Gas chromatography-based method for on-line detection of volatile organic compounds in air

Abstract

The application discloses an online detection method for volatile organic compounds based on gas chromatography, and relates to the technical field of gas chromatography analysis.The method is characterized in that: the air to be detected is introduced into a pretreatment pipeline with a built-in spiral condenser for low-temperature enrichment, and then the concentrate is transferred to a resolution chamber through instantaneous heating.Subsequently, the carrier gas carries the resolved material into a gas chromatography separation column, and the components are separated in time sequence through programmed temperature rising.The separated components are detected by a flame ionization detector to generate an original spectrum containing retention time and peak area data.Finally, the original spectrum data are compared with a pre-stored volatile organic compound fingerprint feature library, the target species are matched according to the retention time, and the concentration is directly calculated by using the standard peak area response coefficient in the library to output the detection result.The method improves the enrichment and resolution efficiency through spiral condensation and instantaneous thermal resolution, realizes rapid qualitative and quantitative analysis by using the fingerprint feature library, and improves the automation level and analysis efficiency of online monitoring.

Description

Technical Field

[0001] This invention belongs to the field of gas chromatography analysis technology, specifically a gas chromatography-based online detection method for volatile organic compounds in air. Background Technology

[0002] Monitoring volatile organic compounds (VOCs) in the air is crucial for environmental assessment, health early warning, and pollution source analysis. Currently, online monitoring based on gas chromatography (GC) is the mainstream method, with the core aspects being rapid sample pretreatment and accurate component analysis. For sample pretreatment, conventional techniques often employ room-temperature adsorption using adsorption tubes or conventional cold traps. The former has limited efficiency in capturing low-boiling-point components, and the adsorbent may introduce background interference or undergo irreversible adsorption. While the latter can achieve low-temperature concentration, common straight-tube or simple coil-type cold traps have limited condensation surface area and gas path length, leaving room for improvement in enrichment efficiency for trace VOCs in complex air matrices. More importantly, when transferring the enriched target analytes to the chromatographic system, conventional programmed temperature ramping and resolution methods are slow, potentially leading to peak broadening and affecting subsequent rapid analysis with high resolution. In continuous online monitoring scenarios, resolution speed becomes one of the bottlenecks limiting analytical throughput.

[0003] In data analysis and quantification, existing methods typically rely on external or internal standard methods. Each detection requires instrument calibration using standard gases of known concentrations to establish retention time windows and peak area-concentration correction curves for each target analyte. This process necessitates frequent preparation and use of standard substances, making it cumbersome and increasing maintenance costs and complexity. For applications requiring the simultaneous monitoring of dozens or even hundreds of volatile organic compounds, establishing and maintaining a complete calibration system is extremely labor-intensive and struggles to quickly identify unknown peaks. The real-time requirements of online monitoring conflict with the offline, time-consuming calibration process, highlighting the urgent need for a data analysis strategy that can reduce reliance on frequent calibrations and rapidly achieve qualitative and quantitative analysis of multiple components. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems existing in the prior art.

[0005] Therefore, this invention proposes an online detection method for volatile organic compounds in air based on gas chromatography, comprising:

[0006] The air sample to be tested is introduced into the pretreatment pipeline, and the air sample is enriched at low temperature using the spiral condenser tube built into the pretreatment pipeline. After enrichment, the concentrate in the spiral condenser tube is transferred to the desorption chamber by instantaneous heating.

[0007] The concentrate in the analytical chamber is carried into a gas chromatography column along with a carrier gas. The temperature gradient control program of the gas chromatography column is used to achieve time-sequential separation of volatile organic compounds with different boiling points in the gas chromatography column, forming a group of separated component peaks.

[0008] The separated component peaks are scanned one by one using a flame ionization detector, and the retention time and peak area intensity of each component peak are collected to generate raw spectral data.

[0009] The original spectral data is input into a preset volatile organic compound fingerprint feature library for comparison. The volatile organic compound fingerprint feature library stores the standard retention times and corresponding standard peak area response coefficients of known volatile organic compounds.

[0010] Based on the comparison results, target volatile organic compounds (VOCs) whose retention time deviation from the original spectral data is within the allowable range are selected from the VOC fingerprint feature library. Then, using the standard peak area response coefficient corresponding to the target VOC, the concentration inversion calculation is performed on the peak area intensity in the original spectral data, and the final VOC concentration detection result is output.

[0011] Furthermore, the air sample to be tested is enriched at low temperature using the spiral condenser tube built into the pretreatment pipeline, including:

[0012] The semiconductor cooling chip located on the outside of the spiral condenser is activated to reduce the wall temperature of the spiral condenser to a preset low-temperature capture range.

[0013] Open the air inlet valve of the air sample to be tested, and control the air sample to be tested to flow through the internal cavity of the spiral condenser tube at a constant low flow rate, so that the volatile organic molecules in the air are condensed and adsorbed on the inner wall of the spiral condenser tube.

[0014] After the enrichment time reaches the preset duration, the air intake valve is closed and the delivery of the air sample to be tested is stopped;

[0015] The instantaneous heating program of the spiral condenser is initiated, and the pulse heating wire wound around the outer wall of the spiral condenser is used to heat it at the millisecond level, causing the volatile organic compounds adsorbed on the inner wall of the spiral condenser to vaporize and desorb instantly; wherein, the winding density of the pulse heating wire from the sample inlet to the sample outlet of the spiral condenser gradually decreases, so that the heating power density at the sample inlet is greater than that at the sample outlet.

[0016] Simultaneously, the inlet solenoid valve of the desorption chamber is switched to the on state, and the vaporized and desorbed volatile organic compounds are brought into the desorption chamber by the reverse purging of the carrier gas.

[0017] Furthermore, by utilizing the temperature gradient control program of the gas chromatography column, volatile organic compounds with different boiling points are sequentially separated within the gas chromatography column, including:

[0018] Before loading the concentrate in the analytical chamber into the gas chromatography column, the initial temperature of the gas chromatography column oven is set to the lowest value below the boiling point of all target volatile organic compounds.

[0019] After the concentrate enters the gas chromatography column with the carrier gas, the initial temperature is maintained for a fixed period of time to ensure that the low-boiling-point components completely enter the gas chromatography column.

[0020] Subsequently, the column oven of the gas chromatography separation column is linearly heated according to the preset heating rate. During the heating process, volatile organic compounds with different boiling points are pushed forward in the gas chromatography separation column in sequence due to the difference in vaporization rate.

[0021] The gas components flowing through the outlet of the gas chromatography separation column are monitored in real time. When a certain volatile organic compound component is detected to flow out, the column oven temperature and the effluent time point are recorded at this time. The effluent time point is confirmed as the retention time of the volatile organic compound component.

[0022] Continue heating until all volatile organic compounds in the gas chromatography column have eluted, thus ending the separation procedure.

[0023] Furthermore, the step of scanning the separated component peaks one by one using a flame ionization detector and acquiring the retention time and peak area intensity of each component peak includes:

[0024] Hydrogen and air are respectively introduced into the combustion head of the flame ionization detector to ignite and form a stable hydrogen flame;

[0025] The gas components flowing out of the gas chromatography column are directly introduced into the root of the hydrogen flame, causing the organic compounds to undergo an ionization reaction in the high-temperature flame.

[0026] A constant voltage is applied to the collecting electrode of the flame ionization detector to collect the charged ion flow generated by the ionization reaction and convert it into a current signal;

[0027] The current signal is continuously sampled using a high-speed analog-to-digital converter module to obtain a current intensity curve that changes over time.

