A method and system for predicting the degree of pyrolysis of a material based on trace gas monitoring

Abstract

The application discloses a material pyrolysis degree prediction method and system based on micro gas monitoring, and belongs to the technical field of material pyrolysis monitoring. The method comprises the following steps: obtaining time sequence concentration data of target gas through a micro gas sampling port during in-situ forming of a thermoplastic composite material; constructing a gas feature vector, establishing a pyrolysis index and a chemical damage variable based on a calibration experiment; short-time prediction of the pyrolysis degree within a future preset time; mapping of the pyrolysis degree and mechanical properties and quality determination; result output and process regulation; and adjustment suggestions for interface surface temperature process parameters, which are used as closed-loop control instructions for real-time adjustment of the forming process. The application can solve the technical problems of difficult accurate monitoring of the pyrolysis state, unreliable characterization of chemical damage and lack of online correlation between the pyrolysis degree and mechanical properties in the in-situ forming process of the thermoplastic composite material.

Description

Technical Field

[0001] This invention relates to the field of material pyrolysis monitoring technology, and in particular to a method and system for predicting the degree of material pyrolysis based on trace gas monitoring. Background Technology

[0002] Thermoplastic composites have become important lightweight structural materials in the fields of aerospace, advanced rail transportation and high-end equipment manufacturing due to their advantages such as high specific strength, high specific stiffness, impact resistance, weldability and recyclability.

[0003] In-situ forming technology for thermoplastic composites heats the material surface and substrate to a molten state through energy input, and achieves welding and interlayer fusion under pressure. It is an important direction for the efficient and automated forming of thermoplastic composites.

[0004] However, this type of process presents the following key challenges: (1) Narrow temperature window and short but steep thermal history: The local temperature in the heating zone rises and falls rapidly in milliseconds to seconds, and there is a significant temperature gradient inside the material. Existing infrared thermometry or thermal imaging monitoring can only obtain the surface or near-surface temperature, which is difficult to reflect the true pyrolysis / degradation state inside the material. (2) Pyrolysis / degradation is irreversible and covert: At high temperatures, the resin components of thermoplastic composites undergo pyrolysis, oxidation, or hydrolysis, leading to a decrease in molecular weight, changes in melt viscosity, alterations in crystallinity and crystalline structure, and a reduction in interfacial wetting ability. Ultimately, this results in a decline in interlaminar strength, impact toughness, and fatigue life. This process is often difficult to identify directly through geometric appearance or simple temperature indicators in its early stages.

[0005] (3) Existing process monitoring methods focus on "temperature" and "appearance" but lack "chemical state" indicators: Traditional in-situ forming process monitoring of thermoplastic composites mainly employs thermographic temperature measurement, forming trajectory monitoring, and visual inspection of surface defects. In actual production, situations often arise where "the temperature meets the target but the material has already undergone pyrolysis" or "significant differences in the degree of pyrolysis occur due to differences in residence time and oxygen content at the same peak temperature," indicating that relying solely on the temperature field cannot reliably characterize the degree of chemical damage to the material.

[0006] (4) Lack of online correlation between the degree of pyrolysis and mechanical properties: Current quality assessments mainly rely on subsequent mechanical performance tests or destructive testing, which cannot predict the interlaminar strength and remaining life of components in real time during the forming process. This results in conservative process windows, high rework costs, and untraceable quality. Summary of the Invention

[0007] The purpose of this invention is to provide a method and system for predicting the degree of material pyrolysis based on trace gas monitoring, which solves the technical problems of difficulty in accurately monitoring the pyrolysis state, unreliable characterization of chemical damage, and lack of online correlation between the degree of pyrolysis and mechanical properties during the in-situ forming process of thermoplastic composite materials.

[0008] To achieve the above objectives, the present invention provides a method for predicting the degree of material pyrolysis based on trace gas monitoring, comprising the following steps: Step 1, In-situ Gas Collection and Detection: During the in-situ formation of thermoplastic composite materials, the gas generated in the thermally affected area is sampled through a micro gas sampling port to obtain time-series concentration data of the target gas. Step 2: Gas Feature Extraction and Pyrolysis Index Construction: The time series concentration data of the target gas obtained in step one are subjected to noise reduction, baseline correction and normalization. A gas feature vector is constructed that includes the instantaneous concentration of each target gas, the rate of concentration change and the ratio of each pair of gases in the target gas. Based on the calibration experiment, the pyrolysis index PI and the chemical damage variable D are established. The gas feature vector is mapped into a unified index characterizing the degree of pyrolysis of thermoplastic composite materials. Step 3: Estimation and short-term forecast of pyrolysis degree: The process parameters of the interface surface temperature of thermoplastic composites are collected, the approximate temperature history and equivalent heat dose of the interface region of thermoplastic composites are calculated, and the pyrolysis state is estimated by using temperature-related variables and the gas feature vector obtained in step two. The current pyrolysis index PI and pyrolysis rate are obtained in real time, and the degree of pyrolysis within a preset time is predicted in a short time by machine learning. Step 4: Mapping of Pyrolysis Degree with Mechanical Properties and Quality Determination: Based on a pre-established material database, the obtained D and PI are mapped to key performance indicators such as interlaminar shear strength, interlaminar fracture toughness, porosity risk, and interface fusion quality. The degree of pyrolysis and percentage of remaining properties of thermoplastic composites are obtained, the performance confidence interval at the target forming path position is output, and the quality of the current forming area is divided into different levels according to a preset threshold. Step 5: Result Output and Process Control The degree of pyrolysis, percentage of remaining properties, and quality grade of the thermoplastic composite material obtained in step four are visualized, and adjustment suggestions are provided for the interface surface temperature process parameters. These adjustment suggestions are then used directly as closed-loop control commands to adjust the molding process in real time.

[0009] Preferably, in step one, at least one micro gas sampling port is arranged, and a gas sensor is installed in the micro gas sampling port to obtain time series concentration data of the target gas. The formula for calculating the time-series concentration data of the target gas is as follows: ; in, for Time of the first The actual concentration of the target gas; This represents the volume of the sampled gas. For the detected first The concentration of the target gas; This represents the volume of gas used as a blank control. For the sampling environment The concentration of the target gas.

