Atmospheric particle detection method for preventing and treating atmospheric pollution

Abstract

The invention discloses an atmospheric particle detection method for preventing and treating atmospheric pollution, and particularly relates to the technical field of environmental atmospheric monitoring, which comprises the following steps: S1, enriching and preliminarily sensing atmospheric particles by using a bionic particle capturing structure, S2, detecting and analyzing the atmospheric particles treated by the bionic capturing structure by using an optical detection and analysis system, and S3, determining the atmospheric particle concentration and the atmospheric particle concentration. S3, processing and transmitting the detection data by using an intelligent data processing and transmission module; and S4, ensuring the long-term stable operation of the detection system by adopting a self-cleaning and calibration mechanism. The system has the remarkable effects of being high in detection precision, good in system stability, high in real-time performance, capable of achieving early warning in time and assisting pollution prevention and control and the like, and powerful data support and reliable guarantee can be provided for air pollution prevention and control work.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental atmosphere monitoring, and more specifically, to an atmospheric particle detection method for preventing and controlling atmospheric pollution. Background Art

[0002] At present, with the rapid development of global industrialization and urbanization, the problem of air pollution is becoming increasingly serious. It has become one of the important factors threatening human health, destroying ecological balance and restricting sustainable development. The particulate matter in atmospheric pollutants includes dust, heavy metal particulate matter from industrial emissions, carbon-based particulate matter from traffic exhaust, and aerosols formed by various organic pollutants. They have a wide range of sources, complex compositions and different particle sizes, which have many adverse effects on air quality, climate environment and human respiratory system.

[0003] In the work of air pollution prevention and control, it is very important to accurately, timely and comprehensively detect the atmospheric particles. This is the key prerequisite for formulating scientific and reasonable pollution prevention and control strategies and evaluating the effectiveness of governance. Although traditional atmospheric particle detection methods can obtain some particle-related information to a certain extent, they often have many limitations, such as:

[0004] 1. Although some methods based on filter sampling are relatively simple to operate, they can only collect atmospheric particles within a specific period of time. The composition needs to be determined in the laboratory through complex chemical analysis methods. Not only is the detection cycle long, but it is also impossible to achieve real-time monitoring and it is difficult to timely feedback the dynamic changes of air pollution, resulting in the inability to quickly respond to preventive measures;

[0005] 2. Although some detection instruments based on the principle of optical scattering can obtain some physical information such as particle size in real time, their ability to analyze in-depth information such as the specific chemical composition and source of particles is limited, and they cannot accurately provide sufficiently detailed data for pollution source tracing and targeted governance;

[0006] 3. In the long-term operation of existing detection equipment, due to the lack of effective self-cleaning and precise calibration mechanisms, problems such as decreased detection sensitivity due to particle adsorption and accumulation, or detection errors due to equipment aging and environmental factors may occur, which affects the accuracy and reliability of the test results, and thus interferes with the scientific decision-making and effective implementation of air pollution prevention and control work.

[0007] In view of the above situation, the present invention provides an atmospheric particle detection method for preventing and controlling air pollution. Summary of the invention

[0008] In order to overcome the above-mentioned defects of the prior art, the present invention provides an atmospheric particle detection method for preventing and controlling air pollution, so as to solve the problems raised in the above-mentioned background technology.

[0009] To achieve the above object, the present invention provides the following technical solution: an atmospheric particle detection method for preventing and controlling air pollution, specifically comprising the following steps:

[0010] S1. Enrich and preliminarily sense atmospheric particles using a bionic particle capture structure, wherein the bionic particle capture structure includes a capture net made by imitating the characteristics of spider silk and a sensing unit designed by imitating the sensing function of insect antennae, wherein:

[0011] The capture net made by imitating the characteristics of spider silk uses nanomaterial technology to synthesize silk materials and weave them into a three-dimensional mesh structure with a specific pore structure to absorb particles in the atmosphere;