[0028] The current intensity curve is subjected to digital filtering and noise reduction processing to identify and extract the peak region. The time span between the start point and the end point of each peak is recorded as the retention time of the corresponding component peak. The area integral value enclosed by the peak is determined as the peak area intensity of the corresponding component peak.

[0029] The original spectral data includes the acquired retention time and peak area intensity.

[0030] Further, the original spectral data is input into a pre-set volatile organic compound fingerprint feature database for comparison, including:

[0031] The retention times of all component peaks are extracted from the original spectral data and sorted according to the order of the retention times to generate a retention time sequence to be matched.

[0032] Traverse each standard record in the volatile organic compound fingerprint feature library, wherein the standard record contains the standard retention time and standard peak area response coefficient of a single volatile organic compound;

[0033] Calculate the absolute value of the difference between each retention time in the retention time series to be matched and the standard retention time in the standard record;

[0034] Determine whether the absolute value of the difference is less than a preset time tolerance threshold. If multiple consecutive absolute values ​​of the difference are less than the time tolerance threshold, then determine that the type of volatile organic compound corresponding to the standard record is a potential matching object.

[0035] All volatile organic compounds identified as potential match objects and their corresponding standard peak area response coefficients are packaged and output as input parameters for subsequent concentration inversion calculations.

[0036] Furthermore, using the standard peak area response coefficients corresponding to the target volatile organic compound species, concentration inversion calculations are performed on the peak area intensities in the original spectral data, including:

[0037] For each of the selected target volatile organic compounds, the corresponding standard peak area response coefficient is read from the volatile organic compound fingerprint feature library. The standard peak area response coefficient characterizes the peak area production capacity of a unit concentration of volatile organic compounds under specific detector conditions.

[0038] Find the component peak that matches the target volatile organic compound in the original spectral data, and read the peak area intensity of the matched component peak;

[0039] The measured concentration of the target volatile organic compound is calculated by dividing the peak area intensity by the standard peak area response coefficient.

[0040] The measured concentration values ​​of all target volatile organic compounds are summarized to generate a dataset of detection results containing the concentrations of all detected volatile organic compounds.

[0041] Furthermore, the steps for constructing the volatile organic compound fingerprint feature library include:

[0042] Several representative known volatile organic compound standard samples were selected, and the concentration range of each standard sample covered the actual concentration level of the expected detection environment.

[0043] Under the same gas chromatography detection conditions, each standard sample was enriched, separated, and detected by flame ionization to obtain the corresponding standard spectral data.

[0044] The retention time and peak area intensity of each volatile organic compound component were extracted from the standard spectral data, and the ratio of peak area intensity to concentration at the same concentration was used as the standard peak area response coefficient of the volatile organic compound component.

[0045] The standard retention time, standard peak area response coefficient, and corresponding chemical name or number of each volatile organic compound are associated and stored to form a standard record;

[0046] All standard records are categorized and indexed, sorted in ascending order by retention time, and appended with version number and generation date to generate the volatile organic compound fingerprint feature library.

[0047] Furthermore, the method also includes a step of periodically self-checking the performance of the gas chromatography separation column:

[0048] Before each formal testing process begins, the air intake path of the air sample to be tested is cut off, and instead, a calibration sample gas containing trace amounts of propane gas is introduced into the gas chromatography separation column.

[0049] Initiate a simplified gas chromatography separation procedure and record the retention time of the propane component in the calibration sample gas at the outlet of the gas chromatography separation column;

[0050] The retention time is compared with the standard propane retention time stored in the system, and the difference between the retention time and the standard propane retention time is calculated.

[0051] If the difference exceeds the preset performance drift threshold, a column aging and cleaning command is triggered, extending the subsequent detection cycle until the difference returns to the normal range.

[0052] Furthermore, the method also includes an adaptive adjustment step for the sampled flow rate based on meteorological parameters:

[0053] The ambient temperature and atmospheric pressure values ​​at the deployment location are collected in real time, and the ambient temperature and atmospheric pressure values ​​are substituted into the gas state equation conversion model to calculate the current air density correction factor.

[0054] Based on the air density correction factor, the rotation speed of the air intake sampling pump for the air sample to be tested is dynamically adjusted to maintain a constant air quality entering the system per unit time.

[0055] Simultaneously, the average concentration of volatile organic compounds obtained from the previous cycle is read. If the average concentration of volatile organic compounds is higher than the high concentration warning threshold, the intake sampling flow rate of the air sample to be tested is temporarily reduced, and the single enrichment time is extended to ensure detection accuracy.

[0056] After completing the high-concentration detection, the intake sampling flow rate of the air sample to be tested is restored to the normal set value.

[0057] Furthermore, the method also includes a step of drift calibration of the detection data:

[0058] Standard gases of known concentrations are periodically drawn into the detection system to perform a complete process from enrichment to detection, and a set of measured peak areas and intensities at standard concentrations are obtained.

[0059] Calculate the ratio of the measured peak area intensity to the theoretical peak area intensity to generate the current detector sensitivity drift coefficient;

[0060] Retrieve the most recently saved historical detector sensitivity drift coefficient and calculate the relative change between the two coefficients;

[0061] The relative change is compared with a preset drift alarm threshold. If it exceeds the drift alarm threshold, the currently generated detector sensitivity drift coefficient is applied to all subsequent concentration inversion calculations, replacing the original coefficient.

[0062] If the drift alarm threshold is not exceeded, the original coefficient remains unchanged, and only the calibration data for this calibration is recorded.

[0063] Furthermore, after comparing the retention time with the standard propane retention time stored in the system and calculating the difference between the retention time and the standard propane retention time, the method further includes a step of dynamically correcting the fingerprint feature database comparison process based on the difference:

[0064] If the difference does not exceed the preset performance drift threshold, the preset volatile organic compound fingerprint feature library is used directly for subsequent comparison.

[0065] If the difference exceeds the preset performance drift threshold but is within an acceptable correction range, a corrected comparison time series will be temporarily generated when the step of inputting the original spectral data into the preset volatile organic compound fingerprint feature library for comparison is subsequently performed.

[0066] Generating the corrected alignment time series includes:

[0067] The standard retention time of each standard record is read from the volatile organic compound fingerprint feature library, and the difference is added to each standard retention time to generate a corresponding temporary corrected retention time. All the temporary corrected retention times constitute the corrected comparison time series.

[0068] The absolute value of the difference between the time series to be matched extracted from the original spectral data and the time tolerance threshold are calculated and judged to complete the screening of target volatile organic compounds.

[0069] After the column aging and cleaning command is completed and subsequent self-test confirms that the difference has returned to the normal range, the use of the corrected comparison time series is stopped, and the original volatile organic compound fingerprint feature library is resumed for comparison.

[0070] Compared with the prior art, the beneficial effects of the present invention are:

[0071] Low-temperature enrichment is achieved using a spiral condenser. Its spiral structure significantly increases the surface area of ​​the condenser tube in contact with the gas, enabling more efficient heat exchange within a limited volume. As the sample air flows through, the condensation efficiency of volatile organic compounds on the tube wall is significantly improved, enhancing the capture capacity of trace components. After enrichment, the spiral condenser is rapidly heated using an instantaneous heating method. This thermal desorption method concentrates energy, allowing the tube wall temperature to exceed the boiling point of the target analyte in a very short time, prompting the rapid and concentrated release of the concentrate in a narrow band. Compared to traditional slow temperature-programmed desorption, this effectively reduces the diffusion of the target analyte during transport, resulting in a narrower sample band entering the chromatographic column. This creates conditions for obtaining sharp, highly resolved chromatographic peaks, improving chromatographic separation efficiency and detection sensitivity, and shortening the cycle time for a single analysis.

