source installed at the inlet of the separation chamber and a dual-wavelength laser monitoring component installed at the outlet, directly connected to the vehicle's CAN bus via a controller. Overcoming the lag of traditional single-frequency blind scanning, it proposes a feedforward initial frequency-feedback frequency modulation closed-loop control strategy: millisecond-level feedforward initial frequency preset is achieved by analyzing engine speed and load signals using a programmed MAP library; red and blue dual-wavelength lasers are used to track the transmittance changes of large and small particles respectively, and a fractional-order update law with an anti-overshoot exponent is introduced to perform microsecond-level dynamic closed-loop optimization of the sound frequency; the power of the fundamental frequency and high-frequency harmonics is dynamically allocated according to the differences in particle concentration of different particle sizes. It possesses functions of prior perception, rapid response, and prevention of getting trapped in local optima, improving the separation efficiency of a wide range of particulate matter of all sizes and effectively reducing the risk of clogging and replacement cycle of particulate filters.

Description

Technical Field

[0001] This invention relates to the field of particulate matter separation technology, specifically to a diesel vehicle exhaust particulate separation device and dynamic optimization method based on a dual-frequency sound field. Background Technology

[0002] Mining explosion-proof diesel trucks are the main power equipment for trackless auxiliary transportation in my country's coal mines. Due to the special requirements of explosion-proof safety modifications (such as the addition of intake flame arresters and exhaust cooling systems), their emission performance is relatively deteriorated, with high concentrations of particulate matter (PM) in the exhaust gas, and containing a large amount of submicron-sized fine particulate matter. Meanwhile, the engine's operating condition is also a key factor affecting the amount and characteristics of particulate matter emissions in the exhaust gas. Particulate matter (PM) in diesel engine exhaust mainly includes soot, hydrocarbons, soluble organic matter, and sulfides. PM particle size is generally below 10 μm, of which fine particulate matter (PM2.5) with a particle size ≤2.5 μm accounts for more than 95% of the total particulate matter emitted from diesel engine exhaust. The particle size determines its impact on atmospheric visibility and human health. Nuclear modal particles with a diameter less than 50 nm can not only reach the lungs directly through respiration and deposit there, but can also enter the bloodstream and flow to the brain and heart, causing serious harm to the human respiratory and immune systems. Particulate matter larger than 50 nm in diameter is a significant factor contributing to reduced atmospheric visibility. Currently, particulate filters (DPFs) are a mandatory requirement for controlling particulate matter emissions from mine explosion-proof diesel vehicles. Separating particulate matter from exhaust gases is crucial for improving equipment efficiency, protecting worker safety and health, and protecting the environment.

[0003] Acoustic agglomeration separation technology, as a pretreatment method, uses a high-intensity sound field to cause fine particulate matter in exhaust gas to collide and agglomerate into larger particles. It has been applied in industrial dust removal and environmental defogging, but its use in diesel vehicle exhaust particulate separation is still limited. Furthermore, most existing acoustic agglomeration separation devices employ fixed parameters or simple fixed-step single-frequency scanning (e.g., increasing the frequency by 100Hz every 20 seconds and observing changes in transmittance). This type of control logic has fatal flaws: First, it suffers from severe lag. The operating conditions of mining diesel vehicles (such as gear shifting, hill climbing, and starting) change rapidly, with exhaust gas velocity, particulate concentration, and particle size distribution changing on the order of seconds or even milliseconds. Traditional 20-second / step scanning algorithms often fail to find the optimal frequency before the engine operating conditions change, resulting in sound field parameters that are always half a beat behind, or even having a counterproductive effect. Second, it is prone to getting trapped in local optima. Simple gradient hill climbing algorithms are easily misled by local concentration decreases when particulate concentration fluctuates, stopping at non-optimal frequencies. Finally, it suffers from the limitations of single-frequency operation. The mass particle size distribution of diesel engine exhaust particles exhibits a bimodal distribution (nuclear mode and agglomeration mode). Single-frequency sound waves can usually only produce the best entrainment effect on particles within a specific size range, making it difficult to achieve efficient agglomeration and separation of particles with a wide range of sizes.

[0004] Therefore, how to design an intelligent sound field separation particulate matter device with prior perception capability, ultra-fast response speed and dual-frequency collaborative control to significantly reduce the blockage of DPF micropores by fine particles, thereby reducing exhaust back pressure and extending the DPF regeneration cycle, is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the core objective of this invention is to provide a diesel vehicle exhaust particulate separation device and dynamic optimization method based on a dual-frequency sound field. This invention breaks through the traditional single optical feedback mechanism by introducing engine operating condition feedforward control and innovatively designing a fractional-order gradient variable-step-size optimization algorithm, which can accurately and rapidly track and adjust the optimal agglomeration and separation sound field.

