Vehicle refrigerant analysis method based on Fourier transform technology

Abstract

The invention relates to the technical field of vehicle refrigerant analysis, and discloses a vehicle refrigerant analysis method based on a Fourier transform technology, which comprises the following steps: acquiring an infrared absorption spectrum of a calibration gas with known concentration; the method comprises the following steps: determining a model range, preparing calibration gas, preselecting a quantitative wave band, preprocessing and calibrating an instrument, collecting spectral data, fitting a verification model and measuring a quantitative limit, so as to construct a quantitative relation model of a refrigerant light absorption signal response value and concentration; based on the quantitative relation model, introducing a gas sample in the leakage test cabin into a Fourier infrared gas analyzer, collecting spectrums in real time according to modeling consistent parameters, and recording moments to realize real-time online monitoring analysis of the refrigerant; the technical problems that in the prior art, quantitative precision is low, and real-time online monitoring cannot be achieved are solved.

Description

Technical Field

[0001] This invention relates to the field of automotive refrigerant analysis technology, specifically to an automotive refrigerant analysis method based on Fourier transform infrared technology. Background Technology

[0002] Automotive air conditioning refrigerants are fluorocarbons (FCCs). The main source of consumption and emissions of refrigerant is the refrigerant. Its leakage will not only cause excessive greenhouse gas emissions and exacerbate environmental pressure, but also reduce the operating efficiency of the air conditioning system and affect the safety of vehicle driving. The carbon emissions caused by refrigerant leakage account for about 10% of the carbon footprint of the whole life cycle of the vehicle.

[0003] Currently, the main methods for detecting and analyzing automotive refrigerants include portable electronic leak detectors, gas chromatography-mass spectrometry (GC-MS) offline analysis, and traditional infrared spectroscopy. Portable electronic leak detectors are primarily used for point-based leak detection during maintenance, only able to determine the presence of obvious leaks, but unable to achieve precise quantitative analysis and continuous online monitoring of concentration. While GC-MS offline analysis offers high quantitative accuracy, it requires sample collection and pretreatment, resulting in a long testing cycle and failing to meet the real-time dynamic monitoring requirements of leak testing. Furthermore, the testing process involves refrigerant sample loss, which can easily lead to data deviation. Summary of the Invention

[0004] The present invention aims to provide an analytical method for automotive refrigerants based on Fourier transform infrared technology, in order to solve the technical problems of low quantitative accuracy and inability to monitor in real time in the prior art.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for analyzing automotive refrigerants based on Fourier transform infrared technology, comprising: S1. By collecting the infrared absorption spectra of calibration gases with known concentrations, a quantitative relationship model between the light absorption signal response value and concentration of the refrigerant to be tested is constructed. S101. Estimate the concentration range of the refrigerant to be tested in the leakage test chamber, determine the working range of the quantitative model and the concentration gradient of the calibration gas. The working range must include the zero point concentration. The calibration gas is prepared using high-purity nitrogen as the base gas, and the concentration of the calibration gas is within ±2% of the target concentration value. S102. Select an appropriate quantitative band based on the standard infrared absorption spectrum of the refrigerant to be tested, the target concentration, and the interfering components. S103. Perform a gas path system sealing check on the Fourier transform infrared gas analyzer, verify the instrument operating parameters, heat the sampling system and gas cell to the specified working temperature, purge the gas path system with high-purity nitrogen until the spectral signal is stable, and complete the zero-point calibration of the interferometer and background spectrum acquisition. S104. Introduce calibration gas in order from zero concentration to high concentration. After the gas in the gas cell stabilizes, collect infrared absorption spectra of each concentration and record the absorbance signal response value in the quantitative band. Repeat the injection for each concentration 3 to 5 times. The relative deviation of the absorbance signal response value at the same concentration is ≤5%, and the average value is taken as the modeling data. S105. A working curve is to be plotted with the absorbance signal response value as the ordinate and the nominal concentration of the calibrated gas as the abscissa, and a quantitative model is to be established. After the model is established, the accuracy is to be verified. If the verification fails, the model is to be re-modeled. S106. Using the blank sample spiking method, repeatedly inject the zero-point calibration gas more than 10 times, calculate the instrument's limit of quantitation for the refrigerant, and establish a low-concentration calibration curve by injecting more than 5 sets of low-concentration calibration gases, according to the formula:

[0006] in, The limit of quantitation of the instrument. ; Factors related to confidence level; The standard deviation of the light absorption signal response value for background noise; The slope of the calibration curve within the concentration range; S2. Based on the established quantitative model, conduct real-time online monitoring and analysis of the automotive refrigerant in the leakage test chamber; introduce the gas sample in the leakage test chamber into the Fourier transform infrared gas analyzer through the sampling system, select the quantitative model corresponding to the refrigerant, and use the same operating parameters as the modeling to collect the infrared absorption spectrum in real time and record the sampling time.

[0007] The principle and advantages of this solution are as follows: In practical applications, this solution relies on the spectral feature identification and quantitative analysis principle of Fourier transform technology. First, a dedicated quantitative model is constructed by calibrating the gas. Then, real-time online monitoring of the refrigerant is carried out based on the model. The core is to achieve accurate and real-time analysis of automotive refrigerant concentration by combining the construction of a dedicated quantitative model with a standardized online monitoring process.

