A particle detection method and system

Abstract

The present application belongs to the field of detection, and particularly relates to a particle detection method for solving the problem of large detection error. The particle detection method comprises: applying multiple laser pulses to a particle flow to excite plasma by heat accumulation; obtaining spectral data according to plasma light signals; detecting particle indexes based on the spectral data; and the single-pulse energy is less than a preset energy threshold and the pulse interval is less than the particle thermal relaxation time. The high-frequency low-energy laser pulse is used for plasma excitation, the particle flow is used as the excitation form to solve the technical problems encountered in excitation, and a new type of plasma is obtained through a new heat accumulation excitation mode. The new type of plasma has good position stability and form stability, so that the plasma light signal has good stability and consistency; and the single-pulse energy is low, so that the air is not broken and the shock wave effect is weak, so that the detection error is small and the detection stability is good.

Description

Technical Field

[0001] This application belongs to the field of detection technology, and in particular relates to a particle detection method and system. Background Technology

[0002] Laser-induced breakdown spectroscopy (LIBS) is an atomic emission spectroscopic analysis technique. The LIBS analysis process typically involves the following key steps: First, a solid-state laser (such as an Nd:YAG laser with a wavelength of 1064 nm) focuses a single high-energy laser pulse (single pulse energy greater than 10 millijoules) onto the sample surface. Through the inverse bremsstrahlung effect, extremely high energy is rapidly deposited in the focused region, causing intense ablation and phase transition in the sample material, leading to breakdown ionization and the formation of a high-temperature, high-density plasma. Next, atoms and ions in the plasma, in their excited states, emit light signals with characteristic wavelengths as they transition to lower energy states. Then, the emitted light signals are collected by a light collector and transmitted via optical fiber to a spectrometer. The spectrometer disperses the composite light into a spectrum and records the intensity of light at different wavelengths, forming spectral data. Finally, the computer system processes the spectral data, analyzing the wavelengths and intensities of atomic and ionic characteristic spectral lines to detect the types and contents of elements in the sample. In addition, it can also detect other indicators of the sample. Taking coal as an example, other indicators include one or more of the following: calorific value, ash content, volatile matter, moisture, ash fusion point, and true density.

[0003] Currently, when using LIBS technology to detect sample indicators, the main problem is that the detection results have relatively large errors. Summary of the Invention

[0004] Existing technologies using LIBS to detect sample parameters suffer from significant errors in the detection results.

[0005] To address the aforementioned technical problems, this application provides a particle detection method and system according to some embodiments.

[0006] According to one aspect of the present invention, a particle detection method is provided, comprising: Multiple laser pulses are applied to a particle stream to excite plasma through thermal accumulation. Spectral data is obtained based on the light signals emitted by the plasma; Based on the spectral data, at least one indicator of the particles is detected; The laser pulse is a low-energy pulse with a single pulse energy less than a preset energy threshold, and its pulse interval is less than the thermal relaxation time of the particle.

[0007] Optionally, the plasma is excited by: Multiple plasmas are excited within a preset integration time; where the preset integration time is the time for acquiring optical signals. Obtain spectral data, including: Spectral data is obtained by accumulating the light signals emitted by multiple plasmas.

[0008] Optionally, spectral data is obtained, including: Obtain spectral data containing molecular characteristic lines, atomic characteristic lines, and ionic characteristic lines; At least one indicator for detecting particles includes: Different indicators of particles are detected based on molecular characteristic spectral lines, atomic characteristic spectral lines, and / or ionic characteristic spectral lines.

[0009] Optionally, the percentage P of the integrated area of ​​the molecular characteristic spectral lines relative to the total integrated area of ​​the net emission spectrum ranges from 2% to 20%, where ; Where A is the integral area of ​​the molecular characteristic spectral line, A total It represents the total integral area of ​​the net emission spectrum.

[0010] Optionally, multiple laser pulses are applied to the particle flow, including: Multiple laser pulses are applied to a dilute phase particle flow, wherein the volume concentration of particles in the dilute phase particle flow ranges from 0.01% to 5%, and the particle size is less than 0.2 mm.

[0011] Optionally, the spatial length of the excited plasma ranges from 0.5 to 1 millimeter.

[0012] Optionally, the preset energy threshold is 1 millijoule, and the repetition frequency of the laser pulse is greater than 20 kHz and the pulse width is less than 30 nanoseconds.

[0013] According to another aspect of the present invention, a particle detection system is provided, comprising: A laser is used to generate multiple laser pulses and apply them to a particle stream to excite plasma through thermal accumulation; wherein the laser pulses are low-energy pulses with single-pulse energy less than a preset energy threshold, and the pulse interval is less than the thermal relaxation time of the particles. A spectrometer is used to collect the light signals emitted by the plasma and obtain spectral data; A detection component for detecting at least one indicator of particles based on the spectral data.

[0014] Optionally, a laser is used to excite multiple plasmas within a preset integration time; wherein the preset integration time is the time for acquiring the optical signal; A spectrometer is used to obtain spectral data based on the accumulation of light signals emitted by multiple plasmas.

[0015] Optionally, a spectrometer is used to obtain spectral data including molecular characteristic spectral lines, atomic characteristic spectral lines and ionic characteristic spectral lines; The detection component is used to detect different indicators of particles based on molecular characteristic spectral lines, atomic characteristic spectral lines, and / or ionic characteristic spectral lines.

[0016] Optionally, the percentage P of the integrated area of ​​the molecular characteristic spectral lines relative to the total integrated area of ​​the net emission spectrum ranges from 2% to 20%, where ; Where A is the integral area of ​​the molecular characteristic spectral line, A total It represents the total integral area of ​​the net emission spectrum.

[0017] Optionally, a laser is used to apply multiple laser pulses to a dilute phase particle flow, wherein the volume concentration of particles in the dilute phase particle flow ranges from 0.01% to 5%, and the particle size is less than 0.2 mm.

[0018] Optionally, the spatial length of the laser-excited plasma ranges from 0.5 to 1 millimeter.

[0019] Optionally, the preset energy threshold is 1 millijoule, and the repetition frequency of the laser pulse is greater than 20 kHz and the pulse width is less than 30 nanoseconds.

[0020] Alternatively, the laser is a fiber laser.

[0021] The above-described technical solution of the present invention has at least the following beneficial technical effects: Through long-term experimentation and research, the inventors discovered that the large detection error in existing LIBS technology for detecting sample parameters is due to the high energy of the single-pulse ablation excitation plasma, which can cause air breakdown and strong shock waves, resulting in significant detection errors. Based on this discovery, the inventors, after overcoming numerous difficulties through further experimentation and research, proposed a new solution: using high-frequency, low-energy laser pulses for plasma excitation and employing a particle flow as the sample excitation mode to solve the technical problem of difficulty in exciting plasma, which can easily induce sample combustion. Through a newly explored thermal accumulation excitation mode, a novel, more compact, and stable plasma is obtained. Because the new plasma has advantages such as good positional and morphological stability, the emitted optical signal exhibits better stability and consistency; moreover, due to the low single-pulse energy, air breakdown is not caused, and the shock wave effect is very weak, resulting in a smaller detection error in this invention. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic flowchart of the particle detection method in the embodiments of this application; Figure 2A This is a schematic diagram of the structure of plasma obtained by high-energy single-pulse ablation excitation in existing technology; Figure 2B This is a schematic diagram of the plasma obtained by low-energy multi-pulse thermal accumulation excitation in the embodiments of this application; Figure 3 This is a comparison chart of the spectral data fluctuations between the embodiments of this application and the prior art; Figure 4A It is a spectral diagram in the existing technology; Figure 4B This is a spectral diagram from an embodiment of this application; Figures 4C to 4G This is a comparison diagram of molecular characteristic spectra of embodiments of this application and prior art; Figure 4H This is a comparison diagram of atomic characteristic spectral lines between embodiments of this application and prior art; Figure 4I and Figure 4J This is a comparison diagram of the ion characteristic spectra of the embodiments of this application and the prior art; Figure 4K This is a comparison chart of the intensity ratios of carbon-related molecular characteristic spectral lines and total spectral data intensity between the embodiments of this application and the prior art; Figure 5 This is a schematic diagram of the particle detection system in an embodiment of this application; Figure 6 This is a schematic diagram of the online coal quality detection system in the embodiments of this application; Figure 7 This is a cross-sectional view of the measuring chamber in an embodiment of this application; Figure 8 This is a schematic diagram of the light-receiving structure in an embodiment of this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the various embodiments of this application will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details are presented in the various embodiments of this application to facilitate a better understanding of the application. However, the technical solutions claimed in this application can be implemented even without these technical details and various variations and modifications based on the following embodiments. The division of the various embodiments below is for ease of description and should not constitute any limitation on the specific implementation of this application. The various embodiments can be combined with and referenced by each other without contradiction.