[0012] The sensing units designed to imitate the sensing function of insect antennae are made of a flexible, bendable organic semiconductor material as a substrate and the surface is covered with a functional coating that has specific adsorption and electrical response to common components in atmospheric particles, and are installed at different positions of the capture net;

[0013] S2. Detect and analyze the atmospheric particles processed by the bionic capture structure through an optical detection and analysis system, wherein the optical detection and analysis system comprises a laser scanning imaging module and a spectral analysis subsystem, wherein:

[0014] The laser scanning imaging module uses multiple sets of high-precision laser scanning devices to scan the area where the particles are located at a specific angle and frequency;

[0015] The spectral analysis subsystem uses a high-resolution Raman spectrometer and infrared spectrometer to perform spectral analysis on the particles irradiated by the laser;

[0016] S3, using intelligent data processing and transmission module to process and transmit the detection data, the intelligent data processing and transmission module includes artificial intelligence algorithm recognition and classification and real-time data transmission and early warning mechanism, wherein,

[0017] Artificial intelligence algorithm recognition and classification: The image data and spectral data obtained by the optical detection system are input into the atmospheric particle recognition model constructed by the deep learning algorithm;

[0018] The real-time data transmission and early warning mechanism transmits the detected atmospheric particle data to the air pollution prevention and control monitoring center and related mobile applications in real time through the wireless communication module;

[0019] S4. Use a self-cleaning and calibration mechanism to ensure the long-term stable operation of the detection system. The self-cleaning and calibration mechanism includes a self-cleaning module and a calibration unit, wherein:

[0020] The self-cleaning module sprays a specially formulated cleaning fluid onto the bionic capture net to dissolve and remove particles adsorbed on the capture net;

[0021] The calibration unit uses a built-in standard atmospheric particle sample storage device and calibration equipment such as a calibration light source and a standard electrical signal generator to calibrate the detection system to ensure the accuracy and reliability of the detection data.

[0022] Preferably, in the capture net manufactured by imitating the properties of spider silk in step S1, the filamentous material is based on polyimide nanofibers doped with 5%-10% by mass of graphene quantum dots and prepared by an electrospinning process. The pores of the woven three-dimensional network structure are between 0.1-10 μm to adapt to the adsorption of atmospheric particles of different particle sizes. The specific preparation steps are as follows:

[0023] Dissolve polyimide powder in a mixed solvent of N,N-dimethylformamide and N-methylpyrrolidone. The mass-volume ratio of polyimide to the mixed solvent is 1:10-1:15. Stir evenly to form a homogeneous polyimide solution. Add graphene quantum dots and ultrasonically disperse for 2-4 h to ensure that the graphene quantum dots are evenly dispersed in the polyimide solution. Load the above solution into a syringe with a metal needle, connect a high-voltage power supply, set the voltage to 15-25 kV, and the receiving distance to 10-20 cm. Perform electrospinning on a grounded collection device to collect the filamentous material. The collected filamentous material is formed into a three-dimensional network structure with a specific pore structure through a weaving process or a specific mold.

[0024] Preferably, in the capture net manufactured by imitating the properties of spider silk, the surface of the filamentous material has nano-scale sticky protrusions. The average height of the sticky protrusions is 100 nm-500 nm, and the spacing is between 500 nm-2000 nm, which is used to enhance the adsorption effect on atmospheric particles;

[0025] The specific preparation method is: after the filamentous material is prepared, through plasma etching, using reaction gases such as oxygen or methane, under the conditions of a power of 100-300 W and a time of 10-30 min, a nano-scale sticky protrusion structure is manufactured on the surface of the filamentous material.

[0026] Preferably, in the sensing unit designed by imitating the sensing function of insect antennae in step S1, the organic semiconductor material is poly(3,4-ethylenedioxythiophene)-polystyrenesulfonic acid;

[0027] The functional coating contains 15%-20% by mass of metal-organic framework material MOFs for specifically adsorbing heavy metal ions, and 10%-15% by mass of molecularly imprinted polymer MIPs for identifying specific organic pollutant components. The type and approximate concentration range of the particles are initially judged by the change in electrical parameters caused by the approach of the sensing particles.