[0008] By utilizing the characteristic absorption properties of different refrigerant molecules to infrared light of specific wavelengths, and by collecting the infrared absorption spectra of calibration gases with known concentrations, a quantitative correspondence between the light absorption signal response value and the refrigerant concentration is established, forming a dedicated quantitative model. During online monitoring, the infrared absorption spectrum of the refrigerant sample in the leak test chamber is substituted into the model, and the real-time refrigerant concentration is directly calculated from the light absorption signal response value of the characteristic band. At the same time, through full-process instrument calibration, parameter matching, and operation standardization, the model's adaptability and the accuracy of spectral acquisition are ensured.

[0009] By customizing a quantitative model specifically for automotive refrigerants, the refrigerant concentration range within the leak test chamber was first estimated to determine the appropriate model range. Then, based on the refrigerant concentration and interfering components, suitable quantitative characteristic bands were selected. Simultaneously, multiple repeated tests were conducted on calibration gas samples, and the average value was used to model the model. Furthermore, the instrument's quantitation limit was accurately measured and controlled to be ≤0.1 using a blank sample spiking method. This improves quantitative accuracy from the source of model construction, effectively avoiding quantitative deviations caused by low-concentration signals being submerged by noise, high-concentration spectral saturation, and interference components. Based on the established dedicated quantitative model, the refrigerant sample in the leak test chamber is directly introduced into the Fourier transform infrared gas analyzer through the sampling system. The infrared absorption spectrum is collected in real time using the same operating parameters as the model. The real-time concentration data is obtained by rapid conversion through the model. No sample pretreatment or offline detection is required, which realizes continuous dynamic online monitoring of refrigerant concentration and meets the real-time monitoring needs of automotive air conditioning refrigerant leak testing. Before model building, the Fourier transform infrared gas analyzer underwent standardized preprocessing, including gas path sealing check, operating parameter verification, system purging, and zero-point calibration. During online monitoring, the instrument parameters used during modeling were strictly followed, ensuring the stability of the instrument's operating status and the consistency of spectral acquisition. This effectively reduced data deviations caused by fluctuations in instrument parameters, gas path leakage, and background interference, and improved the repeatability and reliability of the analysis results. Clear requirements are set for the preparation accuracy of the calibration gas to ensure that the concentration is within ±2% of the target value. After modeling, the model is double-validated by zero-point calibration gas and known concentration calibration gas not involved in the modeling. If the validation fails, the model is remodeled to ensure the accuracy and adaptability of the quantitative model and provide a reliable benchmark for subsequent online quantitative analysis. A low-concentration calibration curve was specifically established using the blank sample spiking method, and the instrument's limit of quantitation was calculated. The instrument's limit of quantitation was rigorously verified to be within 0.1. Within this range, the instrument's detection sensitivity to low concentrations of refrigerant in the leak test chamber has been significantly improved, enabling it to accurately capture concentration changes after a trace refrigerant leak, thus meeting the analytical needs of trace refrigerant leak testing in automotive air conditioning systems.

[0010] Preferably, as an improvement, S2 further includes: S201. Before the first test each day, use the pressurization and pressure holding method or the vacuum pressure holding method to check the sealing of the internal and external gas paths of the Fourier transform infrared gas analyzer, verify the instrument operating parameters and ensure that they are consistent with those during modeling, and heat the sampling system and gas cell to the specified operating temperature. S202. Purge the gas path system of the analyzer with high-purity nitrogen until the spectral signal is stable, complete the zero-point calibration of the interferometer and the acquisition of the background spectrum and subtract the blank background, verify the accuracy of the quantitative model, and maintain the optical chamber purging with high-purity nitrogen throughout the process. S203. The gas sample in the leakage test chamber is introduced into the Fourier transform infrared gas analyzer through the sampling system. The temperature of the sampling system changes synchronously with the test chamber and the control accuracy is ±1℃. The quantitative model of the corresponding refrigerant is selected, and the infrared absorption spectrum is collected in real time and the sampling time is recorded using the same operating parameters as the modeling. S204. The absorbance signal response value is converted into refrigerant concentration and output in real time through the quantitative model. The flow rate, temperature and pressure in the gas cell are monitored synchronously. When the results are inconsistent with the modeling steady state, temperature and pressure compensation is performed and the quantitative model is corrected. When wavenumber drift or injection changes affect the results, the quantitative model is corrected or re-established in a timely manner. S205. The analyzed gas sample is returned to the leak test chamber. After the test, the sampling pipeline is purged with high-purity nitrogen. The shutdown operation is completed according to the instruction manual. The relevant test data is recorded and integrated through the supporting software to form a permanent record.

[0011] The beneficial effects of this improvement are: by checking the gas path sealing, verifying parameters, and preheating the system before daily testing, analytical errors caused by gas path leaks and parameter deviations are avoided from the source of detection, ensuring that the instrument is always in the standard operating condition used during modeling; the entire process is purged with high-purity nitrogen and zero-point calibration and background subtraction are completed, effectively eliminating... , Background interference is eliminated, and the model accuracy is verified in real time to ensure that the model is effective and usable during the monitoring process; The sampling system is synchronized with the temperature of the test chamber and the temperature control accuracy reaches ±1℃, which avoids the concentration deviation caused by temperature changes during sample transmission and ensures the representativeness of the sample. By monitoring the gas cell conditions in real time and compensating for temperature and pressure, and correcting the model in a timely manner, the effects of fluctuations in operating conditions and equipment drift during the test can be dynamically offset, and the accuracy of quantitative analysis can be continuously guaranteed. The sample reflux design avoids refrigerant loss and environmental pollution. The system purging and data integration recording after the test not only protect the instruments and equipment, but also make the test data traceable and verifiable, meeting the management requirements of standardized testing.