[0025] Figure 1 This is a schematic flowchart of the particle detection method in the embodiments of this application.

[0026] like Figure 1 As shown, a preferred embodiment of this application provides a particle detection method, comprising the following steps: Step 102: Apply multiple laser pulses to the particle flow to excite plasma through thermal accumulation; Step 104: Obtain spectral data based on the light signal emitted by the plasma; Step 106: Based on spectral data, detect at least one indicator of the particles; The laser pulse is a low-energy pulse with a single pulse energy less than a preset energy threshold, and its pulse interval is less than the thermal relaxation time of the particle.

[0027] The principle of plasma excitation through thermal accumulation is as follows: The laser pulse is a low-energy pulse with a single pulse energy less than a preset energy threshold. Its instantaneous power density does not reach the breakdown threshold of the particles, so it is impossible to excite high-temperature, high-density plasma with a single laser pulse. However, the pulse interval of the laser pulse (e.g., on the order of nanoseconds to microseconds) is less than the thermal relaxation time of the particles. When multiple laser pulses act on the particle flow, the heat inside the particles will accumulate successively under the continuous action of multiple laser pulses. Correspondingly, the particles will heat up successively, forming a thermal accumulation effect. As the temperature continues to rise, the particles will successively undergo processes such as water evaporation, devolatile matter removal, gasification, and ionization, forming a new plasma different from high-temperature, high-density plasma.

[0028] For ease of understanding, the research and development process of this invention will be described in detail below, mainly involving the identification of technical problems, investigation of causes, design of solutions, technical difficulties encountered and corresponding solutions, new solutions and their resulting technical effects, etc.

[0029] Through numerous experiments and studies, the inventors discovered that in existing technologies, high-energy single pulses, while inducing ablation and generating plasma, are accompanied by air breakdown and strong shock waves, leading to large errors in indicator detection, poor repeatability, and long processing times. The specific analysis is as follows: 1. High single-pulse energy (typically greater than 10 millijoules) can cause air breakdown. Specifically, the high energy of the single pulse results in a power density in the focused region that is much higher than the air breakdown threshold. While exciting a high-temperature, high-density plasma from the sample, it also breaks down the air surrounding the focused region, forming accompanying air plasma, thus significantly increasing the spatial size of the plasma. Taking coal particles as an example, the plasma obtained by high-energy single-pulse ablation excitation is as follows... Figure 2A As shown, the spatial length of the entire plasma is generally between 3.5 and 4.5 mm.

[0030] Air breakdown can lead to problems such as large detection errors, poor repeatability, and long detection times. Specifically: ① Since the main components of air are oxygen and nitrogen, which are generally different from the main components of the sample (e.g., coal samples mainly consist of carbon, silicon, and calcium), the accompanying air plasma can interfere with the optical signal of the sample plasma, resulting in incorrect detection results and large detection errors. ② Because air breakdown consumes pulse energy, it competes with the sample for pulse energy, weakening the effective energy coupling between the pulse and the sample. This may cause insufficient sample excitation, leading to a decrease in the intensity of characteristic spectral lines and large detection errors. Furthermore, it may generate invalid spectral data, resulting in long detection times. ③ Due to the uncertainty of air breakdown (it may break down before or after the sample), and the uncertainty of the air breakdown point, the position and morphology of the plasma fluctuate, resulting in poor stability and repeatability of the plasma optical signal, leading to large detection errors and poor repeatability.

[0031] 2. Single-pulse energy is relatively high, and it also generates a strong shock wave, further disturbing the plasma and causing significant uncertainty in its evolution process. This further exacerbates the fluctuations in the plasma's position and morphology. Taking coal particles as an example, the position and morphology of the plasma obtained by high-energy single-pulse excitation are as follows: Figure 2A As shown, the stability and repeatability of the plasma's optical signal become worse at this time, resulting in larger detection errors and worse repeatability.

[0032] Based on this discovery, the inventors realized that it was necessary to reduce the single-pulse energy. However, if the single-pulse energy was reduced to a certain level, although it would not cause air breakdown, it would also fail to effectively excite the plasma. In other words, simply reducing the single-pulse energy could not solve the problem. Faced with this challenge, the inventors proposed through various explorations that the excitation mode could be changed from ablation excitation based on high-energy single pulses to thermal accumulation excitation based on low-energy multi-pulses, thus perfectly solving the problem. This would prevent air breakdown and effectively excite the plasma.

[0033] Based on low-energy multi-pulse thermal accumulation excitation, in addition to requiring the single-pulse energy to be less than a preset energy threshold, the pulse interval must also be less than the thermal relaxation time of the sample to achieve multi-pulse thermal accumulation. Since solid-state lasers are usually used to emit low-frequency and high-energy laser pulses (conventional parameters are: repetition frequency less than 20 Hz, single-pulse energy greater than 10 mJ, and pulse width less than 15 nanoseconds), they are not suitable for the embodiments of this invention. To address this problem, the inventors have discovered through long-term research that setting the fiber laser to emit high-frequency and low-energy laser pulses, and by setting parameters, such as setting the repetition frequency to be greater than 20 kHz, the single-pulse energy to be less than 1 mJ, and the pulse width to be less than 30 nanoseconds, can meet the plasma excitation requirements of this invention.

[0034] Another technical challenge exists in the process of exciting plasma with high-frequency low-energy laser pulses. When high-frequency low-energy laser pulses are applied to the sample, it is difficult to excite plasma. Instead, it is easy to induce the sample to burn. This means that only the continuous radiation background of the burning flame can be recorded, and the characteristic spectrum reflecting the elemental composition cannot be obtained, making it impossible to detect the indicators. Through long-term research, the inventors discovered that sample morphology also affects plasma excitation. Previously used dense sample morphologies (such as compressed or stacked samples) were highly susceptible to combustion. This is primarily due to the competition between heat transport and heat accumulation within dense samples. Specifically, in dense samples, particles are in close contact, forming a continuous heat conduction network. When a laser pulse acts on the focused area, the particles absorb energy and their temperature rises rapidly. Due to the good thermal contact between particles, the heat at the focal point diffuses quickly to surrounding particles via heat conduction, leading to two consequences: First, the heat in the focused area cannot accumulate effectively at a single point, and the energy density is insufficient to reach the ionization threshold of the sample, thus preventing plasma formation. Second, the outward diffusion of heat causes the temperature of surrounding particles to rise continuously. When this temperature reaches the sample's ignition point (e.g., the ignition point of coal powder is typically 300 to 500°C), localized combustion occurs. During combustion, the sample undergoes an oxidation reaction, releasing a large amount of heat, further intensifying heat diffusion and creating a positive feedback loop. Ultimately, this causes the focused area to completely enter a combustion state rather than a plasma-excited state.

[0035] Based on this discovery, the inventors, through experiments and research, found that particle flow can be used as the sample excitation mode to solve this problem. Specifically, in the particle flow mode, the heat conduction path between particles is effectively blocked, and the heat transfer mode changes from solid-state heat conduction to a weak coupling mode dominated by convection and radiation. The heat dissipation rate is significantly reduced. When a high-frequency, low-energy laser pulse sequence acts on the particles, each particle absorbs the pulse energy as an independent thermodynamic system. Due to the gaps between particles and the weak thermal contact, most of the heat absorbed is not diffused to the surroundings through solid-state heat conduction, but accumulates inside the particles. Under these conditions, the rate at which particles absorb energy is much faster than the rate at which heat is dissipated, causing the particles to heat up rapidly in a very short time (usually within several pulse cycles), successively passing through stages such as water evaporation, volatile matter removal, and vaporization, and finally reaching the ionization threshold to form plasma. That is, by using the particle flow mode, the heat conduction environment between particles can be reconstructed, and the energy coupling mode changes from heat diffusion-dominated to heat accumulation-dominated, thus realizing the effective excitation of particle flow by high-frequency, low-energy laser pulses. Moreover, since the heat is highly localized within a single particle throughout the process, it will not induce combustion of neighboring particles, thus effectively avoiding combustion interference in dense sample morphology and ensuring reliable acquisition of plasma optical signals, laying the foundation for the reliable application of this technology in index detection.

[0036] After solving the above technical problems, the embodiments of the present invention achieve the goal of using low-energy multi-pulse thermal accumulation excitation of particle flow to obtain a novel plasma that is significantly different from high-temperature and high-density plasma.

