Wind energy collection and air purification system based on highway vehicle driving airflow

Abstract

The present application relates to the technical field of wind energy collection and air purification, and discloses a wind energy collection and air purification system based on highway vehicle driving airflow. The system simulates a bidirectional vehicle driving airflow field by obtaining the lane structure and historical traffic flow data of a target road section. According to the airflow field, wind energy collection devices are symmetrically deployed on both sides of the highway. The deployed devices collect real-time bidirectional vehicle driving airflow data and dynamically correct the simulated airflow field based on the data, thereby obtaining a corrected airflow field. The system dynamically adjusts the operating parameters of each wind energy collection device according to the corrected airflow field, so as to efficiently collect vehicle driving airflow energy and convert it into electric energy. The obtained electric energy is used to drive an array of air purification devices deployed on the road section to purify the air. The present application improves the collection efficiency and stability of vehicle driving airflow energy and enables the continuous and self-powered operation of the air purification devices.

Description

Technical Field

[0001] This invention relates to the field of wind energy harvesting and air purification technology, specifically a wind energy harvesting and air purification system based on the airflow from vehicles traveling on highways. Background Technology

[0002] Existing highway wind energy harvesting technologies are typically designed based on general wind field models or natural environmental wind conditions. These methods fail to adequately consider the induced airflow generated by vehicles traveling at high speeds, and even less so to specifically model the complex airflow interactions caused by vehicles traveling in opposite directions on two-way lanes. Due to the lack of analysis of historical traffic flow data, it is impossible to construct an airflow field that reflects the actual vehicle motion and disturbance effects. As a result, the deployment and operation parameter settings of wind energy harvesting devices often deviate from the actual high-efficiency capture area, leaving considerable room for improvement in overall energy harvesting efficiency.

[0003] Existing wind energy harvesting systems mostly operate with fixed parameters, making them unable to adapt to the dynamically fluctuating airflow field caused by real-time changes in traffic volume and speed. This results in an unstable energy capture process and large fluctuations in output power. Such an unstable energy supply makes it difficult to effectively drive the subsequent air purification unit to work continuously and efficiently, limiting the system's practical effectiveness in improving local air quality on highways.

[0004] There is a need for a technical solution that can accurately simulate the airflow characteristics of vehicles traveling on a two-way road and dynamically optimize the energy harvesting strategy based on real-time airflow conditions, so as to improve the stable and efficient capture and conversion of energy from vehicle airflow, thereby providing reliable power for applications such as air purification along the road. Summary of the Invention

[0005] The purpose of this invention is to provide a wind energy harvesting and air purification system based on the airflow of vehicles traveling on highways, so as to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides a wind energy harvesting and air purification system based on the airflow from vehicles traveling on highways, the system comprising:

[0007] The airflow simulation module acquires lane structure parameters and historical traffic flow data of the target road segment, and performs airflow simulation of two-way vehicle travel based on the historical traffic flow data to generate a simulated two-way vehicle travel airflow field.

[0008] The device deployment module, based on the simulated bidirectional vehicle airflow field and the lane structure parameters, symmetrically deploys multiple wind energy collection devices on both sides of the target section of the highway. The multiple wind energy collection devices include multiple wind energy collection devices on the same side and multiple wind energy collection devices on opposite sides.

[0009] The data acquisition module uses the multiple same-side wind energy collection devices and the multiple opposite-side wind energy collection devices to collect real-time data on the actual two-way vehicle airflow on the target section of the highway within the target time period.

[0010] The airflow field correction module dynamically corrects the simulated two-way vehicle airflow field based on the actual two-way vehicle airflow data to obtain the corrected two-way vehicle airflow field.

[0011] The energy conversion module dynamically adjusts the operating parameters of the multiple wind energy collection devices on the same side and the multiple wind energy collection devices on opposite sides according to the modified bidirectional vehicle airflow field, so as to collect the airflow energy of the vehicle and convert it into electrical energy, and obtain the converted electrical energy data stream.

[0012] The air purification module uses the converted electrical energy data stream to drive the air purification device array deployed on the target section of the highway to purify the air on the target section of the highway.

[0013] Preferably, based on the historical traffic flow data, a two-way lane vehicle airflow simulation is performed to generate a simulated two-way vehicle airflow field, including:

[0014] Multiple historical traffic flow features are extracted from the historical traffic flow data, including historical traffic volume sequences, historical average vehicle speed sequences, and historical vehicle type distribution data.

[0015] The multiple historical traffic flow features are input into a pre-trained airflow simulation model, which is constructed based on computational fluid dynamics principles;

[0016] The airflow simulation model is used to calculate the airflow velocity distribution and airflow pressure distribution generated by vehicle driving disturbance in the two-way lane, and to generate the initial two-way vehicle driving airflow field.

[0017] The lane structure parameters are input as boundary conditions into the airflow simulation model to constrain and correct the initial bidirectional vehicle airflow field, thereby generating the simulated bidirectional vehicle airflow field.

[0018] Preferably, the multiple same-side wind energy harvesting devices and the multiple opposite-side wind energy harvesting devices are used to collect real-time data on actual two-way vehicle airflow on the target section of the highway within the target time period, including:

[0019] The airflow sensors built into the multiple wind energy harvesting devices on the same side collect actual vehicle airflow speed data and actual vehicle airflow pressure data in the same lane, forming a set of actual airflow data on the same side.

[0020] The actual airflow speed and pressure data of vehicles traveling on the opposite lane are collected by the airflow sensors built into the multiple opposite wind energy harvesting devices to form an actual airflow data set on the opposite side.

[0021] The actual airflow data set on the same side and the actual airflow data set on the opposite side are time-stamped and merged to form the actual bidirectional vehicle airflow data.

[0022] Preferably, the simulated bidirectional vehicle airflow field is dynamically corrected based on the actual bidirectional vehicle airflow data to obtain a corrected bidirectional vehicle airflow field, including:

[0023] Extract a snapshot of the actual airflow field for the current time slice from the actual bidirectional vehicle airflow data. The actual airflow field snapshot includes the actual airflow velocity distribution and the actual airflow pressure distribution.

[0024] Calculate the airflow field difference between the actual airflow field snapshot and the simulated bidirectional vehicle airflow field at the corresponding time slice prediction value;

[0025] The airflow field difference is input into a pre-trained airflow field correction network, which is constructed based on a recurrent neural network structure.

[0026] The airflow field correction amount is output through the airflow field correction network, and the predicted value of the next time slice of the simulated bidirectional vehicle airflow field is adjusted in real time using the airflow field correction amount to obtain the real-time updated corrected bidirectional vehicle airflow field.

[0027] Preferably, the operating parameters of the plurality of wind energy harvesting devices on the same side and the plurality of wind energy harvesting devices on opposite sides are dynamically adjusted according to the modified bidirectional vehicle airflow field, including:

[0028] The predicted airflow velocity and predicted airflow direction of the multiple same-side wind energy harvesting devices and the multiple opposite-side wind energy harvesting devices at their corresponding deployment locations are extracted from the modified bidirectional vehicle airflow field.