[0010] Preferably, in step two, the formula for calculating the rate of change of the target gas concentration is as follows: ; in, for Time of the first The rate of change in the concentration of the target gas; for Time of the first The actual concentration of the target gas; for Time of the first The actual concentration of the target gas; The sampling time interval; The formula for calculating the target gas ratio is as follows: ; in, for Time of the first species and first The logarithmic ratio of the concentrations of the target gas; for Time of the first The actual concentration of the target gas; Small positive values ​​are 0.1 × 10⁻⁶. -6 ; The target gas ratio, the instantaneous concentration of each target gas measured by the gas sensor, and the calculated rate of concentration change together constitute the gas feature vector. The calculation formula is as follows: ; in, For their respective target gases at time The normalized instantaneous concentration vector; This represents the vector of the rate of change of the concentration of each target gas; The ratio vector formed by the pairwise ratio characteristics of the target gas This is a material-specific fracture fragment fingerprint feature vector; The number of target gas types, The number of material-specific fragmentation characteristics, where the first... Normalized relative intensity of fingerprint fragments: ; in, For the first Individual fingerprint fragment strength; for The sum of the strengths of each fingerprint fragment; Small positive quantities; This is a review of fingerprint fragment features.

[0011] Preferably, in step two, the formula for calculating the pyrolysis index PI is as follows: ; in, Gas eigenvectors The total feature dimension; The first eigenvector of the gas The eigenvalues ​​of each component after normalization mapping; These are the weighting coefficients obtained through the pyrolysis calibration experiment of thermoplastic composites; This is the weight vector; For feature mapping vectors; ; in, Gas eigenvectors The One component; The slope coefficient; For threshold parameters; The formula for calculating the chemical damage variable D is as follows: =1- ; in As a chemical damage variable, The pyrolysis rate constant of thermoplastic composites, for The pyrolysis index at any given time. for Equivalent heat dose at time, This represents the total time of thermal action. The calibration experiment is used to establish the correspondence between the characteristics of the target gas and the degree of pyrolysis of the thermoplastic composite material, and to determine the weighting coefficients and threshold parameters required for calculating the pyrolysis index PI and the chemical damage variable D.

[0012] Preferably, in step three, the formula for calculating the approximate temperature history is as follows: ; in, for Time, distance from the surface depth of thermoplastic composite material The interface temperature at that location; Ambient temperature; The surface heat flux density provided for the heat source acting on the thermoplastic composite; The thermal conductivity of the material; The thermal diffusivity of the thermoplastic composite material is... ; Density of thermoplastic composite materials; Specific heat capacity of thermoplastic composite materials; ; The formula for calculating equivalent heat dose is as follows: ; in, for The equivalent heat dose at any given time; Melting temperature of thermoplastic composites Preferably, in step three, the temperature-related variables include the interface temperature. Temperature change rate Equivalent heat dose and total heat treatment time ; Based on the interface surface temperature process parameters and temperature-related variables, the approximate temperature history and equivalent heat dose at the interface of the heat source action zone are rapidly calculated as mechanistic inputs for preliminary pyrolysis degree estimation. The quantitative relationship between the formation rate of target gas components and the chemical damage variable D under different temperatures and oxygen partial pressures is described, and the preliminary estimation results are kinetically corrected. Based on the above calculations, the predicted gas values ​​for the pyrolysis process are calculated using computer transfer learning, the systematic deviation between the predicted values ​​and the actual gas eigenvectors is solved, and adaptive corrections are performed for different batches of thermoplastic composite materials. The current pyrolysis index PI and pyrolysis rate are obtained in real time. And by combining the interface surface temperature process parameters with a long short-term memory network LSTM or gated recurrent unit GRU, short-term rolling predictions are made for the pyrolysis index PI and chemical damage variable D within a preset time period. The degree of pyrolysis is quantitatively characterized by the pyrolysis index PI and the chemical damage variable D. The pyrolysis index PI is obtained by normalizing and weighting the gas feature vector and is used to reflect the current pyrolysis state of the thermoplastic composite material in real time. The chemical damage variable D is used to characterize the degree of cumulative irreversible chemical damage caused by pyrolysis.

[0013] Preferably, in step four, the performance mapping relationship between D, PI, and interlaminar shear strength is as follows: ; in, Interlaminar shear strength retention rate; The performance mapping relationship between D, PI and interlaminar fracture toughness is as follows: ; in, This represents the interlaminar fracture toughness retention rate. The performance mapping relationship between D, PI, and porosity risk is as follows: ; in, Porosity risk; The performance mapping relationship between D, PI and interface blending quality is as follows: ; in, Score the interface integration quality; Represents auxiliary correction variables, used to eliminate interference from non-pyrolysis factors of thermoplastic composite materials' processes, environment, and the material itself, and to correct the mapping relationship between PI, D and mechanical properties and quality indicators; Remaining performance percentage .

[0014] Preferably, for a given performance prediction value Its confidence interval is expressed as: ; in, To predict the standard error, The standard normal quantiles corresponding to the confidence level; Based on the range of PI values, process quality is divided into three levels: PI-A: Pyrolysis Safety Zone; PI-B: Pyrolysis boundary region; PI-C: Pyrolysis risk zone.

[0015] Preferably, in step five, the pyrolysis index PI, chemical damage variable D, and remaining performance percentage of the thermoplastic composite material at the current moment are compared with preset thresholds to determine whether the current forming state belongs to the pyrolysis safe zone, pyrolysis boundary zone, or pyrolysis risk zone. Based on the short-term prediction results, it is determined whether there is an upward trend or risk of exceeding the limit in the degree of pyrolysis within the future preset time range. If there is an upward trend or risk of exceeding the limit in pyrolysis, the parameters to be adjusted first and the direction of adjustment are determined according to the sensitivity relationship between the pyrolysis state and the interface surface temperature process parameters. Under the premise of meeting the lower limit of the forming temperature, the interface fusion quality requirements and the equipment constraints, adjustment suggestions for the interface surface temperature process parameters are generated.