[0012] Preferably, as an improvement, the analysis method further includes S3 implementing full-process quality assurance and control of the analysis process: S301. The detector dark current, infrared light source, and background noise generated by electronic thermal noise should be tested at least every six months, and the test should be performed immediately after the instrument is repaired. S302. When the gas to be tested contains known interfering components, correct or subtract the infrared absorption spectrum contribution of the interfering components. S303. The performance parameters and moisture content of the Fourier transform infrared gas analyzer shall be calibrated once a year using standard materials. The zero point and range of the instrument shall be recalibrated before each test. The calibration curve shall be fitted using the least squares method. The number of calibration points shall be determined according to the order of the fitted polynomial. The relative deviation between the fitted value of the calibration point and the nominal value shall be ≤2%. S304. Analyzers that are frequently used should be purged with high-purity nitrogen gas at a flow rate of ≥300mL / min for a long time. Analyzers that are not used for a long time should have a desiccant inside that is within its expiration date and be replaced in time. Before turning on the instrument, remove the desiccant and purge it with high-purity nitrogen gas for more than 1 day. After repairing or replacing important parts of the instrument, return it to the factory for recalibration. S305. Regularly clean the sampling head and sampling pipeline. The sampling system components can be easily replaced and cleaned. Ensure good airtightness at the connection points.

[0013] The benefits of this improvement are: regular background noise testing and retesting after instrument maintenance can keep track of the operating status of the instrument's core components, ensuring long-term instrument stability and avoiding signal distortion caused by excessive noise; spectral correction or subtraction of interfering components effectively eliminates the influence of impurities in the gas being tested on the characteristic absorption peaks of the refrigerant, improving the quantitative accuracy under complex operating conditions. The annual metrological calibration is combined with the zero-point / range calibration before each test, and the calibration curve fitting requirements are strictly standardized. This ensures the instrument's detection accuracy from the perspective of traceability and avoids systematic errors caused by equipment drift. Differentiated maintenance plans are developed for different instrument usage conditions. Important components are returned to the factory for calibration after repair, which can extend the instrument's service life and continuously ensure that the instrument's performance meets the standards. The sampling system is regularly maintained and its airtightness and ease of maintenance are ensured, avoiding sample loss or distortion caused by problems such as sampling head blockage, pipeline adsorption, and connection leakage, thus ensuring the reliability of the sampling process.

[0014] Preferably, as an improvement, in step S103, the gas system sealing test adopts the pressure boosting and pressure holding method or the vacuum pressure holding method; after purging, the interference signal in the single beam spectrum is stable or disappears, the full spectrum light intensity signal remains stable and the spectral shape is normal, and the blank absorption spectrum fluctuates flatly near zero value without obvious absorption peaks.

[0015] The beneficial effects of this improvement are: it clearly defines the use of pressurization and pressure holding methods or vacuum pressure holding methods to check the gas path sealing, ensuring no leaks and avoiding spectral acquisition deviations caused by sample leakage or external gas infiltration; it establishes criteria for judging spectral signals after purging, ensuring that residual refrigerant, water vapor, and other components in the gas path system are properly controlled. Once the interfering substances are completely removed, the full-spectrum light intensity is stable and the spectral shape is normal, ensuring the purity of subsequent spectral acquisition; the criteria for determining the blank absorption spectrum are clearly defined to ensure thorough background subtraction and avoid background signals being superimposed on the sample signals, which would cause distortion of the absorbance signal response value and improve the original accuracy of the modeling data.

[0016] Preferably, as an improvement, in step S105, the relevant parameters of the quantitative model and the gas cell include operating temperature, operating pressure, optical path length, and the resolution, number of scans, apodization function, phase correction, and gain strength of the spectrometer. When the relevant parameters change, the quantitative model is re-established; the quantitative model and monitoring analysis use the same instrument to collect absorption spectra. The model validation criteria are: when zero-point calibration gas is introduced, the model output concentration is zero; when a known concentration calibration gas not involved in the modeling is introduced, the relative deviation between the model output concentration and the nominal concentration is ≤5%.

[0017] The beneficial effects of this improvement are: it clarifies the strong correlation between the quantitative model and the core parameters of the gas cell and spectrometer, and stipulates that modeling should be re-established after parameter changes, thus avoiding model failure caused by parameter fluctuations and ensuring that the model always matches the actual working state of the instrument. The quantitative model and monitoring analysis are required to use the same instrument, eliminating systematic errors caused by performance differences between different instruments and ensuring consistency between modeling and monitoring. A dual model validation standard is established: zero-point calibration gas validation ensures that the model has no zero-point drift, while calibration gas validation, which is not involved in modeling, can comprehensively verify the quantitative accuracy of the model and control the relative deviation to ≤5%. The model quality is strictly controlled from the results level to ensure that the model can accurately convert the correspondence between light absorption signal and refrigerant concentration.

[0018] Preferably, as an improvement, the instrument operating parameters in step S201 include spectral range, spectral resolution, number of scans, aperture, moving mirror scanning frequency, apodization function, phase correction, and gain; In step S204, temperature and pressure compensation are performed using the accompanying software or the ideal gas law.

[0019] The beneficial effects of this improvement are: it clarifies the specific range of instrument operating parameters, realizes standardized control of all core parameters of spectral acquisition, avoids distortion of spectral shape and absorbance signal response values ​​caused by deviation of any parameter, and ensures the consistency between the sampled spectrum and the modeled spectrum. The regulations stipulate the use of supporting software or ideal gas law for temperature and pressure compensation, providing a scientific and practical solution for operating condition fluctuations. This can accurately offset the impact of temperature, pressure, and flow fluctuations in the gas pool on the absorption characteristics of refrigerant molecules, correct concentration calculation results according to ideal gas law, and make quantitative analysis results more consistent with actual operating conditions, thereby improving the quantitative accuracy in the dynamic monitoring process.