[0028] Preferably, the electrical parameters include resistance and capacitance, and the atmospheric particle situation is judged by detecting the changes in resistance and capacitance, wherein the resistance change range between 100Ω-1000Ω corresponds to the adsorption of different types of particles, and the capacitance change range between 10pF-100pF is used to assist in concentration range judgment, and when the resistance change rate exceeds 10% or the capacitance change rate exceeds 20%, an instruction for further detection and analysis of the particles is triggered.

[0029] Preferably, in the laser scanning imaging module of step S2, the laser scanning device uses a green laser with a wavelength of 532nm, the scanning angle range of the laser beam is 30°-150°, the scanning frequency is 50Hz, and a high-speed camera is used to capture the optical image generated by the interaction between the laser and the particles to obtain the size, shape, distribution state of the particles and perform preliminary component differentiation based on the difference in optical response. The frame rate of the high-speed camera is ≥1000 frames / second, and the resolution reaches 1920×1080 pixels.

[0030] Preferably, in the spectral analysis subsystem of step S2, the spectral resolution of the Raman spectrometer reaches 1 cm -1 , the excitation wavelength is 785nm, the power is 300mW;

[0031] The spectral resolution of the infrared spectrometer reaches 0.5cm -1 , scanning range is 4000cm -1 -400cm -1 .

[0032] Preferably, in the artificial intelligence algorithm identification and classification in step S3, the deep learning algorithm is an algorithm combining a convolutional neural network algorithm and a recurrent neural network algorithm, wherein;

[0033] The convolutional neural network part uses the ResNet-50 network architecture as its basis, and the recurrent neural network part uses the long short-term memory network. It is trained through a large amount of labeled atmospheric particle sample data under different environments and different pollution levels. The number of training samples is no less than 10,000 groups, which are used to automatically and accurately identify the types, concentrations, sources of atmospheric particles and track their dynamic changes.

[0034] Preferably, in the real-time data transmission and early warning mechanism of step S3, the wireless communication module adopts 4G / 5G communication technology with a transmission rate of ≥100Mbps, and triggers the early warning mechanism to send early warning information when the concentration of key pollutant particles exceeds the standard based on the preset pollution threshold. The pollution threshold is set according to the atmospheric environment quality standards of different regions, for example, the concentration threshold for PM2.5 is set to 35μg / m 3 The sulfur dioxide concentration threshold is set at 150 μg / m 3 .

[0035] Preferably, in the self-cleaning module of step S4, the composition of the cleaning liquid includes 30%-40% by mass of ethanol as an organic solvent for dissolving some organic pollutant particles, 10%-15% by mass of sodium dodecylbenzene sulfonate as a surfactant for enhancing the cleaning effect, and 5%-10% by mass of benzotriazole as an additive with an anti-corrosion effect for preventing corrosion to the capture net and the sensing unit, the injection pressure of the cleaning liquid is between 0.1MPa-0.5MPa, and the flow rate is 10ml / min-50ml / min;

[0036] In the calibration unit, the standard atmospheric particle sample storage device can simulate a variety of atmospheric environments with different pollution levels and types, and release standard particle samples with a particle size of 0.01μm-100μm and different compositions for calibrating the detection system. The wavelength accuracy of the calibration light source is within ±1nm, and the output signal accuracy of the standard electrical signal generator is within ±0.1%.

[0037] Technical effects and advantages of the present invention:

[0038] 1. The present invention adopts a bionic particle capture structure combined with an advanced optical detection and analysis system and intelligent data processing technology. In the bionic capture structure, the capture net made by imitating the characteristics of spider silk can effectively enrich atmospheric particles of different particle sizes. The sensing unit designed by imitating the sensing function of insect antennae can preliminarily judge the particle category and approximate concentration range based on the change of electrical parameters caused by the approach of particles. On this basis, the high-precision laser scanning imaging module and the high-resolution spectral analysis subsystem in the optical detection and analysis system work together to accurately obtain the morphological information of the particles, determine their specific chemical composition and content ratio, and then further accurately identify the type, concentration, source and other multi-dimensional information of atmospheric particles through the well-trained deep learning algorithm model in the intelligent data processing and transmission module. The cooperation of various technical means greatly improves the accuracy of atmospheric particle detection, helps to accurately grasp the atmospheric pollution situation, and provides a reliable basis for subsequent precise governance;

