An Adaptive Dynamic Airflow Enhancement LIBS System and Method Based on Real-Time Spectral Feedback

Abstract

This invention belongs to the field of laser-induced breakdown spectroscopy (LIBS) technology and discloses an adaptive dynamic airflow-enhanced LIBS method based on real-time spectral feedback. The method includes: presetting airflow control parameters, including purge velocity and purge time, and simultaneously setting a characteristic spectral line action threshold as a feedback signal; emitting a laser to act on the sample surface while simultaneously acquiring the full-band spectrum; extracting the CN molecular band or N / O atomic spectral lines and calculating the integral value of the spectral intensity within that band, then calculating the deviation *e* between this value and the characteristic spectral line action threshold; adaptively adjusting the airflow control parameters based on the deviation *e*; and calculating the average value using multiple characteristic spectral line intensity values ​​under compliant conditions. The invention also discloses a corresponding system. Through this invention, the adaptive adjustment of airflow parameters can be executed more accurately, ensuring that LIBS detection is always performed in an optimized inert gas environment, thereby effectively improving the signal stability and quantitative analysis accuracy of LIBS detection.

Description

Technical Field

[0001] This invention belongs to the field of laser-induced breakdown spectroscopy related technology, and more specifically, relates to an adaptive dynamic airflow enhancement LIBS system and method based on real-time spectral feedback. Background Technology

[0002] Laser-induced breakdown spectroscopy (LIBS) is an advanced technique for analyzing the elemental composition of materials. Its basic principle involves using a high-energy pulsed laser to ablate the sample, generating transient plasma. By analyzing the characteristic spectra emitted during the plasma's cooling process, qualitative and quantitative analysis of the sample's elements can be achieved. LIBS technology has gained widespread application in various fields due to its advantages, including eliminating the need for complex sample preparation, simultaneously analyzing multiple elements, and enabling remote and in-situ detection.

[0003] In LIBS detection, to reduce interference from nitrogen, oxygen, and other gases in the air on the plasma, an inert gas (such as argon) is often purged to the sample analysis point to create a localized, clean detection environment. In traditional dynamic airflow LIBS systems, the airflow parameters (such as airflow velocity and purging sequence) are usually preset and kept fixed based on experience.

[0004] However, in actual testing, factors such as the physical morphology of the sample surface (e.g., roughness, porosity), the distribution of chemical composition, and even minor environmental fluctuations can significantly affect the formation and stability of the local gas environment. A fixed airflow program cannot adapt to these dynamic changes: insufficient airflow parameters lead to incomplete air removal, strong environmental interference signals, and high background noise; excessively high airflow parameters may over-disturb the plasma, even dispersing it, resulting in attenuation and instability of characteristic signals. This uncertainty caused by fixed airflow parameters severely restricts further improvements in the accuracy, repeatability, and reliability of LIBS technology analysis.

[0005] A search revealed several existing solutions aimed at improving the LIBS detection environment. For example, patent document CN111398253A discloses an atmosphere-adjustable LIBS device, but its atmosphere adjustment mainly relies on preset pressure and gas flow rate, lacking a mechanism for flexible feedback adjustment based on real-time plasma conditions. Patent document CN115839941A integrates multiple sensors to detect airflow parameters, but its purpose is to correct the measured dust concentration; the airflow itself is not the object of its adaptive control. None of these solutions achieve a closed-loop adaptive airflow control system with the direct goal of optimizing the plasma generation environment in real time.

[0006] Accordingly, further research and improvements are urgently needed in this field in order to better meet the detection needs of LIBS technology under various complex working conditions. Summary of the Invention

[0007] To address the aforementioned deficiencies or needs of existing technologies, the present invention aims to provide an adaptive dynamic airflow-enhanced LIBS system and method based on real-time spectral feedback. This system constructs a real-time closed loop of "spectral sensing-intelligent decision-making-airflow execution" around a selected wavelength band, automatically adapting to dynamic changes in different samples and detection environments. This enables fine-tuning of airflow parameters, including airflow velocity and purging sequence, ensuring that LIBS detection is always performed in an optimized inert gas environment, effectively improving signal stability and quantitative analysis accuracy.

[0008] To achieve the above objectives, according to one aspect of the present invention, an adaptive dynamic airflow enhancement LIBS method based on real-time spectral feedback is provided, characterized in that the method includes the following steps:

[0009] Step 1: System Initialization and Parameter Preset

[0010] Start the LIBS system and set the initial airflow control parameters, including the basic purge velocity V_base and the basic purge time T_base. At the same time, set the characteristic spectral line action threshold I_th as the feedback signal.

[0011] Step 2: Laser Excitation and Spectral Acquisition

[0012] The LIBS system is controlled to emit laser light, which acts on the sample analysis point and simultaneously excites plasma; the full-band spectrum of plasma emission is acquired synchronously.