[0037] Because energy coupling is more selective in the thermal accumulation excitation mode, low-energy pulses can be used. Their power density is insufficient to break down air, and the energy competition between air and the particle flow is negligible. The pulse energy is mainly used for thermal accumulation excitation of the particle flow to obtain plasma. Furthermore, due to the low energy of a single pulse, the shock wave effect is also very weak, and the plasma is minimally disturbed, resulting in a more compact and stable plasma. Specifically, this includes: ① a significantly smaller spatial size. Taking pulverized coal particle flow as an example, the plasma obtained by low-energy multi-pulse thermal accumulation excitation is as follows: Figure 2B As shown, Figure 2B and Figure 2A By comparing images of the same size, it can be seen that the spatial size of the plasma in this embodiment of the invention is significantly smaller. The spatial length of the plasma is generally between 0.5 and 1 mm, such as 0.6, 0.8, or 0.9 mm; ② It has good spatial position and morphological stability. Taking pulverized coal particle flow as an example, for instance... Figure 2B As shown, the position of the plasma is generally stable in the particle flow stream region, and the shape of the plasma is generally a stable ellipse.

[0038] The plasma in this embodiment exhibits good spatial position and morphological stability, ensuring the stability and consistency of the emitted optical signal. Stable optical signals facilitate the detection of elemental and molecular information of particles, improving detection accuracy and reducing errors. Furthermore, the consistency of the optical signal results in good detection stability and repeatability, meaning that the deviation in indicators is relatively small when repeatedly detecting the same sample. Additionally, since air breakdown does not occur, there is no interference from accompanying air plasma, and the pulse energy competition between air and particle flow is negligible, further reducing detection errors and, to some extent, addressing the problem of long detection times caused by invalid spectral data. In summary, this embodiment of the invention has several positive effects, such as small detection errors, good detection repeatability, and a reduction in detection time to some extent.

[0039] The implementation method of the laser pulse in step 102 will be described in detail below.

[0040] According to the working principle of thermally accumulated excited plasma, the single pulse energy of the laser pulse in this embodiment of the invention is less than the preset energy threshold and the pulse interval is less than the thermal relaxation time of the particle.

[0041] It should be noted that the preset energy threshold value only needs to ensure that the power density of the single pulse energy does not reach the breakdown threshold of particles and air. The specific value can be determined according to needs, experience and / or experiments.

[0042] To make it easier to understand, several methods for optionally determining the preset energy threshold value are listed below.

[0043] 1. Determine the value of the preset energy threshold based on the breakdown thresholds of particles and air. For example, assuming that a single pulse energy of less than 20 millijoules will not cause particle breakdown, and a single pulse energy of less than 5 millijoules will not cause air breakdown, then the preset energy threshold value is determined to be 5 millijoules.

[0044] 2. Determine the value of the preset energy threshold based on the breakdown thresholds of particles and air, as well as the shock wave. For example, assuming that a single pulse energy of less than 5 millijoules will not cause particle breakdown or air breakdown, and that the shock wave effect is controllable when the single pulse energy is less than 5 millijoules and negligible when the single pulse energy is less than 1 millijoule, then the value of the preset energy threshold can be determined as needed. If a controllable shock wave effect is required, then the preset energy threshold value is set to 5 millijoules; if a negligible shock wave effect is required, then the preset energy threshold value is set to 1 millijoules.

[0045] III. Preferably, for precise excitation, the preset energy threshold value can be determined based on the energy required for the microscopic particle energy level transitions of the particle elements. As a preferred example, this can be achieved by: identifying the most difficult-to-excite element among the main elements contained in the particle (e.g., carbon in coal particles); determining the single-pulse energy value required for multi-pulse thermal accumulation excitation based on the energy required for the microscopic particle transitions of the most difficult-to-excite element; and adjusting the single-pulse energy value based on empirical values ​​to obtain the preset energy threshold value, since other elements in the particle are easier to excite. To further improve accuracy, after adjusting the single-pulse energy value based on empirical values, further experiments can be conducted to determine the preset energy threshold value.

[0046] The embodiments of the present invention can be used for the detection of various substances, such as coal, cement, or biomass. Optionally, to reduce the complexity of detection, different substances are set with the same preset energy threshold, for example, the preset energy threshold for different substances is 2 millijoules. Preferably, to improve the detection accuracy, different substances are set with different preset energy thresholds, for example, the preset energy threshold for coal is 0.5 millijoules and the preset energy threshold for cement is 1 millijoule.

[0047] Preferably, the single pulse energy is less than 1 millijoules. Taking pulverized coal particle flow as an example, preferably, the single pulse energy is less than 0.5 millijoules, for example, 0.3 millijoules, and more preferably 0.2 millijoules.

[0048] Optionally, the pulse interval can be made smaller than the thermal relaxation time of the particles by setting the repetition frequency parameter of the laser pulse. Based on the thermal relaxation time of the particles, the repetition frequency of the laser pulse is generally greater than 20 kHz. Taking pulverized coal particle flow as an example, preferably, the repetition frequency is greater than 40 kHz, such as 50 kHz, and more preferably 100 kHz.

[0049] Optionally, the pulse width of the laser pulse is generally less than 30 nanoseconds. This can avoid the problem of background noise increase caused by the violent combustion of coal powder particles due to excessively long pulse width, and also avoid the problem of poor spectral data quality caused by short plasma lifetime due to excessively short pulse width. Taking coal powder particles as an example, preferably, the pulse width is less than 15 nanoseconds, such as 10 nanoseconds, and more preferably 9 nanoseconds.

[0050] The implementation method of the granular flow in step 102 will be described in detail below.

[0051] Preferably, to minimize combustion interference, the particle flow is a dilute phase particle flow, and multiple laser pulses are applied to the dilute phase particle flow to excite plasma through thermal accumulation.

[0052] Among them, dilute phase particle flow refers to a sparse particle system in a dispersed flow state, in which the particle size is relatively small and the particles maintain a large distance between each other.

[0053] Preferably, the particle size is less than 0.2 mm, such as 0.05, 0.1, 0.15 or 0.19 mm, which can greatly avoid combustion interference, obtain reliable spectral data and improve the accuracy of detection.

[0054] Preferably, the volume concentration of particles (i.e., the ratio of the total volume of particles to the total volume of the mixture) ranges from 0.01% to 5%, for example, 0.05% or 0.1% (i.e., 10...). -3 ), 0.5%, 1%, 1.5%, 2%, 2.5%, 3%, 3.5%, 4% or 4.5%. At this point, although the particles are sparse, their number is not extremely small. This can greatly avoid combustion interference to obtain reliable spectral data, and also ensure that a sufficient number of particles are excited to provide a basis for obtaining sufficiently strong spectral data, thereby improving the accuracy of detection.

[0055] It should be noted that the embodiments of this application are limited to using granular flow, but not to the specific implementation of granular flow. For relevant content, please refer to the implementation methods in related technologies.

[0056] To facilitate the reproduction of the embodiments of the present invention by those skilled in the art, several specific implementation methods of granular flow are exemplarily listed below.

[0057] I. Gravity-driven particle flow. This involves controlling the mass flow rate of the particles to allow them to fall freely under gravity. For example, in the case of pulverized coal particle flow, the mass flow rate of the particles can be controlled by a vibrating feeder.

[0058] II. Fluid Dynamics Dominated Particle Flow. Taking pneumatic fluid dynamics as an example, particles can be dispersed and flowed in a carrier gas using pneumatic conveying. Specific implementation methods are diverse, including: ① Sampling from a gas-solid two-phase flow pipeline; for example, in pulverized coal particle flow, sampling can be taken from an air-powder pipeline; ② Using negative pressure suction or positive pressure conveying to draw particles and gas into or blow them into the pipeline, controlling the particle dispersion flow by adjusting the particle mass flow rate and gas volume flow rate. In this embodiment of the invention, the particle flow pipeline setup only needs to ensure that the pulse energy can act on the particle flow and that the light signal emitted by the plasma can be collected. For example, a quartz window is typically installed on the pipeline, or the pipeline is connected to a measurement chamber. For pneumatically dominated particle flows, since the heat conduction path between particles is further blocked by the airflow, combustion interference can be further avoided.

[0059] The implementation method of step 104 will be described in detail below.

[0060] When obtaining spectral data based on the optical signal of plasma, new problems were encountered, as follows: In the existing technology, under the ablation excitation mode, due to the high energy of the single pulse, the generated plasma radiation intensity is large. Therefore, sufficient signal-to-noise ratio spectral data can be obtained based on the optical signal of a single plasma. Thus, the "single pulse-single plasma-single spectrum" processing method is adopted, that is, one laser pulse excites one plasma, and a set of spectral data is obtained based on the optical signal of one plasma, and then displayed as a complete spectrum.

[0061] In the thermal accumulation excitation mode of this invention, due to the low energy of a single pulse, the radiation intensity of a single plasma is weak. If the light signal intensity is too weak, it will be masked by the dark noise of the instrument (i.e., the execution subject of step 104, such as a spectrometer), resulting in the loss of feature information, i.e., light signal distortion. Processing such as light signal amplification and / or dark noise removal is required. In addition, in the thermal accumulation excitation mode, the repetition frequency of the laser pulse is generally as high as 20 kHz or more, far exceeding the processing capability of the spectrometer (usually the repetition frequency is less than 1 kHz). Therefore, it is impossible to achieve spectral analysis of a single plasma, i.e., it is impossible to obtain a set of spectral data from the light signal of a single plasma. The problem of limited processing capability needs to be solved by means such as setting up a cluster.