[0039] 2. The present invention is provided with a self-cleaning module, which can effectively dissolve and take away the particles adsorbed on the capture net by regularly spraying a specially formulated cleaning liquid to the bionic capture net, thereby preventing the particles from long-term accumulation and affecting the adsorption performance of the capture net and the detection effect of the sensing unit. The calibration unit uses calibration equipment such as a built-in standard atmospheric particle sample storage device to calibrate the entire detection system at preset time intervals, thereby ensuring that the detection system can operate stably for a long time, reducing the detection data deviation caused by equipment failure or performance degradation, and continuously providing stable data support for air pollution prevention and control work;

[0040] 3. The present invention utilizes the real-time data transmission and early warning mechanism in the intelligent data processing and transmission module, and with the help of 4G / 5G communication technology, can transmit the detected atmospheric particle data in real time to the atmospheric pollution prevention and control monitoring center and related mobile applications, ensuring that relevant departments and personnel can obtain the latest atmospheric pollution data in the first time. Moreover, when the concentration of key pollutant particles exceeds the standard, the early warning mechanism will be triggered immediately to send early warning information to surrounding residents, environmental protection departments and related enterprises, reminding all parties to take corresponding pollution prevention and control measures such as limiting industrial emissions and advocating green travel in a timely manner, so as to achieve timely intervention in atmospheric pollution, minimize the harm caused by atmospheric pollution to the environment and human health, and effectively improve the timeliness and effectiveness of atmospheric pollution prevention and control. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is the overall flow chart of the present invention.

[0042] Figure 2 It is a system module diagram of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0044] As attached Figure 1-2 As shown, the present invention provides an atmospheric particle detection method for preventing and controlling atmospheric pollution, which mainly adopts the atmospheric particle detection system constructed by the present invention for detection, and specifically includes the following steps:

[0045] S1. Construction and deployment of biomimetic particle capture structures

[0046] Mimicking the properties of spider silk to create a capture net:

[0047] First, prepare the filamentous material: take an appropriate amount of polyimide powder, dissolve it in a mixed solvent of N,N-dimethylformamide (DMF) and N-methylpyrrolidone (NMP), so that the mass volume ratio of polyimide to the mixed solvent reaches 1:12, and use a magnetic stirrer to stir at room temperature for 3 hours to form a uniform polyimide solution;

[0048] Next, graphene quantum dots were accurately weighed at a mass fraction of 8% and added to the above solution, and then the solution was placed in an ultrasonic cleaner for ultrasonic dispersion for 3 hours to ensure that the graphene quantum dots were evenly dispersed in the polyimide solution;

[0049] Then, the dispersed solution is loaded into a syringe with a metal needle, the syringe is connected to a high-voltage power supply, the voltage is set to 20kV, the receiving distance is 15cm, and the electrospinning operation is performed on a grounded collection device (such as a roller covered with aluminum foil). After several hours of spinning process, the filamentous material is collected.

[0050] Finally, the collected filamentary materials were woven into a three-dimensional mesh structure with a specific pore structure using a special weaving mold. The average pore size of the mesh structure was measured to be 5 μm, which can effectively adsorb atmospheric particles of different sizes in the passing airflow. In addition, in order to enhance the adsorption effect, after the preparation of the filamentary materials, a nano-scale sticky protrusion structure was manufactured through a plasma etching process. In the specific operation, oxygen was used as the reaction gas, the power of the plasma etching equipment was set to 200 W, and the processing time was 20 minutes, so that sticky protrusions with an average height of 300 nm and a spacing of about 1000 nm were formed on the surface of the filamentary material.