[0013] Step 3: Feedback Signal Extraction and Analysis

[0014] From the collected spectral data, extract the CN molecular band or N / O atomic spectral line and calculate the integral value of the spectral intensity in this band; then use it as the characteristic spectral line intensity I_CN representing environmental interference, and calculate the deviation e between the characteristic spectral line intensity I_CN and the characteristic spectral line action threshold I_th;

[0015] Step 4: Control Decision and Airflow Parameter Adjustment

[0016] Based on the deviation e, the adjustment amount of the airflow control parameters is calculated to obtain the purging time T_purge and / or purging flow rate V_flow for the next measurement and the adjustment is performed accordingly; wherein, when the deviation e > 0, it is determined that the current environment does not meet the standard, and the intensity of the characteristic spectral line I_Element of the element to be measured obtained in this measurement is assigned a first mark; while when the deviation e ≤ 0, it is determined that the current environment meets the standard, and the intensity of the characteristic spectral line I_Element' of the element to be measured obtained in this measurement is assigned a second mark;

[0017] Step 5: Loop Execution and Sample Information Output

[0018] Repeat steps two to four to form a closed-loop control; after the required number of measurements is reached, only the characteristic spectral line I_Element' of the element to be measured with the second label is used to calculate its average value; in this way, the average value is used as the final measurement result of the characteristic spectral line of the element to be measured and output synchronously.

[0019] More preferably, in step one, the characteristic spectral line action threshold I_th is set by the following operation:

[0020] S11. During the LIBS system calibration phase, ensure that the sample analysis point is in a purified inert gas environment. Under this condition, emit a laser to continuously act on the sample analysis point more than 10 times and simultaneously collect the corresponding number of full-band spectra to obtain background spectral data.

[0021] S12. From each synchronously acquired full-band spectrum, extract the preset characteristic spectral lines that are consistent with the bands of the CN molecular band or N / O atomic spectral lines, calculate the integral value of the spectral intensity within the band range, thereby obtaining a set of background spectral intensity sequences; then, calculate the average value I_min_a and the standard deviation S_a of the background spectral intensity sequences, where the standard deviation S_a is used to characterize the background noise fluctuation level of the preset characteristic spectral lines under ideal pure environment;

[0022] S13. Calculate the action threshold I_th of the characteristic spectral line using the following formula:

[0023] I_th = I_min_a + k × S_a

[0024] Where k is a preset signal-to-noise ratio coefficient, and its value ranges from 3 to 5.

[0025] More preferably, in step three, the center line of the CN molecular spectral band is set to 388.3 nm, and the N / O atomic spectral lines are set to 746.8 nm and 777.2 nm.

[0026] More preferably, in step four, the control decision and airflow parameter adjustment process is implemented using a fuzzy PID algorithm, wherein:

[0027] First, the deviation e is used as the input quantity; then, by analyzing the fuzzy relationship between the deviation e and the deviation change rate ec, the proportional, integral, and derivative algorithm parameters are dynamically adjusted; finally, based on the adjustment amount of the algorithm parameters, the adjustment amount of the corresponding airflow control parameters is calculated, thereby obtaining the purging time T_purge and / or purging flow rate V_flow for the next measurement.

[0028] More preferably, in step four, the control decision and airflow parameter adjustment process is implemented by combining fuzzy PID algorithm, fixed step size algorithm and conventional PID algorithm, wherein:

[0029] When the control effect of the LIBS system during the initial calibration or after multiple consecutive measurement cycles does not meet the preset performance indicators, the algorithm evaluation process is initiated. In this process, the performance indicators of the candidate fuzzy PID algorithm, fixed step size algorithm and conventional PID algorithm are calculated one by one within the evaluation cycle, including the number of iterations required for system convergence and the predicted fluctuation variance. Then, the candidate algorithm with the relatively optimal performance indicator calculation results is selected as the current working algorithm, and its algorithm parameters are locked and put into use.

[0030] More preferably, in step five, as an alternative method for calculating the average value, the following operations are included:

[0031] The characteristic spectral line I_Element' of the element to be tested, which is assigned the second label, is multiplied by a high weight value, while the characteristic spectral line I_Element of the element to be tested, which is assigned the first label, is multiplied by a low weight value, and then their average values ​​are calculated together.

[0032] To achieve the above objectives, according to another aspect of the present invention, a corresponding adaptive dynamic airflow enhancement LIBS system is also provided, characterized in that the system comprises:

[0033] The LIBS main unit includes a pulsed laser, a sample stage, a spectrometer, and an optical system for laser focusing and signal collection. The laser emitted by the pulsed laser is deflected and focused by the optical system to form plasma on the sample surface on the sample stage, and the full-band spectrum emitted by the plasma is simultaneously collected by the spectrometer.