[0006] In a first aspect, embodiments of this application provide a diesel vehicle exhaust particulate separation device based on a dual-frequency sound field, comprising: The separation chamber (2) is connected at one end to the diesel engine exhaust port and at the other end to the particulate filter (4), serving as a pre-separation chamber before the exhaust gas enters the particulate filter; A particulate filter (4) is located downstream of the separation chamber (2) and is used to capture particulate matter in the exhaust gas; A sound source (1) is installed on the side wall at the entrance of the separation chamber (2), with the sound-emitting end facing the interior of the separation chamber (2), and is used to generate a high-intensity adjustable sound field in the separation chamber (2); the sound source (1) is a dual-frequency broadband sound source, and simultaneously outputs a fundamental frequency. and high frequency harmonics It also supports independent adjustment of the sound power of each component; The laser monitoring component includes a light source module (3) and a power meter module (5) mounted on the cavity wall of the separation chamber (2) near the particulate filter (4), with its optical path traversing the airflow channel; The control, adjustment and power supply module (6) is electrically connected to the laser monitoring component and the sound source (1), respectively; The vehicle status perception interface connects to the on-board CAN bus of the diesel vehicle to obtain the engine's current speed and load signals in real time. These characteristics determine the amount and properties of particulate matter in the exhaust gas, providing an effective reference for sound field adjustment. The control and power supply module (6) has a built-in feedforward-feedback dual-loop dynamic optimization algorithm, which is used to extract the rotation speed and load signals to generate feedforward frequency commands, and combined with the transmittance data fed back by the laser monitoring component, the fractional gradient variable step size algorithm is used to perform closed-loop dynamic optimization of the working parameters of the sound source (1).

[0007] Preferably, the cross-sectional area of ​​the separation chamber (2) is larger than that of the exhaust pipe, and an expansion structure is formed inside; the inner wall surface is provided with a polytetrafluoroethylene anti-stick coating; the bottom of the separation chamber (2) is provided with an openable and closable ash discharge port for periodically cleaning large particle agglomerates that have condensed and settled under the action of the sound field.

[0008] Preferably, the light source module (3) is a multi-wavelength multiplexed laser that emits red and blue light, and the power meter module (5) is a corresponding multi-channel spectral power meter used to simultaneously monitor the transmittance at different wavelengths in order to distinguish the concentration changes of submicron particles and micron-sized particles.

[0009] Preferably, the sound source (1) is a Hartmann whistle pneumatic sound source or a compression driver with dual resonant cavities, and its total sound frequency range covers 500Hz-10kHz, and the total sound pressure level generated in the separation chamber (2) is not less than 140dB.

[0010] Preferably, the control, regulation, and power supply module (6) includes: The signal processing unit is used to convert optical electrical signals into transmittance values ​​and parse CAN messages; The storage unit is pre-stored with a 3D MAP diagram of the feedforward frequency and related information based on different operating conditions, as well as a dynamic optimization algorithm. The decision-making unit executes a feedforward-feedback dual-loop dynamic optimization algorithm to generate frequency and power adjustment commands for the dual-frequency sound source; The drive unit drives the sound source to produce sound according to the instructions of the decision unit; The power supply unit provides power to all components.

[0011] Secondly, embodiments of this application provide a dynamic optimization method for enhancing the performance of a diesel vehicle particulate filter using the device described in the first aspect, characterized by comprising the following steps: S1, Feedforward Preset: Before the diesel engine exhaust gas enters the separator, the control, regulation, and power supply module reads the engine speed at that moment via the CAN bus. and load By querying a pre-calibrated three-dimensional feedforward MAP diagram with speed and load as inputs, the a priori optimal frequency pair under the current operating conditions is obtained, and the fundamental frequency of the sound source is set accordingly. High-frequency harmonics and initial total power; S2. Performance Evaluation: The laser monitoring component is evaluated based on the sampling period. The system collects transmittance data, controls the adjustment and power supply modules to calculate the transmittance improvement rate, and constructs a separation performance evaluation function. ; S3, Dynamic Fine-tuning: If If the preset threshold is not met, based on the separation performance evaluation function The gradient is used to introduce a fractional update law with an anti-overshoot exponent, and the frequency correction step size is calculated for the fundamental frequency. The preset threshold is obtained from the 3D MAP map based on the current operating conditions. S4. Dual-frequency allocation: The fundamental frequency is dynamically adjusted based on the difference in transmittance between the red and blue wavelengths. With the high frequency harmonics The power allocation ratio; S5. Interception and Cleaning: After the exhaust gas is separated by the dual-frequency dynamic sound field, some large-diameter particles settle at the bottom of the separation chamber, while the rest enter the particulate filter with the airflow and are efficiently captured. The settled matter is cleaned regularly through the ash discharge port.