[0051] The sensing unit is designed based on the sensing function of insect antennae

[0052] Preparation of organic semiconductor material poly (3,4-ethylenedioxythiophene) - polystyrene sulfonic acid:

[0053] 3,4-ethylenedioxythiophene (EDOT) monomer is placed in an aqueous solution containing an appropriate amount of ammonium persulfate (APS) as an initiator, and the temperature of the reaction system is controlled at 3°C ​​under nitrogen protection, and the polymerization reaction is continued for 9 hours to obtain a PEDOT polymer. Subsequently, the PEDOT polymer is mixed with polystyrene sulfonic acid in a mass ratio of 1:2, and stirred for 2 hours using a high-speed stirrer to uniformly mix them to obtain a PEDOT:PSS solution. The solution is coated on a flexible plastic substrate by a spin coating process to form a flexible, bendable substrate, and then dried for later use;

[0054] Preparation of metal organic framework materials (MOFs) for functional coatings:

[0055] Zinc nitrate was selected as a metal ion source and terephthalic acid as an organic ligand, and the mixture was mixed in an organic solvent of N,N-dimethylformamide (DMF) at a molar ratio of 1:1.5, and the mixture was reacted under heating reflux for 18 hours. After the reaction, the mixture was centrifuged, washed with ethanol for multiple times, and dried in vacuum at 60°C to obtain MOFs powder. The MOFs powder was dispersed in ethanol to prepare a solution with a concentration of 15 mg / mL, and the solution was coated on the prepared PEDOT:PSS substrate by dipping, and dried for later use.

[0056] Molecularly Imprinted Polymers (MIPs) for Preparation of Functional Coatings:

[0057] Benzopyrene in polycyclic aromatic hydrocarbons was used as the target organic pollutant molecule (template molecule), methyl methacrylate was selected as the functional monomer, ethylene glycol dimethacrylate was selected as the crosslinker, and azobisisobutyronitrile was selected as the initiator. After mixing in a certain proportion in a chloroform organic solvent, the mixture was prepolymerized at 0°C for 1.5h, and then the temperature was raised to 60°C for polymerization for 9h. After the reaction, the template molecule was removed by Soxhlet extraction to obtain MIPs powder. The MIPs powder was dispersed in toluene to prepare a solution with a concentration of 8mg / mL, and the solution was also coated on the PEDOT:PSS substrate coated with MOFs by dipping, and finally a sensing unit with a functional coating was formed.

[0058] The prepared multiple sensing units are installed at different positions of the woven capture net to ensure that they can effectively sense the changes in electrical parameters caused by the approach of particles.

[0059] S2. Installation and debugging of optical detection and analysis system

[0060] Laser scanning imaging module:

[0061] At the appropriate position behind the installed bionic particle capture structure, multiple sets of high-precision laser scanning devices are installed. The laser wavelength of the laser scanning device is set to a 532nm green laser, and the parameters are set according to the scanning angle range of 90° and the scanning frequency of 50Hz. At the same time, a high-speed camera with a frame rate of 1000 frames per second and a resolution of 1920×1080 pixels is equipped to enable it to accurately capture the optical image generated by the interaction between the laser and the particles, and then obtain information such as the size, shape, and distribution state of the particles. Based on the differences in the optical response of particles of different materials to the laser, the particle composition is preliminarily distinguished, such as distinguishing between mineral particles that may come from industrial emissions and carbon-based particles from traffic exhaust.

[0062] Spectral analysis subsystem:

[0063] Install high-resolution Raman spectrometer and infrared spectrometer, and adjust the spectral resolution of Raman spectrometer to 1cm -1 The excitation wavelength was set to 785 nm, the power was set to 300 mW, and the spectral resolution of the infrared spectrometer was adjusted to 0.5 cm - 1. The scanning range is set to 4000cm -1 -400cm -1 ， The two spectrometers can perform precise spectral analysis on particles irradiated by lasers, thereby accurately determining the specific chemical components contained in the particles and the proportion of each component. For example, they can accurately detect the content of acidic particulate matter formed by sulfur dioxide and nitrogen oxides, as well as organic pollutants such as polycyclic aromatic hydrocarbons.

