Method for detecting film thickness and moisture content based on double narrow-band near-infrared transmission

Abstract

This invention belongs to the field of near-infrared spectroscopy detection and signal processing technology, specifically relating to a method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission. This invention can perform thickness or scattering compensation for moisture content detection without prior thickness measurement, adapting to online transmission scenarios with large thickness variations. A single transmission measurement simultaneously outputs thickness and moisture content, improving detection efficiency and reducing alignment and coupling errors between multiple devices. Through dual-wavelength differential elimination of material absorption ratios or thickness parameter compensation, it significantly reduces moisture content drift caused by thickness and scattering variations, improving long-term stability. It provides a reference intensity recalibration mechanism, error source control, and applicable range determination strategies, allowing the method to be bound to specific materials, equipment, and operating conditions, facilitating industrial deployment and maintenance. It can also be combined with anomaly identification and validity label output to improve reliability under complex operating conditions such as conveyor belt wrinkles, edge light leakage, and contamination.

Description

Technical Field

[0001] This invention belongs to the field of near-infrared spectroscopy detection and signal processing technology, specifically relating to a method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission. Background Technology

[0002] Near-infrared spectroscopy (NIIR) offers advantages such as being non-contact, rapid, non-destructive, and suitable for online applications, leading to its widespread use in material composition analysis and moisture detection. For transmission-type NIIR detection, the measured signal is typically the light intensity transmitted through the sample, which is influenced by multiple factors including sample thickness, bulk material absorption, moisture absorption, and microstructure scattering. Particularly for microporous films, coated films, or materials with pronounced surface textures, variations in scattering and transport conditions can cause significant fluctuations in the transmission signal, resulting in superimposed interference with the absorption characteristics of the moisture-sensitive band and leading to inaccuracies in moisture content estimation.

[0003] From a physical mechanism perspective, for a thin film in which the material and water are mixed in a nearly uniform manner, at a wavelength... The transmission intensity at that point can approximately satisfy the Beer-Lambert law form:

[0004] ;

[0005] in, For reference light intensity, The linear absorption coefficient of the material. The water absorption coefficient. For film thickness, For moisture content (mass fraction), absorbance is usually defined logarithmically in engineering implementation to reduce the influence of light source drift on the results. For ease of calibration and differential processing, the absorption coefficient in the following logarithmic expression is redefined by dividing the corresponding coefficient in the above natural exponential formula by ln(10). For simplification, the original symbols are still used, and all are decimal absorbance coefficients:

[0006] ;

[0007] However, in industrial settings, in addition to absorption, the transmitted signal is often affected by scattering loss, interface reflection, and the equivalent optical path change caused by micropores. In engineering, these can be equivalently incorporated into the absorbance expression:

[0008] ;

[0009] in It is the scattering equivalent attenuation coefficient (related to aperture structure / texture, receiving field of view, and alignment state). Because of interface reflection and system baseline terms (which can drift with contamination, wrinkles or changes in incident angle), in transmission detection, thickness changes and scattering / baseline changes will introduce "pseudo-moisture changes" in the moisture-sensitive band. Moisture changes will also affect the absorbance of the thickness channel, resulting in residual errors in the thickness output. Without a decoupling compensation mechanism, the moisture content and thickness output will be unstable, making it difficult to meet the requirements of high-speed and high-precision detection.

[0010] Existing solutions typically include:

[0011] (1) Relying on external thickness measurement (laser displacement, white light interferometry, contact thickness measurement, etc.) to separate the influence of thickness and moisture, but the system cost is high, the assembly and maintenance are complicated and online alignment is difficult;

[0012] (2) The thickness variation range can be reduced by mechanical thickness limiting or clamping structure, but this may introduce contact contamination, tension disturbance or inability to adapt to multiple specifications of films;

[0013] (3) A full-spectrum spectrometer is used and "absorption and scattering are integrated" through multivariate modeling, but the data volume is large, the calculation delay is high, and it is sensitive to model migration and field drift, which is not conducive to high-frequency real-time output and low-maintenance operation.

[0014] Based on this, this solution provides a method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission to address the aforementioned problems. Summary of the Invention

[0015] The purpose of this invention is to provide a method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission, which can realize the simultaneous online detection of two parameters: film thickness and moisture content.