[0012] Preferably, in step S1, the feedforward algorithm is as follows: the control adjustment and power supply module uses an interpolation algorithm to query the built-in prior 3D MAP map. The corresponding coordinates directly output the current theoretically optimal fundamental frequency. and high frequency harmonics As the initial sound parameters of the sound source.

[0013] Preferably, in step S2, the separation performance evaluation function The calculation formula is: ; in, for The transmittance of red light at any given moment. for The transmittance of blue light at any given time. To increase the rate of red light transmittance, , The weighting coefficients and , This is a correction and compensation term for contamination attenuation in optical lenses. Preferably, the fractional-order update law with an anti-overshoot index term is as follows: First, calculate the gradient of the performance evaluation function with respect to the fundamental frequency. ; Next, update the base frequency using the following formula: ; According to high frequency harmonics With fundamental frequency Update it according to the preset multiple relationship; in, Based on the step size magnification factor, For fractional order and This is used to increase the step size when the gradient is extremely small to prevent getting trapped in local optima; The overshoot attenuation coefficient is used to suppress step size divergence when the gradient is large. It is a symbolic function.

[0014] Preferably, the rule for dual-frequency allocation in step S4 is as follows: when the rate of increase in blue light transmittance is lower than the rate of increase in red light transmittance, the controller increases the high-frequency harmonics while maintaining the total acoustic power unchanged. The power allocation weights are adjusted, and the fundamental frequency is reduced. The power allocation weight is adjusted accordingly; otherwise, it is adjusted in the opposite direction.

[0015] Compared with the prior art, the present invention has the following significant advantages: (1) Innovatively, a feedforward control architecture with vehicle state perception interface and three-dimensional MAP is introduced to directly obtain engine load and speed from CAN bus. In the instant when the concentration and flow rate of exhaust particulate matter change suddenly (within 20ms), the sound source frequency can be adjusted to the theoretical optimal range, completely eliminating the dead time of the system. (2) A fractional update law with an anti-overshoot exponent is introduced. When the gradient is very small (suspected local optimum), the amplification characteristic of the fractional order is used to give the system a very strong ability. When the gradient is very large, the exponential decay term is used to strongly suppress the step size, ensuring that the system smoothly converges to the global optimum frequency, so that the separation efficiency is always kept at the extreme state of more than 90%. (3) In response to the bimodal distribution characteristics of particulate matter in exhaust gas, which exhibits both "accumulation mode" and "nuclear mode", the system employs simultaneous monitoring with dual wavelengths of red light (sensitive to large particles) and blue light (sensitive to small particles), and uses a dual-resonant cavity sound source to simultaneously output the fundamental frequency and high-frequency harmonics. The system can dynamically determine the separation bottleneck of large and small particles based on the difference in transmittance between red and blue light, and intelligently reallocate the ratio of high and low frequency sound power, thus overcoming the limitation that single-frequency sound waves can only be effective for a single particle size range. (4) The separation chamber of this application not only adopts an expansion and speed reduction design to extend the residence time of particles in the sound field, but also innovatively uses a polytetrafluoroethylene (PTFE) anti-stick coating in conjunction with the bottom ash discharge port. Large particles settle directly under gravity and are discharged through the ash discharge port, which greatly reduces the physical interception load of the downstream particulate filter (DPF), extends the regeneration cycle of the DPF by more than 300%, and effectively reduces the overall operation and maintenance costs of diesel vehicles. Attached Figure Description

[0016] Exemplary embodiments of the present invention can be more fully understood by referring to the accompanying drawings. The drawings are provided to further illustrate the embodiments of this application and form part of the specification. They are used together with the embodiments of this application to explain the present invention and do not constitute a limitation thereof. In the drawings, the same reference numerals generally represent the same components or steps.

[0017] Figure 1 A schematic diagram of the structure of a diesel vehicle exhaust particulate separator based on a dual-frequency sound field provided for an exemplary embodiment of this application; Figure 2 A block diagram illustrating a dynamic optimization method for a diesel vehicle exhaust particulate separator based on a dual-frequency sound field, provided as an exemplary embodiment of this application.

[0018] Explanation of reference numerals in the attached figures 1-Sound source; 2-Separation chamber; 3-Light source module; 4-Particulate filter; 5-Power meter module; 6-Control and power supply module. Detailed Implementation

[0019] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0020] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0021] Furthermore, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0022] This embodiment provides a diesel vehicle exhaust sound field agglomeration separation device based on dynamic optimization. See also: Figure 1 The main structure includes: exhaust pipe, separation chamber 2, particulate filter 4, sound source 1, laser monitoring component (including light source module 3 and power meter module 5) and control and power supply module 6.