[0029] The predicted airflow velocity and the predicted airflow direction are input into a preset wind energy harvesting device parameter mapping table to obtain the corresponding wind energy harvesting device operating parameters, which include the blade windward angle and the generator excitation current.

[0030] The operating parameters of the wind energy harvesting device are sent to the corresponding multiple wind energy harvesting devices on the same side and the multiple wind energy harvesting devices on opposite sides to control their dynamic adjustment.

[0031] Preferably, the operating parameters of the plurality of wind energy harvesting devices on the same side and the plurality of wind energy harvesting devices on opposite sides are dynamically adjusted to collect the airflow energy from vehicle travel and convert it into electrical energy, obtaining a data stream of converted electrical energy, including:

[0032] The multiple wind energy collection devices on the same side and the multiple wind energy collection devices on opposite sides operate according to the adjusted operating parameters. Their wind turbines rotate under the action of the airflow from the vehicle, driving the built-in permanent magnet synchronous generator to generate alternating current.

[0033] The AC power generated by the multiple wind energy harvesting devices on the same side and the multiple wind energy harvesting devices on opposite sides is input into their respective corresponding power conversion and conditioning circuits. After rectification, filtering and voltage regulation, DC power is output.

[0034] The DC power output from the multiple wind energy harvesting devices on the same side and the multiple wind energy harvesting devices on opposite sides is combined to form the total converted electrical energy data stream.

[0035] Preferably, the operation of an air purification device array deployed on a target section of a highway is driven by the converted electrical energy data stream, including:

[0036] The converted power data stream is transmitted to the power management unit, which distributes the power to each air purifier in the air purifier array according to a preset power distribution strategy.

[0037] Each air purification device receives the allocated electrical energy to drive its internal high-voltage electrostatic dust collection module and ultraviolet catalytic oxidation module to start operation.

[0038] The high-voltage electrostatic dust collection module adsorbs dust and particulate matter from the highway air flowing through the air inlet of the air purification device.

[0039] The ultraviolet catalytic oxidation module performs catalytic oxidation and decomposition of harmful gases in the exhaust gas of the air flowing through the high-voltage electrostatic dust collection module.

[0040] Preferably, the high-voltage electrostatic dust collection module adsorbs dust and particulate matter from the highway air flowing through the air inlet of the air purification device, including:

[0041] When the air on the highway enters the air intake of the air purification device under the airflow of the vehicle, it first flows through the charged area of ​​the high-voltage electrostatic dust collection module, and the dust and particulate matter in the air gain charge under the action of the high-voltage electric field.

[0042] Charged dust and particulate matter enter the dust collection area of ​​the high-voltage electrostatic dust collection module with the airflow, and are adsorbed onto the dust collection plate with opposite polarity under the action of high-voltage electric field force.

[0043] Preferably, the ultraviolet catalytic oxidation module performs catalytic oxidation decomposition of harmful gases in the exhaust gas of the air flowing through the high-voltage electrostatic dust collection module, including:

[0044] The air processed by the high-voltage electrostatic dust collection module enters the reaction chamber of the ultraviolet catalytic oxidation module;

[0045] The inner wall of the reaction chamber is coated with nanoscale photocatalytic material, and ultraviolet lamps are arranged inside the chamber.

[0046] The ultraviolet light emitted by the ultraviolet lamp activates the nanoscale photocatalytic material, generating highly oxidizing hydroxyl radicals;

[0047] The harmful exhaust gases in the air undergo an oxidation-reduction reaction with the hydroxyl radicals, and are decomposed into carbon dioxide and water.

[0048] Preferably, the system further includes:

[0049] Deploy an air quality sensor network on the target section of the highway to monitor air quality data before and after purification in real time.

[0050] The air quality data before purification and the air quality data after purification are fed back to the power management unit;

[0051] The power management unit dynamically adjusts the power distribution strategy based on the difference between the air quality data before and after purification, so as to regulate the operating power of the air purification device array.

[0052] Compared with the prior art, the beneficial effects of the present invention are:

[0053] Based on historical traffic flow data of the target road segment, airflow simulation of two-way lane vehicle travel is performed to generate an initial airflow field that includes vehicle interactions. This technology uses lane structure, traffic volume, vehicle speed, and vehicle type distribution as key input parameters, enabling the simulated airflow field to reflect the wake superposition effect of same-direction vehicle flows and the collision and interference of airflow from opposite vehicles. Based on this high-precision simulated field, the site selection and orientation of wind energy harvesting devices are determined, allowing the device inlet to more accurately align with the mainstream energy zone and vortex zone generated by vehicle travel. This enhances the device's potential for capturing airflow energy and overcomes the energy capture blind spots caused by deployment based on experience or natural wind models.

[0054] Real-time airflow data is collected by wind energy harvesting devices deployed on the same side and opposite side, and this data is used to dynamically correct the initial simulated field. This process continuously updates the airflow field model to approximate the actual state under real-time changes in traffic flow. Based on the corrected airflow field, the system dynamically adjusts the operating parameters of each wind energy harvesting device. The device's operating state is thus matched with the real-time changes in airflow speed and direction, avoiding idling losses when traffic is sparse and improving capture efficiency when traffic is dense, thereby ensuring that the conversion of airflow energy into electrical energy remains efficient and stable.

[0055] The stable, converted electrical data stream provides a continuous and reliable driving force for the air purification unit array. The operating intensity of the purification unit can be adaptively matched with the real-time power generation capacity, avoiding frequent start-stop cycles or insufficient power due to fluctuations in energy supply. The system achieves closed-loop operation from capturing airflow energy from vehicle traffic to converting it into electrical energy, and then consuming that electrical energy for air purification. The improved energy harvesting efficiency directly enhances the continuous operating capability of the air purification unit, providing an effective technical approach to improving particulate matter pollution in localized areas of highways. Attached Figure Description

[0056] Figure 1 This is a timing diagram of the wind energy harvesting and air purification system based on the airflow of vehicles traveling on highways, as described in this invention.