[0016] The specific technical solution adopted by this invention is as follows:

[0017] Methods for detecting film thickness and moisture content based on dual narrowband near-infrared transmission include:

[0018] Two near-infrared narrowband transmission signals with different center wavelengths were simultaneously acquired in the direction of film transport. One band was the reference band with weak moisture absorption, and the other band was the band with strong moisture absorption.

[0019] Dark current subtraction and no-load normalization were performed on the transmission signals of the weak absorption reference band and the strong absorption band of moisture, respectively, to obtain the transmittance, and absorbance characteristics were obtained based on the transmittance conversion.

[0020] The absorbance characteristics of the weak absorption reference band of water are extracted as thickness-related parameters, and the mapping relationship between the thickness-related parameters and the geometric thickness of the film is established through dry film calibration, and the film thickness value is output.

[0021] The absorbance characteristics of the strong absorption band of moisture are extracted, and thickness-related parameters are introduced as compensation variables into the moisture calculation. This is achieved through dual-wavelength difference analysis. eliminate The coupling effect of material thickness variation and scattering interference on moisture absorption characteristics was investigated, and the moisture content of the decoupled output film was determined.

[0022] The output synchronously detected film thickness and moisture content values ​​are combined with the consistency and fluctuation status of the dual-channel signals to identify and mark abnormal operating conditions.

[0023] In a preferred embodiment, the step of acquiring two near-infrared narrowband transmission signals with different center wavelengths includes:

[0024] The system employs dual near-infrared light sources, each passing through a corresponding narrowband filter, to output near-infrared light at the corresponding center wavelength.

[0025] After the light is irradiated onto the moving film to be tested, the transmitted light energy is simultaneously received by the corresponding photodetector on the other side of the film, and the optical signal is converted into an electrical signal for output, thus obtaining near-infrared narrowband transmission signals corresponding to two center wavelengths.

[0026] In a preferred embodiment, the output process of the absorbance characteristic includes:

[0027] Reference light intensity was collected under sample-free air conditions, and dark current signal was collected under shading conditions.

[0028] The measured transmission signal and the dark current signal of the corresponding band are differentially processed to obtain the net transmitted light intensity that eliminates the influence of dark current.

[0029] The reference light intensity and the dark current signal are differentially processed to obtain the net reference light intensity that eliminates the influence of dark current.

[0030] The net transmitted light intensity is compared with the net reference light intensity, and the standardized transmittance of the corresponding band is output.

[0031] By taking the negative of the logarithm of the standardized transmittance, the absorbance characteristics of each wavelength band are obtained.

[0032] In a preferred embodiment, the extraction of absorbance characteristics in the strong absorption band of water further includes, optionally, the construction of water-related features:

[0033] Moisture-related characteristics are directly set to absorbance amplitude in the strong absorption band;

[0034] The difference between the absorbance in the strong absorption band and the absorbance in the reference band is taken as the moisture-related characteristic;

[0035] The ratio of the transmission signals of the weak absorption reference band and the strong absorption band of moisture is taken as the moisture correlation characteristic.

[0036] In a preferred embodiment, the step of establishing a mapping relationship between thickness-related parameters and film geometric thickness through dry film calibration, and outputting the film thickness value, includes:

[0037] Prepare multiple dry film samples of different thicknesses to cover the measurement range, and use a contact thickness gauge to obtain the reference thickness value of each dry film sample.

[0038] The absorbance of the weak absorption reference band of moisture was used as the independent variable and the reference thickness was used as the dependent variable to fit the data and obtain the fitting coefficient.

[0039] During online operation, the absorbance of the reference band is substituted into the fitting relationship to output the film thickness value.

[0040] In a preferred embodiment, the step of outputting the moisture content includes:

[0041] The thickness-related parameters are used as compensation terms. The linear weighted value of the thickness-related parameters is subtracted from the moisture-related features extracted from the strong absorption band of moisture, and the result is output as a moisture feature.

[0042] The moisture characteristics are fitted to the calibrated moisture content to establish a mapping relationship between the thickness-compensated moisture characteristics and the actual moisture content.

[0043] During online operation, the extracted moisture characteristics are input and the corresponding membrane moisture content is directly output.

[0044] In a preferred embodiment, the step of outputting the moisture content further includes:

[0045] Using dry film samples of different thicknesses, linear fitting was performed with absorbance in the strong absorption band of water as the horizontal axis and absorbance in the weak absorption reference band of water as the vertical axis to obtain the ratio of the absorption coefficients of the non-aqueous components of the material in the two bands.

[0046] The ratio of absorption coefficients is multiplied by the absorbance in the strong absorption band to obtain the contribution of the absorbance in the reference band to the absorbance in the strong absorption band.