[0023] To enable those skilled in the art to fully understand the synergistic design of this invention in terms of mechanics, acoustics, optics, and algorithms, the structural details, material selection, and physical functions of each component are described in detail below: Specifically, since diesel vehicle exhaust velocities are typically high (up to 20-50 m / s), directly applying a sound field would result in insufficient residence time for particulate matter. In this embodiment, the cross-sectional area of ​​the separation chamber 2 is set to be a stainless steel expansion cylinder that is 2.5 times larger than the upstream exhaust pipe. According to the continuity equation, this sudden increase in cross-sectional area significantly reduces the exhaust gas velocity, thereby extending the residence time of exhaust gas containing high concentrations of fine particulate matter within the chamber by a corresponding factor (or 2.5 times). This provides a sufficient time window for the orthokinetic interaction of sound waves and the acoustic wake effect.

[0024] Furthermore, considering that carbon soot particles easily drift and adhere to the pipe wall under strong sound field conditions, the inner wall of separation chamber 2 is coated with a high-temperature resistant polytetrafluoroethylene (PTFE) anti-stick coating. The extremely low surface energy of PTFE prevents the formation of an adhesion layer on the chamber wall by large, agglomerated carbon black chain aggregates, thus preventing carbon on the wall from absorbing sound energy and causing a decrease in sound field intensity. Simultaneously, a funnel-shaped ash discharge port with a controlled automatic ball valve is located at the bottom of separation chamber 2. Large particles (>10μm) that grow under the influence of the sound field settle directly here and are discharged due to gravity, significantly reducing the load on the downstream filter.

[0025] Specifically, due to the inherent airflow background noise within the exhaust pipe, if the sound pressure level is below 130dB, the external sound field will be submerged by the airflow noise, making it impossible to overcome the electrostatic repulsion and van der Waals forces between particles; only a strong sound field above 130dB can effectively excite the nonlinear acoustic effects of the fluid medium. This embodiment takes this into consideration, installing the sound source 1 on the side wall of the inlet of the separation chamber 2, with the emitting surface facing downstream at a certain angle. This embodiment preferably uses a pneumatic Hartmann whistle with a dual-resonant cavity structure or a dual-array compression driver. This design allows the sound source to have a wide total frequency range covering 500Hz to 10kHz, and the total sound pressure level (SPL) generated within the separation chamber 2 is strictly guaranteed to be no less than 130dB. Simultaneously, the sound source 1 supports simultaneous output of the fundamental frequency via a built-in proportional control valve or dual-path PWM drive. (e.g., 1kHz-3kHz) and high-frequency harmonics (e.g., 8kHz-10kHz); where the fundamental frequency mainly acts on particles with larger accumulation mode diameters, causing them to oscillate in the sound field and increasing the probability of collision and coalescence; the high-frequency harmonics mainly act on nanoscale nuclear mode particles, using acoustic radiation force to cause them to drift and enrich in a directional manner, so as to promote the coalescence of fine particles and larger particles. The acoustic power weights of the two can be adjusted independently and continuously.

[0026] Specifically, the multi-wavelength laser monitoring component (including the light source module 3 and the power meter module 5) is installed at the end of the separation chamber near the particulate filter 4 (i.e., DPF, in this embodiment, a cordierite honeycomb ceramic filter element)

[0027] Specifically, the light source module 3 adopts beam recombination technology to coaxially emit a red laser with a wavelength of 650nm and a blue laser with a wavelength of 450nm, and its beams cross the exhaust gas flow channel.

[0028] Specifically, the power meter module 5 employs a multi-channel silicon photodetector with a dichroic mirror for spectral dispersion. According to Mie scattering and Rayleigh scattering theories, particles of different sizes exhibit strong scattering selectivity for different wavelengths of light. The 650nm red light wavelength is longer and more sensitive to concentration changes in micrometer-sized (>1μm) particles; the 450nm blue light wavelength is shorter and more sensitive to concentration changes in submicrometer and nanometer-sized nuclear modal particles. By simultaneously acquiring these two optical signals and controlling the system, the separation efficiency of both large and small particles can be quantified in real time.

[0029] Specifically, the control, regulation, and power supply module 6 in this embodiment has prior sensing capability. This module adopts a 32-bit automotive-grade microprocessor (such as the NXP S32K series or Infineon AURIX series) with a high-speed CAN communication interface, and its internal hardware integrates the following units: Signal processing unit: converts the high-frequency weak current of the photodetector into a smooth digital signal with transmittance, and is also responsible for decoding the J1939 protocol message on the vehicle CAN bus and extracting speed (RPM) and load data.

[0030] Storage unit: The internal storage unit contains a three-dimensional MAP diagram of feedforward frequency based on the characteristics of particulate matter generation in exhaust gas under various operating conditions, calibrated based on a large number of engine bench tests, as well as the core fractional optimization algorithm program.