[0057] Figure 2 A flowchart for generating a simulated airflow field for bidirectional vehicle travel;

[0058] Figure 3 A flowchart for collecting actual bidirectional vehicle airflow data;

[0059] Figure 4 A double-line comparison of the airflow velocity of vehicles traveling on highways over time;

[0060] Figure 5 A radar chart showing the efficiency of air pollutant purification. Detailed Implementation

[0061] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0062] Please see Figure 1This invention provides a wind energy harvesting and air purification system based on the airflow from vehicles traveling on a highway. The system includes: an airflow simulation module that acquires lane structure parameters and historical traffic flow data for a target road segment, and simulates the airflow from vehicles traveling in both directions based on the historical traffic flow data, generating a simulated two-way vehicle airflow field. A device deployment module symmetrically deploys multiple wind energy harvesting devices on both sides of the target highway segment according to the simulated two-way vehicle airflow field and the lane structure parameters. These devices include multiple wind energy harvesting devices on the same side and multiple wind energy harvesting devices on opposite sides. A data acquisition module uses the deployed wind energy harvesting devices to collect real-time data on the actual two-way vehicle airflow from vehicles traveling on the target highway segment within a target time period. An airflow field correction module dynamically corrects the previously generated simulated two-way vehicle airflow field based on the collected actual two-way vehicle airflow data, thereby obtaining a corrected two-way vehicle airflow field that better reflects real-time traffic conditions. An energy conversion module dynamically adjusts the operating parameters of all wind energy harvesting devices according to the corrected two-way vehicle airflow field to efficiently collect vehicle airflow energy and convert it into electrical energy, generating a stable data stream of converted electrical energy. The air purification module uses the converted electrical energy data stream to drive the air purification device array deployed on the same road section to purify the air of the target section of the highway.

[0063] In one embodiment of the present invention, see [reference] Figure 2 Multiple historical traffic flow features are extracted from the historical traffic flow data, including historical traffic volume sequences, historical average vehicle speed sequences, and historical vehicle type distribution data. These extracted features are then input into a pre-trained airflow simulation model, which is built based on computational fluid dynamics principles. The airflow simulation model calculates the airflow velocity and pressure distributions generated by vehicle disturbances in both directions of the road, generating an initial two-way vehicle airflow field. The lane structure parameters are then input as boundary conditions into the airflow simulation model to constrain and correct the initial two-way vehicle airflow field, ultimately generating the simulated two-way vehicle airflow field.

[0064] In practice, the process of extracting multiple historical traffic flow features from historical traffic flow data involves querying and format conversion of the original database. These features include historical traffic volume sequences, historical average vehicle speed sequences, and historical vehicle type distribution data. The historical traffic volume sequences record the number of vehicles passing through the target road segment at fixed time intervals, forming a time series array. The historical average vehicle speed sequences calculate the arithmetic mean of instantaneous speed data collected by speed measurement equipment within each time interval, also forming a time series array. The historical vehicle type distribution data is obtained by statistically analyzing the frequency of occurrence of various vehicle types from video recognition or electronic tag systems and converting it into percentage or probability distribution vectors. These historical traffic flow features are then input into a pre-trained airflow simulation model. This model is built based on computational fluid dynamics principles, and its internal solver discretizes the Navier-Stokes equations and couples them with vehicle motion source terms, numerically solving for airflow using the finite volume method or the lattice Boltzmann method.

[0065] In some embodiments, the computational domain of the airflow simulation model covers the target road segment and a certain range of air on both sides. The computational domain is divided into hexahedral or tetrahedral grid cells, each storing airflow velocity, pressure, and turbulence parameters. Vehicle driving disturbances are introduced into the model via momentum source terms. The magnitude and direction of these momentum source terms are dynamically calculated based on historical average vehicle speed sequences and historical vehicle type distribution data. The historical traffic flow sequence determines the number and temporal distribution of activations of the momentum source terms within the computational domain. The airflow simulation model calculates the airflow velocity and pressure distributions generated by vehicle driving disturbances on both directions of the road, generating an initial two-way vehicle driving airflow field. This initial two-way vehicle driving airflow field stores the airflow velocity components and pressure values ​​at each grid center point at each simulation time step in a three-dimensional matrix format.

[0066] In practical implementation, lane structure parameters are input as boundary conditions to the airflow simulation model. These parameters include the number of lanes, the width of each lane, shoulder dimensions, the geometry and height of the guardrail, and the width and material properties of the median strip. Optionally, lane structure parameters can be constrained by modifying the boundary mesh properties of the computational domain, such as setting the guardrail surface as a no-slip wall boundary condition and the road surface as a rough wall boundary condition. The initial bidirectional vehicle airflow field is constrained and corrected to generate a simulated bidirectional vehicle airflow field. The correction process involves numerical iteration to ensure that the airflow field solution satisfies all boundary conditions. The output of the simulated bidirectional vehicle airflow field is a multidimensional dataset containing time, space, and physical variable dimensions.

[0067] It is understandable that the pre-training of the airflow simulation model uses a large amount of historical traffic flow data and corresponding meteorological monitoring data as input, and high-precision simulation results generated by computational fluid dynamics software as training labels, optimizing the model parameters through backpropagation. In specific implementation, a formula is introduced to characterize the mathematical relationship between airflow velocity distribution and input features in the airflow simulation model, the formula being:

[0068]

[0069] in: Indicates during simulation time The complete airflow field state (including velocity and pressure). The parameter is airflow simulation model function, Indicates the historical traffic flow sequence in time scalar value, This represents the historical average vehicle speed sequence over time. scalar value, This represents a vector of historical vehicle type distribution data. This represents the lane structure parameter vector.

[0070] In one embodiment of the present invention, see [reference] Figure 3 The system collects actual vehicle airflow velocity and pressure data for the same-side lane using airflow sensors built into multiple wind energy harvesting devices on the same side, forming a set of actual airflow data for the same side. It also collects actual vehicle airflow velocity and pressure data for the opposite-side lane using airflow sensors built into multiple wind energy harvesting devices on the opposite side, forming a set of actual airflow data for the opposite side. The system then aligns and merges the actual airflow data sets for the same side and the opposite side to form the actual bidirectional vehicle airflow data. A snapshot of the actual airflow field for the current time slice is extracted from the actual bidirectional vehicle airflow data; this snapshot includes the actual airflow velocity and pressure distributions. The airflow field difference between the actual airflow field snapshot and the predicted value of the simulated bidirectional vehicle airflow field for the corresponding time slice is calculated. This airflow field difference is input into a pre-trained airflow field correction network, which is constructed based on a recurrent neural network structure. The airflow field correction amount is output through the airflow field correction network, and the predicted value of the next time slice of the simulated bidirectional vehicle airflow field is adjusted in real time using the airflow field correction amount to obtain the real-time updated corrected bidirectional vehicle airflow field.

[0071] In practice, multiple wind energy harvesting devices on the same side operate with their built-in airflow sensors at a fixed sampling frequency. The airflow sensors use hot-wire or ultrasonic anemometers to simultaneously measure the vector velocity and static pressure of the airflow. The wind energy harvesting devices are deployed on one side of the guardrail or shoulder of the target section of the highway. The probe head of the airflow sensor of each wind energy harvesting device is exposed to the airflow from the vehicles. The actual airflow velocity and pressure data of the vehicles in the same lane are collected by the built-in airflow sensors of the wind energy harvesting devices. The collected data is temporarily stored in the local cache of the wind energy harvesting devices in the form of an array. The data array includes timestamp, three-dimensional velocity components, and pressure value fields. The system periodically collects the cached data from all wind energy harvesting devices on the same side through wired or wireless communication networks, forming a set of actual airflow data on the same side on the server side. The actual airflow speed and pressure data of vehicles traveling on the opposite lane are collected by airflow sensors built into multiple opposite wind energy harvesting devices. The opposite wind energy harvesting devices are deployed at symmetrical positions on the other side of the target section of the highway. Their data acquisition process is the same as that of the wind energy harvesting devices on the same side, and finally a set of actual airflow data on the opposite side is formed.