[0062] Based on this problem, the inventors believe that since the plasma and its optical signals obtained under the two excitation modes are fundamentally different, a differentiated processing method is required. After experiments and research, the following solution was proposed.

[0063] Preferably, in step 102, exciting the plasma includes: Multiple plasmas are excited within a preset integration time. In step 104, spectral data is obtained, including: Spectral data is obtained by accumulating the light signals emitted by multiple plasmas.

[0064] The preset integration time is the time for acquiring the optical signal. Taking the spectrometer as an example, the execution subject of step 104 is a spectrometer. In related technologies, the preset integration time generally refers to the photosensitivity time of the spectrometer's detector, that is, the duration of the detector's electronic shutter being open.

[0065] The preset integration time is a system parameter that can be set as needed, such as 1 second. Preferably, the optimal range for the preset integration time is from tens of milliseconds to a few seconds, which ensures that the obtained spectral data is of sufficient quality while also preventing the time taken from becoming too long.

[0066] In specific implementation, since the laser pulse repetition frequency of the present invention is high, a large amount of plasma can be excited within a preset integration time, so that spectral data can be obtained based on the accumulation of multiple plasma optical signals.

[0067] It should be noted that the improvement of this embodiment of the invention is that by setting a preset integration time parameter, the optical signals of multiple plasmas are accumulated, rather than the optical signal of a single plasma. As for how the optical signals are accumulated and how to obtain spectral data based on the accumulation of optical signals, please refer to the relevant technology, which will not be repeated here.

[0068] For ease of understanding, taking a spectrometer as an example of the execution subject of step 104, an exemplary implementation of optical signal accumulation is given.

[0069] After the light signal is split by the spectrometer's beam splitter, it illuminates the pixels of the spectrometer's detector. The incident photons excite the pixels to generate photoelectrons, which are collected and stored in the corresponding potential wells. All photoelectrons generated within the preset integration time are accumulated. After the preset integration time ends, other components of the spectrometer read the accumulated charge of each pixel and perform subsequent processing to obtain spectral data. In essence, "accumulation" is the accumulation of photogenerated charge.

[0070] As can be seen from the above analysis, the embodiments of the present invention adopt a processing method different from the prior art of "multi-pulse-multi-plasma-single spectrum", that is, multiple laser pulses excite one plasma, and a set of spectral data is obtained based on the accumulation of the optical signals of multiple plasmas, which can also be displayed in a single spectral graph.

[0071] In practice, spectral data is obtained by accumulating the optical signals of multiple plasmas, and the quality of the spectral data can be improved in at least two ways: On the one hand, it effectively enhances the intensity of spectral data, at which point the influence of instrument dark noise becomes extremely weak, thereby improving the signal-to-noise ratio.

[0072] On the other hand, although the plasma stability of the embodiments of the present invention is relatively good, the plasma still has relatively small fluctuations due to the energy fluctuations between pulses and the randomness of particles, which will cause fluctuations in the optical signal. In related technologies, the fluctuations in the optical signal are mainly reduced by averaging multiple sets of spectral data. However, since the spectral data of the embodiments of the present invention is obtained based on the accumulation of multiple plasma optical signals, it is similar to the average value of multiple sets of spectral data in related technologies. It belongs to the statistical value containing multiple optical signal fluctuations, so the fluctuation is small, the stability and repeatability are good, and its reliability is far superior to the spectral data obtained by the "single pulse-single plasma-single spectrum" method in the prior art.

[0073] To clearly demonstrate the volatility of the spectral data in this embodiment of the invention, five different coal samples were selected. The spectral data obtained were compared with the "multi-pulse-multi-plasma-single-spectrum" method in the prior art using the "single-pulse-single-plasma-single-spectrum" method in this embodiment of the invention. The RSD (relative standard deviation) of the carbon atom spectral line intensity at wavelength 247.86 nm was used to measure the volatility of the spectral data. The final experimental results of the spectral data volatility are as follows: Figure 3 As shown, from Figure 3 It can be seen that, regardless of the type of coal sample, the spectral data fluctuation of the present invention is very small, basically around 2%, while the spectral data fluctuation of the prior art is very large, with the smallest being close to 8%, and the spectral data fluctuation of different coal samples also varies greatly.

[0074] Optionally, since the spectral data of the embodiments of the present invention has very small fluctuations, it is possible to average multiple sets of spectral data regardless of whether averaging is performed. If averaging is not performed, hardware, software and time resources can be saved. If averaging is performed, fluctuations can be further reduced, and detection accuracy and repeatability can be improved.

[0075] During implementation, the improved quality of spectral data will increase detection accuracy and reduce detection errors.

[0076] In practice, due to the good repeatability of spectral data, the repeatability of the test results is better than that of existing technologies when the same sample is tested multiple times.

[0077] In practice, the detection speed of this invention will also be significantly improved, achieving second-level detection, which is far superior to the minute-level detection of existing technologies. For ease of comparison, examples are provided below for detailed analysis: Current detection time: Assuming a 20 Hz laser pulse is used, 20 plasmas can be generated per second, resulting in 20 sets of spectral data. Due to the fluctuation of spectral data, 200 to 300 sets of spectral data are needed for averaging to obtain relatively stable spectral data with an RSD of less than 5%, thus taking at least 10 seconds. Furthermore, because the high energy of a single pulse results in a strong shock wave that disperses the particle flow, the repetition frequency is generally less than 5 Hz, thus taking at least 40 seconds. In addition, the time for averaging the spectral data and the time for index detection based on the spectral data must be added, taking approximately 1 to 1.5 minutes, i.e., minute-level detection.

[0078] The detection time of this invention is as follows: Assuming a 20 kHz laser pulse is used, 20,000 laser pulses can be output per second. Even if 10 pulses generate 1 plasma (the fastest 3 pulses can generate 1 plasma), 2,000 plasmas can be obtained. Thus, a very stable set of spectral data corresponding to the accumulation of 2,000 plasma optical signals is obtained. Since the spectral data is very stable, there is no need to average the spectral data. The time for index detection based on spectral data is about 5 to 10 seconds. Therefore, the detection time of this invention can be less than 10 seconds, achieving second-level detection.

[0079] The implementation method of step 106 will be described in detail below.

[0080] Through experiments and research, the inventors have made a new discovery: different excitation modes yield different spectral data, as detailed below: In existing ablation excitation modes, high-energy-density single laser pulses (power density typically greater than 10) are used. 10 When a laser (watts per square centimeter) acts on a particle, the energy density in the focused region is extremely high, far exceeding the particle's ionization threshold. The laser breaks down the particle in a very short time, causing ionization. The strong electric field in the focused region directly induces multiphoton ionization and avalanche ionization, stripping most molecules and atoms of their outer electrons, forming a highly ionized plasma. The spectral data obtained from the plasma's emitted light signal at this point is as follows: Figure 4A As shown, the specific characteristics are: the resonance spectrum of primary ionized ions (such as carbon ions C+, silicon ions Si+, aluminum ions Al+, etc.) is dominant, the ion characteristic spectrum is strong, the atomic characteristic spectrum is relatively weak, and because the molecular bonds are suddenly broken, the intensity of the obtained molecular characteristic spectrum is particularly weak, which cannot be distinguished from the background noise. It is generally regarded as noise and cannot be used as effective information for particle detection.

[0081] In the thermal accumulation excitation mode of this invention embodiment, a low-energy, high-frequency laser pulse is used to gradually heat the particles through the thermal accumulation effect. During this process, the energy density in the laser focusing area is relatively low and cannot reach the ionization threshold of the particles. After absorbing laser energy, the particles do not ionize directly but are first converted into molecular bond energy. Therefore, the pulse energy first excites the vibrational and rotational energy levels of the molecules in the particles, generating strong molecular emission bands (such as diatomic carbon Swan bands C2Swan Bands, cyanide bands CN Violet Bands, etc.). As the temperature continues to rise to the vaporization point, neutral atoms dissociate from the particle surface, and atomic emission lines begin to appear. Finally, ionization occurs. Due to the relatively low plasma temperature (approximately 5000 to 8000 Kelvin) and limited ionization degree, the intensity is slightly weaker compared to the ion emission lines in the ablation excitation mode. At this time, the spectrum corresponding to the spectral data obtained from the plasma emission light signal is as follows: Figure 4B As shown, the specific characteristics are as follows: due to the gradual breaking of molecular bonds, compared to the sudden breaking of molecular bonds in existing technologies, the excitation in the molecular spectrum is stronger, resulting in better molecular characteristic spectral lines, which can be used as effective information for particle detection. The entire spectrum exhibits the coexistence of molecular characteristic spectral lines, atomic characteristic spectral lines, and ionic characteristic spectral lines, all of which are clearly observable. Furthermore, as particle indicators change, the intensities of molecular characteristic spectral lines, atomic characteristic spectral lines, and / or ionic characteristic spectral lines also change accordingly.