[0072] The system compares raw spectral data using a pre-built fingerprint feature library of volatile organic compounds (VOCs). This library integrates standard retention times and corresponding standard peak area response coefficients for various known organic compounds. During detection, the system quickly matches the retention times of each component in the measured spectrum with the feature library, filtering out candidates with retention time deviations within a preset range for automatic qualitative analysis. Based on this, the system directly calls the standard peak area response coefficients corresponding to the candidate compounds to perform concentration inversion calculations on the measured peak areas. This method eliminates the need to plot working curves on-site with standard gases before each analysis, simplifying complex quantitative calculations into rapid retrieval and computation based on feature library data. This significantly reduces the real-time dependence on standard substances and the burden of frequent calibration, enabling the system to stably and quickly output qualitative and quantitative results for various VOCs in unattended online monitoring mode, enhancing the automation level and long-term operational convenience of the method. Attached Figure Description

[0073] Figure 1 This is a flowchart illustrating the steps of the online detection method for volatile organic compounds in air based on gas chromatography according to the present invention.

[0074] Figure 2 This is a flowchart illustrating the retention time, peak area, and intensity of component peaks scanned and acquired using a flame ionization detector. Detailed Implementation

[0075] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0076] This invention provides an online detection method for volatile organic compounds (VOCs) in air based on gas chromatography. (See reference...) Figure 1The method involves introducing an air sample into a pretreatment pipeline and using a spiral condenser built into the pretreatment pipeline to perform low-temperature enrichment of the air sample. After enrichment, the concentrate in the spiral condenser is transferred to the desorption chamber by instantaneous heating. Subsequently, the concentrate in the desorption chamber is loaded into a gas chromatography column along with a carrier gas. Using a temperature gradient control program of the gas chromatography column, volatile organic compounds with different boiling points are sequentially separated within the column, forming a group of separated component peaks. Then, the separated component peaks are scanned one by one using a flame ionization detector, and the retention time and peak area intensity of each component peak are collected to generate raw spectral data. The raw spectral data is then input into a pre-set volatile organic compound fingerprint feature database for comparison. This database stores the standard retention times and corresponding standard peak area response coefficients of known volatile organic compounds. Finally, based on the comparison results, target volatile organic compounds (VOCs) whose retention time deviation from the original spectral data is within the allowable range are selected from the feature library. Then, using the standard peak area response coefficients corresponding to these target VOCs, the concentration inversion calculation is performed on the peak area intensity in the original spectral data, and the final VOC concentration detection results are output.

[0077] In one embodiment of the present invention, the operation of low-temperature enrichment of the air sample to be tested using a spiral condenser tube built into the pretreatment pipeline is as follows: The semiconductor cooling chip located on the outside of the spiral condenser tube is activated to lower the tube wall temperature to a preset low-temperature capture range. The inlet valve of the air sample to be tested is opened, controlling the air sample to flow through the internal cavity of the spiral condenser tube at a constant low flow rate, causing volatile organic compound molecules in the air to condense and adsorb on the inner wall of the spiral condenser tube. After the enrichment time reaches a preset duration, the inlet valve is closed and the delivery of the air sample to be tested is stopped. The instantaneous heating program of the spiral condenser tube is activated, using a pulse heating wire wound around the outer wall of the spiral condenser tube to heat it at millisecond levels, causing the volatile organic compounds adsorbed on the inner wall of the spiral condenser tube to vaporize and desorb instantaneously. Simultaneously, the inlet solenoid valve of the desorption chamber is switched to the conducting state, and the vaporized and desorbed volatile organic compounds are carried into the desorption chamber by the reverse purging of the carrier gas. It is worth mentioning that the winding density of the pulse heating wire gradually decreases from the inlet end to the outlet end of the spiral condenser, so that the heating power density at the inlet end is greater than that at the outlet end, thereby improving the detection peak shape.

[0078] In practice, the low-temperature enrichment of the air sample to be tested using the spiral condenser tube built into the pretreatment pipeline includes the following operations: activating the semiconductor cooling chip located on the outside of the spiral condenser tube to lower the tube wall temperature to a preset low-temperature capture range. The preset low-temperature capture range is set to a fixed value according to the properties of the target volatile organic compounds, such as -20℃ or -10℃. Opening the air inlet valve of the air sample to be tested and controlling the air sample to be tested to flow through the internal cavity of the spiral condenser tube at a constant low flow rate. The constant low flow rate is maintained at a specific value by the flow controller, such as 100mL / min or 200mL / min, so that the volatile organic compound molecules in the air are condensed and adsorbed on the inner wall of the spiral condenser tube. In some embodiments, after the enrichment time reaches a preset duration, the inlet valve is closed and the delivery of the air sample to be tested is stopped. The preset duration is set to a fixed value according to the detection requirements, such as 10 minutes or 20 minutes. The instantaneous heating program of the spiral condenser is then initiated, using a pulse heating wire wound around the outer wall of the spiral condenser to heat it at millisecond levels. The millisecond heating time is set to a fixed value, such as 50 milliseconds or 100 milliseconds, causing the volatile organic compounds adsorbed on the inner wall of the spiral condenser to vaporize and desorb instantaneously. Optionally, the inlet solenoid valve of the desorption chamber is simultaneously switched to the on state, and the vaporized and desorbed volatile organic compounds are carried into the desorption chamber by the back purge of the carrier gas. The back purge flow rate of the carrier gas is set to a fixed value, such as 30 mL / min or 50 mL / min. It is understood that the low-temperature enrichment process involves parameter coordination; the preset enrichment time is related to the internal cavity volume of the spiral condenser and the sampling flow rate. It can be done through the formula:

[0079]

[0080] in Indicates the preset duration of enrichment timing, in minutes. This indicates the volume of the internal cavity of the spiral condenser, in milliliters. This indicates a constant low flow rate of the air sample being tested, expressed in milliliters per minute.

[0081] In specific implementations, the cooling operation of the semiconductor refrigeration chip is performed at a linear rate, set to a fixed value, such as 5°C per second, until the wall temperature of the spiral condenser reaches a preset low-temperature capture range. The heating operation of the pulse heating wire is performed in a step manner, with a fixed heating rate, such as 1000°C per second, to achieve millisecond-level heating. In some embodiments, the preset low-temperature capture range is selected based on the condensation characteristics of volatile organic compounds. In the example scenario, for low-boiling-point components, the low-temperature capture range is set to -30°C, and for medium-boiling-point components, it is set to -20°C. The constant low flow rate is adjusted according to the sampling environment. In the example scenario, the constant low flow rate is set to 100 mL / min for indoor air detection and 200 mL / min for outdoor air detection. Optionally, the preset enrichment timing duration is related to the sampling concentration. In the example scenario, when the expected concentration is high, the preset enrichment timing duration is set to 5 minutes, and when the expected concentration is low, the preset enrichment timing duration is set to 30 minutes. It is understandable that the execution of the instantaneous heating program is synchronized with the switching of the desorption chamber. At the same time as the pulse heating wire is started, the air intake solenoid valve of the desorption chamber switches to the conducting state, and the reverse purging of the carrier gas starts immediately when the heating begins.