[0034] The feedback signal detection unit is used to extract the CN molecular band or N / O atomic spectral line from the spectral data collected by the spectrometer and calculate the integral value of the spectral intensity in the band, and then use it as the characteristic spectral line intensity I_CN representing environmental interference.

[0035] The dynamic airflow generating unit includes an inert gas source, a conical nozzle, and a first branch and a second branch connected between the two. The first branch is sequentially connected to a first pressure regulating valve, a flow meter, and a first high-speed solenoid valve. The second branch is sequentially connected to a second pressure regulating valve and a second high-speed solenoid valve. The two branches then merge at an airflow confluencer and flow into the conical nozzle.

[0036] A real-time control and processing unit is connected to the LIBS main unit, the feedback signal detection unit, and the dynamic airflow generation unit. It calculates the deviation e between the characteristic spectral line intensity I_CN and the preset characteristic spectral line action threshold I_th, and simultaneously calculates the adjustment amount of the airflow control parameters based on the deviation e. The adjustment of the purge flow rate is achieved through the coordinated opening and closing of the first and second high-speed solenoid valves, while the adjustment of the purge time is achieved independently through the control of the opening duration of the first high-speed solenoid valve.

[0037] More preferably, the first high-speed solenoid valve is used to provide basic airflow and is a normally closed solenoid valve with millisecond-level response, made of aluminum alloy or stainless steel; the second high-speed solenoid valve is used to provide pulse-enhanced airflow and is a normally closed solenoid valve with millisecond-level response, made of aluminum alloy or stainless steel.

[0038] More preferably, the conical nozzle is a converging conical nozzle with a gradually narrowing internal channel, which can accelerate and rectify the airflow, thereby concentrating and stably covering the laser application point of the sample with the airflow ejected from the conical nozzle, and forming a local protective environment.

[0039] More preferably, the system further includes a display unit for outputting multiple sequences of the airflow control parameters, as well as the final measurement results of the characteristic spectral lines of the element to be measured and their converted elemental concentrations.

[0040] In summary, the technical solutions conceived by this invention have the following main technical advantages compared with the prior art:

[0041] (1) This invention uses a specific band of appropriate wavelength to sense the state of the plasma environment in real time, and on this basis, adjusts the airflow control parameters, including the purge flow rate and purge time, in an intelligent and dynamic manner. Compared with the prior art, it can more accurately perform adaptive adjustment of airflow parameters, get rid of the dependence on preset parameters and operator experience, and ensure that LIBS detection is always in an optimized inert gas environment, thereby effectively improving signal stability and quantitative analysis accuracy.

[0042] (2) By binding the plasma environment state with the quality of detection data, this invention can automatically screen high-reliability detection data for final calculation, and can automatically adapt to changes in different samples and detection environments, thereby improving the accuracy and reliability of elemental quantitative analysis results from the source. In addition, this invention has made further optimization design for the setting process of characteristic spectral line action threshold. A lot of actual measurements show that it can more accurately guide the calculation of deviation and help improve the accuracy of gas flow parameter adjustment.

[0043] (3) The present invention further makes targeted designs for decision-making algorithms and algorithm combinations, algorithm evaluation and other aspects in the process of airflow parameter adjustment. Among them, by adopting the fuzzy PID algorithm, it can combine the rapid combined response of multiple high-speed solenoid valves in the relevant hardware system to realize millisecond-level intelligent adjustment of airflow parameters. In addition, by adopting multiple algorithm combinations and introducing algorithm evaluation, it can ensure that even under complex working conditions, it can automatically find and lock the best strategy, showing better robustness and improving adaptability.

[0044] (4) The present invention further makes targeted designs on the related hardware system from the overall structure composition, especially the working mechanism and specific setting of key components such as dynamic airflow generation unit and feedback signal detection unit. The obtained overall system does not need to add complex optical splitters and special detectors. It can realize the corresponding spectral sensing and adaptive adjustment functions with high precision and high efficiency in a compact and easy-to-operate manner, and has high application flexibility.

[0045] (5) The entire adaptive dynamic airflow enhancement LIBS system of the present invention can automatically tend to use the minimum necessary gas consumption based on the airflow parameter adjustment results, while ensuring that the environment meets the standards, thereby reducing the consumable cost of long-term operation. Attached Figure Description

[0046] Figure 1 This is an overall process flow diagram of the adaptive dynamic airflow enhancement LIBS method according to the present invention;

[0047] Figure 2 This is a logic flowchart used to exemplify and explain the adaptive dynamic airflow enhancement and sample information processing of the present invention;

[0048] Figure 3 This is a schematic diagram of the overall structure of the adaptive dynamic airflow enhancement LIBS system designed according to a preferred embodiment of the present invention;

[0049] Figure 4 This is a flowchart illustrating the algorithm evaluation module designed according to a preferred embodiment of the present invention.