[0047] The contribution value is subtracted from the absorbance of the reference band to obtain the net absorbance of the corresponding moisture absorption.

[0048] The net absorbed light intensity is mapped to the water surface density through a calibration coefficient, and the water content is obtained by combining the film thickness value with the material density.

[0049] In a preferred embodiment, the output process of the film thickness value and moisture content further includes applicable range determination and iterative decoupling. By setting a preset upper limit for moisture content, when the online measured moisture content exceeds the upper limit, an alarm is output, and the iterative decoupling process is activated. Specifically, this includes:

[0050] The initial thickness value is obtained from the absorbance of the reference wavelength band;

[0051] The surface density of water was obtained from the net absorbed light intensity and then converted to the initial water content.

[0052] The corrected absorbance is obtained by subtracting the moisture contribution from the absorbance in the reference band, and the film thickness value is updated accordingly.

[0053] Repeat the process of recalculating the moisture contribution based on the corrected and updated film thickness value, updating the moisture areal density and water content results, until the film thickness value and water content converge, and output the final film thickness value and water content.

[0054] In a preferred embodiment, the identification and marking of abnormal operating conditions by combining the consistency and fluctuation status of the dual-channel signals specifically includes the following criteria:

[0055] The transmittance of any band exceeds the preset saturation or underrange range.

[0056] The short-time variance and jump amplitude of transmittance exceed the preset threshold.

[0057] The ratio or net absorbance of the two channels deviates from historical statistical characteristics;

[0058] An alarm was triggered when multiple sampling points showed abnormalities.

[0059] Data deemed abnormal is subject to deweighting, removal, or invalidation, and the events are logged for diagnostic maintenance.

[0060] The present invention also provides a thin film thickness and moisture content detection system based on dual narrowband near-infrared transmission, which uses the above-mentioned method for detecting thin film thickness and moisture content based on dual narrowband near-infrared transmission, including:

[0061] The signal classification module is used to simultaneously acquire two near-infrared narrowband transmission signals with different center wavelengths in the film's transport direction. One band is a weak moisture absorption reference band, and the other band is a strong moisture absorption band.

[0062] The feature extraction module is used to perform dark current subtraction and no-load normalization on the transmission signals of the weak absorption reference band and the strong absorption band of water, respectively, to obtain the transmittance, and to obtain the absorbance feature based on the transmittance conversion.

[0063] The thickness output module is used to extract the absorbance characteristics of the weak absorption reference band of moisture as a thickness-related parameter, and to establish the mapping relationship between the thickness-related parameter and the geometric thickness of the film through dry film calibration, and output the film thickness value.

[0064] The moisture content output module is used to extract the absorbance characteristics of the strong absorption band of moisture and introduce thickness-related parameters as compensation variables into the moisture calculation. By eliminating the coupling effect of material thickness variation and scattering interference on moisture absorption characteristics through dual-wavelength difference, the moisture content of the thin film is decoupled and output.

[0065] The operating condition identification module is used to output the film thickness and moisture content values ​​obtained by synchronous detection, and to identify and mark abnormal operating conditions by combining the consistency and fluctuation status of the dual-channel signals.

[0066] The technical effects achieved by this invention are as follows:

[0067] This invention enables thickness or scattering compensation for moisture content detection without prior thickness measurement, making it suitable for online transmission scenarios with large thickness variations. A single transmission measurement simultaneously outputs thickness and moisture content (or water surface density), improving detection efficiency and reducing alignment and coupling errors between multiple devices. By eliminating moisture content drift caused by thickness and scattering variations through dual-wavelength differential elimination of material absorption ratios or thickness parameter compensation, it significantly reduces moisture content drift caused by thickness and scattering changes, improving long-term stability. It provides a reference light intensity recalibration mechanism, error source control, and applicable range determination strategies, allowing the method to be bound to specific materials, equipment, and operating conditions, facilitating industrial deployment and maintenance. Combined with anomaly identification and validity label output, it improves reliability under complex operating conditions such as conveyor belt wrinkles, edge light leakage, and contamination. Attached Figure Description

[0068] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0069] Figure 2 This is a schematic diagram of the system modules of the present invention. Detailed Implementation

[0070] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0071] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0072] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in a preferred embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that mutually excludes other embodiments.