[0082] To clearly demonstrate the differences between the embodiments of the present invention and the prior art in terms of molecular, atomic, and ionic characteristic spectral lines, taking pulverized coal particle flow as an example, spectra were obtained under ablation excitation mode and thermal accumulation excitation mode, respectively. Molecular, atomic, and ionic characteristic spectral lines were extracted from the two spectra for comparative analysis. Due to the complex composition of coal, which contains multiple elements such as carbon (C), hydrogen (H), oxygen (O), nitrogen (N), silicon (Si), calcium (Ca), and magnesium (Mg), a comprehensive comparative analysis is difficult. Therefore, representative non-metallic elements (C, Si) and metallic elements (Ca) in coal were selected for analysis, as detailed below: 1. Differences in molecular characteristic spectral lines: Figures 4C to 4G The differences between the embodiments of the present invention and the prior art in terms of molecular characteristic spectral lines are demonstrated. Specifically, Figure 4C and Figure 4D This demonstrates the differences in the characteristic spectral lines of CN molecules. Figures 4E to 4G The diagram shows the differences in the characteristic spectral lines of the C2 molecule, from... Figures 4C to 4G It can be seen that in the existing technology, the intensity of most carbon-related molecular characteristic spectral lines in the ablation excitation mode is weak and basically unobservable. They cannot be distinguished from background noise and will be regarded as noise, and cannot be used as effective information for coal quality detection. In contrast, the thermal accumulation excitation mode in the embodiment of the present invention has a stronger excitation of carbon-related molecular spectra (such as CN and C2, and related to volatile matter and calorific value, etc.), which can obtain better molecular characteristic spectral lines, which can be clearly observed and can be used as effective information for coal quality detection.

[0083] 2. Differences in atomic characteristic spectral lines: Figure 4H This invention demonstrates the differences between the embodiments of the present invention and the prior art in terms of atomic characteristic spectral lines. Specifically, it shows the differences between the characteristic spectral lines of C atoms and the characteristic spectral lines of Si atoms. Figure 4H It can be seen that, compared with the atomic characteristic spectral lines in the ablation excitation mode of the prior art, the thermal accumulation excitation mode in the embodiment of the present invention has a stronger excitation of atomic emission lines and can obtain better atomic characteristic spectral lines.

[0084] 3. Differences in characteristic ionic spectral lines: Figure 4I and Figure 4JThis invention demonstrates the differences between the embodiments of the present invention and the prior art in terms of ion characteristic spectral lines. Specifically, it demonstrates the differences in the characteristic spectral lines of Ca ions. Figure 4I and Figure 4J It can be seen that, compared with the ion characteristic spectral lines in the ablation excitation mode of the prior art, the thermal accumulation excitation mode in the embodiment of the present invention has a weaker excitation of ion emission lines, but it can still obtain relatively good ion characteristic spectral lines.

[0085] 4. Difference in intensity of molecular characteristic spectral lines: In order to give those skilled in the art an intuitive impression of the intensity of molecular characteristic spectral lines in the embodiments of the present invention, two parameters are used for comparison and explanation.

[0086] ① Define parameter P to characterize the weight or proportion of the molecular net emission intensity in the total net emission intensity of the spectrum. The specific formula is as follows: ; Where A is the intensity of the molecular characteristic spectral line, generally referring to the integrated area of ​​the molecular characteristic spectral line (or characteristic band). total The intensity of all spectral data generally refers to the total integrated area of ​​the net emission spectrum of the full spectrum, and both can be measured by instruments.

[0087] In the prior art, the parameter P is generally greater than 0 and less than 2%. In the optimal case, the parameter P is less than 4.5%.

[0088] In the embodiments of the present invention, the parameter P may range from 2% to 20%, such as 2.5%, 3%, 3.5% or 4%; generally, the parameter P is greater than 4.5% and less than 19%, such as 4.6%, 4.8% or 4.9%; preferably, the parameter P is greater than 5% and less than 20%, such as 8%, 10%, 12%, 14%, 15%, 16% or 18%.

[0089] To clearly demonstrate the differences in parameter P values ​​between the embodiments of the present invention and the prior art, five different coal types (including bituminous coal, anthracite, and lignite) were selected for a comparative experiment. The final experimental results are as follows: Figure 4K As shown, since the molecular characteristic spectral lines of coal are mainly carbon-related molecular characteristic lines (mainly C2 and CN), the intensity of the carbon-related molecular characteristic spectral lines is used as A in the formula for calculation. Figure 4K The parameter P in the text, from Figure 4K It can be seen that, regardless of the type of coal sample, the parameter P in the prior art is consistently greater than 0 and less than 2%, while in the embodiments of the present invention, the parameter P is greater than 6% and less than 9%, and its value changes significantly with the coal sample. This indicates that there is a significant difference between the embodiments of the present invention and the prior art in terms of net molecular emission intensity.

[0090] ② The parameter I is set as follows: I = molecular characteristic spectral line intensity / ion characteristic spectral line intensity. Since C accounts for a large proportion of non-metallic elements and Ca accounts for a large proportion of metallic elements in coal, the CN molecular characteristic spectral line at a wavelength of 388.27 nm is selected as the numerator, and the Ca ion characteristic spectral line at a wavelength of 393.54 nm is selected as the denominator to obtain the ratio I. According to experiments, in the thermal accumulation excitation mode of this embodiment, I = 0.968, while in the ablation excitation mode of the prior art, I = 0.048. Based on this comparison of I values, it can be seen that the intensity of the molecular characteristic spectral line in this embodiment is relatively large, almost the same as the intensity of the ion characteristic spectral line, while the intensity of the molecular characteristic spectral line in the prior art is too weak, much smaller than the intensity of the ion characteristic spectral line.

[0091] In summary, compared with the prior art, the spectral data obtained by the embodiments of the present invention have more information in the dimension of molecular characteristic spectral lines, and the molecular characteristic spectral lines are related to particle indices.

[0092] Therefore, based on this new discovery, molecular characteristic spectral line information can be applied when detecting particle indicators to improve the accuracy of detection, as detailed below.

[0093] Preferably, in step 104, obtaining spectral data includes: Obtain spectral data containing molecular characteristic lines, atomic characteristic lines, and ionic characteristic lines; In step 106, at least one indicator of the particles is detected, including: Different indicators of particles are detected based on molecular characteristic spectral lines, atomic characteristic spectral lines, and / or ionic characteristic spectral lines.

[0094] In practice, the intensity of molecular characteristic spectral lines in existing technologies is always very weak and is treated as noise, making them unsuitable for particle index detection. However, in this invention, the intensity of molecular characteristic spectral lines varies depending on the particle index, thus making them suitable for particle index detection. Furthermore, since the particle index detection incorporates information from the molecular characteristic spectral line dimension as a reference, the accuracy of detection can be improved, especially for indicators strongly correlated with molecular characteristic spectral lines, such as volatile matter, calorific value (i.e., heat value), and ash content in coal, where the detection accuracy is even higher.

[0095] In practice, when the scheme of "obtaining spectral data based on the accumulation of multiple plasma optical signals" in this embodiment of the invention is combined with the scheme of "detecting different indicators of particles based on molecular characteristic spectral lines, atomic characteristic spectral lines and / or ionic characteristic spectral lines", the accuracy of detection can be significantly improved. Taking coal as an example, the detection errors of the combined scheme in this embodiment of the invention and the scheme in the prior art CN116660248A are compared in the following table.

[0096]

[0097] As can be seen from the table above, the detection errors of the combination scheme of the present invention in terms of volatile matter, calorific value and ash content are significantly smaller than those in CN116660248A. However, the detection error of carbon content is not disclosed in CN116660248A. According to the inventors' understanding, the minimum detection error of carbon content in the prior art is 1.133. In comparison, the detection error of carbon content in the combination scheme of the present invention is also reduced considerably.

[0098] Since volatile matter and calorific value are important indicators that power plants are concerned about, volatile matter is related to safety, and too high a volatile matter content can lead to excessively rapid combustion, which may even cause boiler explosions. Calorific value reflects combustion efficiency and is related to economic benefits. The solution of this invention can significantly improve the detection accuracy of indicators such as volatile matter, calorific value and ash content, providing an important reference for whether to adjust combustion parameters, and therefore has significant commercial value.