[0050] In all the accompanying drawings, the same reference numerals are used to denote the same elements or structures, wherein:

[0051] 1-Pulsed laser; 2-Reflector; 3-Focusing lens; 4-Sample; 5-Plasma; 6-Acquisition probe; 7-Fiber optic cable; 8-Spectrometer; 9-Real-time control and processing unit; 10-Inert gas source; 11-First pressure regulating valve; 12-Flow meter; 13-First high-speed solenoid valve; 14-Second pressure regulating valve; 15-Second high-speed solenoid valve; 16-Airflow confluencer; 17-Conical nozzle; 18-Airflow; 19-Display unit. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] Figure 1 This is an overall process flow diagram of the adaptive dynamic airflow enhanced LIBS method according to the present invention. Figure 2 This is a logic flowchart illustrating the adaptive dynamic airflow enhancement and sample information processing of the present invention. See below. Figure 1 and Figure 2 To provide a more specific explanation.

[0054] like Figure 1 and Figure 2 As shown, the adaptive dynamic airflow enhancement LIBS method of the present invention mainly includes the following steps:

[0055] Step 1: System Initialization and Parameter Preset

[0056] Start the LIBS system and set the initial airflow control parameters, including the basic purge velocity V_base and the basic purge time T_base. At the same time, set the characteristic spectral line action threshold I_th as the feedback signal. The characteristic spectral line action threshold I_th can be set based on multiple experimental measurements and combined with expert experience.

[0057] More specifically, according to a preferred embodiment of the invention, the characteristic spectral line action threshold I_th is preferably set by the following operation:

[0058] First, during the LIBS system calibration phase, ensure that the sample analysis point is in a purified inert gas environment. Under this condition, emit a laser to continuously act on the sample analysis point more than 10 times and simultaneously collect the corresponding number of full-band spectra to obtain background spectral data.

[0059] Secondly, from each synchronously acquired full-band spectrum, a preset characteristic spectral line that is consistent with the band of the CN molecular band or the N / O atomic spectral line is extracted, and the integral value of the spectral intensity within the band range is calculated to obtain a set of background spectral intensity sequences. Then, the average value I_min_a and the standard deviation S_a of the background spectral intensity sequence are calculated, where the standard deviation S_a is used to characterize the background noise fluctuation level of the preset characteristic spectral line in an ideal pure environment.

[0060] Finally, the action threshold I_th of the characteristic spectral line is calculated using the following formula:

[0061] I_th = I_min_a + k × S_a

[0062] Wherein, k is a preset signal-to-noise ratio coefficient, and its value ranges from 3 to 5; when k=3, it means that the system determines that the environmental interference is significant and airflow parameters need to be adjusted when the intensity of the characteristic spectral line I_CN exceeds three times the standard deviation of the background average.

[0063] Step 2: Laser Excitation and Spectral Acquisition

[0064] The LIBS system is controlled to emit laser light, which acts on the sample analysis point and simultaneously excites plasma; the full-band spectrum of plasma emission is acquired synchronously.

[0065] Step 3: Feedback Signal Extraction and Analysis

[0066] From the collected spectral data, extract the CN molecular band or N / O atomic spectral line and calculate the integral value of the spectral intensity in this band. Then, use it as the characteristic spectral line intensity I_CN representing environmental interference, and calculate the deviation e between the characteristic spectral line intensity I_CN and the characteristic spectral line action threshold I_th.

[0067] More specifically, in this step, the center line of the CN molecular band is preferably set to 388.3 nm, and the N / O atomic spectral lines are preferably set to 746.8 nm and 777.2 nm. The selection is based on the fact that the CN molecular band is generated by the reaction of nitrogen in the air with carbon in the sample, and is extremely sensitive to trace amounts of air contamination. Furthermore, the CN molecular band has high signal intensity and is easily identifiable, with an excellent signal-to-noise ratio; therefore, it is used in this invention as an indicator to characterize the state of an inert gas environment. For the N / O atomic spectral lines, the selection is based on the fact that nitrogen (N) and oxygen (O) are the main components of ambient air, and their atomic spectral line signal intensity is directly related to the amount of air contamination. More specifically, the selected Ni 746.8 nm spectral line is a strong characteristic line with a high transition probability and signal-to-noise ratio, and can sensitively reflect changes in the partial pressure of nitrogen in the environment; while the selected O I 777.2 nm spectral line intensity is extremely sensitive to oxygen content. Both spectral lines are within the detection range (200-900 nm) of the spectrometer and have little overlap or interference with the characteristic spectral lines of common analytes (such as Fe).