[0020] Preferably, as an improvement, in step S303, the calibration curve includes at least 5 equidistant calibration points, and the nominal value of the highest concentration calibration point is ≥ 80% of the working range; If the order of the fitted polynomial is greater than 3, then the number of calibration points shall be greater than or equal to the order of the polynomial plus 2. A calibration curve table shall be drawn based on the polynomial coefficients fitted to the calibration curve. The table step size shall not be greater than 1% of the full scale and shall include the calibration date, zero point, range, calibration gas reference data, and the percentage deviation between the nominal value and the fitted value.

[0021] The beneficial effects of this improvement are: it requires the calibration curve to include at least 5 equidistant calibration points and the highest concentration to be ≥80% of the working range, ensuring that the calibration curve can cover the full concentration range of actual detection and avoiding concentration conversion deviations caused by insufficient calibration points or incomplete coverage. For high-order polynomial fitting, a requirement for the number of calibration points is set to ensure that the fitting curve can accurately fit the concentration-signal response relationship of the standard gas and avoid curve distortion caused by insufficient calibration points during high-order fitting. The requirement is to create a detailed calibration curve table containing complete calibration information, which ensures the accuracy of concentration conversion and enables full traceability of the calibration process. This facilitates subsequent verification of calibration quality and analysis of equipment drift, and meets the needs of standardized and regulated testing management. Attached Figure Description

[0022] Figure 1 This is a flowchart of an embodiment of the present invention. Detailed Implementation

[0023] The following detailed description illustrates the specific implementation method: Example Fourier transform infrared technology, also known as Fourier infrared transformation technology, includes related equipment such as Fourier transform infrared gas analyzers (including core components such as gas cells, interferometers, detectors, control and analysis software), sampling systems, standard leaks, and auxiliary equipment such as SUMMA canisters; The gas cell, the core component of a Fourier transform infrared gas analyzer, is a sealed cavity in which the gas sample to be tested resides and interacts with infrared light. When infrared light passes through the gas cell, the refrigerant molecules in the sample absorb infrared light of a specific wavelength, forming a characteristic absorption spectrum. This is a key step in achieving concentration analysis and requires materials that are inert and have good airtightness.

[0024] The interferometer, the core optical component of the Fourier transform infrared analyzer, converts continuous light emitted from an infrared source into interference light. The interference light passes through a gas cell and is received by a detector, then converted into an infrared absorption spectrum through a Fourier transform, which is the basis for achieving spectral analysis.

[0025] The sampling system, consisting of a sampling head, dust filter, insulated and heat-traced pipeline, and flow controller, is designed to collect representative gas samples from the leak test chamber and deliver them at a constant temperature and without loss to the gas cell of the infrared analyzer. The pipeline is designed with insulation and heat tracing to prevent concentration deviations caused by sample temperature changes.

[0026] The sampling head, the front-end component of the sampling system, is used to extract gas samples from the test chamber. It should preferably be made of stainless steel. The collected samples should be representative of the overall gas concentration inside the chamber and should be cleaned regularly to prevent the sampling holes from becoming clogged.

[0027] Standard leak orifices are auxiliary equipment for infrared analysis and test chamber calibration. The leak rate is adjustable (1.0~60 g / year) with an accuracy of ±0.1 g / year. Their core function is to provide a quantitative source of refrigerant leakage for the leak test chamber and infrared analyzer for equipment calibration and performance verification. Standard leak orifices for different refrigerants cannot be mixed.

[0028] The SUMMA canister is a high-precision gas sampling container that connects to the sampling port of the test chamber for intermittent sampling and analysis of the gas inside the chamber. It can save the sample for offline testing, supplementing the shortcomings of online monitoring, and is suitable for accurately verifying concentration data.

[0029] The spectral range of a Fourier transform infrared gas analyzer should ideally be 4500. ~800 The instrument's limit of molar refrigerant concentration should not exceed 0.1. The spectral resolution should ideally be around 0.5. ~4 Within this range, the accuracy of the wavenumber should preferably not exceed ±1. .

[0030] Under the same conditions, the repeatability of the absorption signal response value of continuously acquired absorption spectra of the same gas sample should not exceed 0.5%, and the wavenumber repeatability should not exceed 0.5%. When using zero-point gas and range gas for testing, the zero-point drift within 24 hours shall not exceed ±1% of full scale, and the range drift within 24 hours shall not exceed ±1% of full scale.

[0031] In 1 The spectral acquisition rate at the specified resolution should be no less than 1 acquisition per second. Dedicated control and analysis software should include functions such as instrument control, online diagnostics, anomaly detection, and self-testing procedures. It should support continuous quantitative analysis of gas sample concentrations and preferably have a built-in standard spectral library of refrigerants such as R134B, R1234yf, and R152B.

[0032] Leakage test related equipment includes a leakage test chamber, a gas mixing device, a gas purging device, a pressure compensation device, and an air conditioning test component support.

[0033] The leak test chamber is a sealed enclosure device. Its core function is to simulate the temperature and humidity environment of a real automotive air conditioning system, collect leaked refrigerant from air conditioning components, and provide a sealed and controllable test space for leak testing; a volume of ≥3 cubic meters is required. It is equipped with temperature / pressure regulation, gas mixing / purging, sampling ports and other devices. The chamber material is impermeable and does not react with refrigerant.