[0072] It is understandable that the timestamp alignment and merging operations are completed on the data center server. The server receives all data packets from the same-side and opposite-side actual airflow data sets, parses the timestamp field in each data packet (timestamps are uniformly calibrated by the GPS clock or network time protocol), and uses the minimum sampling interval as the time window to group all same-side and opposite-side data points falling within the same time window into the same data frame. The merging operation integrates the measured values ​​of the same-side and opposite-side actual airflow data sets belonging to the same data frame into a structured data object based on spatial location encoding. This structured data object is the actual bidirectional vehicle airflow data. In specific implementation, the actual bidirectional vehicle airflow data can be organized as a four-dimensional tensor, with the four dimensions being time index, device spatial location index, lane marking, and physical quantity type.

[0073] In some embodiments, a snapshot of the actual airflow field for the current time slice is extracted from actual bidirectional vehicle airflow data. The extraction process involves slicing a four-dimensional tensor. The time index is fixed at the current moment. The device spatial location index and lane markings jointly define a two-dimensional spatial grid. Two two-dimensional matrices, the actual airflow velocity distribution and the actual airflow pressure distribution, are extracted from the physical quantity type dimension. The actual airflow field snapshot is the set of these two two-dimensional matrices. The airflow field difference between the actual airflow field snapshot and the predicted value of the simulated bidirectional vehicle airflow field at the corresponding time slice is calculated. The predicted value of the simulated bidirectional vehicle airflow field at the corresponding time slice is obtained by indexing according to the simulation time axis. The calculation of the airflow field difference involves element-wise subtraction of the actual airflow velocity distribution matrix and the predicted airflow velocity distribution matrix, and element-wise subtraction of the actual airflow pressure distribution matrix and the predicted airflow pressure distribution matrix to generate a velocity difference matrix and a pressure difference matrix.

[0074] In practice, the velocity difference matrix and pressure difference matrix are concatenated into a multi-channel feature map, which is then input into a pre-trained airflow field correction network. This network is built upon a recurrent neural network (RNN) structure, with its core containing long short-term memory (LSTM) units or gated recurrent units to capture the dynamic dependencies of the airflow field over time. The input layer of the network receives the airflow field difference features of the current time slice, while simultaneously receiving the hidden state from the previous time step within the network. Through layer-by-layer transmission and transformation within the RNN structure, an airflow field correction is generated at the output layer. This correction is a tensor with the same spatial and physical dimensions as the simulated bidirectional vehicle airflow field. The network outputs the correction, which is then used to adjust the predicted value of the simulated bidirectional vehicle airflow field for the next time slice in real time. The adjustment operation involves element-wise addition of the correction tensor to the predicted value tensor for the next time slice, resulting in a real-time updated corrected bidirectional vehicle airflow field.

[0075] Optionally, the pre-training of the airflow field correction network uses historically collected actual bidirectional vehicle airflow data and corresponding high-precision simulated bidirectional vehicle airflow field data as training pairs. The training objective is to minimize the difference between the corrected airflow field output by the network and the high-precision simulated airflow field. In some embodiments, a formula is introduced to characterize the calculation process of the airflow field correction amount, the formula being:

[0076]

[0077] in: Indicates application to the next time slice The airflow field correction tensor. The parameter is The airflow field correction network function, Indicates the current time slice The airflow field difference tensor (including the velocity and pressure difference). This indicates that the airflow field correction network was at the previous time step. The hidden state vector. This formula describes the mechanism by which recurrent neural networks generate future corrections using the current difference and historical memory.

[0078] In practical implementation, a specific example scenario involves data comparison. For instance, at a certain sampling time, the actual vehicle airflow velocity recorded at a certain location in the same side's actual airflow data set is 1.5 m / s along the lane direction. The predicted value of the simulated bidirectional vehicle airflow field at the corresponding location and time is 1.2 m / s. The velocity difference at this point is then +0.3 m / s. All such differences at spatial points constitute a velocity difference matrix, and the pressure difference matrix is ​​generated in a similar manner. After these matrices are input into the airflow field correction network, the output airflow field correction may include a +0.25 m / s velocity correction at the corresponding location. This correction is added to the original predicted value of 1.3 m / s for the next time slice of the simulated bidirectional vehicle airflow field, resulting in a corrected predicted value of 1.55 m / s, thus making the prediction closer to the upcoming actual airflow conditions. Optionally, the hidden state vector of the airflow field correction network... The algorithm updates and transmits the data after each calculation, serving as a memory for the next calculation, thereby enabling the learning and tracking of the dynamic evolution trend of the airflow field.

[0079] In one embodiment of the present invention, predicted airflow velocity and predicted airflow direction are extracted from the modified bidirectional vehicle airflow field, representing the deployment locations of the plurality of same-side and opposite-side wind energy harvesting devices. The predicted airflow velocity and predicted airflow direction are input into a preset wind energy harvesting device parameter mapping table to obtain the corresponding wind energy harvesting device operating parameters, including blade angle of attack and generator excitation current. These operating parameters are then sent to the corresponding plurality of same-side and opposite-side wind energy harvesting devices to control their dynamic adjustment. The plurality of same-side and opposite-side wind energy harvesting devices operate according to the adjusted operating parameters, their rotors rotating under the influence of the vehicle airflow, driving their built-in permanent magnet synchronous generators to generate alternating current (AC). The AC generated by the plurality of same-side and opposite-side wind energy harvesting devices is input to their respective corresponding power conversion and conditioning circuits, and after rectification, filtering, and voltage regulation, direct current (DC) is output. The DC power output from the multiple wind energy harvesting devices on the same side and the multiple wind energy harvesting devices on opposite sides is combined to form the total converted electrical energy data stream.