[0016] This invention also provides a system for predicting the degree of material pyrolysis based on trace gas monitoring, comprising: The gas sampling module is used to extract gas from the heat-affected area at a constant flow rate through a micro gas sampling port during the in-situ formation of thermoplastic composite materials, and to heat and maintain the temperature of the gas and filter particles. The multi-sensor gas detection module detects the target gas through a gas sensor in a miniature gas sampling port; The process data acquisition module is used to synchronously acquire process parameters such as the interface surface temperature of thermoplastic composite materials. The pyrolysis degree index calculation module is used to preprocess and extract features from the concentration data of each target gas, calculate the gas feature vector, and convert it into an index of the degree of pyrolysis, namely the pyrolysis index PI and the chemical damage variable D. The pyrolysis state estimation and prediction module is used to perform online soft measurement of the current pyrolysis degree of thermoplastic composite materials and to make short-term predictions of the pyrolysis degree within a preset time period in the future. The performance mapping and quality assessment module is used to map the pyrolysis degree index to interlayer mechanical properties, porosity risk and interface quality level, and output the remaining performance percentage and quality level. The process control interface module is used to connect the quality judgment results with the real-time adjustment strategy of the interface surface temperature process parameters to realize closed-loop control guided by the degree of pyrolysis.

[0017] The advantages and positive effects of the material pyrolysis degree prediction method and system based on trace gas monitoring described in this invention are as follows: 1. Achieve precise monitoring of pyrolysis state and solve the problem of identifying hidden pyrolysis: By collecting and detecting trace gases in situ, it breaks through the limitation of monitoring only the surface temperature. It can accurately capture the chemical signals generated by pyrolysis inside the material. Combined with gas feature extraction and pyrolysis index construction, it can effectively identify and quantitatively characterize early hidden pyrolysis and chemical damage. 2. Establish an online correlation between pyrolysis degree and mechanical properties to achieve real-time quality prediction: Based on the material database, construct a mapping relationship between the pyrolysis index PI, chemical damage variable D and key mechanical properties such as interlaminar shear strength and interlaminar fracture toughness. It can output the percentage of remaining material properties and the performance confidence interval in real time during the forming process, replacing the traditional destructive inspection, realizing online prediction of component quality and reducing rework costs; 3. Achieve short-term forecasting of pyrolysis degree and closed-loop control of forming process: By combining temperature-related variables and gas feature vectors, the short-term forecasting of pyrolysis degree is achieved through model building and machine learning. At the same time, process parameter adjustment suggestions are output and converted into closed-loop control commands to adjust the forming process in real time, optimize the process window, and improve the stability and consistency of thermoplastic composite material forming quality. 4. Improve the monitoring system and enhance the level of process intelligence: make up for the lack of chemical state indicators in existing monitoring methods, and build a full-process monitoring and control system that includes gas collection, feature extraction, pyrolysis prediction, performance mapping and process control. This will enable intelligent monitoring, prediction and control of the in-situ forming process of thermoplastic composite materials, and help the high-quality forming of lightweight materials in the field of high-end equipment manufacturing.

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating the steps of an embodiment of the material pyrolysis degree prediction method based on trace gas monitoring according to the present invention. Figure 2 This is a schematic diagram of the micro-gas inlet arrangement in a system embodiment of a material pyrolysis degree prediction method based on trace gas monitoring according to the present invention; Figure 3 This is a schematic diagram of the actual arrangement of micro gas ports in a system embodiment of a material pyrolysis degree prediction method based on trace gas monitoring according to the present invention.

[0020] Figure label: 1. Material tray; 2. Continuous carbon fiber reinforced thermoplastic composite material; 3. Laser; 4. Pressure roller; 5. Miniature gas sampling port; 6. Gas sensor; 7. Gas monitor. Detailed Implementation

[0021] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product is in use. They are used only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," and "connect" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0022] In this application, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. In case of any inconsistency, the meaning set forth in this specification or derived from the content described herein shall prevail. Furthermore, the terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit the scope of this application.

[0023] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0024] Example: like Figure 1 As shown, this invention discloses a method for predicting the degree of material pyrolysis based on trace gas monitoring. This method is applicable to the pyrolysis process of thermoplastic composite materials such as nylon, PP, carbon fiber, reinforced glass fiber, reinforced basalt fiber, and reinforced basalt fiber. The pyrolysis process includes, but is not limited to, laser, ultrasonic, and 3D printing. This embodiment uses continuous carbon fiber laser pyrolysis as an example to specifically illustrate the method of this invention. The method in this embodiment includes the following steps: Step 1, In-situ Gas Collection and Detection: like Figure 2 , Figure 3As shown, during the laser-assisted automatic layup / winding in-situ formation of continuous carbon fiber reinforced thermoplastic composite material 2, at least one micro gas sampling port 5 is arranged behind the layup / winding head (material tray 1 in the figure) and the heating point of laser 3, or before and after the pressure roller 4. The micro gas sampling port 5 extracts and samples trace amounts of gas generated near the laser 3's action area. Gas concentration data is obtained through a gas sensor 6 within the micro gas sampling port 5 and monitored and analyzed by a gas monitor 7 (or other gas detection system). After heating and particle filtration, the gas sampled by the micro gas sampling port 5 is used to obtain time-series concentration data of the target gas through the gas sensor 6 (NDIR infrared CO2 sensor, CO sensor, humidity sensor, and micro mass spectrometer or micro FTIR sensor). The target gas includes CO, CO2, H2O, volatile organic compounds (VOCs), and material-specific pyrolysis fragment fingerprint spectra.

[0025] When the micro gas sampling port 5 draws in the sampled gas, it simultaneously monitors CO, CO2, H2O (water vapor) and volatile organic compounds (VOCs), and constructs at least one or more characteristic ratios among CO / CO2, VOC / CO2, and H2O / VOC to distinguish between volatilization caused by pyrolysis and the release of moisture from materials and environmental background interference.

[0026] A miniature mass spectrometer or miniature FTIR sensor is configured to detect material-specific fingerprint gas components in continuous carbon fiber reinforced thermoplastic composites containing different thermoplastic resins. The time-series concentration data of the target gas are calculated using the following formula: ; in, for Time of the first The actual concentration of the target gas; This represents the volume of the sampled gas. For the detected first The concentration of the target gas; This represents the volume of gas used as a blank control. For the sampling environment The concentration of the target gas.

[0027] During the calculation, the micro gas sampling port 5 collects the detected target gas concentration in real time. The time series concentration data of each target gas is obtained by subtracting the target gas concentration interference in the sampling environment through the calculation formula of the target gas time series concentration data.