[0082] In one embodiment of the present invention, the process of sequentially separating volatile organic compounds (VOCs) with different boiling points within a gas chromatography column using a temperature gradient control program is as follows: Before loading the concentrate from the desorption chamber into the gas chromatography column, the initial temperature of the column oven is set to a value lower than the lowest value of the boiling points of all target VOCs. After the concentrate enters the gas chromatography column with the carrier gas, this initial temperature is maintained for a fixed period to ensure that the low-boiling-point components completely enter the gas chromatography column. Subsequently, the column oven is linearly heated according to a preset heating rate. During the heating process, VOCs with different boiling points are sequentially pushed forward within the gas chromatography column due to differences in their vaporization rates. The gas components flowing through the outlet of the gas chromatography column are monitored in real time. When a certain VOC component is detected to be eluting, the column oven temperature and elution time are recorded, and this elution time is identified as the retention time of that VOC component. Continue heating until all volatile organic compounds in the gas chromatography column have eluted, thus ending the separation procedure.

[0083] In practical implementation, the temperature gradient control program of the gas chromatography column is used to achieve time-sequential separation of volatile organic compounds with different boiling points. This includes the following operations: Before loading the concentrate in the desorption chamber into the gas chromatography column, the initial temperature of the gas chromatography column oven is set to the lowest value below the boiling point of all target volatile organic compounds. For example, if the target volatile organic compounds include acetone with a boiling point of 56°C and benzene with a boiling point of 80°C, the initial temperature of the column oven is set to 40°C. After the concentrate enters the gas chromatography column with the carrier gas, the initial temperature is maintained for a fixed period of time. The fixed period is set to 2 minutes according to the column length and the carrier gas flow rate to ensure that the low-boiling-point components completely enter the gas chromatography column. The column oven of the gas chromatography column was then linearly heated according to a preset heating rate of 10°C per minute. During the heating process, volatile organic compounds with different boiling points moved forward sequentially within the gas chromatography column due to their different vaporization rates. For example, acetone began to elute when the column oven temperature reached 60°C, and benzene began to elute when the column oven temperature reached 85°C. The gas components flowing through the outlet of the gas chromatography column were monitored in real time. When the eluent of a certain volatile organic compound was detected, the column oven temperature and the eluent time were recorded. The eluent time was identified as the retention time of the volatile organic compound. The heating continued until all volatile organic compounds in the gas chromatography column had eluted, at which point the separation program ended. The end temperature was set at 200°C.

[0084] In some embodiments, the initial temperature of the column oven is dynamically adjusted according to the lowest boiling point of the target volatile organic compound (VOC). For example, if the lowest boiling point of the target VOC is 30°C, the initial temperature of the column oven is set to 20°C. The fixing time is adjusted according to the inner diameter of the chromatographic column. For example, the fixing time is set to 1 minute for a column with an inner diameter of 0.32 mm, and 2 minutes for a column with an inner diameter of 0.53 mm. Optionally, the preset heating rate adopts a multi-stage program. In the example scenario, a heating rate of 15°C per minute is used from the initial temperature to 100°C, and a heating rate of 5°C per minute is used above 100°C to optimize the separation effect of high-boiling-point components. It can be understood that the relationship between column oven temperature and time during linear heating follows the formula:

[0085] in This indicates the real-time temperature of the column oven, in degrees Celsius. This indicates the initial temperature of the column oven, in degrees Celsius. This indicates the preset heating rate, expressed in degrees Celsius per minute. This indicates the time taken from the start of the temperature rise, in minutes.

[0086] In some embodiments, real-time monitoring is performed using the signal triggering of a flame ionization detector. The elution time is recorded as the time point when the signal intensity exceeds a set threshold, and the retention time is stored as the time interval from the start of injection to the elution time point. In the example scenario, the retention time of acetone is recorded as 3.5 minutes, and the retention time of benzene is recorded as 6.2 minutes. Optionally, the termination of the separation program is determined based on temperature and time. The separation program automatically terminates when the column oven temperature reaches the preset termination temperature and the detection signal returns to the baseline for more than 1 minute. It is understood that the temperature gradient control program is coordinated with the carrier gas flow rate, which is kept constant at 1 mL per minute to ensure a consistent migration rate of volatile organic compounds within the gas chromatography column.

[0087] In one embodiment of the present invention, see [reference] Figure 2 The process involves scanning the separated component peaks one by one using a flame ionization detector (FIDD) to acquire the retention time and peak area intensity of each component peak. This includes introducing hydrogen and air separately into the burner head of the FIDD to ignite a stable hydrogen flame. Gas components emanating from the gas chromatography column are directly introduced into the root of the hydrogen flame, causing the organic compounds to ionize in the high-temperature flame. A constant voltage is applied to the collector electrode of the FIDD to collect the charged ion current generated by the ionization reaction and convert it into a current signal. The current signal is continuously sampled using a high-speed analog-to-digital converter (ADC) to obtain a current intensity curve over time. The current intensity curve is digitally filtered to remove noise, and peak regions are identified and extracted. The time span from the start point to the end point of each peak is recorded as the retention time of the corresponding component peak, and the integral value of the area enclosed by the peak is determined as the peak area intensity of the corresponding component peak. The generated raw spectral data includes these acquired retention times and peak area intensities.

[0088] In practice, the separation of component peaks is scanned one by one using a flame ionization detector (FID), and the retention time and peak area intensity of each component peak are collected. The process includes the following steps: Hydrogen and air are introduced separately into the burner head of the FID, with the hydrogen flow rate set to 30 mL / min and the air flow rate set to 300 mL / min. The mixed gas is ignited to form a stable hydrogen flame. The gaseous components effluent from the gas chromatography column are directly introduced into the root of the hydrogen flame, causing the organic compounds to ionize in the high-temperature flame. A constant voltage of +200 volts is applied to the collecting electrode of the FID to collect the charged ion current generated by the ionization reaction. This charged ion current is converted into a current signal by a signal conversion circuit. The current signal is continuously sampled using a high-speed analog-to-digital converter at a sampling frequency of 100 points per second to obtain a current intensity curve over time. The current intensity curve is digitally filtered and denoised using a low-pass filter with a cutoff frequency of 10 Hz. The peak regions are identified and extracted. The time span between the start and end points of each peak is recorded as the retention time of the corresponding component peak. The area integral value enclosed by the peak is determined as the peak area intensity of the corresponding component peak. The generated original spectrum data includes all the acquired retention times and peak area intensities.

[0089] In some embodiments, the hydrogen to air flow rate ratio is adjusted according to the carrier gas type. When nitrogen is used as the carrier gas, the hydrogen flow rate is set to 30 mL / min and the air flow rate to 300 mL / min; when hydrogen is used as the carrier gas, the hydrogen flow rate is set to 40 mL / min and the air flow rate to 400 mL / min. Optionally, the constant voltage applied to the collecting electrode of the flame ionization detector is set according to the detection sensitivity requirements. In the example scenario, for trace analysis, the constant voltage is set to ±250 volts, and for constant analysis, the constant voltage is set to ±150 volts. It is understood that the signal-to-noise ratio of the current intensity curve is related to the filtering parameters. Evaluation can be performed using a formula:

[0090]

[0091] in The signal-to-noise ratio (SNR) of the current intensity curve is expressed in decibels (dB). This represents the power of the signal component in the current intensity curve. This represents the power of the noise component in the current intensity curve.