[0080] In practice, predicted airflow velocity and predicted airflow direction are extracted from the modified bidirectional vehicle airflow field, along with the corresponding deployment locations of multiple wind energy harvesting devices on the same side and multiple wind energy harvesting devices on opposite sides. The modified bidirectional vehicle airflow field is stored as a database containing spatial grid point information. Each grid point is associated with three-dimensional coordinates, a predicted airflow velocity vector, and a predicted airflow pressure value. The deployment location of each wind energy harvesting device has known three-dimensional geographic coordinates. A coordinate matching algorithm maps the device location to the nearest grid point in the modified bidirectional vehicle airflow field. The algorithm reads the predicted airflow velocity vector stored at the grid point and decomposes it into magnitude and direction. The predicted airflow velocity scalar value is obtained by calculating the magnitude of the velocity vector, and the predicted airflow direction is obtained by calculating the angle between the projection of the velocity vector on the horizontal plane and the due north direction. The predicted airflow velocity and predicted airflow direction are input into a preset wind energy harvesting device parameter mapping table. The corresponding wind energy harvesting device operating parameters are then retrieved. The wind energy harvesting device parameter mapping table is a two-dimensional lookup table that has been pre-calibrated through wind tunnel experiments or computational fluid dynamics simulations. Its row index is a discretized range of predicted airflow velocity, and its column index is a discretized range of predicted airflow direction. Each cell stores a set of optimal wind energy harvesting device operating parameters, including the blade angle of attack and the generator excitation current.

[0081] In some embodiments, the query process employs bilinear interpolation to improve accuracy. For any input predicted airflow velocity and predicted airflow direction, the four cells surrounding the point in the wind energy harvesting device parameter mapping table are located. Based on the distance weights between the point and the center points of the four cells, a weighted average is used to calculate the continuous values ​​of the blade angle of attack and the generator excitation current. The operating parameters of the wind energy harvesting device are sent to multiple wind energy harvesting devices on the same side and multiple wind energy harvesting devices on opposite sides to control their dynamic adjustment. The sending operation is completed through a low-power wide-area network communication protocol. The control unit of each wind energy harvesting device receives a parameter instruction packet containing its own device identifier and parses out the set values ​​of the blade angle of attack and the generator excitation current. It can be understood that the blade angle of attack is adjusted by a stepper motor or servo motor in the drive device to adjust the angle between the wind turbine blades and the incoming flow direction, and the generator excitation current is achieved by adjusting the field winding current of the permanent magnet synchronous generator.

[0082] Multiple wind energy harvesting devices on the same side and multiple wind energy harvesting devices on opposite sides operate according to adjusted operating parameters. Their wind turbines rotate under the influence of the airflow from the moving vehicles. The wind turbines adopt a vertical or horizontal axis design, and the blades capture airflow kinetic energy and convert it into rotational mechanical energy according to the set blade angle. The rotational mechanical energy is transmitted to the built-in permanent magnet synchronous generator through the main shaft and gearbox. The permanent magnet synchronous generator operates under a set generator excitation current, cutting magnetic field lines to generate three-phase alternating current. The alternating current generated by the multiple wind energy harvesting devices on the same side and multiple wind energy harvesting devices on opposite sides is input to their respective power conversion and conditioning circuits. The power conversion and conditioning circuits include a three-phase bridge rectifier, an LC filter, and a DC-DC voltage regulator module. The three-phase alternating current is first converted into pulsating DC current by the three-phase bridge rectifier. The pulsating DC current is smoothed into DC current with less ripple by the LC filter. The DC-DC voltage regulator module stabilizes the voltage to the nominal value required by the system and outputs DC current.

[0083] In practical implementation, a specific example scenario involves data comparison. For instance, for a wind energy harvesting device numbered A-05 on the same side, the predicted airflow velocity at its location point is extracted from the corrected bidirectional vehicle traffic airflow field as 3.5 m / s, and the predicted airflow direction is 45 degrees east of north. After querying the wind energy harvesting device parameter mapping table, the corresponding blade angle of attack is found to be 30 degrees, and the generator excitation current is 5 amps. The control unit adjusts the blade angle of attack from 20 degrees to 30 degrees and the generator excitation current from 4 amps to 5 amps. After the adjustment, the actual vehicle traffic airflow velocity at the location of the device increases to 3.8 m / s due to the change in traffic flow, and the output AC line voltage of its permanent magnet synchronous generator increases from 200 volts to 210 volts. The DC power outputs from multiple wind energy harvesting devices on the same side and multiple wind energy harvesting devices on opposite sides are combined to form a total converted power data stream. The combination is achieved by connecting the DC output terminals of all power conversion and conditioning circuits to a common DC bus in parallel. The common DC bus is equipped with voltage and current sensors to continuously monitor and upload the total converted power data stream, which includes information on total voltage, total current, and total power.

[0084] Optionally, a formula is introduced to abstractly describe the mathematical essence of the parameter mapping table for wind energy harvesting devices, the formula being:

[0085]

[0086] in: This represents the vector of operating parameters for the wind energy harvesting device (including the blade angle of attack and the generator excitation current). This function represents the parameter mapping table for wind energy harvesting devices. This represents the predicted airflow velocity scalar value extracted from the modified bidirectional vehicle airflow field. This represents the predicted airflow direction angle value extracted from the corrected bidirectional vehicle airflow field. This formula summarizes the mapping relationship from airflow state to device control parameters. It can be understood that in actual deployment, the adjustment of multiple wind energy harvesting devices on the same side and multiple wind energy harvesting devices on opposite sides is parallel and independent. Each device optimizes only based on the predicted airflow state at its own location, thereby achieving an overall improvement in wind energy harvesting efficiency across the entire road segment.

[0087] In one embodiment of the present invention, the converted electrical energy data stream is transmitted to a power management unit, which distributes the electrical energy to each air purifier in the air purifier array according to a preset power distribution strategy. Each air purifier receives the distributed electrical energy and drives its internal high-voltage electrostatic dust collection module and ultraviolet catalytic oxidation module to start operation. The high-voltage electrostatic dust collection module adsorbs dust and particulate matter from the highway air flowing through the air inlet of the air purifier. Specifically, when the highway air enters the air inlet of the air purifier under the influence of the airflow from the vehicle, it first flows through the charged area of ​​the high-voltage electrostatic dust collection module. The dust and particulate matter in the air gain charge under the action of the high-voltage electric field. The charged dust and particulate matter enter the dust collection area of ​​the high-voltage electrostatic dust collection module with the airflow and are adsorbed onto the dust collection plate with opposite polarity under the action of the high-voltage electric field. The ultraviolet catalytic oxidation module performs catalytic oxidation and decomposition of harmful gases in the exhaust gas of the air flowing through the high-voltage electrostatic dust collection module. Specifically, the air treated by the high-voltage electrostatic dust collection module enters the reaction chamber of the ultraviolet catalytic oxidation module. The inner wall of the reaction chamber is coated with nanoscale photocatalytic material, and an ultraviolet lamp is arranged in the chamber. The ultraviolet light emitted by the ultraviolet lamp activates the nanoscale photocatalytic material, generating highly oxidizing hydroxyl radicals. The harmful gases in the exhaust gas in the air undergo an oxidation-reduction reaction with the hydroxyl radicals and are decomposed into carbon dioxide and water.