[0028] Step 2: Gas Feature Extraction and Pyrolysis Index Construction: Denoise, baseline correction, and normalization are performed on the time-series concentration data of the target gases obtained in Step 1. A gas feature vector is constructed that includes the instantaneous concentration of each target gas, the concentration change rate, and the ratios of CO / CO2, VOC / CO2, and H2O / VOC. Based on the calibration experiment, the pyrolysis index PI and the chemical damage variable D are established, and the gas feature vector is mapped to a unified index representing the degree of pyrolysis of the thermoplastic composite material.

[0029] The denoising process is carried out using a Savitzky-Golay smoothing filter. The window size is set to 11 sampling points (corresponding to 1.1 s, matching the 0.1 s sampling interval), the polynomial order is taken as 2, and the filtering weights are calculated by local least-squares fitting to retain the peak and trend characteristics of the concentration signal while effectively suppressing high-frequency noise interference. Subsequently, baseline correction is performed using the asymmetric least-squares (ALS) smoothing algorithm with a smoothing parameter λ = 10 5 , an asymmetric weight p = 0.001, and the number of iterations is set to 10. The specific operation process is as follows: First, initialize the weight vector w = 1, and perform a weighted least-squares fit on the original concentration signal y and the baseline to be estimated z1 to obtain a preliminary baseline; then update the weight w = p·(y > z1)+(1 - p)·(y < z1) according to the residual; repeat the iteration until the baseline converges or the number of iterations reaches the upper limit; finally, subtract the estimated baseline from the corrected signal to obtain the net concentration data after removing the drift and low-frequency background interference.

[0030] The normalization process uses the min-max normalization method to map the time-series concentration data of each target gas after denoising and baseline correction to the [0, 1] interval. The calculation formula is: ; where, is the actual concentration value of the th target gas at time , and are the minimum and maximum values of this gas in the entire time series, respectively.

[0031] The calculation formula for the concentration change rate of the target gas is as follows: ; where, is the concentration change rate of the th target gas at time (unit: ppm / s); is the actual concentration of the th target gas at time ; is the actual concentration of the th target gas at time The actual concentration of the target gas; The sampling time interval is set to 0.1s to match the sampling frequency.

[0032] During the calculation, the sliding window method was used to perform differential calculation on the time series concentration data of the target gas. The rate of change of the target gas concentration was obtained by using the time series concentration data of the target gas measured before and after two measurements. The sliding window size was set to 5 sampling points (0.5s) to reduce noise interference.

[0033] The formula for calculating the target gas ratio is as follows: ; in, for Time of the first species and first The logarithmic ratio of the concentrations of the target gas; for Time of the first The actual concentration of the target gas; Small positive values ​​are 0.1 × 10⁻⁶. -6 .

[0034] The target gas ratio, the instantaneous concentration of each target gas measured by the gas sensor, and the calculated rate of concentration change together constitute the gas feature vector. The calculation formula is as follows: ; in, For their respective target gases at time The normalized instantaneous concentration vector; This represents the vector of the rate of change of the concentration of each target gas; The ratio vector formed by the pairwise ratio characteristics of the target gas This is a material-specific fracture fragment fingerprint feature vector; The number of target gas types, The number of material-specific fragmentation characteristics, where the first... Normalized relative intensity of fingerprint fragments: ; in, For the first Individual fingerprint fragment strength; for The sum of the strengths of each fingerprint fragment; Small positive quantities; This is a review of fingerprint fragment features.

[0035] The formula for calculating the pyrolysis index PI is as follows: ; in, Gas eigenvectors The total feature dimension; The first eigenvector of the gas The eigenvalues ​​of each component after normalization mapping; These are the weighting coefficients obtained through the pyrolysis calibration experiment of thermoplastic composites; This is the weight vector; This is the feature mapping vector.

[0036] ; in, Gas eigenvectors The One component; The slope coefficient; This is the threshold parameter.

[0037] The formula for calculating the chemical damage variable D is as follows: =1- ; in As a chemical damage variable, The pyrolysis rate constant of thermoplastic composites, for The pyrolysis index at any given time. for Equivalent heat dose at time, This represents the total time of thermal action. The calibration experiment is used to establish the correspondence between the characteristics of the target gas and the degree of pyrolysis of the thermoplastic composite material, and to determine the weighting coefficients and threshold parameters required for calculating the pyrolysis index PI and the chemical damage variable D.

[0038] The calibration experiment is preferably conducted by combining controlled heat exposure test and in-situ process test.

[0039] In controlled heat exposure testing, thermoplastic composite material samples to be processed are selected for heating calibration. These samples are preferably sheet-like, strip-like, or laminated samples, with dimensions determined based on material type, thickness, and heating method. It is preferable to maintain consistency in the sample's material type, reinforcement type, fiber volume fraction, laminate structure, thickness, initial moisture content, and batch composition to minimize interference from non-target factors on the gas signal.

[0040] During the test, a predetermined thermal history is applied to the sample, which includes one or more variables such as peak temperature, heating rate, high-temperature residence time, heat flux density, oxygen content, protective gas flow rate, ambient humidity, and initial moisture content. Preferably, orthogonal experimental design, partial factorial design, or response surface methodology is used to construct multiple sets of thermal exposure conditions to cover the material's non-pyrolysis, slight pyrolysis, moderate pyrolysis, and significant pyrolysis ranges. For each set of conditions, at least three parallel samples are preferably used to improve the repeatability and statistical stability of the calibration results.

[0041] During calibration, time-series concentration data and temperature-related process data of the target gas are collected simultaneously. The target gas includes conventional pyrolysis product gases and material-specific pyrolysis fragment signals. The temperature-related process data includes one or more parameters selected from surface temperature, approximate interface temperature, input power, thermal exposure time, oxygen content, protective gas flow rate, and ambient temperature and humidity. Preferably, the gas and temperature signals are synchronized with a unified timestamp to establish a temporal correspondence between gas release behavior and thermal history.

[0042] The collected target gas time-series concentration data are subjected to noise reduction, baseline correction, and normalization to construct a gas feature vector. The calculation parameters of the pyrolysis index (PI) are then supervised and calibrated by combining one or more reference labels from preset thermal exposure levels, offline chemical structure characterization results, thermal stability characterization results, and mechanical property test results. The offline characterization results include one or more of the following: molecular weight retention rate, characteristic functional group change rate, crystal structure change rate, thermogravimetric characteristic parameters, and mechanical property retention rate.