[0064] S3, configuration and operation of intelligent data processing and transmission module

[0065] Artificial intelligence algorithm recognition and classification:

[0066] Construct an atmospheric particle recognition model based on the combination of convolutional neural network algorithm and recurrent neural network, where the convolutional neural network part uses the ResNet-50 network architecture as the basis, and the recurrent neural network part uses the long short-term memory network. Collect and organize no less than 10,000 sets of labeled atmospheric particle sample data from different environments, including different functional areas of the city, different seasons, and different pollution levels. Use these data to train the model. After multiple iterations of training, the accuracy of the model on the validation set reached 92%, and the recall rate reached 88%, meeting the expected requirements. Afterwards, the image data and spectral data obtained by the optical detection system are input into the trained model in real time to automatically and accurately identify the type, concentration, source and other information of atmospheric particles, and track their dynamic change trends, such as real-time grasp of the increase and decrease of industrial emission particle concentrations in different time periods;

[0067] Real-time data transmission and early warning mechanism:

[0068] 5G communication technology is used as the wireless communication module to ensure that the data transmission rate is stable at more than 100Mbps. According to the atmospheric environment quality standards formulated by the local environmental protection department, the corresponding pollution threshold is set. For example, the concentration threshold for PM2.5 is set at 35μg / m 3 The sulfur dioxide concentration threshold is set at 150 μg / m 3 When the concentration of key pollutant particles detected exceeds these preset thresholds, the system immediately triggers the early warning mechanism and sends early warning information to surrounding residents, relevant environmental protection departments and nearby industrial enterprises through text messages, mobile application push, etc., reminding all parties to take corresponding pollution prevention and control measures in a timely manner, such as residents reducing outdoor activities and enterprises adjusting production processes or limiting production.

[0069] S4, self-cleaning and calibration mechanism setting and operation

[0070] Self-cleaning module:

[0071] Prepare a cleaning solution, using 35% ethanol as an organic solvent to dissolve some organic pollutant particles, 12% sodium dodecylbenzene sulfonate as a surfactant to enhance the cleaning effect, and 8% benzotriazole as an anti-corrosion additive to prevent corrosion to the capture net and the sensing unit. Load the cleaning solution into a spray system with a pressure regulator and flow control device, set the spray pressure of the cleaning solution to 0.3MPa and the flow rate to 30ml / min, and use an intelligent control system to automatically spray the cleaning solution onto the bionic capture net every 24g to dissolve and take away the particles adsorbed on the capture net, thereby ensuring the adsorption performance of the capture net and the sensitivity of the sensing unit. At the same time, the system can dynamically adjust the spray time interval through a built-in intelligent algorithm according to the actual particle adsorption amount of the capture net, and the adjustment period shall not exceed 6h.

[0072] Calibration Unit:

[0073] Utilizing the built-in standard atmospheric particle sample storage device, according to different calibration requirements, a variety of different pollution levels and types of atmospheric environments are simulated, and standard particle samples with a particle size of 0.01μm-100μm and different compositions are released for calibration of the detection system. During the calibration process, the wavelength accuracy of the calibration light source is controlled within ±1nm, and the output signal accuracy of the standard electrical signal generator is controlled within ±0.1%, ensuring that the calibration error of the optical detection system is controlled within ±5%, and the calibration error of the electrical sensing unit is controlled within ±3%. The relevant records of each calibration operation will be kept in detail for no less than 3 years, which is convenient for subsequent tracing and analysis of system performance changes, and timely discovery and resolution of potential detection accuracy issues.

[0074] In summary, the atmospheric particle detection method of the present invention can operate stably in actual atmospheric environment monitoring, accurately detect atmospheric particle conditions, and provide powerful data support and decision-making basis for atmospheric pollution prevention and control work.