[0068] Step 4: Control Decision and Airflow Parameter Adjustment

[0069] Based on the deviation e, the adjustment amount of the airflow control parameters is calculated to obtain the purging time T_purge and / or purging flow rate V_flow for the next measurement and the adjustment is performed accordingly; wherein, when the deviation e > 0, it is determined that the current environment is not up to standard, and the intensity of the characteristic spectral line I_Element of the element to be measured obtained in this measurement is assigned a first label (e.g., a low confidence label); while when the deviation e ≤ 0, it is determined that the current environment is up to standard, and the intensity of the characteristic spectral line I_Element' of the element to be measured obtained in this measurement is assigned a second label (e.g., a high confidence label).

[0070] More specifically, in this step, according to a preferred embodiment of the present invention, the control decision and airflow parameter adjustment process is implemented using a fuzzy PID algorithm, wherein: firstly, the deviation e is used as an input quantity; then, by analyzing the fuzzy relationship between the deviation e and the deviation change rate ec, the proportional, integral, and derivative algorithm parameters are dynamically adjusted; finally, based on the adjustment amount of the algorithm parameters, the corresponding adjustment amount of the airflow control parameters is calculated, thereby obtaining the purging time T_purge and / or purging flow rate V_flow for the next measurement.

[0071] It should be noted that the basic principles and model composition of the fuzzy PID algorithm are well known to those skilled in the art, and therefore will not be elaborated upon here. This invention, by employing the fuzzy PID algorithm, can effectively utilize the deviation e obtained in the preceding steps, and, combined with the rapid combined response of multiple high-speed solenoid valves in the relevant hardware system, achieve millisecond-level intelligent adjustment of airflow parameters.

[0072] More specifically, in this step, see Figure 4 According to another preferred embodiment of the present invention, the control decision and airflow parameter adjustment process can also be implemented by combining fuzzy PID algorithm, fixed step size algorithm and conventional PID algorithm, wherein: when the control effect of multiple consecutive measurement cycles does not meet the preset performance index, the algorithm evaluation process is initiated; in this process, firstly, for the candidate fuzzy PID algorithm, fixed step size algorithm and conventional PID algorithm, their performance index in the evaluation cycle is calculated one by one, including the number of iterations required for system convergence and the predicted fluctuation variance; then, the candidate algorithm with the relatively optimal performance index calculation result is selected as the current working algorithm, and its algorithm parameters are locked for use.

[0073] It should be noted that the basic principles and model composition of fixed-step algorithms and conventional PID algorithms are well known to those skilled in the art, and therefore will not be elaborated upon here. This invention, by employing a combination of multiple algorithms and introducing algorithm evaluation, ensures that the optimal strategy can be automatically found and locked even under complex operating conditions, exhibiting superior robustness and improved adaptability. Simultaneously, it can continuously monitor relevant performance indicators throughout the entire process; if the performance indicators drop below a threshold, the algorithm evaluation process can be retried.

[0074] Step 5: Loop Execution and Sample Information Output

[0075] Repeat steps two to four above to form a closed-loop control; after the required number of measurements is reached, only the characteristic spectral line I_Element' of the element to be measured, which has been given a second label (that is, a high confidence label), is used to calculate its average value; in this way, the average value is used as the final measurement result of the characteristic spectral line of the element to be measured and output synchronously.

[0076] More specifically, according to another preferred embodiment of the present invention, as an alternative method for calculating the average value, the following operation may be included: multiplying the spectral line I_Element' of the element to be tested, which is assigned a second label (high confidence label), by a high weight value, and simultaneously multiplying the spectral line I_Element of the element to be tested, which is assigned a first label (low confidence label), by a low weight value, and then calculating their average value together.

[0077] Figure 3 This is a schematic diagram of the overall structural composition of the adaptive dynamic airflow enhancement LIBS system designed according to a preferred embodiment of the present invention. See below for further details. Figure 3 To explain the relevant hardware system.

[0078] The adaptive dynamic airflow enhancement LIBS system of the present invention mainly includes functional modules such as LIBS main unit, feedback signal detection unit, dynamic airflow generation unit and real-time control and processing unit. In addition, it may also include other functional modules such as display unit, which will be explained in detail below.

[0079] like Figure 3 As shown, for the LIBS main unit, the LIBS main unit includes a pulsed laser 1, a sample stage, a spectrometer 8, and an optical system for laser focusing and signal collection; wherein the laser emitted by the pulsed laser 1 is deflected and focused by the optical system to form plasma 5 on the surface of sample 4 on the sample stage, and the full-band spectrum emitted by the plasma 5 is synchronously collected by the spectrometer 8.