[0031] Decision unit: Runs the feedforward initial frequency-feedback frequency modulation closed-loop algorithm and outputs the optimal frequency value and power duty cycle command for the dual-frequency sound source.

[0032] Drive and power supply unit: Converts commands into high-voltage drive signals to control the sound source to produce sound, and provides a regulated power supply for the entire system.

[0033] Specifically, in this embodiment, the calibration method for obtaining the "3D MAP of Feedforward Frequency of Particulate Matter Generation Characteristics in Exhaust Gas Based on Operating Conditions" is as follows: The MAP (Magnetic Particulate Spectrometer) is not a theoretical calculation but is pre-calibrated through extensive engine bench tests. The specific calibration process is as follows: Connect the diesel vehicle to a dynamometer, and use a particulate size spectrometer (such as EEPS-3090) as the standard measuring device at the outlet of the separation unit. Select representative grid nodes across the entire engine operating range (the grid consists of speed range × load percentage range, for example, one node is selected for every 200 RPM × 10% load; the data in these grid nodes determine the amount and characteristics of particulate matter in the exhaust gas, providing a valid reference for sound field adjustment). At each steady-state node, independently adjust the fundamental frequency using a traversal frequency sweep method. (Covering the entire 500Hz-10kHz operating frequency band), the fractional-order optimization algorithm in step S3 is used to find the fundamental frequency that minimizes the number concentration of micron-sized particles at the separation chamber outlet and converges the separation performance evaluation function to its optimal value. This fundamental frequency is recorded as the optimal feedforward fundamental frequency corresponding to the node (N,L). High-frequency harmonics Then, according to the preset multiple relationship (2 times in this embodiment, i.e.) =2× Automatically determined.

[0034] After the above calibration, a three-dimensional data structure can be obtained, with the following three axes: engine speed N (unit: RPM), engine load L (unit: %), and optimal feedforward base frequency. (Unit: Hz). Corresponding to this fundamental frequency, the accompanying optimal high-frequency harmonics... satisfy =2× The three-dimensional map is pre-programmed into the storage unit of the control, regulation and power supply module (6) in the form of a multidimensional array or lookup table.

[0035] In actual vehicle operation, when the operating conditions undergo transient changes, the control and power supply modules no longer wait for feedback signals from downstream optical sensors. Instead, they directly use the current engine speed N(t) and load L(t) as inputs, and instantly obtain the a priori optimal frequency pair suitable for the new operating conditions by looking up a table in the three-dimensional MAP using a bilinear interpolation algorithm. , This enables millisecond-level feedforward presets, thereby fundamentally eliminating the response dead time of traditional pure feedback control systems.

[0036] For the aforementioned hardware, the control, regulation and power supply module 6 of this embodiment incorporates a feedforward-feedback dual-loop dynamic optimization algorithm. Figure 2 The specific flowchart of the algorithm is given. To illustrate the method's high-speed tracking and overshoot prevention capabilities in practical applications, the following scenario is assumed: an explosion-proof diesel vehicle is traveling at a constant speed on flat ground underground. The driver then needs to climb a slope, pressing the accelerator hard, and the vehicle suddenly switches to a heavy-load climbing condition. The sampling period of the control algorithm is set to... =0.5 seconds. The following is combined with... Figure 2 The algorithm flow is explained in detail: S1, Feedforward preset When a vehicle transitions from flat ground to an uphill slope, the moment the driver presses the accelerator, the exhaust velocity and carbon emissions increase exponentially. If relying on traditional optical feedback for optimization, the light transmittance drops sharply at this point, and the system needs to undergo multiple trials to find the appropriate frequency for the new operating condition. The adjustment lag can be as long as tens of seconds, during which a large number of unseparated small particles will directly flow into the DPF, causing blockage.

[0037] In this embodiment, feedforward intervention is triggered. Within 20ms, the control and adjustment module detects message changes via the vehicle's CAN bus and parses the data to determine that the engine load has abruptly changed from 30% to 85%, and the engine speed has jumped from 1200 RPM to 2000 RPM. Next, a lookup table calculation is performed. Specifically, the control and adjustment module uses a bilinear interpolation algorithm to query the coordinates (2000, 85%) in the built-in 3D MAP map. The system directly outputs the theoretically optimal initial sound parameters (e.g., fundamental frequency). High-frequency harmonics Total power Sound source 1 transitions to this frequency band and emits sound within 50ms.

[0038] Through the aforementioned feedforward mechanism, the system achieves the instantaneous approximation of the sound generation parameters to the vicinity of the global optimal solution under the current operating conditions.

[0039] S2, Performance Evaluation Due to uncontrollable factors such as exhaust pipe temperature, air pressure, diesel quality, and engine aging, the feedforward frequency given by the MAP diagram is not absolutely perfect. Therefore, it is necessary to use laser components for closed-loop fine-tuning.