[0073] Please see Figure 1 As shown, this invention provides a method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission, comprising:

[0074] S1. Simultaneously acquire two near-infrared narrowband transmission signals with different center wavelengths in the film's transport direction. One band is the reference band for weak moisture absorption, and the other band is the band for strong moisture absorption.

[0075] In step S1, near-infrared spectroscopy has advantages such as being non-contact, rapid, non-destructive, and suitable for online applications. It is widely used in material composition analysis and moisture detection. In practical applications, the transmission signals of the thin film are acquired under at least two near-infrared narrowbands with different center wavelengths. Preferably, the transmission signals of the weakly absorbed moisture reference band and the strongly absorbed moisture band are acquired synchronously with a parallel InGaAs detector using parallel narrowband filters to reduce timing errors caused by belt jitter. The sampling frequency is preferably ≥1000Hz to meet the requirements of high-speed online detection. The acquired signals may include: measurement signals. , and the reference light intensity updated periodically under sample-free conditions. , The step of acquiring two near-infrared narrowband transmission signals with different center wavelengths includes:

[0076] Dual near-infrared light sources are used to transmit light through corresponding narrowband filters and output near-infrared light with corresponding center wavelengths. The center wavelengths of the narrowband filters are set to 1720nm and 1940nm respectively. The center wavelength of the reference band is about 1720nm and the center wavelength of the moisture-sensitive band is about 1940nm, so as to match the absorption characteristics of moisture in the corresponding bands.

[0077] After the light is irradiated onto the moving film to be tested, the transmitted light energy is simultaneously received by the corresponding photodetector on the other side of the film. The light signal is converted into an electrical signal and output, resulting in near-infrared narrowband transmission signals corresponding to two center wavelengths. Specifically, the two photodetectors correspond to a near-infrared light source and a narrowband filter, respectively. The detection positions are kept aligned along the film's travel direction to ensure that the two transmission measurements are performed on the same position of the film, avoiding additional measurement errors caused by positional deviations. The original electrical signals output by the detectors are preprocessed by an amplification and filtering circuit and converted into digital signals for subsequent feature extraction.

[0078] S2. Dark current subtraction and no-load normalization are performed on the transmission signals of the weak absorption reference band and the strong absorption band of moisture, respectively, to obtain the transmittance, and absorbance characteristics are obtained based on the transmittance conversion.

[0079] In step S2, the original electrical signal output by the detector contains a dark current component (i.e., the thermal noise output of the detector itself under no-light conditions). If this component is not subtracted, it will introduce a significant proportional error in low-transmittance (high-absorbance) scenarios. Simultaneously, the light source intensity drifts slowly with aging, temperature, and usage time. If absorbance is calculated directly from the original signal, this light source drift will be misinterpreted as a change in sample absorption. Therefore, it is necessary to first eliminate the detector zero-point offset through dark current subtraction, then eliminate the influence of the absolute value of the light source intensity through no-load normalization (i.e., dividing by the reference light intensity under no-sample air conditions), and finally convert the multiplicative error into an additive error through logarithmic transformation, facilitating subsequent linear fitting and difference elimination. The output process of the absorbance characteristic includes:

[0080] The measurement signals were collected during normal operation of the thin film. The reference light intensity was collected under the condition of no sample air, and the dark current signal was collected under the condition of shading.

[0081] The measured transmission signal and the dark current signal in the corresponding band are differentially processed to obtain the net transmitted light intensity after eliminating the influence of dark current, i.e.:

[0082] ;

[0083] In the formula, Indicates the effect of eliminating dark current. Net transmitted light intensity in the band express Measured transmission signal in the band, express Dark current signal in the band, Choose 1720nm or 1940nm;

[0084] By performing differential processing on the reference light intensity and the dark current signal, the net reference light intensity, which eliminates the influence of dark current, is obtained, i.e.:

[0085] ;

[0086] In the formula, Indicates the effect of eliminating dark current. Band net reference light intensity, express Reference light intensity in the band under sample-free conditions;

[0087] The net transmitted light intensity is compared with the net reference light intensity, and the normalized transmittance for the corresponding wavelength band is output, i.e.:

[0088] ;

[0089] In the formula, express The normalized transmittance of the band ranges from (0,1).

[0090] Taking the negative of the logarithm of the normalized transmittance yields the absorbance characteristics for each wavelength band, i.e.:

[0091] ;

[0092] In the formula, express Absorbance characteristics of the band, For the substrate material of the thin film to be tested in Absorption coefficient of the band, Let be the geometric thickness of the film to be tested. For moisture in Absorption coefficient of the band, The water content of the film to be tested is given by the formula. and These refer to the net transmitted light intensity and net reference light intensity after subtracting the dark signal, respectively. Therefore, absorbance can also be written as... .