[0099] It should be noted that the specific process for detecting particle indicators based on spectral data in step 106 is not particularly improved in this embodiment of the invention. Please refer to the implementation methods in related technologies, and it will not be repeated here.

[0100] To facilitate those skilled in the art to reproduce the embodiments of the present invention, the specific process of detecting particle indicators based on spectral data will be briefly described below by way of example.

[0101] Optionally, based on spectral data, particle indicators are detected, including: Step S1: Preprocess the spectral data; Alternatively, common preprocessing methods include: Normalization. For example, normalization of molecular characteristic spectral line intensities. To eliminate the influence of laser pulse energy fluctuations and particle density variations on spectral intensity, the total area of ​​molecular bands composed of molecular characteristic spectral lines or specific internal standard lines (such as the integrated intensity of the C2 band) can be normalized to make the spectral intensities more quantitatively comparable.

[0102] Baseline correction. Since the spectral data is obtained from the accumulation of multiple plasma optical signals, it is a statistical average of multiple excitations. Although it already possesses high stability, residual background disturbances caused by factors such as particle size distribution and airflow fluctuations still need to be considered. Adaptive iterative weighted penalized least squares AirPLS or wavelet transform methods can be used for baseline correction to eliminate continuous background interference.

[0103] Spectral shift correction. Standard normal variable transformation (SNV) or multivariate scattering correction (MSC) can be used to eliminate the spectral shift caused by particle scattering effects and perform spectral shift correction.

[0104] Step S2: Based on the preprocessed spectral data, perform feature engineering processing, that is, screen out the set of feature wavelength points that are most correlated with the particle index from the full spectrum data. Preferably, when screening the characteristic wavelength points most relevant to particle indicators, the wavelength points corresponding to molecular characteristic spectral lines are selected; for example, when screening the characteristic wavelength points most relevant to the volatile matter of coal, the wavelength points corresponding to molecular characteristic spectral lines are generally selected.

[0105] Preferably, in order to reduce data dimensionality, avoid overfitting, and improve detection accuracy, a competitive adaptive reweighted sampling CARS method can be used to screen the set of feature wavelength points that are most correlated with particle indicators from the full spectrum data.

[0106] Step S3: Input the intensity data corresponding to the set of characteristic wavelength points into the detection model to obtain the detection value of the particle index.

[0107] The detection model is constructed based on the quantitative relationship between the intensity data corresponding to the set of characteristic wavelength points and the particle indicators.

[0108] Preferably, in order to reduce data dimensionality, avoid overfitting, and improve detection accuracy, partial least squares (PLS) can be used to establish the detection model.

[0109] Figure 5 This is a schematic diagram of the particle detection system in an embodiment of this application.

[0110] like Figure 5 As shown, a preferred embodiment of this application provides a particle detection system, which includes the following components.

[0111] Laser 501 is used to apply multiple laser pulses to a particle stream to excite plasma through thermal accumulation; wherein, the laser pulses are low-energy pulses with single-pulse energy less than a preset energy threshold, and their pulse interval is less than the thermal relaxation time of the particles. Spectrometer 502 is used to obtain spectral data based on the light signals emitted by the plasma; Detection component 503 is used to detect at least one indicator of particles based on spectral data.

[0112] During implementation, the inventors, through long-term experiments and research, discovered that the reason for the relatively large detection error when using LIBS technology to detect sample indicators in existing technologies is that when using single-pulse ablation to excite plasma, the high energy of the single pulse can cause air breakdown and strong shock waves, resulting in a large detection error. Based on this discovery, the inventors, after long-term experiments and research and overcoming multiple difficulties, proposed a new solution: using high-frequency, low-energy laser pulses for plasma excitation and employing particle flow as the sample excitation mode to solve the technical problem of difficulty in exciting plasma and easy induction of sample combustion during the excitation process. Through the newly explored thermal accumulation excitation mode, a new type of more compact and stable plasma is obtained. Because the new plasma has advantages such as good position and morphological stability, the light signal emitted by the plasma has good stability and consistency; moreover, because the single pulse energy is low, it will not cause air breakdown, and the shock wave effect is also very weak. Therefore, the detection error of the embodiments of the present invention is relatively small.

[0113] Preferably, the laser 501 is used to excite multiple plasmas within a preset integration time; wherein, the preset integration time is the time for acquiring optical signals; Spectrometer 502 is used to obtain spectral data based on the accumulation of light signals emitted by multiple plasmas.

[0114] Preferably, the spectrometer 502 is used to obtain spectral data including molecular characteristic spectral lines, atomic characteristic spectral lines and ionic characteristic spectral lines; The detection component 503 is used to detect different indicators of particles based on molecular characteristic spectral lines, atomic characteristic spectral lines and / or ionic characteristic spectral lines.

[0115] Optionally, the percentage P of the integrated area of ​​the molecular characteristic spectral lines relative to the total integrated area of ​​the net emission spectrum ranges from 2% to 20%, where ; Where A is the integral area of ​​the molecular characteristic spectral line, A total It represents the total integral area of ​​the net emission spectrum.

[0116] Preferably, the laser 501 is used to apply multiple laser pulses to a dilute phase particle flow, wherein the volume concentration of particles in the dilute phase particle flow ranges from 0.01% to 5%, and the particle size is less than 0.2 mm.

[0117] Preferably, the laser 501 excites plasma with a spatial length ranging from 0.5 to 1 millimeter.

[0118] Preferably, the preset energy threshold is 1 millijoules, the repetition frequency of the laser pulse is greater than 20 kHz and the pulse width is less than 30 nanoseconds.

[0119] Optionally, the laser 501 can be any type of laser whose single-pulse energy and pulse interval parameters meet the requirements of the embodiments of the present invention. Preferably, the laser 501 is a fiber laser.

[0120] It should be noted that the particle detection system can execute the particle detection method provided in any embodiment of the present invention and bring corresponding beneficial effects. Therefore, technical details not described in detail in the system embodiment of the present invention can be found in the method embodiment of the present invention, and will not be repeated here.

[0121] Optionally, the detection component 503 may be implemented entirely or partially by software, hardware, or a combination thereof.

[0122] When implemented in hardware, it can be implemented entirely or partially as a processor to implement all or part of the processes or functions of the embodiments of this application; it can also be implemented entirely or partially as an electronic device including a processor to implement all or part of the processes or functions of the embodiments of this application. The processor can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processors include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), microprocessors (MPUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), various artificial intelligence processors, various processors running machine learning model algorithms, and any suitable processor, controller, microcontroller, etc. The electronic device can be various computing devices capable of detecting particle indicators based on spectral data. Some examples of electronic devices include, but are not limited to, computing devices with different computing power, such as terminals, industrial control computers, servers, cloud computing nodes, and other suitable computing devices. When the electronic device is configured as a control host computer in the system architecture, it is also used to issue control commands to the laser 501 and the spectrometer 502, such as control commands for starting / stopping and operating parameters.

[0123] The following example uses a fiber laser (501), a coal powder particle flow, a multi-channel spectrometer (502), and a control host computer (503) as an example. Figures 6 to 8 An exemplary complete solution for online detection of coal quality entering the furnace is provided.

[0124] Figure 6 The structure of the online coal quality detection system in the embodiment of this application is shown, such as... Figure 6 As shown, the online coal quality detection system for furnace feed in this embodiment of the application includes a sampling device and a coal quality detection device.

[0125] like Figure 6As shown, the sampling device includes a proportional valve 6, a compressed air source 7, a first nozzle 81, a second nozzle 82, and a connecting pipe 9. One end of the proportional valve 6 is connected to the compressed air source 7 via the connecting pipe 9 to regulate the flow rate of the compressed air output from the compressed air source 7. The other end of the proportional valve 6 is connected to the first ends of the first nozzle 81 and the second nozzle 82 via the connecting pipe 9, respectively, to control the magnitude of the negative pressure generated in the first nozzle 81 and the second nozzle 82 by adjusting the compressed air flow rate. The second and third ends of the first nozzle 81 are connected to the air-coal duct and the top of the measuring chamber 2 of the coal quality testing device via the connecting pipe 9, respectively, to introduce the coal powder particle flow from the air-coal duct into the measuring chamber 2 under negative pressure for coal quality testing. The mass flow rate of the introduced coal powder particle flow varies depending on the magnitude of the negative pressure. The second and third ends of the second nozzle 82 are connected to the air-coal duct and the bottom of the measuring chamber 2 via the connecting pipe 9, respectively, to guide the coal powder particle flow from the measuring chamber 2 back into the air-coal duct under negative pressure after coal quality testing. Figure 6 In the diagram, solid arrows indicate the flow direction of pulverized coal particles, while dashed arrows indicate the flow direction of compressed air.