[0034] The gas mixing device, an auxiliary device of the leak test chamber, is mostly a circulating fan. Its function is to fully mix the leaked refrigerant with the air in the test chamber, ensure that the refrigerant concentration in the chamber is uniform, avoid local concentration deviations that could lead to inaccurate analysis data, and the airflow must not directly blow on the test components.

[0035] Gas purging device, an auxiliary device of the test chamber, including a fan, blower or nitrogen purging equipment, is turned on before / after the test to reduce the refrigerant concentration in the chamber to below the ambient concentration in the test room, so as to avoid refrigerant residue interfering with subsequent tests and ensure the safety of test operation.

[0036] Pressure compensation devices, mostly gas-impermeable expansion bags, are designed to automatically adjust the chamber's internal volume based on temperature changes and gas sampling operations. This maintains the pressure difference between the chamber and atmospheric pressure within ±500 Pa, ensuring the entire test is conducted under isobaric conditions and eliminating the impact of pressure changes on refrigerant concentration measurements.

[0037] Air conditioning test component brackets are tooling used to fix air conditioning compressors, evaporators, condensers and other components. They are required to ensure that there is no relative displacement of the components during the test, to fit the actual vehicle installation state, and that the material is consistent with the inner wall of the test chamber, and that it is non-permeable and has low refrigerant adsorption.

[0038] The temperature control of the entire sampling system should be synchronized with the leak test chamber, with a temperature control accuracy within ±1℃, to ensure that the gas sample enters the Fourier transform infrared gas analyzer under constant temperature conditions. All materials in the sampling system that come into direct contact with the gas sample should be inert materials that are corrosion-resistant, impermeable, contamination-resistant, adsorption-resistant, and do not chemically react with the refrigerant being tested. Materials that have undergone passivation, coating, polishing, or other surface treatments are preferable. Components in the sampling system should be easily replaceable and cleanable, and connections should ensure good airtightness.

[0039] As attached Figure 1 As shown, the method for analyzing automotive refrigerants based on Fourier transform infrared technology includes: S1. Establish a quantitative model specific to automotive refrigerants. By collecting the infrared absorption spectra of calibration gases with known concentrations, construct a quantitative relationship model between the light absorption signal response value and concentration of the refrigerant to be tested, providing a benchmark for subsequent quantitative analysis.

[0040] S101. Determine the model range and prepare the calibration gas, estimate the concentration range of the refrigerant to be tested in the leakage test chamber, determine the working range of the quantitative model and the concentration gradient of the calibration gas, and the working range of the quantitative model must include the zero-point concentration. Using high-purity nitrogen as the base gas, a certain amount of the refrigerant to be tested is added to prepare the calibration gas, ensuring that the gas is dry, free from pollution and interference, and that the concentration is within the nominal value, i.e., the target concentration value, ±2%.

[0041] Pure gases used for equipment calibration and testing operations include: Standard refrigerants: 1,1,1,2-tetrafluoroethane (R134a), 2,2,2,3-tetrafluoropropylene (R1234yf) and 1,1-difluoroethane (R152a), with a purity ≥99.5%, stored in steel cylinders and kept in accordance with the product instructions.

[0042] High-purity nitrogen ( The purity should be no less than 99.999%, and it should undergo the corresponding metrological certification.

[0043] The zero-point calibration gas should be high-purity nitrogen gas that does not contain the refrigerant being tested. The range calibration gas should be a mixture of standard refrigerant and high-purity nitrogen gas, and its concentration should be within ±2% of the nominal value.

[0044] S102. Pre-select quantitative characteristic bands. Based on the standard infrared absorption spectrum of the refrigerant to be tested, and in combination with the target concentration and interfering components, select appropriate quantitative bands. Quantitative band selection rules include: When the concentration of the refrigerant to be measured is low, the quantitative band should preferably be a characteristic absorption band with a strong absorption peak intensity to avoid the absorption peak being submerged by baseline noise. When the concentration of the refrigerant to be tested is high, the quantitative band should be selected from the characteristic absorption band with a weaker absorption peak intensity to avoid spectral saturation of the absorption peak. When other interfering components are present in the refrigerant to be tested, the quantitative band should be selected from the characteristic absorption bands that have less interference, higher absorption peak intensity, and better signal-to-noise ratio.

[0045] Specifically, for example, for low-concentration scenarios where the concentration of the refrigerant to be tested in the test chamber is within 5 times the limit of quantitation, the characteristic absorption band with the largest light absorption signal response value in its standard infrared absorption spectrum is selected. For scenarios where the concentration of the refrigerant to be tested in the test chamber is more than 50 times the limit of quantitation, the characteristic absorption band in its standard infrared absorption spectrum that has no maximum light absorption signal response value and no spectral saturation is selected. For scenarios where known interfering components exist in the gas to be tested, a characteristic absorption band in the standard infrared absorption spectrum of the refrigerant to be tested is selected, which has no overlap with the absorption peak of the interfering component, an absorption signal response value ≥ 30% of the maximum response value, and a signal-to-noise ratio ≥ 10.

[0046] S103. Instrument preprocessing and parameter calibration: Check the gas path system sealing of the Fourier transform infrared gas analyzer, and verify the operating parameters such as spectral range, resolution, number of scans, and apodization function by using the pressurization and pressure holding method or the vacuum pressure holding method. The sampling system and gas cell of the Fourier transform infrared gas analyzer are heated to the specified operating temperature to remove residual gas from the inner wall of the gas path; The optical chamber, sampling lines, gas cell, and other gas path systems were purged with high-purity nitrogen, and single-beam spectra were continuously acquired until the single-beam spectrum reached the target value. , Wait until the interference signal stabilizes or disappears, the full-spectrum light intensity signal remains stable, and the spectral shape is normal to ensure that the instrument purging has reached a stable state.