[0043] Furthermore, based on the changes in chemical structure, thermal stability, and mechanical property degradation of samples under different thermal exposure conditions, a normalized damage reference value is constructed, and the damage evolution parameters of the chemical damage variable D are determined accordingly to establish a mapping relationship between gas characteristics, thermal history parameters, and the degree of chemical damage to the material.

[0044] In the in-situ process test, target gas signals and process parameters are collected simultaneously during the actual in-situ forming process of thermoplastic composite materials. The calibration results obtained from the controlled heat exposure test are corrected and verified to improve the applicability and prediction accuracy of the pyrolysis index PI and chemical damage variable D under real process conditions.

[0045] Step 3: Estimation and short-term forecast of pyrolysis degree: The interface surface temperature process parameters of the continuous carbon fiber reinforced thermoplastic composite 2 are collected, and the approximate temperature history and equivalent heat dose of the interface region of the continuous carbon fiber reinforced thermoplastic composite 2 are calculated. The pyrolysis state is estimated by using the laser power corresponding to temperature and other relevant variables together with the gas feature vector obtained in step two, and the current pyrolysis index PI and pyrolysis rate are obtained in real time. The degree of pyrolysis within a preset time is then predicted in the short term through model building and machine learning.

[0046] The interface surface temperature process parameters include laser power, spot size, laying speed, pressure roller pressure, protective gas flow rate, and surface temperature obtained by infrared thermometry.

[0047] By attaching an appropriate number of K-type thermocouples (fine wire thermocouples with a diameter of less than 0.08 mm are recommended to reduce thermal inertia) to the material surface, and welding or embedding the thermocouples into the material surface, appropriate spacing is maintained before and after the laser action area along the laying path (typical spacing is 5-20 mm, determined according to the laying speed and laser spot size) to record the temperature history data of the material interface region in real time. To ensure the accuracy and reliability of the measurement data, multiple repeated experiments are conducted (no less than 3-5 times), and abnormal fluctuations or unstable measurement points (such as outlier data caused by momentary poor contact or environmental interference) are eliminated. The effective data are then arithmetically averaged to obtain stable interface surface temperature process parameters (including laser power, spot size, laying speed, pressure roller pressure, protective gas flow rate, and surface temperature obtained from infrared thermometry), which serve as inputs for subsequent approximate temperature history and equivalent heat dose calculations. During calculation, the interface surface temperature process parameters measured and averaged by the above multi-point thermocouples are used.

[0048] The approximate temperature history calculation formula is as follows: ; in, For a moment Depth of distance from the surface of thermoplastic composite material Interface temperature at the location (unit: °C); Ambient temperature (unit: °C); Surface heat flux density provided to the heat source (unit: W / m²) 2 (Determined by specific forming process parameters). Thermal conductivity of the material (unit: W / (m·K)); Thermal diffusivity of the material (unit: m) 2 / s), ; Material density (unit: kg / m³) 3 ); Specific heat capacity of the material (unit: J / (kg·℃)); , Represents the integral variable; The heat flux density of the heat source The equivalent heat flux density is the amount of heat generated by any heat source (laser, ultrasound, 3D printing, etc.) acting on the material surface during the in-situ forming process of thermoplastic composites. It can be calculated in real time based on the collected interface surface temperature process parameters. During laser-assisted forming: ( For laser power, (where the light spot radius is...) During ultrasound-assisted molding: It is calculated from parameters such as ultrasonic amplitude, frequency, and pressure roller pressure; When other heat sources are present: It is determined by the corresponding process parameters (power, speed, heat input density, etc.).

[0049] The formula for calculating equivalent heat dose is as follows: ; in, The equivalent heat dose at any given time; This represents the melting temperature of the thermoplastic composite material.

[0050] Temperature-related variables include interface temperature. Temperature change rate Equivalent heat dose and total heat treatment time .

[0051] Based on interface surface temperature process parameters and temperature-related variables, the approximate temperature history and equivalent heat dose at the laser / heat source interface are rapidly calculated as mechanistic inputs for preliminary pyrolysis degree estimation. The quantitative relationship between the target gas component generation rate and the chemical damage variable D under different temperatures and oxygen partial pressures is described, and the preliminary estimation results are kinetically corrected. Based on the above calculations, computer transfer learning (using existing algorithms, trained and fine-tuned with extensive data to improve prediction accuracy) is used to calculate the predicted gas values ​​for the pyrolysis process. The systematic deviation between the predicted values ​​and the actual gas feature vectors is solved, and adaptive corrections are performed for different batches of thermoplastic composite materials. The current pyrolysis index PI and pyrolysis rate are obtained in real time. Furthermore, by combining existing algorithms such as Long Short-Term Memory Network (LSTM) or Gated Recurrent Unit (GRU) with interface surface temperature process parameters, short-term rolling predictions are made for the pyrolysis index PI and chemical damage variable D within a preset time (5~30s).

[0052] The degree of pyrolysis is quantitatively characterized by the pyrolysis index PI and the chemical damage variable D. The pyrolysis index PI is obtained by normalizing and weighting the gas feature vector and is used to reflect the current pyrolysis state of the thermoplastic composite material in real time. The chemical damage variable D is used to characterize the degree of cumulative irreversible chemical damage caused by pyrolysis.

[0053] Step 4: Mapping of Pyrolysis Degree with Mechanical Properties and Quality Determination: Based on a pre-established material database, the obtained D and PI values ​​are mapped to key performance indicators such as interlaminar shear strength, interlaminar fracture toughness, porosity risk, and interface fusion quality, yielding the degree of pyrolysis and the percentage of remaining properties of the thermoplastic composite material. The performance confidence interval at the target forming path location is output, and the quality of the current forming area is classified into different levels according to preset thresholds.

[0054] The materials database is constructed by combining calibration experimental data, in-situ forming experimental data, and offline performance test data. The sample types in the database include one or more of the following: controlled heat-exposed specimens, actual in-situ formed specimens, specimens prepared under different process parameters, and specimens from different material batches and with different layup structures.