[0126] Figure 7 The internal structure of measuring chamber 2 is shown exemplarily, such as... Figure 7As shown, the measuring chamber 2 includes a housing 21, a pulverized coal pipe assembly 22 arranged axially within the housing 21 (a gas flow zone is formed between the pulverized coal pipe assembly 22 and the inner wall of the housing 21), a gas interface 23 located at the top of the housing 21 and communicating with the gas flow zone, and an exhaust port 24 located at the bottom of the housing 21. The pulverized coal pipe assembly 22 includes an interface pipe 221, an upper constricted pipe 222, a lower constricted pipe 223, and a pressure regulating pipe 224 arranged coaxially from top to bottom. The bottom end of the interface pipe 221 is connected to the top end of the upper constricted pipe 222, and the bottom end of the lower constricted pipe 223 is connected to the top end of the pressure regulating pipe 224. The bottom end of the upper constricted pipe 222 and the top end of the lower constricted pipe 223 are constricted structures, and there is a gap between the bottom end of the upper constricted pipe 222 and the top end of the lower constricted pipe 223 to form a focusing area. The bottom end of the pressure regulating pipe 224 is located inside the exhaust port 24. A two-layer flow structure is formed within the shell 21. The inner flow structure is the internal space of the pulverized coal pipe assembly 22, referred to as the inner layer. The outer flow structure is the gas flow zone between the pulverized coal pipe assembly 22 and the shell 21, referred to as the outer layer. The inner and outer layers are connected at the focusing area. Pulverized coal particles are input at the interface pipe 221. Under negative pressure, the pulverized coal particles flow from top to bottom within the pulverized coal pipe assembly 22. Compressed gas is input at the gas interface 23. Under negative pressure, the compressed gas flows from top to bottom along the gas flow zone. Due to the downward contraction... The constriction structure at the top of the nozzle 223 and the design of the pressure regulating pipe 224 (both the bottom ends of the inner and outer layers are connected to the second nozzle 82) allow for Bernoulli's principle, which states that the pressure in the inner layer near the inlet surface of the constriction in the lower constriction pipe 223 is lower than the pressure in the outer layer. The constriction position of the lower constriction pipe 223 not only draws in the coal powder particle flow output from the constriction inlet of the upper constriction pipe 222 but also draws in some of the outer layer gas. Therefore, a gas flow from the outer layer to the inner layer is formed in the focusing region. The inflowing gas from the outer layer constrains the flow shape of the inner coal powder particle flow. Because the flow shape of the coal powder particle flow can be constrained, the plasma position can be stabilized within the coal powder particle flow flow region, thereby improving spectral stability. Furthermore, it can prevent the coal powder particle flow from scattering and increasing the number of excited plasma particles, thus improving spectral intensity.

[0127] like Figure 6As shown, the coal quality testing device includes a measuring chamber 2, a fiber laser 501, a multi-channel spectrometer 502, a detection component 503 (i.e., a control host computer), and an optical fiber 4. The detection component 503 is connected to both the fiber laser 501 and the multi-channel spectrometer 502, controlling their start / stop and operating parameters (e.g., laser pulse repetition frequency, single pulse energy, pulse width, preset integration time, and number of spectra). To avoid the adverse effects of gravity on the flow shape of the coal powder particle stream, the measuring chamber 2 is generally vertically positioned, while the horizontally positioned fiber laser 501 is typically located on one side of the measuring chamber 2, with its output end aligned with the measuring chamber 2. This allows it to apply multiple laser pulses to the coal powder particle stream according to the configured parameters, stimulating plasma through thermal accumulation (the stimulated plasma is like...). Figure 7 (As shown); the multi-channel spectrometer 502 is connected to the measurement chamber 2 via optical fiber 4. It is used to collect light signals emitted by multiple plasmas within a preset integration time, and to acquire a spectrum based on the accumulation of multiple plasma light signals. The process of acquiring the spectrum is repeated until the number of acquired spectra reaches the configuration requirement. The detection component 503 is connected to the multi-channel spectrometer 502. It is used to receive the spectrum and detect coal quality indicators based on the molecular characteristic spectral lines, atomic characteristic spectral lines and ion characteristic spectral lines contained in the spectrum.

[0128] Preferably, such as Figure 7 and Figure 8 As shown, a first focusing lens 31 can be installed on the side of the measuring chamber 2 facing the fiber laser 501 to focus the laser pulses emitted by the fiber laser 501, thereby increasing the laser energy density. Figure 8 As shown, a collimating lens 32 and a second focusing lens 33 can also be arranged circumferentially along the side of the shell 21 of the measuring chamber 2; wherein, the collimating lens 32 is used to collimate the scattered light signal emitted by the plasma; the second focusing lens 33 is used to converge the collimated scattered light signal; by setting the collimating lens 32 and the second focusing lens 33, the light signal emitted by the plasma can be efficiently converged to the entrance of the optical fiber 4.

[0129] In practice, based on obtaining a sufficient amount of relatively stable plasma, a spectrum with high signal-to-noise ratio and good stability (RSD value around 2%) is obtained by accumulating multiple plasma optical signals. Furthermore, molecular characteristic spectral information is added as a reference when detecting indicators, which can significantly improve the accuracy of detection. In addition, based on the high-frequency characteristics of laser pulses, a stable spectrum is obtained by accumulating multiple plasma optical signals, eliminating the need for multiple spectral acquisitions or averaging of multiple spectra, resulting in a very short detection cycle and enabling second-level online detection. Therefore, the system has high detection efficiency and strong practicality. Since sampling is taken from the air-coal duct connected to the burner, it can reliably reflect the true state of the coal entering the furnace in real time, providing effective data support for combustion optimization. This can meet the needs of rapid load change and deep peak shaving of the new generation of coal-fired power units, and support the flexible, low-carbon and intelligent operation of coal-fired power units.

[0130] Moreover, fiber lasers are small in size and energy-saving. Their horizontal setting is conducive to long-term stable operation and is not easily damaged. In addition, the system integration is relatively high, which makes the system smaller in size, lower in testing cost and more stable. It is also very adaptable to complex environments and can meet the needs of industrial sites for long-term stable operation and flexible deployment.

[0131] Preferably, the fiber laser 501 can be an MOPA (Master Oscillator Power-Amplifier) ​​type pulsed fiber laser. This laser has a pulse repetition frequency of not less than 1 kHz and a pulse width of not more than 350 nanoseconds, featuring high-frequency output and adjustable pulse width. Preferably, the laser pulse emitted by the fiber laser 501 has a repetition frequency of 100 kHz, a single pulse energy of 0.2 millijoules, and a pulse width of 9 nanoseconds, resulting in an average power of 20W. This allows for the quasi-continuous excitation of a large amount of compact and morphologically stable plasma within a very short time, yielding a spectrum with high signal-to-noise ratio and good stability. Preferably, other parameters of the laser pulse can also be set according to specific application requirements, such as setting the wavelength to 1064 nanometers.

[0132] Preferably, the output end of the fiber laser 501 and the first focusing lens 31 are both aligned with the focusing area, so that the emitted laser pulse passes perpendicularly through the optical center of the first focusing lens 31 and is focused on the center of the coal powder stream in the focusing area, thereby exciting plasma in a better position more quickly and obtaining a stronger and more stable spectrum faster. Preferably, a plano-convex lens can be used as the first focusing lens 31, and preferably, its focal length can be set to 30.1 mm.

[0133] Preferably, to excite a sufficient amount of plasma with minimal interference to obtain a sufficiently strong and reliable spectrum, the intake power of the first nozzle 81 can be set to control the mass flow rate of the pulverized coal particle stream. For example, the intake power of the first nozzle 81 can be set to 25 AW (air watts) to control the mass flow rate of the pulverized coal particle stream to 2 g / min. Preferably, to better maintain the pulverized coal stream morphology, the inner diameter of the bottom end of the upper constriction tube 222 is smaller than the inner diameter of the top end of the lower constriction tube 223. For example, the ratio of their inner diameters is 4:5, and the inner diameter of the bottom end of the upper constriction tube 222 is 4 mm, which can form a cylindrical pulverized coal particle stream with a diameter of approximately 4 mm. Preferably, the measuring chamber 2 can adopt a regular octagonal prism structure or other centrally symmetric structures to better excite the plasma and collect the optical signals emitted by the plasma.