[0047] Complete the zero-point calibration of the interferometer and the acquisition of the background spectrum. The blank absorption spectrum should fluctuate flatly near zero and have no obvious absorption peaks.

[0048] S104. Collect calibration gas spectral data. In order from zero concentration to high concentration, the calibration gas is introduced into the analyzer in sequence. After the gas in the gas cell is fully replaced and stabilized, the infrared absorption spectrum of the calibration gas at each concentration is collected, and the light absorption signal response value in the pre-selected calibration band is recorded. Each concentration was injected 3 to 5 times to ensure that the relative deviation of the absorbance signal response value at the same concentration was ≤5%, and the average value was taken as the modeling data.

[0049] S105. Fit, establish and verify the quantitative model, plot the working curve with the absorbance signal response value as the ordinate and the nominal concentration of the calibration gas as the abscissa, evaluate the linearity of the curve, and add calibration points and supplement the spectral data in the nonlinear region. Based on the complexity of the calibration gas, such as overlapping spectral peaks, methods such as univariate linear regression, multivariate linear regression, and spectral fitting are selected to establish a quantitative model. The quantitative model is established by collecting the absorption spectra of a set of calibration gases with known concentrations and establishing a response relationship model between the absorbance signal response value of the analyte and its concentration.

[0050] The quantitative model is strongly correlated with the operating temperature, pressure, and optical path length of the gas cell, and is also related to the spectrometer's resolution, scan count, apodization function, phase correction, and gain settings. If any parameter changes, the quantitative model should be re-established. The quantitative model and monitoring analysis should use the same instrument for absorption spectral acquisition.

[0051] The model should be verified immediately after it is built. Zero-point calibration gas should be introduced and the model output concentration should be zero. If a known concentration calibration gas that was not involved in the modeling is introduced, the relative deviation between the model output concentration and the nominal concentration should be ≤5%. If the verification fails, the model should be rebuilt.

[0052] S106. To determine the instrument's limit of quantitation (LOQ), the blank sample spiking method is used. Zero-point calibration gas is introduced and the sample is injected repeatedly at least 10 times. Then, at least 5 sets of low-concentration calibration gas covering the estimated LOQ are introduced to establish a low-concentration calibration curve. The LOQ of the instrument for this refrigerant is calculated using the formula:

[0053] in, The limit of quantitation of the instrument. ; As a factor related to confidence level, it is set to 10 according to the regulations of the International Union of Pure and Applied Chemistry. This represents the standard deviation of the background noise light absorption signal response value, and its unit is consistent with that of the light absorption signal response value. The slope of the calibration curve within the concentration range, with absorbance signal response values ​​in units of / .

[0054] S2. Conduct online quantitative analysis of automotive refrigerants. Based on the established quantitative model, perform real-time online monitoring and analysis of the automotive refrigerants in the leakage test chamber, ensuring the non-destructive nature of the analysis process and the accuracy of the data throughout. Specific operations are as follows: S201. Pre-test instrument check: Before the first test each day, check the gas path sealing of the Fourier transform infrared gas analyzer and verify the operating parameters to ensure that the parameters are consistent with those used in the modeling.

[0055] The gas path system sealing test includes the external gas path of the sampling system and the internal gas path of the Fourier transform infrared gas analyzer. The gas path system sealing test should be performed using the pressurization and pressure holding method or the vacuum pressure holding method.

[0056] The instrument operating parameters check includes setting the spectral range, spectral resolution, number of scans, aperture, moving mirror scanning frequency, apodization function, phase correction, gain, and other instrument parameters to ensure that the operating parameters are exactly the same as those set when establishing the quantitative model.

[0057] Before starting system purging, heating the sampling system and gas pool to the specified operating temperature helps to remove residual gas adhering to the inner wall of the gas path system.

[0058] S202. Pre-monitoring preparations: Continuously purge the analyzer's optical chamber, sampling pipeline, gas cell, and other gas path systems with high-purity nitrogen until the spectral signal stabilizes. Complete the zero-point calibration of the interferometer and the acquisition of the background spectrum, and subtract the blank background; verify the accuracy of the quantitative model, ensure that the model is in an effective state, and maintain the optical chamber purging with high-purity nitrogen throughout the process.

[0059] S203. Real-time sampling and spectral acquisition: Gas samples from the leakage test chamber are introduced into a Fourier transform infrared gas analyzer through a sampling system, including a dust filter and insulated heating pipeline. The temperature of the sampling system changes synchronously with the test chamber, with a control accuracy of ±1℃. A quantitative model for the corresponding refrigerant is selected, and operating parameters consistent with the modeling are used to acquire the infrared absorption spectrum of the gas sample in real time and record the sampling time.

[0060] S204. Real-time quantification and data correction: The absorbance signal response value is converted into refrigerant concentration through a quantitative model, and the concentration data is output in real time. Simultaneously monitor the flow rate, temperature, and pressure in the gas pool. If they are inconsistent with the steady state during modeling, use the accompanying software or ideal gas law to compensate for the temperature and pressure and correct the quantitative model. When factors such as wavenumber drift and changes in sample injection significantly affect the accuracy of the results, timely correction or re-establishment of the quantitative model is necessary.