[0055] The database sample preferably includes one or more of the following information: material type, resin system, reinforcement type and volume fraction; sample size, laminate structure, layup direction and thickness; process parameters, including energy input intensity, thermally affected area size, relative motion speed, pressure roller pressure, protective gas flow rate and temperature parameters; gas characteristic vector, pyrolysis index PI, chemical damage variable D and pyrolysis rate; interlaminar shear strength, interlaminar fracture toughness, porosity index, interface fusion quality index and quality grade label; environmental parameters, material batch information and test time.

[0056] The performance data are preferably obtained through standardized offline tests, wherein the interlaminar shear strength is preferably obtained through short beam shear tests, the interlaminar fracture toughness is preferably obtained through mode I or mode II fracture toughness tests, the porosity is preferably obtained through microscopic image analysis, ultrasonic testing, density method or X-ray computed tomography, and the interface fusion quality is preferably obtained through cross-sectional morphology analysis, fusion width measurement, interface defect detection and comprehensive evaluation of interlaminar mechanical properties.

[0057] The material database is preferably constructed using a hierarchical structure, including a raw data layer, a feature data layer, a performance label layer, and a model parameter layer. The raw data layer is used to store the original experimental signals, process parameters, and environmental parameters; the feature data layer is used to store gas feature vectors, PI, D, and other intermediate features; the performance label layer is used to store mechanical properties, porosity, interface fusion quality, and quality grade; and the model parameter layer is used to store mapping model parameters, confidence interval parameters, and threshold parameters.

[0058] The update mechanism of the material database includes offline batch updates and online incremental updates. Offline batch updates are used to incorporate newly added calibration experimental data, in-situ process verification data, and offline test results into the database, and periodically retrain or correct the mapping model; online incremental updates are used to continuously write monitoring data, quality test results, and abnormal operating condition records from the actual production process into the database, so as to gradually correct the mapping relationship between different material batches, different equipment states, and different environmental conditions.

[0059] The performance mapping relationship between D, PI and interlaminar shear strength is as follows: ; in, This represents the interlaminar shear strength retention rate.

[0060] Interlaminar shear strength retention rate Defined as: = ; in, The interlaminar shear strength under the current working conditions. This represents the baseline interlaminar shear strength of the corresponding material in its undamaged state.

[0061] The performance mapping relationship between D, PI and interlaminar fracture toughness is as follows: ; in, This represents the interlaminar fracture toughness retention rate.

[0062] The interlaminar fracture toughness retention rate can be defined as: ; in, For the interlaminar fracture toughness under the current working conditions, The interlaminar fracture toughness of the reference layer is undamaged.

[0063] The performance mapping relationship between D, PI, and porosity risk is as follows: ; in, Porosity risk.

[0064] Porosity can be obtained through microscopic image analysis, ultrasonic testing, density method, or X-ray computed tomography. Porosity risk is defined as porosity exceeding a preset threshold. The probability is expressed as: ; Represents porosity; This indicates that the porosity exceeds a preset threshold. The possibility; The preferred interface fusion quality indicators include one or more of the following: fusion width, interface defect area ratio, interlayer continuity index, interface bonding integrity index, or comprehensive quality score. The performance mapping relationship between D, PI, and interface fusion quality is as follows: ; in, For the interface integration quality score, the preferred value range is [0,1]. Represents auxiliary correction variables, used to eliminate interference from non-pyrolysis factors of thermoplastic composite materials' processes, environment, and the material itself, and to correct the mapping relationship between PI, D and mechanical properties and quality indicators.

[0065] Remaining performance percentage .

[0066] The performance confidence interval is used to characterize the range of uncertainty in the performance prediction results for the target region. The performance confidence interval can be calculated based on one or more of the following methods: regression model residual distribution, Bayesian posterior distribution, Bootstrap resampling results, ensemble model output dispersion, or prediction interval model.

[0067] For a certain performance prediction value Its confidence interval is expressed as: ; in, To predict the standard error, This represents the standard normal quantile corresponding to the confidence level.

[0068] The confidence level is preferably set to 90%, 95%, or 99%, more preferably 95%; the setting is based on one or more factors including material quality control requirements, equipment operation stability requirements, the safety level of the target structural component, and the historical quality fluctuation range.

[0069] Furthermore, based on the range of PI values, the process quality is divided into three levels: PI-A (PI approximately equal to 0): Safe window for pyrolysis degree; PI-B (PI between 0.3 and 0.6): Pyrolysis boundary region; PI-C (PI > 0.6): Pyrolysis risk zone.

[0070] The preset threshold preferably includes one or more of the following: pyrolysis index threshold, chemical damage variable threshold, residual performance lower limit threshold, porosity upper limit threshold, and interface fusion quality lower limit threshold. The methods for determining the preset threshold include one or more of the following: determination based on statistical analysis of calibration experimental data, determination based on the minimum allowable value of mechanical properties, determination based on the distribution of historical qualified samples, determination based on the process stability window boundary, or determination based on reliability targets.

[0071] The boundary values ​​of PI and D can be deduced from the performance retention rate threshold. For example, when the interlaminar shear strength retention rate is lower than a preset lower limit, the interlaminar fracture toughness retention rate is lower than a preset lower limit, the porosity is higher than a preset upper limit, or the interface fusion quality score is lower than a preset value, the corresponding PI and D combination region is defined as a risk zone; when the performance is close to the threshold boundary, the corresponding region is defined as a boundary zone; the rest are defined as safe zones. The thresholds are set according to different material systems, thicknesses, layup structures, and application scenarios, and adaptive correction based on newly added sample data is allowed.

[0072] Step 5: Result Output and Process Control The degree of pyrolysis, percentage of remaining properties, and quality grade of the thermoplastic composite material obtained in step four are visualized, and adjustment suggestions are provided for the interface surface temperature process parameters. These adjustment suggestions are then used directly as closed-loop control commands to adjust the molding process in real time.

[0073] The pyrolysis index PI, chemical damage variable D, and remaining performance percentage of the thermoplastic composite material at the current moment are compared with preset thresholds to determine whether the current forming state belongs to the pyrolysis safe zone, pyrolysis boundary zone, or pyrolysis risk zone.

[0074] Based on the short-term prediction results, it is determined whether there is an upward trend or risk of exceeding the limit in the degree of pyrolysis within the future preset time range. If there is an upward trend or risk of exceeding the limit in pyrolysis, the parameters to be adjusted first and the direction of adjustment are determined according to the sensitivity relationship between the pyrolysis state and the interface surface temperature process parameters. Under the premise of meeting the lower limit of the forming temperature, the interface fusion quality requirements and the equipment constraints, adjustment suggestions for the interface surface temperature process parameters are generated.