[0080] More specifically, the main specifications of the pulsed laser 1 include, for example, Nd:YAG, 1064nm, 10Hz, and a maximum energy of 150mJ. The optical system may include a reflector 2 and a focusing lens 3. The laser emitted by the pulsed laser is deflected by the reflector 2 and then focused by the focusing lens 3 onto the surface of the sample 4 to form plasma 5. The acquisition range of the spectrometer 8 may be 200nm to 900nm.

[0081] The feedback signal detection unit may include a acquisition probe 6 and be equipped with an optical fiber 7 connected to the spectrometer 8. The spectrometer 8 is equipped with an enhanced charge-coupled device camera (ICCD). The feedback signal detection unit can extract the CN molecular band or N / O atomic spectral line from the spectral data acquired by the spectrometer 8 and calculate the integral value of the spectral intensity in that band, and then use it as the characteristic spectral line intensity I_CN representing environmental interference.

[0082] The dynamic airflow generating unit includes an inert gas source 10, a conical nozzle 17, and a first branch and a second branch connected between the two. The first branch is connected in sequence to a first pressure regulating valve 11, a flow meter 12, and a first high-speed solenoid valve 13. The second branch is connected in sequence to a second pressure regulating valve 14 and a second high-speed solenoid valve 15. The two branches then merge at the airflow confluencer 16 and lead to the conical nozzle 17.

[0083] More specifically, the inert gas source 10, such as high-purity argon, will split into two independent branches at the outlet. The first branch is used to sequentially connect to the first pressure regulating valve 11 (preferably made of stainless steel, suitable for various inert gases), the flow meter 12 (digital flow meter, range 0-50 L / min), and the first high-speed solenoid valve 13 (responsible for the basic airflow, normally closed, millisecond-level response, preferably made of aluminum alloy or stainless steel). The second branch connects to the second pressure regulating valve 14 and then directly to the second high-speed solenoid valve 15 (responsible for the pulse-enhanced airflow, normally closed, millisecond-level response, preferably made of aluminum alloy or stainless steel). These two airflows merge at the airflow confluencer 16 and are finally directed towards the laser-induced focal region through the conical nozzle 17.

[0084] According to another preferred embodiment of the present invention, the internal channel of the conical nozzle 17 is a tapered section, which can accelerate and rectify the airflow, so that the airflow 18 flowing out of the nozzle outlet is concentrated and stably covered on the laser analysis point, forming an effective local protective environment.

[0085] Furthermore, according to another preferred embodiment of the present invention, the purging flow rate of the inert gas is preferably adjusted by controlling the coordinated opening and closing state of the first high-speed solenoid valve 13 and the second high-speed solenoid valve 15 to provide at least two discrete and stable flow rate levels; while the control of the purging time is achieved independently by the opening duration of the first high-speed solenoid valve 13.

[0086] For the real-time control and processing unit, the real-time control and processing unit 9 maintains signal connection with the LIBS main unit, the feedback signal detection unit and the dynamic airflow generation unit respectively, and calculates the deviation e between the characteristic spectral line intensity I_CN and the preset characteristic spectral line action threshold I_th, and calculates the adjustment amount of the airflow control parameters based on the deviation e.

[0087] As for the display unit, it is used to output multiple sequences of the airflow control parameters, as well as the final measurement results of the characteristic spectral lines of the element to be measured and the converted element concentration, etc.

[0088] The following is a specific example according to the invention, in order to illustrate the invention more intuitively and clearly.

[0089] Specific example 1

[0090] (I) System Initialization

[0091] Start the system and set the initial parameters: V_base = 8 L / min, T_base = 4s, I_th (set via pre-calibration). At this point, the algorithm evaluation process can begin. After evaluation, the fuzzy PID algorithm will be selected as the working algorithm.

[0092] (ii) First measurement

[0093] The real-time control and processing unit 9 opens the first high-speed solenoid valve 13 according to the initial parameters. After purging for 4 seconds, the laser 1 is triggered to excite, and the spectrometer 8 collects the full spectrum.

[0094] The real-time control and processing unit 9 extracts the 388.3 nm band from the full spectrum data and integrates it to obtain I_CN(1). Calculations show that I_CN(1) is much larger than I_th, and the deviation e(1) is positive. At this time, the fuzzy PID controller calculates the control command based on e(1) and ec(1) (ec(1) is 0 at this time) through fuzzy inference and defuzzification: extend the purge time T_purge before the next measurement to 7 seconds, and simultaneously use the first high-speed solenoid valve 13 and the second high-speed solenoid valve 15 (i.e., enhanced airflow mode) to increase the total flow rate V_flow to 12 L / min. At the same time, the Fe element characteristic line intensity I_Fe(1) obtained in this measurement (also extracted from the full spectrum) is marked as "low confidence".