[0134] Preferably, when the measuring chamber 2 is a regular octagonal prism structure, multiple sets of optical fibers can be installed; for example, four sets. Figure 8 As shown, four sets of collimating lenses 32 and a second focusing lens 33 are used. Each set of collimating lenses 32 and the second focusing lens 33 are aligned with the focal area and set on the same optical axis. The optical center of the first focusing lens 31 is in the same horizontal plane as the optical centers of the collimating lenses 32 and the second focusing lens 33. The light-receiving probe of each optical fiber 4 is aligned with the optical center of each set of collimating lenses 32 and the second focusing lens 33. Taking the line connecting the optical center of the first focusing lens 31 and the plasma as the starting line, the first focusing lens 31 is set at a position with an angle of 0°. Preferably, the four sets of collimating lenses 32 and the second focusing lens 33 can be set at positions with angles of 45°, 135°, 225° and 315° respectively, so as to achieve comprehensive acquisition of the light signal emitted by the plasma and obtain a more stable and stronger spectrum. Preferably, both the collimating lens 32 and the second focusing lens 33 are plano-convex lenses coated with an anti-reflection film of 245 to 400 nanometers. Preferably, the focal lengths can be set to 30.0 mm and 20 mm, respectively, which can significantly improve the light flux in the 245 to 400 nanometer band while ensuring efficient collection of plasma light signals.

[0135] Optionally, the connection method between the first focusing lens 31, the collimating lens 32, the second focusing lens 33 and the optical fiber 4 and the measuring chamber 2 can be set as needed. For example, the first focusing lens 31, the collimating lens 32 and the second focusing lens 33 can be embedded on the side of the measuring chamber 2, and the optical fiber 4 can be connected to the measuring chamber 2 through a connector (such as a bracket).

[0136] Optionally, the preset integration time and the number of spectra can be set as needed, for example, the preset integration time can be set to 1.5 seconds and the number of spectra can be set to 4.

[0137] Alternatively, the implementation of the connecting pipe 9 can refer to related technologies, for example, it can be achieved by using a combination of stainless steel pipe and wear-resistant flexible hose.

[0138] To fully demonstrate the embodiments of the present invention, the following is an exemplary description of the method and process for coal quality testing using the above-described online coal quality testing system for furnace feed.

[0139] Optionally, the online coal quality detection method for furnace feed according to embodiments of the present invention includes: M1. Start the compressed air source 7, open the proportional valve 6, and under negative pressure, the first nozzle 81 draws coal powder particles from the air-coal pipeline through the connecting pipe 9 and introduces them into the measuring chamber 2. M2 and the detection component 503 control the activation of the fiber laser 501 and the multi-channel spectrometer 502, configuring the fiber laser 501 with a wavelength of 1064 nm, a repetition frequency of 100 kHz, a single pulse energy of 0.2 mJ and a pulse width of 9 nanoseconds, and configuring the multi-channel spectrometer 502 with a preset integration time of 1.5 seconds and a number of spectral images of 4. M3 and multiple laser pulses emitted from the fiber laser 501 pass vertically through the optical center of the first focusing lens 31 and are focused on the coal powder particles at the center of the coal powder stream, thereby stimulating plasma through thermal accumulation. M4, the multi-channel spectrometer 502 starts to collect optical signals. The optical signals emitted by the plasma are collimated by the collimating lens 32 and converged by the second focusing lens 33 before entering the probe of the fiber optic 4. The fiber optic 4 transmits the plasma optical signals to the multi-channel spectrometer 502. M5. After the acquisition time reaches 1.5 seconds, the multi-channel spectrometer 502 stops acquiring optical signals and obtains a spectrum based on the accumulation of multiple acquired plasma optical signals. M6. Repeat steps M4 and M5 until four spectra are obtained. Then, the multi-channel spectrometer 502 transmits the four spectra to the detection component 503. After screening the validity of the four spectra, M7 and detection component 503 perform normalization, averaging, baseline correction, and spectral shift correction on the valid spectra to obtain spectral data representing the coal powder sample. Based on the processed spectral data, the set of characteristic wavelength points with the strongest correlation to coal quality indicators is selected from the full spectrum data. The intensity data corresponding to the set of characteristic wavelength points is input into the detection model to obtain the detection values ​​of coal quality indicators (such as industrial analysis, elemental analysis, process properties, etc.). The detection model is a quantitative analysis model built by combining multiple linear regression, machine learning, and deep learning (such as partial least squares regression, support vector machine, random forest, decision tree, etc.).

[0140] Optionally, in step M7, if the detection component 503 fails to filter out a valid spectrum, it will adjust the preset integration time until a valid spectrum is obtained.

[0141] In the description of this invention, the term "multiple" refers to two or more. Unless otherwise explicitly defined, the terms "upper," "lower," etc., indicate the orientation or positional relationship based on the accompanying drawings, and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. The terms "connection," "installation," "fixing," etc., should be interpreted broadly. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a direct connection or an indirect connection through an intermediate medium. For those skilled in the art, the specific meaning of the above terms in this invention can be understood according to the specific circumstances.

[0142] It should be understood that the specific embodiments described above are merely illustrative or explanatory of the principles of this application and do not constitute a limitation thereof. Therefore, any modifications, equivalent substitutions, improvements, etc., made without departing from the spirit and scope of this application should be included within the protection scope of this application. Furthermore, the appended claims are intended to cover all variations and modifications falling within the scope and boundaries of the appended claims, or equivalent forms of such scope and boundaries.

Claims

1. A particle detection method, characterized in that, include: Multiple laser pulses are applied to a particle stream to excite plasma through thermal accumulation. Spectral data is obtained based on the light signals emitted by the plasma; Based on the spectral data, at least one indicator of the particles is detected; The laser pulse is a low-energy pulse with a single pulse energy less than a preset energy threshold, and its pulse interval is less than the thermal relaxation time of the particle.

2. The method as described in claim 1, characterized in that, The excited plasma includes: Multiple plasmas are excited within a preset integration time; wherein, the preset integration time is the time for acquiring optical signals; The obtained spectral data includes: The spectral data is obtained by accumulating the light signals emitted by the multiple plasmas.

3. The method as described in claim 1 or 2, characterized in that, The obtained spectral data includes: Obtain spectral data containing molecular characteristic lines, atomic characteristic lines, and ionic characteristic lines; At least one indicator for detecting particles includes: Different indicators of the particles are detected based on the molecular characteristic spectral lines, atomic characteristic spectral lines, and / or ionic characteristic spectral lines.

4. The method as described in claim 3, characterized in that, The percentage (P) of the integral area of ​​molecular characteristic spectral lines relative to the total integral area of ​​the net emission spectrum ranges from 2% to 20%. ； Where A is the integral area of ​​the molecular characteristic spectral line, A total It represents the total integral area of ​​the net emission spectrum.

5. The method as described in claim 1 or 2, characterized in that, The process of applying multiple laser pulses to a particle stream includes: Multiple laser pulses are applied to a dilute phase particle flow, wherein the volume concentration of particles in the dilute phase particle flow ranges from 0.01% to 5%, and the particle size is less than 0.2 mm.

6. The method as described in claim 1 or 2, characterized in that, The spatial length of the excited plasma ranges from 0.5 to 1 millimeter.

7. The method as described in claim 1 or 2, characterized in that, The preset energy threshold is 1 millijoules, and the repetition frequency of the laser pulse is greater than 20 kHz and the pulse width is less than 30 nanoseconds.

8. A particle detection system, characterized in that, include: A laser is used to generate multiple laser pulses and apply them to a particle stream to excite plasma through thermal accumulation; wherein the laser pulses are low-energy pulses with single-pulse energy less than a preset energy threshold, and the pulse interval is less than the thermal relaxation time of the particles. A spectrometer is used to collect the light signals emitted by the plasma and obtain spectral data; A detection component for detecting at least one indicator of particles based on the spectral data.

9. The system as described in claim 8, characterized in that, The laser is used to excite multiple plasmas within a preset integration time; wherein, the preset integration time is the time for acquiring optical signals; The spectrometer is used to obtain the spectral data based on the accumulation of light signals emitted by the plurality of plasmas.

10. The system as described in claim 8 or 9, characterized in that, The spectrometer is used to obtain spectral data including molecular characteristic spectral lines, atomic characteristic spectral lines and ionic characteristic spectral lines; The detection component is used to detect different indicators of the particles based on the molecular characteristic spectral lines, atomic characteristic spectral lines and / or ionic characteristic spectral lines.

11. The system as claimed in claim 10, characterized in that, The percentage P of the integral area of ​​the molecular characteristic spectral lines relative to the total integral area of ​​the net emission spectrum ranges from 2% to 20%. ； Where A is the integral area of ​​the molecular characteristic spectral line, A total It represents the total integral area of ​​the net emission spectrum.

12. The system as described in claim 8 or 9, characterized in that, The laser is used to apply multiple laser pulses to a dilute phase particle flow, wherein the volume concentration of particles in the dilute phase particle flow ranges from 0.01% to 5%, and the particle size is less than 0.2 mm.

13. The system as described in claim 8 or 9, characterized in that, The spatial length of the plasma excited by the laser ranges from 0.5 to 1 millimeter.

14. The system as described in claim 8 or 9, characterized in that, The preset energy threshold of the laser is 1 millijoule, and the repetition frequency of the laser pulse is greater than 20 kHz and the pulse width is less than 30 nanoseconds.

Images