[0093] S3. Extract the absorbance characteristics of the weak absorption reference band of water as a thickness-related parameter, and establish the mapping relationship between the thickness-related parameter and the geometric thickness of the film through dry film calibration, and output the film thickness value.

[0094] In step S3, since the absorption of moisture in the weak absorption reference band is relatively weak, its absorbance characteristics are mainly affected by the film's geometric thickness, bulk absorption, and scattering attenuation. Therefore, it can be used as a thickness-related parameter to characterize the film's thickness information. Dry film calibration can establish a mapping relationship between this thickness-related parameter and the film's geometric thickness, thereby outputting an accurate thickness value during online operation. The step of establishing the mapping relationship between the thickness-related parameter and the film's geometric thickness through dry film calibration and outputting the film thickness value includes:

[0095] Prepare multiple dry film samples of different thicknesses to cover the measurement range, and use a contact thickness gauge to obtain the reference thickness value of each dry film sample.

[0096] The absorbance of the weak absorption reference band of moisture was used as the independent variable and the reference thickness was used as the dependent variable to fit the data and obtain the fitting coefficient.

[0097] During online operation, the absorbance of the reference band is substituted into the fitting relationship to output the film thickness value;

[0098] Specifically, the moisture content of the dry film sample should be lower than the preset upper limit. To avoid moisture absorption introducing errors into thickness calibration, at least eight calibration points are used, covering the entire measurement range. Linear least squares fitting is employed, with a correlation coefficient R² preferably greater than 0.999, and the residual plot showing no systematic trend. During online operation, the thickness output formula is:

[0099] ;

[0100] in, , These are the fitting coefficients. The units should be consistent with the calibration (e.g., μm or mm). The 1720nm channel may still contain moisture residuals; therefore, in high-moisture scenarios, the thickness can be corrected using subsequent iterative decoupling processes. It should be noted that the absorption coefficient of the reference band (1720nm) for moisture is extremely low (about 1 / 50 of that of the 1940nm band). In dry film calibration, even if there is a trace amount of moisture, its impact on the absorbance of the 1720nm band is much smaller than the impact of the material's bulk absorption and scattering. By limiting the upper limit of the moisture content, it can be regarded as a systematic error and absorbed in the fitting intercept.

[0101] S4. Extract the absorbance characteristics of the strong absorption band of moisture and introduce the thickness-related parameters as compensation variables into the moisture calculation. Eliminate the coupling effect of material thickness variation and scattering interference on moisture absorption characteristics through dual-wavelength difference, and decouple the output of the film moisture content.

[0102] In step S4, the strong absorption band of moisture (e.g., 1940nm) has strong absorption characteristics for moisture. However, the absorbance of this band is simultaneously affected by the coupling effects of film thickness variation, material bulk absorption, and scattering interference. Furthermore, since the two near-infrared narrowband wavelengths are close (1720nm and 1940nm), the scattering attenuation coefficients of the film microstructure for both are approximately equal. In subsequent dual-wavelength difference or ratio calculations, the scattering term, as common-mode interference, is largely canceled out, and no significant error is introduced. To accurately decouple the output moisture content, thickness-related parameters need to be introduced as compensation variables into the moisture calculation. The coupling effects of thickness variation and scattering interference are deducted through dual-wavelength difference elimination or thickness compensation. The moisture content output step includes:

[0103] First, the thickness-related parameters are used as compensation terms. The linear weighted value of the thickness-related parameters is subtracted from the moisture-related features extracted from the strong absorption band of moisture, and the result is output as a moisture feature.

[0104] The moisture characteristics are fitted to the calibrated moisture content to establish a mapping relationship between the thickness-compensated moisture characteristics and the actual moisture content.