[0040] In this step, the control adjustment module operates periodically. Read the current red light transmittance and blue light transmittance ; Calculate the rate of increase in red light transmittance ; To balance absolute transmittance and the trend of transmittance improvement, a separation performance evaluation function is constructed: ; in, for The transmittance of red light at any given moment. for The transmittance of blue light at any given time. The rate of increase in red light transmittance reflects the change in the large particle scattering signal relative to the sampling period at the current moment, and indirectly characterizes the transient trend of decreasing large particle concentration. , The weighting coefficients are and satisfy the following conditions: In this embodiment , ; This is a correction and compensation term for optical lens contamination attenuation, obtained through dynamic fitting of the optical power attenuation model. It is used to compensate for the decrease in transmittance due to dirt on the laser window, ensuring... It accurately reflects the particulate matter separation efficiency rather than the degree of window contamination. The value is obtained through dynamic fitting of the reference transmittance attenuation coefficient acquired periodically. It is an adaptive correction value and does not require a preset fixed range; its specific value is determined online by the optical power attenuation model during calibration. Based on the above formula, The larger the value, the better the effect of large particle agglomeration and sedimentation separation at the current base frequency, and the more positive the particle size conversion trend implied by the difference in red and blue transmittance.

[0041] S3, Dynamic Fine-tuning In this embodiment, a judgment needs to be made before fine-tuning. The relationship between the preset threshold and the target frequency is determined. If the preset threshold is not reached, step S3 is executed, i.e., dynamic fine-tuning is performed. In this embodiment, a value greater than or equal to the preset threshold is defined as reaching the target frequency. The preset threshold is determined by continuously recording data under standard operating conditions after the device has found the globally optimal frequency through the aforementioned bench calibration method and operated stably, under conditions without external interference. The value is calculated by taking the lower limit of the 95% confidence interval of the steady-state value, and is expressed as... Among them, the preset threshold It is also stored along with the operating condition data in the corresponding operating condition node of the MAP diagram. Therefore, when the actual operating conditions change, the evaluation function threshold... It is also updated according to the working conditions.

[0042] The following explanation focuses solely on the closed-loop optimization update of the fundamental frequency. High-frequency harmonics automatically follow the preset multiples, thus avoiding convergence difficulties arising from multi-dimensional optimization. The system calculates the gradient difference of the current evaluation function with respect to the fundamental frequency: Current base frequency The fundamental frequency of the previous cycle The gradient is calculated. At this point, the gradient is minimal (close to the optimal extreme point); if the traditional gradient climbing method is used ( When the optimization approaches a certain extreme point, g(t)≈0, the step size approaches 0, and the algorithm will stagnate at 4800Hz, easily getting trapped in a local non-optimal extreme point (i.e., a false optimum). To overcome this predicament, this application introduces the following formula to calculate the next correction step size and update the base frequency: ; in, The base step size amplification factor is preferably 4000. For fractional order and In this embodiment, the step size is set to 0.5 to prevent getting trapped in local optima when the gradient is extremely small. It is a symbolic function.

[0043] Specifically, when gradient Extremely small (e.g., calculated) When the fractional power is 0.006, the weak gradient is magnified by tens of times because 0.006^0.5 ≈ 0.077. This is then multiplied by the magnification factor. This endows the system with enormous corrective kinetic energy, which can then be calculated. This enables it to traverse current local valleys or flat areas and continue exploring towards the global optimal peak. This results in the updated base frequency. 4800 + 289 = 5089 Hz. Simultaneously, high-frequency harmonics are automatically updated. . in the formula For the exponentially decaying term ( To prevent overshoot, the attenuation factor, also known as the penalty factor, is set to 10 in this embodiment. If the gradient is measured under other operating conditions... If the step size is too large, relying solely on fractional amplification would result in an excessively large step size, causing the system to collapse and diverge. In this case, the negative exponent term rapidly approaches 0, strongly suppressing the step size output, thus ensuring the robustness and convergence stability of the system.

[0044] S4, Dual-frequency allocation Limited by the vehicle's power supply and driver heat dissipation, the total power of the sound field has an upper limit. How to rationally allocate energy between low and high frequencies is crucial to determining the overall separation efficiency. In this embodiment, the control module compares the dual-channel data in real time: if, after a few seconds of optimization, a significant increase in red light transmittance is observed (indicating rapid collision and aggregation of large particles), but the blue light transmittance increase curve flattens out, according to the principle of optical particle size sensitivity, this means that most of the current sound field energy is consumed by the basic low frequencies, and the entrainment resonance energy (high-frequency harmonics) for nanoscale particles is severely insufficient.