[0092] In some embodiments, peak region identification is based on a threshold method. A threshold condition is set where the absolute value of the rate of change of current intensity exceeds 0.1 microamps per second, and a threshold condition is set where the current intensity value falls back to the baseline level and remains there for 0.5 seconds, indicating the peak termination point. Optionally, peak area intensity is calculated using a numerical integration method. A trapezoidal integral is performed on the area enclosed by the current intensity curve and the time axis within the interval from the peak start point to the termination point. The integration result is recorded in microamps multiplied by seconds. It can be understood that the original spectral data is stored in a two-dimensional array. Each row of the array corresponds to an identified component peak, containing two data fields: retention time and peak area intensity. In the example scenario, the benzene component peak is stored in the array with a retention time of 6.2 minutes and a peak area intensity of 350 microamps multiplied by seconds.

[0093] In one embodiment of the present invention, the operation of inputting the original spectral data into a preset volatile organic compound (VOC) fingerprint feature library for comparison is as follows: The retention times of all component peaks are extracted from the original spectral data and sorted according to the order of retention times to generate a retention time sequence to be matched. Each standard record in the VOC fingerprint feature library is traversed; this standard record contains the standard retention time and standard peak area response coefficient of a single VOC. The absolute value of the difference between each retention time in the retention time sequence to be matched and the standard retention time in the standard record is calculated. It is determined whether the absolute value of the difference is less than a preset time tolerance threshold. If multiple consecutive absolute values ​​of the difference are less than the time tolerance threshold, the VOC species corresponding to the standard record is determined to be a potential matching object. All VOC species determined to be potential matching objects and their corresponding standard peak area response coefficients are packaged and output as input parameters for subsequent concentration inversion calculations.

[0094] The process of calculating the concentration inversion of peak area intensity in the original spectral data using the standard peak area response coefficient corresponding to the target volatile organic compound (VOC) species involves: for each selected target VOC species, retrieving its corresponding standard peak area response coefficient from the VOC fingerprint feature database. This standard peak area response coefficient characterizes the peak area production capacity of a unit concentration of VOC under specific detector conditions. Finding the component peak matching the target VOC species in the original spectral data and reading its peak area intensity. Dividing this peak area intensity by its corresponding standard peak area response coefficient to calculate the measured concentration value of the target VOC species. Summarizing the measured concentration values ​​of all target VOC species generates a detection result dataset containing the concentrations of all detected VOCs.

[0095] The construction steps of the volatile organic compound (VOC) fingerprint feature library include: selecting several representative known VOC standard samples, with the concentration range of each standard sample covering the actual concentration level of the expected detection environment; performing enrichment, separation, and flame ionization detection on each standard sample under the same gas chromatography detection conditions to obtain corresponding standard spectral data; extracting the retention time and peak area intensity of each VOC component from the standard spectral data, and using the ratio of peak area intensity to concentration at the same concentration as the standard peak area response coefficient for that VOC component; associating and storing the standard retention time, standard peak area response coefficient, and corresponding chemical name or number for each VOC to form a standard record; classifying and indexing all standard records, sorting them in ascending order by retention time, and adding version numbers and generation dates to generate the final VOC fingerprint feature library.

[0096] In practice, the process of inputting the raw spectral data into a pre-set volatile organic compound (VOC) fingerprint feature library for comparison includes the following operations: Extracting the retention times of all component peaks from the raw spectral data and sorting them according to their order of retention time to generate a retention time sequence to be matched. For example, the retention time sequence extracted from a single detection might be [3.5, 6.2, 8.7, 12.1] minutes. Traversing each standard record in the VOC fingerprint feature library, each standard record contains the standard retention time and standard peak area response coefficient for a single VOC. Calculating the absolute value of the difference between each retention time in the retention time sequence to be matched and the standard retention time in the standard record. Determining whether the absolute value of the difference is less than a preset time tolerance threshold (set to 0.1 minutes). If multiple consecutive absolute values ​​of the difference are less than the time tolerance threshold, the VOC species corresponding to the standard record is identified as a potential match. Packaging and outputting all VOC species identified as potential matches and their corresponding standard peak area response coefficients as input parameters for subsequent concentration inversion calculations. In some embodiments, the preset time tolerance threshold is dynamically fine-tuned based on the column status. When the column is new, the time tolerance threshold is set to 0.08 minutes; when the column has been used more than 1000 times, the time tolerance threshold is set to 0.15 minutes. Optionally, the standard records in the volatile organic compound fingerprint feature library are stored in the form of a data table. Some examples of standard records are shown in Table 1 below:

[0097] Table 1: Partial Standard Record Table of Volatile Organic Compound Fingerprint Feature Database

[0098]

[0099] It is understandable that the comparison process is a sequential computational logic. For each value in the retention time series to be matched, it is necessary to calculate the difference and make a threshold judgment with the standard retention time of each standard record in the volatile organic compound fingerprint feature library.

[0100] In specific implementation, the concentration inversion calculation of the peak area intensity in the original spectral data using the standard peak area response coefficient corresponding to the target volatile organic compound (VOC) includes the following steps: For each selected target VOC, its corresponding standard peak area response coefficient is read from the VOC fingerprint feature database. The standard peak area response coefficient characterizes the peak area production capacity of a unit concentration of VOC under specific detector conditions. The component peak matching the target VOC is found in the original spectral data, and the peak area intensity of the matched component peak is read. The peak area intensity is divided by the standard peak area response coefficient to calculate the measured concentration value of the target VOC. The measured concentration values ​​of all target VOCs are summarized to generate a detection result dataset containing the concentrations of all detected VOCs. In some embodiments, the measured concentration values... The calculation follows the formula:

[0101]

[0102] in This indicates the measured concentration of the target volatile organic compound, expressed in milligrams per cubic meter. This represents the peak area intensity of the component peak that matches the target volatile organic compound, read from the raw spectral data, in microamperes multiplied by seconds. This represents the standard peak area response coefficient corresponding to the target volatile organic compound (VOC), retrieved from the VOC fingerprint feature database, expressed in microamps per second per milligram per cubic meter (µA / s). Optionally, in the example scenario, the peak area intensity of the component matching benzene in the original spectral data is 350 µA / s. The standard peak area response coefficient for benzene, obtained from the VOC fingerprint feature database, is 189.7 µA / s per milligram per cubic meter (µA / s), resulting in a calculated measured concentration of 1.84 mg / m³. It can be understood that the detection result dataset is presented in a structured list format, where each item contains the chemical name of the VOC and its corresponding measured concentration value.

[0103] In practice, the construction steps of the volatile organic compound (VOC) fingerprint feature library include the following process: First, select several representative known VOC standard samples. The concentration range of each standard sample should cover the actual concentration levels of the expected detection environment. For example, for indoor air detection, the concentration gradient of the benzene standard sample is set to 0.05 mg / m³, 0.5 mg / m³, and 5 mg / m³. Under the same gas chromatography detection conditions, each standard sample is enriched, separated, and subjected to flame ionization detection to obtain the corresponding standard spectral data. The retention time and peak area intensity of each VOC component are extracted from the standard spectral data, and the ratio of peak area intensity to concentration at the same concentration is used as the standard peak area response coefficient of the VOC component. The standard retention time, standard peak area response coefficient, and corresponding chemical name or number of each VOC are associated and stored to form a standard record. All standard records are classified and indexed, arranged in ascending order of retention time, and appended with a version number and generation date to generate the final VOC fingerprint feature library. In some embodiments, the standard retention time is the average of multiple repeated measurements. For example, if a benzene standard sample is measured ten times, the average retention time is 6.2 minutes. Optionally, the standard peak area response coefficient is obtained by linear regression of the measurement results of multiple concentration gradients. Taking benzene as an example, the peak area intensity corresponding to three concentration points of 0.05, 0.5, and 5 mg / m³ is measured, and a linear relationship between peak area intensity and concentration is fitted. The slope of this relationship is the standard peak area response coefficient of benzene. It can be understood that the volatile organic compound fingerprint feature library exists as an independent database file and is updated through version number management. When a new standard substance is added, a new feature library version is generated and the generation date is updated.