[0075] The system for predicting the degree of material pyrolysis based on trace gas monitoring, as described in this invention, comprises: The gas sampling module is used to extract gas at a constant flow rate from the area behind the laser heating point of the laser 3 and the boundary layer area before and after the pressure roller 4 near the laser-assisted automatic laying / winding head of the continuous carbon fiber reinforced thermoplastic composite material 2 through the micro gas sampling port 5, and to heat and keep the gas warm and filter particles. The multi-sensor gas detection module detects the target gas through an NDIR infrared CO2 sensor, a CO sensor, a humidity sensor, and a miniature mass spectrometer or a miniature FTIR sensor. The process data acquisition module is used to synchronously acquire process parameters including laser power, spot size, laying speed, pressure roller pressure, protective gas flow rate, and surface temperature obtained by infrared thermometry, as well as interface surface temperature. The pyrolysis degree index calculation module is used to preprocess and extract features from the concentration data of each target gas, calculate the gas feature vector, and convert it into an index of the degree of pyrolysis, namely the pyrolysis index PI and the chemical damage variable D. The pyrolysis state estimation and prediction module is used to perform online soft measurement of the current pyrolysis degree of the continuous carbon fiber reinforced thermoplastic composite material 2, and to make short-term predictions of the pyrolysis degree within a preset time period in the future. The performance mapping and quality assessment module is used to map the pyrolysis degree index to interlayer mechanical properties, porosity risk and interface quality level, and output the remaining performance percentage and quality level. The process control interface module is used to connect the quality judgment results with the real-time adjustment strategy of the interface surface temperature process parameters to realize closed-loop control guided by the degree of pyrolysis.

[0076] The above methods and systems are applicable to the in-situ forming process of continuous carbon fiber reinforced thermoplastic prepreg tapes such as CF / PEEK, CF / PEKK, CF / PPS, and CF / PAEK in laser-assisted automatic lay-up or laser-assisted winding.

[0077] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. 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 still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for predicting the degree of material pyrolysis based on trace gas monitoring, characterized in that, Includes the following steps: Step 1, In-situ Gas Collection and Detection: During the in-situ formation of thermoplastic composite materials, the gas generated in the thermally affected area is sampled through a micro gas sampling port to obtain time-series concentration data of the target gas. Step 2: Gas Feature Extraction and Pyrolysis Index Construction: The time series concentration data of the target gas obtained in step one are subjected to noise reduction, baseline correction and normalization. A gas feature vector is constructed that includes the instantaneous concentration of each target gas, the rate of concentration change and the ratio of each pair of gases in the target gas. Based on the calibration experiment, the pyrolysis index PI and the chemical damage variable D are established. The gas feature vector is mapped into a unified index characterizing the degree of pyrolysis of thermoplastic composite materials. Step 3: Estimation and short-term forecast of pyrolysis degree: The process parameters of the interface surface temperature of thermoplastic composites are collected, the approximate temperature history and equivalent heat dose of the interface region of thermoplastic composites are calculated, and the pyrolysis state is estimated by using temperature-related variables and the gas feature vector obtained in step two. The current pyrolysis index PI and pyrolysis rate are obtained in real time, and the degree of pyrolysis within a preset time is predicted in a short time by machine learning. Step 4: Mapping of Pyrolysis Degree with Mechanical Properties and Quality Determination: Based on a pre-established material database, the obtained D and PI are mapped to key performance indicators such as interlaminar shear strength, interlaminar fracture toughness, porosity risk, and interface fusion quality. The degree of pyrolysis and percentage of remaining properties of thermoplastic composites are obtained, the performance confidence interval at the target forming path position is output, and the quality of the current forming area is divided into different levels according to a preset threshold. Step 5: Result Output and Process Control The degree of pyrolysis, percentage of remaining properties, and quality grade of the thermoplastic composite material obtained in step four are visualized, and adjustment suggestions are provided for the interface surface temperature process parameters. These adjustment suggestions are then used directly as closed-loop control commands to adjust the molding process in real time.

2. The method for predicting the degree of material pyrolysis based on trace gas monitoring according to claim 1, characterized in that: In step one, at least one miniature gas sampling port is arranged, and a gas sensor is installed inside the miniature gas sampling port to obtain time series concentration data of the target gas. The formula for calculating the time-series concentration data of the target gas is as follows: ； in, for Time of the first The actual concentration of the target gas; This represents the volume of the sampled gas. For the detected first The concentration of the target gas; This represents the volume of gas used as a blank control. For the sampling environment The concentration of the target gas.

3. The method for predicting the degree of material pyrolysis based on trace gas monitoring according to claim 2, characterized in that: In step two, the formula for calculating the rate of change of the target gas concentration is as follows: ； in, for Time of the first The rate of change in the concentration of the target gas; for Time of the first The actual concentration of the target gas; for Time of the first The actual concentration of the target gas; The sampling time interval; The formula for calculating the target gas ratio is as follows: ； in, for Time of the first species and first The logarithmic ratio of the concentrations of the target gas; for Time of the first The actual concentration of the target gas; Small positive values ​​are 0.1 × 10⁻⁶. -6 ; The target gas ratio, the instantaneous concentration of each target gas measured by the gas sensor, and the calculated rate of concentration change together constitute the gas feature vector. The calculation formula is as follows: ； in, For their respective target gases at time The normalized instantaneous concentration vector; This represents the vector of the rate of change of the concentration of each target gas; The logarithmic ratio of concentrations formed by the pairwise ratio characteristics of the target gases This is a material-specific fracture fragment fingerprint feature vector; The number of target gas types, The number of material-specific fragmentation characteristics, where the first... Normalized relative intensity of fingerprint fragments: ； in, For the first Individual fingerprint fragment strength; for The sum of the strengths of each fingerprint fragment; Small positive quantities; This is a review of fingerprint fragment features.