[0045] At this point, while maintaining a total power of 150W, the controller adjusts the PWM duty cycle to execute the dual-frequency matching re-switching rule: adjusting the base frequency... The power output ratio at 5089Hz was smoothly reduced from 70% to 40%, while high-frequency harmonics were also reduced. The power proportion at 10178Hz was increased from 30% to 60%. Under this high-frequency, strong sound field compensation, the originally suspended nanoscale nuclear modal particles underwent violent acoustic radiation drift and adsorbed onto the surface of larger particles, causing the blue light transmittance to rapidly recover. Conversely, if the increase in red light transmittance slowed down, the opposite adjustment would occur. This mechanism ensures that neither submicron nor micron particles can escape the coagulation and separation of the sound field.

[0046] S5, Interception and Cleanup After the aforementioned millisecond-level feedforward intervention, microsecond-level fractional-order fine-tuning, and dual-wavelength frequency domain energy distribution, the exhaust gas flows through the middle and rear sections of separation chamber 2. At this point, a massive amount of nano- / micron-sized fine particles that would normally easily penetrate the filter element have condensed into large-volume separated particles under the influence of complex acoustic waves, like a snowball rolling downhill.

[0047] Due to their increased volume by hundreds of times, some of the oversized particles overcome the airflow's carrying capacity under their own gravity and settle in large quantities at the ash discharge port at the bottom of the separation chamber. The remaining large particles that do not settle enter the downstream particulate filter (DPF) with the exhaust gas. Because the particle size has been significantly amplified by the sound field, they can no longer penetrate the deep micropores of the ceramic filter element and cause dense blockage. Instead, they are directly and physically intercepted on the shallow surface with 100% efficiency. The air permeability of the filter pores is largely preserved, and the exhaust back pressure rises extremely slowly. Maintenance personnel only need to open the ash discharge port valve to clean the bottom ash within the specified cycle and clean the surface intercepted material of the DPF in conjunction with the regular high-temperature regeneration command.

[0048] This application overcomes the hysteresis of traditional single-frequency blind scanning by proposing a dual-loop control strategy of feedforward initial frequency and feedback frequency modulation. First, millisecond-level feedforward frequency preset is achieved by analyzing engine speed and load signals. Then, red and blue dual-wavelength lasers are used to track the transmittance changes of large and small particles respectively, introducing a fractional-order update law with an anti-overshoot exponent to perform microsecond-level dynamic closed-loop optimization of the sound frequency. Furthermore, based on the differences in particle concentration for different particle sizes, the power of the fundamental frequency and high-frequency harmonics is dynamically allocated. This invention possesses the capabilities of prior perception, rapid response, and prevention of getting trapped in local optima, significantly improving the separation efficiency of particulate matter across a wide range of particle sizes, effectively reducing the risk of clogging in particulate filters, and extending their service life.

[0049] It should be noted that 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 this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0050] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0051] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and there may be other division methods in actual implementation. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the coupling or direct coupling or communication connection shown or discussed may be through some communication interface; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0052] 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 units can be selected to achieve the purpose of this embodiment according to actual needs.

[0053] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0054] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and they should all be covered within the scope of the claims and specification of this application.

Claims

1. Diesel vehicle exhaust particulate separation device based on a dual frequency sound field, characterized in that, include: The separation chamber (2) is connected at one end to the diesel engine exhaust port and at the other end to the particulate filter (4), serving as a pretreatment chamber for exhaust gas before it enters the particulate filter. A particulate filter (4) is located downstream of the separation chamber (2) and is used to capture particulate matter in the exhaust gas; A sound source (1) is installed on the side wall at the entrance of the separation chamber (2), with the sound-emitting end facing the interior of the separation chamber (2), and is used to generate a high-intensity adjustable sound field in the separation chamber (2); the sound source (1) is a dual-frequency broadband sound source, and simultaneously outputs a fundamental frequency. and high frequency harmonics It also supports independent adjustment of the sound power of each component; The laser monitoring component includes a light source module (3) and a power meter module (5) mounted on the cavity wall of the separation chamber (2) near the particulate filter (4), with its optical path traversing the airflow channel; The control, adjustment and power supply module (6) is electrically connected to the laser monitoring component and the sound source (1), respectively; The vehicle status perception interface connects to the on-board CAN bus of the diesel vehicle to obtain the engine's current speed and load signals in real time. The control and power supply module (6) has a built-in feedforward-feedback dual-loop dynamic optimization algorithm, which is used to extract the rotation speed and load signals to generate feedforward frequency commands, and combined with the transmittance data fed back by the laser monitoring component, the fractional gradient variable step size algorithm is used to perform closed-loop dynamic optimization of the working parameters of the sound source (1).