[0088] In practice, the process of transmitting the converted power data stream to the power management unit is completed via a DC bus. The converted power data stream contains real-time information on total voltage, total current, and total power. The power management unit has a built-in microprocessor and power distribution circuit. The microprocessor receives the converted power data stream and performs calculations according to a preset power distribution strategy. The power management unit then distributes the power to each air purifier in the air purifier array according to the preset power distribution strategy. The preset power distribution strategy can employ algorithms such as proportional distribution, priority-weighted distribution, or dynamic distribution based on air quality requirements. The calculation results are converted into pulse-width modulation signals or analog voltage signals and output to the power distribution circuit. The power distribution circuit consists of multiple parallel DC-DC converter channels, each connected to one air purifier. The amount of power distributed to each air purifier is precisely controlled by adjusting the duty cycle or output voltage of the channel.

[0089] Each air purification unit receives allocated electrical energy, driving its internal high-voltage electrostatic dust collection module and ultraviolet catalytic oxidation module to start operation. The control circuit inside the air purification unit performs secondary conversion on the input electrical energy: part of the electrical energy is boosted and supplied to the high-voltage electric field of the high-voltage electrostatic dust collection module, and the other part is stabilized and supplied to the ultraviolet lamp tube and its driving circuit of the ultraviolet catalytic oxidation module. The high-voltage electrostatic dust collection module adsorbs dust and particulate matter from the highway air flowing through the air inlet of the air purification unit. The air inlet of the air purification unit is designed with louvers or a mesh structure, facing the direction of traffic flow to utilize the dynamic pressure of the airflow from the vehicles to promote air inflow. When the air on the highway enters the air intake of the air purification device under the airflow of the vehicle, it first flows through the charged area of ​​the high-voltage electrostatic dust collection module. The dust and particulate matter in the air gain charge under the action of the high-voltage electric field. The charged area consists of a series of spaced discharge electrodes and grounding electrodes. The discharge electrodes are subjected to a DC negative high voltage of thousands to tens of thousands of volts, which generates a strong corona discharge around them, ionizing the passing air molecules. The dust and particulate matter collide with negative ions and become negatively charged.

[0090] Charged dust and particulate matter enter the collection area of ​​the high-voltage electrostatic precipitator module with the airflow. Under the action of the high-voltage electric field, they are adsorbed onto the collection plates with opposite polarity charges. The collection area consists of multiple sets of parallel plates, which are alternately subjected to positive high voltage and grounding. Negatively charged dust and particulate matter are attracted by the Coulomb force of the positive plates and captured, adhering to the surface of the collection plates. In some embodiments, the operating parameters of the high-voltage electrostatic precipitator module can be dynamically adjusted by the power management unit. For specific adjustment methods, please refer to the following example data table. Table 1 shows the recommended operating voltage and power distribution ratio of the high-voltage electrostatic precipitator module under different air quality concentration ranges.

[0091] Table 1: Correspondence between operating parameters of high-voltage electrostatic dust collection module and air mass concentration

[0092]

[0093] It is understood that the ultraviolet catalytic oxidation module catalytically oxidizes and decomposes harmful gases in the exhaust gas of the air flowing through the high-voltage electrostatic dust collection module. After flowing out of the dust collection area of ​​the high-voltage electrostatic dust collection module, the air enters the reaction chamber of the ultraviolet catalytic oxidation module through a connecting duct. The inner wall of the reaction chamber is coated with nanoscale photocatalytic materials, primarily titanium dioxide. Ultraviolet lamps are arranged within the chamber. The ultraviolet light emitted by the lamps activates the nanoscale photocatalytic materials, generating highly oxidizing hydroxyl radicals. The wavelength of the ultraviolet lamps is 254 nm or 365 nm, and the light intensity is adjusted according to the input electrical energy. The harmful gases in the exhaust gas undergo an oxidation-reduction reaction with the hydroxyl radicals, decomposing into carbon dioxide and water. The harmful gases in the exhaust gas include carbon monoxide, nitrogen oxides, and hydrocarbons.

[0094] In practical implementation, a specific example scenario involves data comparison. For instance, during the morning traffic peak, the conversion power data stream for the target road segment shows a total power of 5 kW. The air quality sensor network detects a PM2.5 concentration of 110 micrograms per cubic meter before purification, which, according to Table 1, falls within the light pollution range. Based on this, the power management unit executes a power allocation strategy, allocating 65% (3.25 kW) of the total power to the air purification device array. Of this, according to a preset ratio, 2.2 kW is allocated to the high-voltage electrostatic dust collection module (operating voltage set at 12 kV) of each device, and 1.05 kW is allocated to the ultraviolet catalytic oxidation module. During the same period in the afternoon off-peak hours, the total power drops to 2 kW, and the PM2.5 concentration before purification is 50 micrograms per cubic meter, which falls within the good range. The power management unit then allocates 50% (1 kW) of the total power, with 0.7 kW used for the high-voltage electrostatic dust collection module (operating voltage 10 kV) and 0.3 kW used for the ultraviolet catalytic oxidation module. Optionally, a formula is introduced to describe the decision logic of the power allocation strategy, as follows:

[0095]

[0096] in: Indicates time The total power vector allocated to the air purification unit array. This represents the power allocation strategy function. Indicates time The total power of the converted electrical energy data stream, Indicates time PM2.5 concentration monitoring value before purification.

[0097] In some embodiments, the air purification device array is deployed at equal intervals along the highway guardrail. The processing air volume of each air purification device is proportional to the electrical energy it receives. The power management unit can control the purification intensity of each individual air purification device in different sections of the road by adjusting the electrical energy allocated to it. It is understood that the dust collection plate of the high-voltage electrostatic dust collection module requires regular cleaning and maintenance; the maintenance cycle is determined based on the accumulated weight of adsorbed dust or operating time. The nanoscale photocatalytic material coating of the ultraviolet catalytic oxidation module may experience reduced activity after long-term use; its replacement or regeneration cycle is determined based on operating time and degradation efficiency monitoring data.

[0098] See Figure 4This is a double-line graph comparing the airflow velocity of vehicles traveling on a highway over time. The two lines show a highly consistent trend, indicating a high degree of fit between the simulation results and actual measurements. The airflow velocity peaks around 4 PM, consistent with the airflow disturbance characteristics during peak traffic hours. The deviation between the simulated and actual values ​​is small, falling within a reasonable engineering error range. This type of graph is commonly used to validate traffic airflow simulation models. By comparing simulated and measured data, the model's accuracy can be evaluated, providing data support for engineering applications such as wind energy harvesting device deployment and airflow field optimization. The error range between the simulated and actual data can serve as a performance benchmark for the subsequent "airflow field correction module," helping to quantify the optimization effect of the correction algorithm.

[0099] In one embodiment of the present invention, an air quality sensor network is deployed on a target section of a highway to monitor air quality data before and after purification in real time. The air quality data before and after purification is fed back to the power management unit. Based on the differences between the air quality data before and after purification, the power management unit dynamically adjusts the power allocation strategy to regulate the operating power of the air purification device array.