[0061] S205, Sample reflux and instrument termination: The analyzed gas sample is refluxed back to the leak test chamber through the loop pipeline to avoid refrigerant loss and environmental pollution. After the test, the entire sampling pipeline was purged with high-purity nitrogen to ensure that no sample remained. The instrument was then shut down according to the instruction manual. Data such as the temperature, pressure difference, refrigerant concentration, and sampling time of the leak test chamber were recorded and integrated using the analyzer's software to form a permanent record.

[0062] S3. Implement full-process quality assurance and control for the analysis process. Through multi-dimensional quality control measures, ensure the stability of the quantitative model, the accuracy of the analyzer, and the reliability of the analysis results. Specific technical means include: S301 Background noise monitoring: Background noise generated by detector dark current, infrared light source, electronic thermal noise, etc., shall be tested at least every six months, and retested immediately after instrument maintenance to ensure that the noise level meets the analytical requirements and to ensure instrument stability.

[0063] S302. Elimination of interfering components: When the gas sample to be tested contains known interfering components, the infrared absorption spectrum contribution of the interfering components is corrected or subtracted by special software with spectral analysis capabilities, thereby eliminating their influence on the quantitative analysis of refrigerants.

[0064] S303. Instrument Measurement and Calibration: The performance parameters and moisture content of the Fourier Transform Infrared Gas Analyzer shall be calibrated once a year using standard materials. The zero point and range of the analytical equipment are recalibrated before each test. The calibration curve contains at least 5 equidistant calibration points. The nominal value of the highest concentration calibration point is ≥ 80% of the working range, and the relative deviation between the fitted value and the nominal value is ≤ 2%.

[0065] The calibration curve should be fitted using the least squares method. If the order of the fitting polynomial is greater than 3, the number of calibration points should be greater than or equal to the order of the polynomial plus 2. The relative deviation between the fitted value at each calibration point on the calibration curve and the nominal value of its corresponding calibration gas should not exceed ±2%.

[0066] Using the polynomial coefficients obtained from fitting the calibration curve, plot a calibration curve table showing the nominal and fitted values ​​of the calibration gas, with a step size no greater than 1% of the full scale. This table should also include other relevant data, such as: calibration date, zero point, range, reference data for each calibration gas, and the percentage deviation between the nominal and fitted values ​​for each calibration gas.

[0067] S304. Routine maintenance and upkeep of instruments: For analyzers that are used frequently, long-term purging with high-purity nitrogen gas at a flow rate of ≥300mL / min is required. For analyzers that have not been used for a long time, place a desiccant within its expiration date inside and replace it in a timely manner. Before turning on the instrument, remove the desiccant and purge it with high-purity nitrogen for more than one day. After repairing or replacing important parts of the instrument, return it to the factory for recalibration.

[0068] S305. Sampling system maintenance: Regularly clean the sampling head and sampling pipeline to prevent clogging of sampling holes and residual contaminants from affecting analysis results; ensure that sampling system components are easy to replace and clean, and ensure good airtightness at connection points.

[0069] The automotive refrigerant analysis method based on Fourier transform technology achieves qualitative analysis by comparing the characteristic absorption peaks of the infrared absorption spectrum of the refrigerant to be tested with those of the standard substance. Then, quantitative analysis is completed by using a quantitative model of light absorption signal response value and concentration. The entire process covers three major links: quantitative model establishment and accuracy verification, daily analysis execution, and quality assurance and control. The process is coherent and logically closed-loop.

[0070] The above descriptions are merely embodiments of the present invention, and common knowledge such as specific technical solutions and / or characteristics are not described in detail here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the technical solutions of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for analyzing automotive refrigerants based on Fourier transform infrared technology, characterized in that, include: S1. By collecting the infrared absorption spectra of calibration gases with known concentrations, a quantitative relationship model between the light absorption signal response value and concentration of the refrigerant to be tested is constructed. S101. Estimate the concentration range of the refrigerant to be tested in the leakage test chamber, determine the working range of the quantitative model and the concentration gradient of the calibration gas. The working range must include the zero point concentration. The calibration gas is prepared using high-purity nitrogen as the base gas, and the concentration of the calibration gas is within ±2% of the target concentration value. S102. Select an appropriate quantitative band based on the standard infrared absorption spectrum of the refrigerant to be tested, the target concentration, and the interfering components. S103. Perform a gas path system sealing check on the Fourier transform infrared gas analyzer, verify the instrument operating parameters, heat the sampling system and gas cell to the specified working temperature, purge the gas path system with high-purity nitrogen until the spectral signal is stable, and complete the zero-point calibration of the interferometer and background spectrum acquisition. S104. Introduce calibration gas in order from zero concentration to high concentration. After the gas in the gas cell stabilizes, collect infrared absorption spectra of each concentration and record the absorbance signal response value in the quantitative band. Repeat the injection for each concentration 3 to 5 times. The relative deviation of the absorbance signal response value at the same concentration is ≤5%, and the average value is taken as the modeling data. S105. A working curve is to be plotted with the absorbance signal response value as the ordinate and the nominal concentration of the calibrated gas as the abscissa, and a quantitative model is to be established. After the model is established, the accuracy is to be verified. If the verification fails, the model is to be re-modeled. S106. Using the blank sample spiking method, repeatedly inject the zero-point calibration gas more than 10 times, calculate the instrument's limit of quantitation for the refrigerant, and establish a low-concentration calibration curve by injecting more than 5 sets of low-concentration calibration gases, according to the formula: in, The limit of quantitation of the instrument. ; Factors related to confidence level; The standard deviation of the light absorption signal response value for background noise; The slope of the calibration curve within the concentration range; S2. Based on the established quantitative model, conduct real-time online monitoring and analysis of the automotive refrigerant in the leakage test chamber; introduce the gas sample in the leakage test chamber into the Fourier transform infrared gas analyzer through the sampling system, select the quantitative model corresponding to the refrigerant, and use the same operating parameters as the modeling to collect the infrared absorption spectrum in real time and record the sampling time.