[0104] In one embodiment of the present invention, the method includes a step of periodically self-checking the performance of the gas chromatography column. Before each formal testing process begins, the inlet passage of the air sample to be tested is cut off, and a calibration sample gas containing a trace amount of propane is introduced into the gas chromatography column instead. A simplified gas chromatography separation program is started, and the retention time of the propane component in the calibration sample gas at the outlet of the gas chromatography column is recorded. This retention time is compared with the standard retention time of propane stored in the system, and the difference between the two is calculated. If the difference exceeds a preset performance drift threshold, a column aging and cleaning command is triggered, extending the subsequent testing cycle until the difference returns to the normal range.

[0105] This method includes an adaptive adjustment step for sampling flow rate based on meteorological parameters. It involves real-time acquisition of ambient temperature and atmospheric pressure values ​​at the deployment site, substituting these values ​​into a gas state equation conversion model to calculate the current air density correction factor. Based on this correction factor, the rotational speed of the intake sampling pump for the air sample is dynamically adjusted to maintain a constant air quality entering the system per unit time. Simultaneously, the average concentration of volatile organic compounds (VOCs) from the previous period is read. If this average VOC concentration exceeds a high-concentration warning threshold, the intake sampling flow rate of the air sample is temporarily reduced, extending the single enrichment time. After completing the high-concentration detection, the intake sampling flow rate of the air sample is restored to the normal set value.

[0106] This method includes a drift calibration step for detection data. Standard gas of known concentrations is periodically introduced into the detection system, and the entire process from enrichment to detection is executed to obtain a set of measured peak area intensities at standard concentrations. The ratio of the measured peak area intensities to the theoretical peak area intensities is calculated to generate the current detector sensitivity drift coefficient. The most recently saved historical detector sensitivity drift coefficient is retrieved, and the relative change between the two coefficients is calculated. This relative change is compared with a preset drift alarm threshold. If the drift alarm threshold is exceeded, the currently generated detector sensitivity drift coefficient is applied to all subsequent concentration inversion calculations, replacing the original coefficients. If the drift alarm threshold is not exceeded, the original coefficients are maintained, and only the calibration data for this period is recorded.

[0107] In its implementation, this method includes a step of periodically self-checking the performance of the gas chromatography column. Before each formal testing process begins, the inlet path of the air sample to be tested is cut off, and a calibration sample gas containing a trace amount of propane is introduced into the gas chromatography column instead. The concentration of propane in the calibration sample gas is set to 0.1% by volume. A simplified gas chromatography separation program is started. The simplified gas chromatography separation program uses fixed isothermal separation conditions, with the column oven temperature set to 80°C. The retention time of the propane component in the calibration sample gas at the outlet of the gas chromatography column is recorded. The recorded retention time is compared with the propane standard retention time stored in the system. The propane standard retention time is pre-determined to be 2.1 minutes. The difference between the recorded retention time and the propane standard retention time is calculated. If the difference exceeds the preset performance drift threshold (set to 0.1 minutes), a column aging and cleaning command is triggered, extending the subsequent testing cycle until the difference returns to the normal range. In some embodiments, the propane standard retention time is obtained by averaging 10 repeated measurements on a new column. In the example scenario, the average propane retention time measured on the new column is 2.1 minutes, with a standard deviation of 0.02 minutes. Optionally, the simplified gas chromatography separation procedure does not perform temperature gradient ramping, but only maintains isothermal conditions to shorten the self-test time. The aging and cleaning instruction includes a procedure to heat the gas chromatography column to 250°C and maintain it for 30 minutes. It is understood that the performance drift threshold setting is related to the separation accuracy requirements. For routine detection, the performance drift threshold can be set to 0.05 minutes, and for high-precision detection, the performance drift threshold can be set to 0.02 minutes.

[0108] In its implementation, the method includes an adaptive adjustment step for sampling flow rate based on meteorological parameters. It involves real-time acquisition of ambient temperature and atmospheric pressure values ​​at the deployment location. The ambient temperature is measured using a temperature sensor integrated into the device housing, and the atmospheric pressure is measured using a built-in pressure sensor. These values ​​are then substituted into the gas state equation conversion model to calculate the current air density correction factor. Based on this correction factor, the rotational speed of the intake sampling pump for the air sample is dynamically adjusted to maintain a constant air mass entering the system per unit time. Simultaneously, the average concentration of volatile organic compounds (VOCs) from the previous detection cycle is read. If the average VOC concentration exceeds a high-concentration warning threshold (set at 10 mg / m³), the intake sampling flow rate of the air sample is temporarily reduced to extend the single enrichment time and ensure detection accuracy. After completing the high-concentration detection, the intake sampling flow rate is restored to the normal set value. In some embodiments, the gas state equation conversion model is derived using the ideal gas state equation, and the air density correction factor... The calculation formula is:

[0109]

[0110] in Indicates the air density correction factor. This indicates that the standard atmospheric pressure is set at 101.3 kPa. This represents the atmospheric pressure value collected in real time. This represents the real-time ambient temperature value, in Kelvin. This indicates the standard temperature setting is 293 Kelvin. Optionally, the intake sampling pump speed is adjusted proportionally based on an air density correction factor. In the example scenario, the sampling flow rate is normally set to 500 ml / min. When the ambient temperature rises from 20°C to 30°C and the pressure drops from 101.3 kPa to 100.0 kPa, the calculated air density correction factor is approximately 0.94. Therefore, the intake sampling pump speed is adjusted to 1.06 times the original speed to maintain a constant mass flow rate. It can be understood that temporarily reducing the intake sampling flow rate is a step-wise operation. When the average concentration of volatile organic compounds exceeds the high-concentration warning threshold, the flow rate is reduced from 500 ml / min to 250 ml / min, and the single enrichment time is correspondingly doubled.

[0111] In its implementation, this method includes a drift calibration step for the detection data. A standard gas of known concentration is periodically introduced into the detection system. The standard gas is a single-component benzene gas with a concentration of 5 mg / m³. A complete process from enrichment to detection is executed to obtain a set of measured peak area intensities at standard concentrations. The ratio of the measured peak area intensities to the theoretical peak area intensities is calculated. The theoretical peak area intensities are read from the standard values ​​at that concentration in the volatile organic compound fingerprint feature database. The current detector sensitivity drift coefficient is generated. The most recently saved historical detector sensitivity drift coefficient is retrieved, and the relative change between the two coefficients is calculated. The relative change is compared with a preset drift alarm threshold, which is set to 5%. If the drift alarm threshold is exceeded, the currently generated detector sensitivity drift coefficient is applied to all subsequent concentration inversion calculations, replacing the original coefficients. If the drift alarm threshold is not exceeded, the original coefficients are maintained, and only the calibration data for this calibration is recorded. In some embodiments, the calibration cycle is set to be performed automatically every 24 hours. In an example scenario, the historical detector sensitivity drift coefficient is 0.98, and the measured peak area intensity obtained in this calibration is 0.95 times the theoretical value. That is, the currently generated detector sensitivity drift coefficient is 0.95, and the calculated relative change is 3.1%, which does not exceed the 5% drift alarm threshold. Therefore, the original coefficient is maintained. Optionally, the theoretical peak area intensity is calculated based on the standard peak area response coefficient in the volatile organic compound fingerprint feature library. The theoretical peak area intensity is equal to the product of the standard concentration and the standard peak area response coefficient. It can be understood that the relative change... The calculation formula is:

[0112]

[0113] in Indicates relative change. This represents the currently generated detector sensitivity drift coefficient. This represents the sensitivity drift coefficient of the most recently saved historical detector.