[0100] In practice, an air quality sensor network is deployed on the target section of the highway to monitor air quality data before and after purification in real time. The air quality sensor network consists of several sensor nodes distributed upwind, downwind, and at key locations along the target section of the highway. Each sensor node integrates a PM2.5 sensor, a PM10 sensor, a nitrogen dioxide sensor, and a carbon monoxide sensor. The sensor nodes collect air quality parameters at their location at a fixed sampling period and send monitoring data packets containing timestamps, location identifiers, and various concentration values ​​to the central data processing unit via a wireless sensor network. Air quality data before purification is collected by the sensor node group upwind and at the beginning of the section, while air quality data after purification is collected by the sensor node group downwind and at the end of the section. The air quality data before and after purification are fed back to the power management unit. The feedback process is realized through the data bus. The central data processing unit verifies, aligns and spatially interpolates the received raw monitoring data to generate the air quality index sequence before and after purification, which represent the average condition of the entire target road segment. These two sequences are then transmitted in real time to the decision processor of the power management unit as input signals.

[0101] The power management unit dynamically adjusts the power allocation strategy based on the difference between pre- and post-purification air quality data to regulate the operating power of the air purification device array. The decision processor incorporates a control algorithm that uses the difference between pre- and post-purification air quality data as the primary feedback signal, combined with the current total power of the converted power data stream, to calculate a new, optimized power allocation strategy parameter set. This new parameter set includes the proportion of total power allocated to the air purification device array, and the internal power allocation ratio between the high-voltage electrostatic dust collection module and the ultraviolet catalytic oxidation module under this total power. The decision processor sends the new parameter set to the power allocation circuit, which adjusts the voltage or current of each output channel according to the new parameters, thereby changing the power allocated to each air purification device and ultimately regulating the overall operating power of the air purification device array.

[0102] In some embodiments, the node deployment spacing of the air quality sensor network is optimized based on the length of the target road segment and the meteorological diffusion model. Typically, 1-2 nodes are arranged upwind as background monitoring points, and nodes are arranged at the front, middle and rear of the air purification device array coverage area to monitor the longitudinal gradient of the purification effect. The installation height of all nodes is consistent with the height of the air inlet of the air purification device. It's understandable that the dynamic adjustment process is a closed-loop control. A specific example involves data comparison. For instance, in a certain control cycle, the air quality data before purification (average PM2.5 concentration) is 100 micrograms per cubic meter, and the air quality data after purification (average PM2.5 concentration) is 70 micrograms per cubic meter, a difference of 30 micrograms per cubic meter. Simultaneously, the converted power data stream shows a total power of 4 kilowatts. The control algorithm in the power management unit, based on a preset lookup table or function relationship, determines that the current purification efficiency (reflected by the difference) has not met the expected target. Therefore, it dynamically adjusts the power allocation strategy, increasing the proportion of total power allocated to the air purification device array from the current 50% to 70%, i.e., increasing the allocated power from 2 kilowatts to 2.8 kilowatts. In the next control cycle, due to the increased operating power of the air purification device array, the new post-purification air quality data may show a PM2.5 concentration dropping to 60 micrograms per cubic meter, widening the difference from the pre-purification data to 40 micrograms per cubic meter.

[0103] Optionally, a formula is introduced to characterize this feedback-based dynamic adjustment logic, as follows:

[0104]

[0105] in: This represents the parameter set of the new power allocation strategy generated after dynamic adjustment. This represents the dynamic adjustment function built into the power management unit. This indicates the PM2.5 concentration monitoring value before purification. This indicates the PM2.5 concentration monitoring value after purification. This represents the total power converted into electrical energy data streams. The formula summarizes the mathematical relationship between the difference in purification effect and the total available power for updating the operating strategy. In some embodiments, the feedback and adjustment frequency can be set to once every 5 minutes or every 10 minutes to adapt to the changing rhythm of traffic flow and air quality. It can be understood that the difference between the air quality data before purification and the air quality data after purification is the core feedback signal. When the difference value remains below a certain threshold, it indicates that the purification effect is significant, and the power management unit may appropriately reduce the allocated power to save energy; when the difference value narrows or decreases, it triggers an instruction to increase the allocated power.

[0106] See Figure 5 , Figure 5 This is a visualization of pollutant purification efficiency obtained during the R&D process through a combination of on-site testing on open roads and laboratory calibration. It intuitively reflects the system's multi-pollutant purification capabilities in real-world application scenarios. This is a radar chart of air pollutant purification efficiency. The purification efficiency varies for different pollutants, with PM2.5 and PM10 particles showing relatively high efficiency (close to 80%), while gaseous pollutants have a moderate efficiency. The chart comprehensively presents the air purification module's ability to handle multiple types of pollutants, demonstrating the module's multi-dimensional purification characteristics. It visually demonstrates the purification effect of the air purification device on different pollutants, verifying the rationality of its technical solution. For pollutant types with low efficiency, the purification device parameters can be adjusted to improve targeted treatment capabilities. As a core data carrier of the system's environmental benefits, it can be used to illustrate the project's improvement effect on highway air quality.

[0107] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0108] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A wind energy harvesting and air purification system based on airflow from vehicles traveling on highways, characterized in that, The system includes: The airflow simulation module acquires lane structure parameters and historical traffic flow data of the target road segment, and performs airflow simulation of two-way vehicle travel based on the historical traffic flow data to generate a simulated two-way vehicle travel airflow field. The device deployment module, based on the simulated bidirectional vehicle airflow field and the lane structure parameters, symmetrically deploys multiple wind energy collection devices on both sides of the target section of the highway. The multiple wind energy collection devices include multiple wind energy collection devices on the same side and multiple wind energy collection devices on opposite sides. The data acquisition module uses the multiple same-side wind energy collection devices and the multiple opposite-side wind energy collection devices to collect real-time data on the actual two-way vehicle airflow on the target section of the highway within the target time period. The airflow field correction module dynamically corrects the simulated two-way vehicle airflow field based on the actual two-way vehicle airflow data to obtain the corrected two-way vehicle airflow field. The energy conversion module dynamically adjusts the operating parameters of the multiple wind energy collection devices on the same side and the multiple wind energy collection devices on opposite sides according to the modified bidirectional vehicle airflow field, so as to collect the airflow energy of the vehicle and convert it into electrical energy, and obtain the converted electrical energy data stream. The air purification module uses the converted electrical energy data stream to drive the air purification device array deployed on the target section of the highway to purify the air on the target section of the highway. Based on the historical traffic flow data, a two-way lane vehicle airflow simulation is performed to generate a simulated two-way vehicle airflow field, including: Multiple historical traffic flow features are extracted from the historical traffic flow data, including historical traffic volume sequences, historical average vehicle speed sequences, and historical vehicle type distribution data. The multiple historical traffic flow features are input into a pre-trained airflow simulation model, which is constructed based on computational fluid dynamics principles; The airflow simulation model is used to calculate the airflow velocity distribution and airflow pressure distribution generated by vehicle driving disturbance in the two-way lane, and to generate the initial two-way vehicle driving airflow field. The lane structure parameters are input as boundary conditions into the airflow simulation model to constrain and correct the initial bidirectional vehicle airflow field, thereby generating the simulated bidirectional vehicle airflow field.