2. The method for analyzing automotive refrigerants based on Fourier transform technology according to claim 1, characterized in that, S2 further includes: S201. Before the first test each day, use the pressurization and pressure holding method or the vacuum pressure holding method to check the sealing of the internal and external gas paths of the Fourier transform infrared gas analyzer, verify the instrument operating parameters and ensure that they are consistent with those during modeling, and heat the sampling system and gas cell to the specified operating temperature. S202. Purge the gas path system of the analyzer with high-purity nitrogen until the spectral signal is stable, complete the zero-point calibration of the interferometer and the acquisition of the background spectrum and subtract the blank background, verify the accuracy of the quantitative model, and maintain the optical chamber purging with high-purity nitrogen throughout the process. S203. The gas sample in the leakage test chamber is introduced into the Fourier transform infrared gas analyzer through the sampling system. The temperature of the sampling system changes synchronously with the test chamber and the control accuracy is ±1℃. The quantitative model of the corresponding refrigerant is selected, and the infrared absorption spectrum is collected in real time and the sampling time is recorded using the same operating parameters as the modeling. S204. The absorbance signal response value is converted into refrigerant concentration and output in real time through the quantitative model. The flow rate, temperature and pressure in the gas cell are monitored synchronously. When the results are inconsistent with the modeling steady state, temperature and pressure compensation is performed and the quantitative model is corrected. When wavenumber drift or injection changes affect the results, the quantitative model is corrected or re-established in a timely manner. S205. The analyzed gas sample is returned to the leak test chamber. After the test, the sampling pipeline is purged with high-purity nitrogen. The shutdown operation is completed according to the instruction manual. The relevant test data is recorded and integrated through the supporting software to form a permanent record.

3. The method for analyzing automotive refrigerants based on Fourier transform technology according to claim 2, characterized in that, The analytical method also includes S3 implementation of full-process quality assurance and control of the analytical process: S301. The detector dark current, infrared light source, and background noise generated by electronic thermal noise should be tested at least every six months, and the test should be performed immediately after the instrument is repaired. S302. When the gas to be tested contains known interfering components, correct or subtract the infrared absorption spectrum contribution of the interfering components. S303. The performance parameters and moisture content of the Fourier transform infrared gas analyzer shall be calibrated once a year using standard materials. The zero point and range of the instrument shall be recalibrated before each test. The calibration curve shall be fitted using the least squares method. The number of calibration points shall be determined according to the order of the fitted polynomial. The relative deviation between the fitted value of the calibration point and the nominal value shall be ≤2%. S304. Analyzers that are frequently used should be purged with high-purity nitrogen gas at a flow rate of ≥300mL / min for a long time. Analyzers that are not used for a long time should have a desiccant inside that is within its expiration date and be replaced in time. Before turning on the instrument, remove the desiccant and purge it with high-purity nitrogen gas for more than 1 day. After repairing or replacing important parts of the instrument, return it to the factory for recalibration. S305. Regularly clean the sampling head and sampling pipeline. The sampling system components can be easily replaced and cleaned. Ensure good airtightness at the connection points.

4. The method for analyzing automotive refrigerants based on Fourier transform technology according to claim 3, characterized in that: In step S103, the air circuit system sealing is checked by pressurization and pressure holding method or vacuum pressure holding method; after purging, the interference signal in the single beam spectrum is stable or disappears, the full spectrum light intensity signal remains stable and the spectral shape is normal, and the blank absorption spectrum fluctuates flatly near zero value without obvious absorption peaks.

5. The method for analyzing automotive refrigerants based on Fourier transform technology according to claim 4, characterized in that: In step S105, the relevant parameters of the quantitative model and the gas cell include operating temperature, operating pressure, optical path length, and the spectrometer's resolution, number of scans, apodization function, phase correction, and gain strength. The quantitative model is re-established when the relevant parameters change. The quantitative model and monitoring analysis use the same instrument to acquire absorption spectra. The model validation standard is: when zero-point calibration gas is introduced, the model output concentration is zero; When a known concentration calibration gas that is not involved in the modeling is introduced, the relative deviation between the model output concentration and the nominal concentration is ≤5%.

6. The method for analyzing automotive refrigerants based on Fourier transform technology according to claim 5, characterized in that: The instrument operating parameters mentioned in step S201 include spectral range, spectral resolution, number of scans, aperture, moving mirror scanning frequency, apodization function, phase correction, and gain; In step S204, temperature and pressure compensation are performed using the accompanying software or the ideal gas law.

7. The method for analyzing automotive refrigerants based on Fourier transform technology according to claim 6, characterized in that: In step S303, the calibration curve shall include at least 5 equidistant calibration points, and the nominal value of the highest concentration calibration point shall be ≥ 80% of the working range; If the order of the fitted polynomial is greater than 3, then the number of calibration points shall be greater than or equal to the order of the polynomial plus 2. A calibration curve table shall be drawn based on the polynomial coefficients fitted to the calibration curve. The table step size shall not be greater than 1% of the full scale and shall include the calibration date, zero point, range, calibration gas reference data, and the percentage deviation between the nominal value and the fitted value.

Images