[0075] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for detecting atmospheric particles for preventing and controlling air pollution, characterized in that: The specific steps include: S1. Enrich and preliminarily sense atmospheric particles using a bionic particle capture structure, wherein the bionic particle capture structure includes a capture net made by imitating the characteristics of spider silk and a sensing unit designed by imitating the sensing function of insect antennae, wherein: The capture net made by imitating the characteristics of spider silk uses nanomaterial technology to synthesize silk materials and weave them into a three-dimensional mesh structure with a specific pore structure to absorb particles in the atmosphere; The sensing units designed to imitate the sensing function of insect antennae are made of a flexible, bendable organic semiconductor material as a substrate and the surface is covered with a functional coating that has specific adsorption and electrical response to common components in atmospheric particles, and are installed at different positions of the capture net; S2. Detect and analyze the atmospheric particles processed by the bionic capture structure through an optical detection and analysis system, wherein the optical detection and analysis system comprises a laser scanning imaging module and a spectral analysis subsystem, wherein: The laser scanning imaging module uses multiple sets of high-precision laser scanning devices to scan the area where the particles are located at a specific angle and frequency; The spectral analysis subsystem uses a high-resolution Raman spectrometer and infrared spectrometer to perform spectral analysis on the particles irradiated by the laser; S3, using intelligent data processing and transmission module to process and transmit the detection data, the intelligent data processing and transmission module includes artificial intelligence algorithm recognition and classification and real-time data transmission and early warning mechanism, wherein, Artificial intelligence algorithm recognition and classification: The image data and spectral data obtained by the optical detection system are input into the atmospheric particle recognition model constructed by the deep learning algorithm; The real-time data transmission and early warning mechanism transmits the detected atmospheric particle data to the air pollution prevention and control monitoring center and related mobile applications in real time through the wireless communication module; S4. Use a self-cleaning and calibration mechanism to ensure the long-term stable operation of the detection system. The self-cleaning and calibration mechanism includes a self-cleaning module and a calibration unit, wherein: The self-cleaning module sprays a specially formulated cleaning fluid onto the bionic capture net to dissolve and remove particles adsorbed on the capture net; The calibration unit uses the built-in standard atmospheric particle sample storage device and the calibration equipment of the calibration light source and standard electrical signal generator to calibrate the detection system to ensure the accuracy and reliability of the detection data.

2. The atmospheric particle detection method according to claim 1, characterized in that: In the capture net made by imitating the characteristics of spider silk in step S1, the filamentous material is based on polyimide nanofibers, doped with 5%-10% of graphene quantum dots by mass fraction, and prepared by electrospinning process. The pores of the woven three-dimensional mesh structure are between 0.1-10 μm to adapt to the adsorption of atmospheric particles of different particle sizes. The specific preparation steps are as follows: The polyimide powder is dissolved in a mixed solvent of N,N-dimethylformamide and N-methylpyrrolidone, the mass volume ratio of polyimide to the mixed solvent is 1:10-1:15, and the solution is stirred evenly to form a uniform polyimide solution. Graphene quantum dots are added and ultrasonically dispersed for 2-4 hours to ensure that the graphene quantum dots are evenly dispersed in the polyimide solution. The solution is loaded into a syringe with a metal needle, connected to a high-voltage power supply, the voltage is set to 15-25kV, the receiving distance is 10-20cm, and electrostatic spinning is performed on a grounded collection device to collect filamentous materials. The collected filamentous materials are formed into a three-dimensional network structure with a specific pore structure through a weaving process or a specific mold.

3. The atmospheric particle detection method according to claim 2, characterized in that: In the capture net made by imitating the characteristics of spider silk, the surface of the silk material has nano-scale sticky protrusions, the average height of the sticky protrusions is 100nm-500nm, and the spacing is between 500nm-2000nm, which is used to enhance the adsorption effect on atmospheric particles; The specific preparation method is: after the filamentary material is prepared, a nano-scale sticky protrusion structure is manufactured on the surface of the filamentary material through plasma etching using oxygen or methane reaction gas at a power of 100-300W and a time of 10-30 minutes.