[0114] In one embodiment of the present invention, it is assumed that the preset performance drift threshold is ±0.03 minutes, and the acceptable correction range is defined as an absolute difference not exceeding 0.1 minutes. The system performs a periodic self-check step, cuts off the air inlet path of the air sample to be tested, introduces a calibration sample gas containing a trace amount of propane into the gas chromatography column, and starts a simplified gas chromatography separation program. The retention time of the propane component at the outlet of the gas chromatography column is recorded as 2.15 minutes, while the retention time of the propane standard stored in the system is 2.10 minutes, and the calculated difference is +0.05 minutes. Since the difference of 0.05 minutes exceeds the preset performance drift threshold of 0.03 minutes, but is within the acceptable correction range of 0.1 minutes, the system determines that the chromatographic column has performance drift but does not need to be immediately triggered for aging and cleaning, and instead starts the dynamic time correction process.

[0115] In practical implementation, when subsequent detection processes generate raw spectral data and require fingerprint feature database comparison, the system temporarily generates a corrected comparison time series. The operation to generate the corrected comparison time series involves reading the standard retention time from each standard record in the volatile organic compound (VOC) fingerprint feature database, adding the difference calculated in the aforementioned self-inspection steps to each standard retention time to obtain the corresponding temporary corrected retention time. For example, if the standard retention time of a standard record in the VOC fingerprint feature database is 5.60 minutes, then its temporary corrected retention time is calculated as 5.60 minutes + 0.05 minutes = 5.65 minutes. The temporary corrected retention times corresponding to all standard records constitute the corrected comparison time series. This correction process can be understood as follows:

[0116]

[0117] in, This represents the temporary corrected retention time for the i-th volatile organic compound. This represents the standard retention time of the i-th volatile organic compound. This represents the difference between the propane retention time calculated in the self-test step and the standard propane retention time. Subsequently, the system extracts the retention time series to be matched from the original spectral data and performs absolute value calculations and time tolerance threshold checks on the differences with the corrected comparison time series to complete the screening of target volatile organic compounds. Optionally, if the retention time of a peak to be matched extracted from the original spectral data is 5.66 minutes, and the difference between this peak and the temporary corrected retention time of benzene (5.65 minutes) in the corrected comparison time series is 0.01 minutes, which is below the time tolerance threshold, then this peak is matched as benzene.

[0118] In some embodiments, after completing the current test, the system performs another column performance self-test before the next self-test cycle. It is understood that if, after re-introducing the calibration gas, the difference between the measured propane retention time and the standard propane retention time returns to 0.02 minutes, which does not exceed the performance drift threshold of 0.03 minutes, the system determines that the column performance has returned to normal. When subsequently performing the step of inputting the original chromatographic data into a preset volatile organic compound fingerprint feature library for comparison, the system stops using the corrected comparison time series and resumes directly using the standard retention time from the original volatile organic compound fingerprint feature library for chromatographic data comparison. In some embodiments, if the difference calculated by subsequent self-tests consistently exceeds the performance drift threshold, the system will execute a column aging and cleaning command. Before the aging and cleaning are completed and the self-test confirms that the difference has returned to the normal range, the system dynamically generates and uses the corrected comparison time series based on the latest difference for each comparison. Optionally, the implementation of the dynamic time correction process ensures that even when the column experiences slight performance drift but has not yet affected the separation properties, the detection system can still perform accurate qualitative identification of substances based on the corrected time reference, providing the necessary time alignment for subsequent correct concentration inversion calculations.

[0119] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A method for online detection of volatile organic compounds in air based on gas chromatography, characterized in that, include: The air sample to be tested is introduced into the pretreatment pipeline, and the air sample is enriched at low temperature using the spiral condenser tube built into the pretreatment pipeline. After enrichment, the concentrate in the spiral condenser tube is transferred to the desorption chamber by instantaneous heating. The concentrate in the analytical chamber is carried into a gas chromatography column along with a carrier gas. The temperature gradient control program of the gas chromatography column is used to achieve time-sequential separation of volatile organic compounds with different boiling points in the gas chromatography column, forming a group of separated component peaks. The separated component peaks are scanned one by one using a flame ionization detector, and the retention time and peak area intensity of each component peak are collected to generate raw spectral data. The original spectral data is input into a preset volatile organic compound fingerprint feature library for comparison. The volatile organic compound fingerprint feature library stores the standard retention times and corresponding standard peak area response coefficients of known volatile organic compounds. Based on the comparison results, target volatile organic compound (VOC) species with retention time deviations within the allowable range from the VOC fingerprint feature library are selected. Then, using the standard peak area response coefficient corresponding to the target VOC species, the peak area intensity in the original spectral data is used to perform concentration inversion calculation, and the final VOC concentration detection result is output. The original spectral data is input into a pre-set volatile organic compound fingerprint feature database for comparison, including: The retention times of all component peaks are extracted from the original spectral data and sorted according to the order of the retention times to generate a retention time sequence to be matched. Traverse each standard record in the volatile organic compound fingerprint feature library, wherein the standard record contains the standard retention time and standard peak area response coefficient of a single volatile organic compound; Calculate the absolute value of the difference between each retention time in the retention time series to be matched and the standard retention time in the standard record; Determine whether the absolute value of the difference is less than a preset time tolerance threshold. If multiple consecutive absolute values ​​of the difference are less than the time tolerance threshold, then determine that the type of volatile organic compound corresponding to the standard record is a potential matching object. All volatile organic compounds identified as potential match targets and their corresponding standard peak area response coefficients are packaged and output as input parameters for subsequent concentration inversion calculations; The method further includes a step of periodically self-checking the performance of the gas chromatography separation column: Before each formal testing process begins, the air intake path of the air sample to be tested is cut off, and instead, a calibration sample gas containing trace amounts of propane gas is introduced into the gas chromatography separation column. Initiate a simplified gas chromatography separation procedure and record the retention time of the propane component in the calibration sample gas at the outlet of the gas chromatography separation column; The retention time is compared with the standard propane retention time stored in the system, and the difference between the retention time and the standard propane retention time is calculated. If the difference exceeds the preset performance drift threshold, a column aging and cleaning command is triggered to extend the subsequent detection cycle until the difference returns to the normal range. After comparing the retention time with the standard propane retention time stored in the system and calculating the difference between the two, the method further includes a step of dynamically correcting the fingerprint feature database comparison process based on the difference: If the difference does not exceed the preset performance drift threshold, the preset volatile organic compound fingerprint feature library is used directly for subsequent comparison. If the difference exceeds the preset performance drift threshold but is within an acceptable correction range, a corrected comparison time series will be temporarily generated when the step of inputting the original spectral data into the preset volatile organic compound fingerprint feature library for comparison is subsequently performed. Generating the corrected alignment time series includes: The standard retention time of each standard record is read from the volatile organic compound fingerprint feature library, and the difference is added to each standard retention time to generate a corresponding temporary corrected retention time. All the temporary corrected retention times constitute the corrected comparison time series. The absolute value of the difference between the time series to be matched extracted from the original spectral data and the time tolerance threshold are calculated and judged to complete the screening of target volatile organic compounds. After the column aging and cleaning command is completed and subsequent self-test confirms that the difference has returned to the normal range, the use of the corrected comparison time series is stopped, and the original volatile organic compound fingerprint feature library is resumed for comparison.