[0105] During online operation, the extracted moisture characteristics are directly output as the corresponding membrane moisture content;

[0106] Specifically, extracting absorbance characteristics of the strong absorption band of moisture also includes optional forms of constructing moisture-related features: the moisture-related feature is directly set as the absorbance amplitude of the strong absorption band, the difference between the absorbance of the strong absorption band and the absorbance of the reference band is taken as the moisture-related feature, and the ratio of the transmission signal of the weak absorption reference band and the strong absorption band of moisture is taken as the moisture-related feature. The amplitude, difference, and ratio are optional implementations in parallel, rather than having to be executed simultaneously. For example, method one is suitable for scenarios with high moisture content and stable thickness, method two (dual-wavelength difference) is suitable for scenarios with large thickness variations, and method three is suitable for scenarios with strong scattering interference. Dual-wavelength difference can cancel common-mode interference caused by thickness and scattering because the effects of thickness and scattering on the two bands are approximately proportional, while moisture absorption is specific. Thickness-related parameters Moisture is introduced as a compensation item in moisture calculation:

[0107] ;

[0108] in, or Determined by calibration samples, and then The output is mapped to moisture content or water surface density. This method is suitable for scenarios where the material elimination conditions are not fully met, but thickness interference can be approximately linearly compensated. The weighting coefficients in the above difference calculation Compensation function The scaling factor in the ratio calculation is pre-calibrated in the following way: Prepare at least three sets of standard thin film samples with known thickness and known moisture content. Substitute the measured transmission signals into the corresponding formulas and use the least squares method to fit the parameter values ​​that minimize the error between the calculated moisture content and the actual moisture content.

[0109] Secondly, using dry film samples of different thicknesses, linear fitting was performed with the absorbance of the strong absorption band of water as the horizontal axis and the absorbance of the weak absorption reference band of water as the vertical axis to obtain the ratio of the absorption coefficients of the non-aqueous components of the material in the two bands.

[0110] The ratio of absorption coefficients is multiplied by the absorbance in the strong absorption band to obtain the contribution of the absorbance in the reference band to the absorbance in the strong absorption band.

[0111] The contribution value is subtracted from the absorbance of the reference band to obtain the net absorbance of the corresponding moisture absorption.

[0112] The net absorbed light intensity is mapped to the water surface density through a calibration coefficient, and the water content is obtained by combining the film thickness value with the material density.

[0113] Specifically, construct the difference component (this difference component is the net absorbance):

[0114] ;

[0115] in, The physical meaning of is the ratio of the absorption coefficients of the material (non-aqueous) in the two wavelength bands, which is obtained through dry film calibration. Since the two wavelengths are close, the scattering terms are canceled out in the difference calculation. This primarily reflects the absorption characteristics of the material itself. When the sample contains moisture, the absorbance at 1940 nm increases significantly, while the increase at 1720 nm is minimal, leading to... The difference is no longer 0. It exhibits a linear relationship with water surface density, thus enabling the extraction of water characteristics:

[0116] ;

[0117] Under the condition that the material formulation is stable and satisfies approximately linear additivity... This can be considered a material constant, independent of thickness and moisture content, and can be calibrated by selecting samples with different dry film thicknesses:

[0118] by For the horizontal axis, For a linear fit along the vertical axis, its slope is... The fitting intercept should be close to 0. If it deviates, it indicates that the consistency of the baseline or reference light intensity needs to be corrected.

[0119] After the material item is eliminated, the difference component Mainly reflects the moisture content:

[0120] ;

[0121] To obtain the output water content of the project, this invention preferably outputs the water surface density (water mass per unit area) W_{areal}, for example:

[0122] ;

[0123] in , The result was obtained by fitting a standard moisture-containing sample, and the calculation process is as follows: Standard water-bearing samples were selected, and the water surface density of each sample was obtained using an independent reference method. ; Under the same measurement conditions, absorbance was obtained in two bands, and the results were calculated using the original differential formula. To form paired data ( , The target quantity for fitting is the water surface density, not the percentage water content. .

[0124] If the reference data is the moisture content of the sample wet basis Wet film quality and area The corresponding water surface density is calculated using the following formula, where Substitute into dimensionless decimal form: ; Under ordinary least squares linear fitting, a and b can be obtained by the following formula:

[0125] ; in , These are the mean values ​​of each calibration data. Each summation is performed on all calibration samples, and the calibration difference components should have sufficient variation so that the denominator is not zero.

[0126] If you need to output the moisture content It can be combined with material density With thickness Perform the conversion:

[0127] ; in, The effective bulk density of the wet film under test conditions.

[0128] During this process, it is important to pay attention to unit consistency (conversion between μm and mm).

[0129] Thirdly, the output process for film thickness and moisture content also includes applicable range determination and iterative decoupling. By setting a preset upper limit for moisture content, when the online measured moisture content exceeds the upper limit, an alarm is output and the iterative decoupling process is activated, specifically including:

[0130] The initial thickness value is obtained from the absorbance of the reference wavelength band;

[0131] The surface density of water was obtained from the net absorbed light intensity and then converted to the initial water content.