4. The atmospheric particle detection method according to claim 1, characterized in that: In the sensing unit designed to imitate the sensing function of insect antennae in step S1, the organic semiconductor material is poly(3,4-ethylenedioxythiophene)-polystyrene sulfonic acid; The functional coating contains 15%-20% by mass of metal organic framework materials MOFs for specific adsorption of heavy metal ions, and 10%-15% by mass of molecular imprinting polymers MIPs for identifying specific organic pollutant components, and preliminarily judging the type and approximate concentration range of the particles by sensing the changes in electrical parameters caused by the approach of the particles.

5. The atmospheric particle detection method according to claim 4, characterized in that: The electrical parameters include resistance and capacitance, and the atmospheric particle situation is judged by detecting the changes in resistance and capacitance. The resistance change range between 100Ω-1000Ω corresponds to the adsorption of different types of particles, and the capacitance change range between 10pF-100pF is used to assist in concentration range judgment. When the resistance change rate exceeds 10% or the capacitance change rate exceeds 20%, an instruction for further detection and analysis of the particles is triggered.

6. The atmospheric particle detection method according to claim 1, characterized in that: In the laser scanning imaging module of step S2, the laser scanning device uses a green laser with a wavelength of 532nm, the scanning angle range of the laser beam is 30°-150°, the scanning frequency is 50Hz, and a high-speed camera is used to capture the optical image generated by the interaction between the laser and the particles to obtain the size, shape, distribution state of the particles and perform preliminary component differentiation based on the difference in optical response. The frame rate of the high-speed camera is ≥1000 frames / second, and the resolution reaches 1920×1080 pixels.

7. The atmospheric particle detection method according to claim 1, characterized in that: In the spectral analysis subsystem of step S2, the spectral resolution of the Raman spectrometer reaches 1 cm -1 , the excitation wavelength is 785nm, the power is 300mW; The spectral resolution of the infrared spectrometer reaches 0.5cm -1 , scanning range is 4000cm -1 -400cm -1 .

8. The atmospheric particle detection method according to claim 1, characterized in that: In the artificial intelligence algorithm identification and classification in step S3, the deep learning algorithm is an algorithm combining a convolutional neural network algorithm and a recurrent neural network algorithm, wherein; The convolutional neural network part uses the ResNet-50 network architecture as its basis, and the recurrent neural network part uses the long short-term memory network. It is trained through a large amount of labeled atmospheric particle sample data under different environments and different pollution levels. The number of training samples is no less than 10,000 groups, which are used to automatically and accurately identify the types, concentrations, sources of atmospheric particles and track their dynamic changes.

9. The atmospheric particle detection method according to claim 1, characterized in that: In the real-time data transmission and early warning mechanism of step S3, the wireless communication module adopts 4G / 5G communication technology with a transmission rate of ≥100Mbps, and triggers the early warning mechanism to send early warning information when the concentration of key pollutant particles exceeds the standard based on the preset pollution threshold. The pollution threshold is set according to the atmospheric environment quality standards of different regions. For example, the concentration threshold for PM2.5 is set at 35μg / m 3 The sulfur dioxide concentration threshold is set at 150 μg / m 3 .

10. The atmospheric particle detection method according to claim 1, characterized in that: In the self-cleaning module of step S4, the composition of the cleaning liquid includes 30%-40% ethanol as an organic solvent to dissolve some organic pollutant particles, 10%-15% sodium dodecylbenzene sulfonate as a surfactant to enhance the cleaning effect, and 5%-10% benzotriazole as an anti-corrosion additive to prevent corrosion of the capture net and the sensing unit. The injection pressure of the cleaning liquid is between 0.1MPa-0.5MPa, and the flow rate is 10ml / min-50ml / min. In the calibration unit, the standard atmospheric particle sample storage device can simulate a variety of atmospheric environments with different pollution levels and types, and release standard particle samples with a particle size of 0.01μm-100μm and different compositions for calibrating the detection system. The wavelength accuracy of the calibration light source is within ±1nm, and the output signal accuracy of the standard electrical signal generator is within ±0.1%.

Images