[0095] (III) Second Measurement

[0096] After the real-time control and processing unit 9 purges according to the new parameters (T_purge=7s, and start enhanced airflow), it is excited again to collect the full spectrum. I_CN(2) is extracted and it is found that it has decreased significantly, but there is still I_CN(2) greater than I_th, the deviation e(2) is positive and small, and ec(2) is negative and large (rapid decreasing trend). The fuzzy PID controller calculates a milder adjustment command based on the new input: slightly increase T_purge to 7.5 seconds, close the second high-speed solenoid valve 15, and restore the flow rate to 8 L / min. I_Fe(2) is still marked as "low confidence".

[0097] (iv) Third and subsequent measurements

[0098] After several adjustments, the system measured I_CN(k) ≤ I_th, and entered the "Environmental Compliance" branch. The control objective shifted to optimization and maintenance. The algorithm attempted to slightly shorten T_purge, for example, by 0.5 seconds each time, and observed whether I_CN still met the requirements. Under this branch, all measured I_Fe were marked as "high confidence" and included in the averaging calculation.

[0099] Ultimately, the output of the real-time control and processing unit 9 includes: a stable sequence of airflow control parameters, and the average intensity calculated based on all “high confidence” I_Fe values ​​and the converted Fe element concentration.

[0100] In summary, this invention uses appropriate specific wavebands to sense the plasma environment state in real time, and then adjusts airflow control parameters, including purge velocity and purge time, in an intelligent and dynamic manner. Compared with existing technologies, this invention can more accurately perform adaptive adjustment of airflow parameters, ensuring that LIBS detection is always in an optimized inert gas environment. At the same time, by binding the plasma environment state with the quality of detection data, this invention can automatically select high-reliability detection data for final calculation, thereby improving the accuracy and reliability of elemental quantitative analysis results from the source. This effectively enhances the signal stability and quantitative analysis accuracy of LIBS detection, thus possessing broad application prospects.

[0101] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An adaptive dynamic airflow enhancement LIBS method based on real-time spectral feedback, characterized in that, The method includes the following steps: Step 1: System Initialization and Parameter Preset Start the LIBS system and set the initial airflow control parameters, including the basic purge velocity V_base and the basic purge time T_base. At the same time, set the characteristic spectral line action threshold I_th as the feedback signal. Step 2: Laser Excitation and Spectral Acquisition The LIBS system is controlled to emit laser light, which acts on the sample analysis point and simultaneously excites plasma; the full-band spectrum of plasma emission is acquired synchronously. Step 3: Feedback Signal Extraction and Analysis From the collected spectral data, extract the CN molecular band or N / O atomic spectral line and calculate the integral value of the spectral intensity in this band; then use it as the characteristic spectral line intensity I_CN representing environmental interference, and calculate the deviation e between the characteristic spectral line intensity I_CN and the characteristic spectral line action threshold I_th; Step 4: Control Decision and Airflow Parameter Adjustment Based on the deviation e, the adjustment amount of the airflow control parameters is calculated to obtain the purging time T_purge and / or purging flow rate V_flow for the next measurement and the adjustment is performed accordingly; wherein, when the deviation e > 0, it is determined that the current environment does not meet the standard, and the intensity of the characteristic spectral line I_Element of the element to be measured obtained in this measurement is assigned a first mark; while when the deviation e ≤ 0, it is determined that the current environment meets the standard, and the intensity of the characteristic spectral line I_Element' of the element to be measured obtained in this measurement is assigned a second mark; Step 5: Loop Execution and Sample Information Output Repeat steps two to four to form a closed-loop control; after the required number of measurements is reached, only the characteristic spectral line I_Element' of the element to be measured with the second label is used to calculate its average value; in this way, the average value is used as the final measurement result of the characteristic spectral line of the element to be measured and output synchronously.

2. The method as described in claim 1, characterized in that, In step one, the characteristic spectral line action threshold I_th is set through the following operation: S11. During the LIBS system calibration phase, ensure that the sample analysis point is in a purified inert gas environment. Under this condition, emit a laser to continuously act on the sample analysis point more than 10 times and simultaneously collect the corresponding number of full-band spectra to obtain background spectral data. S12. From each synchronously acquired full-band spectrum, extract the preset characteristic spectral lines that are consistent with the bands of the CN molecular band or N / O atomic spectral lines, calculate the integral value of the spectral intensity within the band range, thereby obtaining a set of background spectral intensity sequences; then, calculate the average value I_min_a and the standard deviation S_a of the background spectral intensity sequences, where the standard deviation S_a is used to characterize the background noise fluctuation level of the preset characteristic spectral lines under ideal pure environment; S13. Calculate the action threshold I_th of the characteristic spectral line using the following formula: I_th = I_min_a + k × S_a Where k is a preset signal-to-noise ratio coefficient, and its value ranges from 3 to 5.