[0132] The corrected absorbance is obtained by subtracting the moisture contribution from the absorbance in the reference band, and the film thickness value is updated accordingly.

[0133] Repeat the process of recalculating the moisture contribution based on the corrected and updated film thickness value, updating the moisture areal density and water content results, until the film thickness value and water content converge, and output the final film thickness value and water content.

[0134] Specifically, will As input variables, establish a moisture content mapping model:

[0135] or ;

[0136] Moisture content mapping models can be linear regression, partial least squares, piecewise fitting or other equivalent mapping methods, which can improve robustness when there are large fluctuations in material batches or changes in operating conditions, but require matching calibration data management and drift monitoring strategies.

[0137] Furthermore, this invention provides a mechanism for determining the scope of application:

[0138] Calculated moisture content With preset upper limit In comparison, when The system outputs alarms in real time and enables an iterative decoupling process to ensure that thickness and moisture content are output simultaneously.

[0139] During online operation, if the initial thickness value has residuals due to trace amounts of moisture, the moisture contribution can be subtracted from the absorbance of the reference band through an iterative decoupling process, which can successively approximate the true thickness, thereby eliminating the residual influence of moisture content on thickness calibration.

[0140] S5. Output the film thickness and moisture content values ​​obtained from synchronous detection, and identify and mark abnormal working conditions by combining the consistency and fluctuation status of the dual-channel signals.

[0141] In step S5, the strong absorption band of moisture (e.g., 1940nm) exhibits strong absorption characteristics for moisture. However, the absorbance in this band is simultaneously affected by the coupling effects of film thickness variation, material bulk absorption, and scattering interference. To accurately decouple the output moisture content, thickness-related parameters need to be introduced as compensation variables into the moisture calculation. The coupling effects of thickness variation and scattering interference are deducted through dual-wavelength differential elimination or thickness compensation, thereby obtaining characteristics that only reflect moisture absorption. Furthermore, abnormal operating conditions are identified and marked by combining the consistency and fluctuation status of the dual-channel signals, specifically including the following criteria:

[0142] The transmittance of any band exceeds the preset saturation or underrange range.

[0143] The short-time variance and jump amplitude of transmittance exceed the preset threshold.

[0144] The ratio or net absorbance of the two channels deviates from historical statistical characteristics;

[0145] An alarm was triggered when multiple sampling points showed abnormalities.

[0146] Data deemed abnormal is subject to deweighting, removal, or invalidation, and the events are logged for diagnostic maintenance.

[0147] Specifically, to further improve the stability of online operation, an applicable scope determination mechanism can also be set up:

[0148] The calculated moisture content is compared with a preset upper limit. When c_w > c{w,max}, an alarm is output, and an iterative decoupling process is initiated. Iterative decoupling obtains the initial thickness value from the absorbance of the reference band, obtains the water surface density from the net absorbance, and converts it to the initial moisture content. The moisture contribution is subtracted from the absorbance of the reference band to obtain the corrected absorbance, which updates the film thickness value. This process is repeated to recalculate the moisture contribution from the corrected and updated film thickness value, updating the water surface density and moisture content results, until the film thickness and moisture content converge (e.g., ...). The final film thickness and moisture content are output. Additionally, the reference light intensity can be periodically... Recalibrate or recalibrate when the drift exceeds the threshold to suppress system errors caused by light source aging and temperature drift.

[0149] Please see Figure 2 A film thickness and moisture content detection system based on dual narrowband near-infrared transmission, using the aforementioned method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission, includes:

[0150] The signal classification module is used to simultaneously acquire two near-infrared narrowband transmission signals with different center wavelengths in the film's transport direction. One band is a weak moisture absorption reference band, and the other band is a strong moisture absorption band.

[0151] The feature extraction module is used to perform dark current subtraction and no-load normalization on the transmission signals of the weak absorption reference band and the strong absorption band of water, respectively, to obtain the transmittance, and to obtain the absorbance feature based on the transmittance conversion.

[0152] The thickness output module is used to extract the absorbance characteristics of the weak absorption reference band of moisture as a thickness-related parameter, and to establish the mapping relationship between the thickness-related parameter and the geometric thickness of the film through dry film calibration, and output the film thickness value.