4. The method for predicting the degree of material pyrolysis based on trace gas monitoring according to claim 3, characterized in that: In step two, the formula for calculating the pyrolysis index PI is as follows: ； in, Gas eigenvectors The total feature dimension; The first eigenvector of the gas The eigenvalues ​​of each component after normalization mapping; These are the weighting coefficients obtained through the pyrolysis calibration experiment of thermoplastic composites; This is the weight vector; For feature mapping vectors; ； in, Gas eigenvectors The One component; The slope coefficient; For threshold parameters; The formula for calculating the chemical damage variable D is as follows: =1- ； in As a chemical damage variable, The pyrolysis rate constant of thermoplastic composites, for The pyrolysis index at any given time. for Equivalent heat dose at time, This represents the total time of thermal action. The calibration experiment is used to establish the correspondence between the characteristics of the target gas and the degree of pyrolysis of the thermoplastic composite material, and to determine the weighting coefficients and threshold parameters required for calculating the pyrolysis index PI and the chemical damage variable D.

5. The method for predicting the degree of material pyrolysis based on trace gas monitoring according to claim 4, characterized in that: In step three, the formula for calculating the approximate temperature history is as follows: ； in, for Time, distance from the surface depth of thermoplastic composite material The interface temperature at that location; The ambient temperature; The surface heat flux density provided for the heat source acting on the thermoplastic composite; The thermal conductivity of the material; The thermal diffusivity of the thermoplastic composite material is... ; Density of thermoplastic composite materials; Specific heat capacity of thermoplastic composite materials; ; The formula for calculating equivalent heat dose is as follows: ； in, for The equivalent heat dose at any given time; This represents the melting temperature of the thermoplastic composite material.

6. The method for predicting the degree of material pyrolysis based on trace gas monitoring according to claim 5, characterized in that: In step three, temperature-related variables include interface temperature. Temperature change rate Equivalent heat dose and total heat treatment time ; Based on the interface surface temperature process parameters and temperature-related variables, the approximate temperature history and equivalent heat dose at the interface of the heat source action zone are quickly calculated, serving as the mechanism input for preliminary pyrolysis degree estimation; This study describes the quantitative relationship between the formation rate of target gas components and the chemical damage variable D under different temperature and oxygen partial pressure conditions, and performs kinetic correction on the preliminary estimation results. Based on the above calculations, computer transfer learning is used to calculate the predicted gas values ​​for the pyrolysis process, solve for the systematic deviation between the predicted values ​​and the actual gas eigenvectors, and perform adaptive corrections for different batches of thermoplastic composite materials. The current pyrolysis index PI and pyrolysis rate are obtained in real time. ; The pyrolysis index PI and chemical damage variable D within a preset time are predicted in a short time by combining the interface surface temperature process parameters with a long short-term memory network LSTM or gated recurrent unit GRU. The degree of pyrolysis is quantitatively characterized by the pyrolysis index PI and the chemical damage variable D. The pyrolysis index PI is obtained by normalizing and weighting the gas feature vector and is used to reflect the current pyrolysis state of the thermoplastic composite material in real time. The chemical damage variable D is used to characterize the degree of cumulative irreversible chemical damage caused by pyrolysis.

7. The method for predicting the degree of material pyrolysis based on trace gas monitoring according to claim 6, characterized in that: In step four, the performance mapping relationship between D, PI, and interlaminar shear strength is as follows: ； in, Interlaminar shear strength retention rate; The performance mapping relationship between D, PI and interlaminar fracture toughness is as follows: ； in, This represents the interlaminar fracture toughness retention rate. The performance mapping relationship between D, PI, and porosity risk is as follows: ； in, Porosity risk; The performance mapping relationship between D, PI and interface blending quality is as follows: ； in, Score the interface integration quality; Represents auxiliary correction variables, used to eliminate interference from non-pyrolysis factors of thermoplastic composite materials' processes, environment, and the material itself, and to correct the mapping relationship between PI, D and mechanical properties and quality indicators; Remaining performance percentage .

8. The method for predicting the degree of material pyrolysis based on trace gas monitoring according to claim 7, characterized in that: For a certain performance prediction value Its confidence interval is expressed as: ； in, To predict the standard error, The standard normal quantiles corresponding to the confidence level; Based on the range of PI values, process quality is divided into three levels: PI-A: Pyrolysis Safety Zone; PI-B: Pyrolysis boundary region; PI-C: Pyrolysis risk zone.

9. The method for predicting the degree of material pyrolysis based on trace gas monitoring according to claim 8, characterized in that: In step five, the pyrolysis index PI, chemical damage variable D, and remaining performance percentage of the thermoplastic composite material at the current moment are compared with preset thresholds to determine whether the current forming state belongs to the pyrolysis safe zone, pyrolysis boundary zone, or pyrolysis risk zone. Based on the short-term prediction results, it is determined whether there is an upward trend or risk of exceeding the limit in the degree of pyrolysis within the future preset time range. If there is an upward trend or risk of exceeding the limit in pyrolysis, the parameters to be adjusted first and the direction of adjustment are determined according to the sensitivity relationship between the pyrolysis state and the interface surface temperature process parameters. Under the premise of meeting the lower limit of the forming temperature, the interface fusion quality requirements and the equipment constraints, adjustment suggestions for the interface surface temperature process parameters are generated.

10. The system for predicting the degree of material pyrolysis based on trace gas monitoring according to claim 9, characterized in that, include: The gas sampling module is used to extract gas from the heat-affected area at a constant flow rate through a micro gas sampling port during the in-situ formation of thermoplastic composite materials, and to heat and maintain the temperature of the gas and filter particles. The multi-sensor gas detection module detects the target gas through a gas sensor in a miniature gas sampling port; The process data acquisition module is used to synchronously acquire process parameters such as the interface surface temperature of thermoplastic composite materials. The pyrolysis degree index calculation module is used to preprocess and extract features from the concentration data of each target gas, calculate the gas feature vector, and convert it into an index of the degree of pyrolysis, namely the pyrolysis index PI and the chemical damage variable D. The pyrolysis state estimation and prediction module is used to perform online soft measurement of the current pyrolysis degree of thermoplastic composite materials and to make short-term predictions of the pyrolysis degree within a preset time period in the future. The performance mapping and quality assessment module is used to map the pyrolysis degree index to interlayer mechanical properties, porosity risk and interface quality level, and output the remaining performance percentage and quality level. The process control interface module is used to connect the quality judgment results with the real-time adjustment strategy of the interface surface temperature process parameters to realize closed-loop control guided by the degree of pyrolysis.

Images