3. The method as described in claim 2, characterized in that, In step three, the center line of the CN molecular spectral band is set to 388.3 nm, and the N / O atomic spectral lines are set to 746.8 nm and 777.2 nm.

4. The method according to any one of claims 1 to 3, characterized in that, In step four, the control decision-making and airflow parameter adjustment process is implemented using a fuzzy PID algorithm, wherein: First, the deviation e is used as the input quantity; then, by analyzing the fuzzy relationship between the deviation e and the deviation change rate ec, the proportional, integral, and derivative algorithm parameters are dynamically adjusted; finally, based on the adjustment amount of the algorithm parameters, the adjustment amount of the corresponding airflow control parameters is calculated, that is, the purging time T_purge and / or purging flow rate V_flow for the next measurement.

5. The method according to any one of claims 1 to 3, characterized in that, In step four, the control decision-making and airflow parameter adjustment process is implemented by combining fuzzy PID algorithm, fixed step size algorithm, and conventional PID algorithm, wherein: When the control effect of the LIBS system during the initial calibration or after multiple consecutive measurement cycles does not meet the preset performance indicators, the algorithm evaluation process is initiated. In this process, the performance indicators of the candidate fuzzy PID algorithm, fixed step size algorithm and conventional PID algorithm are calculated one by one within the evaluation cycle, including the number of iterations required for system convergence and the predicted fluctuation variance. Then, the candidate algorithm with the relatively optimal performance indicator calculation results is selected as the current working algorithm, and its algorithm parameters are locked and put into use.

6. The method as described in claim 1, characterized in that, In step five, as an alternative method for calculating the average value, the following operations are included: The characteristic spectral line I_Element' of the element to be tested, which is assigned the second label, is multiplied by a high weight value, while the characteristic spectral line I_Element of the element to be tested, which is assigned the first label, is multiplied by a low weight value, and then their average value is calculated together.

7. An adaptive dynamic airflow enhancement LIBS system based on real-time spectral feedback, characterized in that, The system is used to perform the method as described in any one of claims 1 to 6, and includes: The LIBS main unit includes a pulsed laser (1), a sample stage, a spectrometer (8), and an optical system for laser focusing and signal collection. The laser emitted by the pulsed laser (1) is deflected and focused by the optical system to form plasma (5) on the surface of the sample (4) on the sample stage, and the full-band spectrum emitted by the plasma (5) is synchronously collected by the spectrometer (8). Feedback signal detection unit, which is used to extract CN molecular bands or N / O atomic spectral lines from the spectral data collected by the spectrometer (8) and calculate the integral value of the spectral intensity in the band, and then use it as the characteristic spectral line intensity I_CN representing environmental interference; The dynamic airflow generating unit includes an inert gas source (10), a conical nozzle (17), and a first branch and a second branch connected between the two. The first branch is connected in sequence to a first pressure regulating valve (11), a flow meter (12), and a first high-speed solenoid valve (13). The second branch is connected in sequence to a second pressure regulating valve (14) and a second high-speed solenoid valve (15). The two branches then merge at the airflow confluencer (16) and lead to the conical nozzle (17). The real-time control and processing unit (9) is connected to the LIBS main unit, the feedback signal detection unit and the dynamic airflow generation unit respectively. It calculates the deviation e between the characteristic spectral line intensity I_CN and the preset characteristic spectral line action threshold I_th, and calculates the adjustment amount of the airflow control parameters based on the deviation e. The adjustment of the purging flow rate is achieved by the coordinated opening and closing state of the first high-speed solenoid valve (13) and the second high-speed solenoid valve (15), and the adjustment of the purging time is achieved independently by controlling the opening duration of the first high-speed solenoid valve (13).

8. The system as described in claim 7, characterized in that, The first high-speed solenoid valve (13) is used to provide basic airflow and is a normally closed solenoid valve with millisecond-level response, made of aluminum alloy or stainless steel; the second high-speed solenoid valve (15) is used to provide pulse-enhanced airflow and is a normally closed solenoid valve with millisecond-level response, made of aluminum alloy or stainless steel.

9. The system as described in claim 8, characterized in that, The conical nozzle (17) is a converging conical nozzle with a gradually narrowing internal channel, which can accelerate and rectify the airflow, thereby making the airflow (18) ejected from the conical nozzle (17) concentrated and stably cover the laser action point of the sample (4) and form a local protective environment.

10. The system according to any one of claims 7 to 9, characterized in that, The system also includes a display unit (19) for outputting multiple sequences of the airflow control parameters, as well as the final measurement results of the characteristic spectral lines of the element to be measured and the converted element concentrations.

Images