2. The apparatus according to claim 1, characterized in that, The cross-sectional area of ​​the separation chamber (2) is larger than that of the exhaust pipe, forming an expansion structure inside; its inner wall surface is provided with a polytetrafluoroethylene anti-stick coating; the bottom of the separation chamber (2) is provided with an openable and closable ash discharge port for periodically cleaning large particles of separated material that have condensed and settled under the action of the sound field.

3. The apparatus according to claim 1, characterized in that, The light source module (3) is a multi-wavelength multiplexed laser that emits red and blue light, and the power meter module (5) is a corresponding multi-channel spectral power meter used to simultaneously monitor the transmittance at different wavelengths in order to distinguish the concentration changes of submicron particles and micron particles.

4. The apparatus according to claim 1, characterized in that, The sound source (1) is a Hartmann whistle pneumatic sound source or a compression driver with dual resonant cavities, and its total sound frequency range covers 500Hz-10kHz. The total sound pressure level generated in the separation chamber (2) is not less than 140dB.

5. The apparatus according to claim 1, characterized in that, The control, regulation, and power supply module (6) includes: The signal processing unit is used to convert optical electrical signals into transmittance values ​​and parse CAN messages; The storage unit is pre-stored with a 3D MAP diagram of the feedforward frequency and related information based on different operating conditions, as well as a dynamic optimization algorithm. The decision unit executes the feedforward-feedback dual-loop dynamic optimization algorithm to generate frequency and power adjustment commands for the dual-frequency sound source; The driving unit drives the sound source (1) to emit sound according to the instructions of the decision unit; The power supply unit provides power to all components.

6. A dynamic optimization method for a diesel vehicle exhaust particulate separator based on a dual-frequency sound field as described in any one of claims 1-5, characterized in that, Includes the following steps: S1, Feedforward Preset: Before the diesel engine exhaust gas enters the separator, the control, regulation, and power supply module reads the engine speed at that moment via the CAN bus. and load By querying a pre-calibrated three-dimensional feedforward MAP diagram with speed and load as inputs, the a priori optimal frequency pair under the current operating conditions is obtained, and the fundamental frequency of the sound source is set accordingly. High-frequency harmonics and initial total power; S2. Performance Evaluation: The laser monitoring component is evaluated based on the sampling period. The system collects transmittance data, controls the adjustment and power supply modules to calculate the transmittance improvement rate, and constructs a separation performance evaluation function. ; S3, Dynamic Fine-tuning: If If the preset threshold is not met, based on the separation performance evaluation function The gradient is used to introduce a fractional update law with an anti-overshoot exponent, and the frequency correction step size is calculated for the fundamental frequency. The preset threshold is obtained from the 3D MAP map based on the current operating conditions. S4. Dual-frequency allocation: The fundamental frequency is dynamically adjusted based on the difference in transmittance between the red and blue wavelengths. With the high frequency harmonics The power allocation ratio; S5. Interception and Cleaning: After the exhaust gas is separated by the dual-frequency dynamic sound field, some large-diameter particles settle at the bottom of the separation chamber, while the rest enter the particulate filter with the airflow and are efficiently captured. The settled matter is cleaned regularly through the ash discharge port.

7. The method according to claim 6, characterized in that, In step S1, the feedforward algorithm is as follows: the control adjustment and power supply module uses an interpolation algorithm to query the built-in prior 3D MAP map. The corresponding coordinates directly output the current theoretically optimal fundamental frequency. and high frequency harmonics As the initial sound parameters of the sound source.

8. The method according to claim 6, characterized in that, In step S2, the separation performance evaluation function The calculation formula is: ； in, for The transmittance of red light at any given moment. for The transmittance of blue light at any given time. To increase the rate of red light transmittance, , The weighting coefficients and , This is a correction and compensation term for the degradation caused by contamination of optical lenses.

9. The method according to claim 8, characterized in that, The fractional-order update law with an overshoot prevention exponent is as follows: First, calculate the gradient of the performance evaluation function with respect to the fundamental frequency. ; Next, update the base frequency using the following formula: ； Finally, based on high-frequency harmonics With fundamental frequency Update it according to the preset multiple relationship; in, Based on the step size magnification factor, For fractional order and This is used to increase the step size when the gradient is extremely small to prevent getting trapped in local optima; The overshoot attenuation coefficient is used to suppress step size divergence when the gradient is large. It is a symbolic function.

10. The method according to claim 6, characterized in that, The rule for dual-frequency allocation in step S4 is as follows: when the rate of increase in blue light transmittance is lower than the rate of increase in red light transmittance, the controller increases the high-frequency harmonics while maintaining the total acoustic power constant. The power allocation weights are adjusted, and the fundamental frequency is reduced. The power allocation weight is adjusted accordingly; otherwise, it is adjusted in the opposite direction.

Images