2. The method for online detection of volatile organic compounds in air based on gas chromatography according to claim 1, characterized in that, The air sample to be tested is enriched at low temperature using a spiral condenser tube built into the pretreatment pipeline, including: The semiconductor cooling chip located on the outside of the spiral condenser is activated to reduce the wall temperature of the spiral condenser to a preset low-temperature capture range. Open the air inlet valve of the air sample to be tested, and control the air sample to be tested to flow through the internal cavity of the spiral condenser tube at a constant low flow rate, so that the volatile organic molecules in the air are condensed and adsorbed on the inner wall of the spiral condenser tube. After the enrichment time reaches the preset duration, the air intake valve is closed and the delivery of the air sample to be tested is stopped; The instantaneous heating program of the spiral condenser is initiated, and the pulse heating wire wound around the outer wall of the spiral condenser is used to heat it at the millisecond level, causing the volatile organic compounds adsorbed on the inner wall of the spiral condenser to vaporize and desorb instantly; wherein, the winding density of the pulse heating wire from the sample inlet to the sample outlet of the spiral condenser gradually decreases, so that the heating power density at the sample inlet is greater than that at the sample outlet. Simultaneously, the inlet solenoid valve of the desorption chamber is switched to the on state, and the vaporized and desorbed volatile organic compounds are brought into the desorption chamber by the reverse purging of the carrier gas.

3. The method for online detection of volatile organic compounds in air based on gas chromatography according to claim 1, characterized in that, The temperature gradient control program of the gas chromatography column enables the sequential separation of volatile organic compounds with different boiling points within the gas chromatography column, including: Before loading the concentrate in the analytical chamber into the gas chromatography column, the initial temperature of the gas chromatography column oven is set to the lowest value below the boiling point of all target volatile organic compounds. After the concentrate enters the gas chromatography column with the carrier gas, the initial temperature is maintained for a fixed period of time to ensure that the low-boiling-point components completely enter the gas chromatography column. Subsequently, the column oven of the gas chromatography separation column is linearly heated according to the preset heating rate. During the heating process, volatile organic compounds with different boiling points are pushed forward in the gas chromatography separation column in sequence due to the difference in vaporization rate. The gas components flowing through the outlet of the gas chromatography separation column are monitored in real time. When a certain volatile organic compound component is detected to flow out, the column oven temperature and the effluent time point are recorded at this time. The effluent time point is confirmed as the retention time of the volatile organic compound component. Continue heating until all volatile organic compounds in the gas chromatography column have eluted, thus ending the separation procedure.

4. The method for online detection of volatile organic compounds in air based on gas chromatography according to claim 1, characterized in that, The step of scanning the separated component peaks one by one using a flame ionization detector and acquiring the retention time and peak area intensity of each component peak includes: Hydrogen and air are respectively introduced into the combustion head of the flame ionization detector to ignite and form a stable hydrogen flame; The gas components flowing out of the gas chromatography column are directly introduced into the root of the hydrogen flame, causing the organic compounds to undergo an ionization reaction in the high-temperature flame. A constant voltage is applied to the collecting electrode of the flame ionization detector to collect the charged ion flow generated by the ionization reaction and convert it into a current signal; The current signal is continuously sampled using a high-speed analog-to-digital converter module to obtain a current intensity curve that changes over time. The current intensity curve is subjected to digital filtering and noise reduction processing to identify and extract the peak region. The time span between the start point and the end point of each peak is recorded as the retention time of the corresponding component peak. The area integral value enclosed by the peak is determined as the peak area intensity of the corresponding component peak. The original spectral data includes the acquired retention time and peak area intensity.

5. The method for online detection of volatile organic compounds in air based on gas chromatography according to claim 1, characterized in that, Using the standard peak area response coefficients corresponding to the target volatile organic compound species, concentration inversion calculations are performed on the peak area intensities in the original spectral data, including: For each of the selected target volatile organic compounds, the corresponding standard peak area response coefficient is read from the volatile organic compound fingerprint feature library. The standard peak area response coefficient characterizes the peak area production capacity of a unit concentration of volatile organic compounds under specific detector conditions. Find the component peak that matches the target volatile organic compound in the original spectral data, and read the peak area intensity of the matched component peak; The measured concentration of the target volatile organic compound is calculated by dividing the peak area intensity by the standard peak area response coefficient. The measured concentration values ​​of all target volatile organic compounds are summarized to generate a dataset of detection results containing the concentrations of all detected volatile organic compounds.

6. The method for online detection of volatile organic compounds in air based on gas chromatography according to claim 1, characterized in that, The steps for constructing the volatile organic compound fingerprint feature library include: Several representative known volatile organic compound standard samples were selected, and the concentration range of each standard sample covered the actual concentration level of the expected detection environment. Under the same gas chromatography detection conditions, each standard sample was enriched, separated, and detected by flame ionization to obtain the corresponding standard spectral data. The retention time and peak area intensity of each volatile organic compound component were extracted from the standard spectral data, and the ratio of peak area intensity to concentration at the same concentration was used as the standard peak area response coefficient of the volatile organic compound component. The standard retention time, standard peak area response coefficient, and corresponding chemical name or number of each volatile organic compound are associated and stored to form a standard record; All standard records are categorized and indexed, sorted in ascending order by retention time, and appended with version number and generation date to generate the volatile organic compound fingerprint feature library.

7. The method for online detection of volatile organic compounds in air based on gas chromatography according to claim 1, characterized in that, The method also includes an adaptive adjustment step for the sampled flow rate based on meteorological parameters: The ambient temperature and atmospheric pressure values ​​at the deployment location are collected in real time, and the ambient temperature and atmospheric pressure values ​​are substituted into the gas state equation conversion model to calculate the current air density correction factor. Based on the air density correction factor, the rotation speed of the air intake sampling pump for the air sample to be tested is dynamically adjusted to maintain a constant air quality entering the system per unit time. Simultaneously, the average concentration of volatile organic compounds obtained from the previous cycle is read. If the average concentration of volatile organic compounds is higher than the high concentration warning threshold, the intake sampling flow rate of the air sample to be tested is temporarily reduced, and the single enrichment time is extended to ensure detection accuracy. After completing the high-concentration detection, the intake sampling flow rate of the air sample to be tested is restored to the normal set value.

8. The method for online detection of volatile organic compounds in air based on gas chromatography according to claim 1, characterized in that, The method further includes a drift calibration step for the detection data: Standard gases of known concentrations are periodically drawn into the detection system to perform a complete process from enrichment to detection, and a set of measured peak areas and intensities at standard concentrations are obtained. Calculate the ratio of the measured peak area intensity to the theoretical peak area intensity to generate the current detector sensitivity drift coefficient; Retrieve the most recently saved historical detector sensitivity drift coefficient and calculate the relative change between the two coefficients; The relative change is compared with a preset drift alarm threshold. If it exceeds the drift alarm threshold, the currently generated detector sensitivity drift coefficient is applied to all subsequent concentration inversion calculations, replacing the original coefficient. If the drift alarm threshold is not exceeded, the original coefficient remains unchanged, and only the calibration data for this calibration is recorded.

Images