2. The wind energy harvesting and air purification system based on the airflow of vehicles traveling on highways as described in claim 1, characterized in that, The system utilizes the multiple same-side wind energy harvesting devices and the multiple opposite-side wind energy harvesting devices to collect real-time bidirectional vehicle airflow data for the target section of the highway within a target time period, including: The airflow sensors built into the multiple wind energy harvesting devices on the same side collect actual vehicle airflow speed data and actual vehicle airflow pressure data in the same lane, forming a set of actual airflow data on the same side. The actual airflow speed and pressure data of vehicles traveling on the opposite lane are collected by the airflow sensors built into the multiple opposite wind energy harvesting devices to form an actual airflow data set on the opposite side. The actual airflow data set on the same side and the actual airflow data set on the opposite side are time-stamped and merged to form the actual bidirectional vehicle airflow data.

3. The wind energy harvesting and air purification system based on the airflow of vehicles traveling on highways as described in claim 2, characterized in that, The simulated two-way vehicle airflow field is dynamically corrected based on the actual two-way vehicle airflow data to obtain the corrected two-way vehicle airflow field, including: Extract a snapshot of the actual airflow field for the current time slice from the actual bidirectional vehicle airflow data. The actual airflow field snapshot includes the actual airflow velocity distribution and the actual airflow pressure distribution. Calculate the airflow field difference between the actual airflow field snapshot and the simulated bidirectional vehicle airflow field at the corresponding time slice prediction value; The airflow field difference is input into a pre-trained airflow field correction network, which is constructed based on a recurrent neural network structure. The airflow field correction amount is output through the airflow field correction network, and the predicted value of the next time slice of the simulated bidirectional vehicle airflow field is adjusted in real time using the airflow field correction amount to obtain the real-time updated corrected bidirectional vehicle airflow field.

4. The wind energy harvesting and air purification system based on the airflow of vehicles traveling on highways as described in claim 3, characterized in that, Based on the modified bidirectional vehicle airflow field, the operating parameters of the plurality of same-side wind energy harvesting devices and the plurality of opposite-side wind energy harvesting devices are dynamically adjusted, including: The predicted airflow velocity and predicted airflow direction of the multiple same-side wind energy harvesting devices and the multiple opposite-side wind energy harvesting devices at their corresponding deployment locations are extracted from the modified bidirectional vehicle airflow field. The predicted airflow velocity and the predicted airflow direction are input into a preset wind energy harvesting device parameter mapping table to obtain the corresponding wind energy harvesting device operating parameters, which include the blade windward angle and the generator excitation current. The operating parameters of the wind energy harvesting device are sent to the corresponding multiple wind energy harvesting devices on the same side and the multiple wind energy harvesting devices on opposite sides to control their dynamic adjustment.

5. The wind energy harvesting and air purification system based on the airflow of vehicles traveling on highways as described in claim 4, characterized in that, The operating parameters of the multiple wind energy harvesting devices on the same side and the multiple wind energy harvesting devices on opposite sides are dynamically adjusted to collect the airflow energy from vehicle travel and convert it into electrical energy, obtaining a data stream of converted electrical energy, including: The multiple wind energy collection devices on the same side and the multiple wind energy collection devices on opposite sides operate according to the adjusted operating parameters. Their wind turbines rotate under the action of the airflow from the vehicle, driving the built-in permanent magnet synchronous generator to generate alternating current. The AC power generated by the multiple wind energy harvesting devices on the same side and the multiple wind energy harvesting devices on opposite sides is input into their respective corresponding power conversion and conditioning circuits. After rectification, filtering and voltage regulation, DC power is output. The DC power output from the multiple wind energy harvesting devices on the same side and the multiple wind energy harvesting devices on opposite sides is combined to form the total converted electrical energy data stream.

6. The wind energy harvesting and air purification system based on the airflow of vehicles traveling on highways as described in claim 5, characterized in that, The operation of an air purification device array deployed on a target section of a highway is driven by the converted electrical energy data stream, including: The converted power data stream is transmitted to the power management unit, which distributes the power to each air purifier in the air purifier array according to a preset power distribution strategy. Each air purification device receives the allocated electrical energy to drive its internal high-voltage electrostatic dust collection module and ultraviolet catalytic oxidation module to start operation. The high-voltage electrostatic dust collection module adsorbs dust and particulate matter from the highway air flowing through the air inlet of the air purification device. The ultraviolet catalytic oxidation module performs catalytic oxidation and decomposition of harmful gases in the exhaust gas of the air flowing through the high-voltage electrostatic dust collection module.

7. The wind energy harvesting and air purification system based on highway vehicle airflow as described in claim 6, characterized in that, The high-voltage electrostatic dust collection module adsorbs dust and particulate matter from the highway air flowing through the air inlet of the air purification device, including: When the air on the highway enters the air intake of the air purification device under the airflow of the vehicle, it first flows through the charged area of ​​the high-voltage electrostatic dust collection module, and the dust and particulate matter in the air gain charge under the action of the high-voltage electric field. Charged dust and particulate matter enter the dust collection area of ​​the high-voltage electrostatic dust collection module with the airflow, and are adsorbed onto the dust collection plate with opposite polarity under the action of high-voltage electric field force.

8. The wind energy harvesting and air purification system based on the airflow of vehicles traveling on highways as described in claim 7, characterized in that, The ultraviolet catalytic oxidation module performs catalytic oxidation and decomposition of harmful gases in the exhaust gas of the air flowing through the high-voltage electrostatic dust collection module, including: The air processed by the high-voltage electrostatic dust collection module enters the reaction chamber of the ultraviolet catalytic oxidation module; The inner wall of the reaction chamber is coated with nanoscale photocatalytic material, and ultraviolet lamps are arranged inside the chamber. The ultraviolet light emitted by the ultraviolet lamp activates the nanoscale photocatalytic material, generating highly oxidizing hydroxyl radicals; The harmful exhaust gases in the air undergo an oxidation-reduction reaction with the hydroxyl radicals, and are decomposed into carbon dioxide and water.

9. The wind energy harvesting and air purification system based on the airflow of vehicles traveling on highways as described in claim 8, characterized in that, The system also includes: Deploy an air quality sensor network on the target section of the highway to monitor air quality data before and after purification in real time. The air quality data before purification and the air quality data after purification are fed back to the power management unit; The power management unit dynamically adjusts the power distribution strategy based on the difference between the air quality data before and after purification, so as to regulate the operating power of the air purification device array.

Images