[0153] The moisture content output module is used to extract the absorbance characteristics of the strong absorption band of moisture and introduce thickness-related parameters as compensation variables into the moisture calculation. By eliminating the coupling effect of material thickness variation and scattering interference on moisture absorption characteristics through dual-wavelength difference, the moisture content of the thin film is decoupled and output.

[0154] The operating condition identification module is used to output the film thickness and moisture content values ​​obtained by synchronous detection, and to identify and mark abnormal operating conditions by combining the consistency and fluctuation status of the dual-channel signals.

[0155] The execution process of the above detection system corresponds to the aforementioned method flow, and will not be repeated here.

[0156] The above description is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained in this invention are implemented according to conventional methods in the art unless otherwise specified or limited.

Claims

1. A method for detecting film thickness and water content based on dual narrowband near-infrared transmission, characterized in that: include: Two near-infrared narrowband transmission signals with different center wavelengths were simultaneously acquired in the direction of film transport. One band was the reference band with weak moisture absorption, and the other band was the band with strong moisture absorption. Dark current subtraction and no-load normalization were performed on the transmission signals of the weak absorption reference band and the strong absorption band of moisture, respectively, to obtain the transmittance, and absorbance characteristics were obtained based on the transmittance conversion. The absorbance characteristics of the weak absorption reference band of water are extracted as thickness-related parameters, and the mapping relationship between the thickness-related parameters and the geometric thickness of the film is established through dry film calibration, and the film thickness value is output. The absorbance characteristics of the strong absorption band of moisture are extracted, and thickness-related parameters are introduced into the moisture calculation as compensation variables. The coupling effect of material thickness variation and scattering interference on moisture absorption characteristics is eliminated by dual-wavelength difference, and the moisture content of the thin film is decoupled and output. The output synchronously detected film thickness and moisture content values ​​are combined with the consistency and fluctuation status of the dual-channel signals to identify and mark abnormal operating conditions.

2. The method for film thickness and moisture content detection based on dual-narrowband near-infrared transmission according to claim 1, characterized in that: The step of acquiring two near-infrared narrowband transmission signals with different center wavelengths includes: The system employs dual near-infrared light sources, each passing through a corresponding narrowband filter, to output near-infrared light at the corresponding center wavelength. After the light is irradiated onto the moving film to be tested, the transmitted light energy is simultaneously received by the corresponding photodetector on the other side of the film, and the optical signal is converted into an electrical signal for output, thus obtaining near-infrared narrowband transmission signals corresponding to two center wavelengths.

3. The method for film thickness and moisture content detection based on dual-narrowband near-infrared transmission according to claim 1, characterized in that: The output process of the absorbance feature includes: Reference light intensity was collected under sample-free air conditions, and dark current signal was collected under shading conditions. The measured transmission signal and the dark current signal of the corresponding band are differentially processed to obtain the net transmitted light intensity that eliminates the influence of dark current. The reference light intensity and the dark current signal are differentially processed to obtain the net reference light intensity that eliminates the influence of dark current. The net transmitted light intensity is compared with the net reference light intensity, and the standardized transmittance of the corresponding band is output. By taking the negative of the logarithm of the standardized transmittance, the absorbance characteristics of each wavelength band are obtained.

4. The method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission according to claim 1, characterized in that: The extraction of absorbance characteristics in the strong absorption band of water also includes optional forms of constructing water-related features: Moisture-related characteristics are directly set to absorbance amplitude in the strong absorption band; The difference between the absorbance in the strong absorption band and the absorbance in the reference band is taken as the moisture-related characteristic; The ratio of the transmission signals of the weak absorption reference band and the strong absorption band of moisture is taken as the moisture correlation characteristic.

5. The method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission according to claim 1, characterized in that: The step of establishing the mapping relationship between thickness-related parameters and film geometric thickness through dry film calibration, and outputting the film thickness value, includes: Prepare multiple dry film samples of different thicknesses to cover the measurement range, and use a contact thickness gauge to obtain the reference thickness value of each dry film sample. The absorbance of the weak absorption reference band of moisture was used as the independent variable and the reference thickness was used as the dependent variable to fit the data and obtain the fitting coefficient. During online operation, the absorbance of the reference band is substituted into the fitting relationship to output the film thickness value.

6. The method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission according to claim 1, characterized in that: The step of outputting the moisture content includes: The thickness-related parameters are used as compensation terms. The linear weighted value of the thickness-related parameters is subtracted from the moisture-related features extracted from the strong absorption band of moisture, and the result is output as a moisture feature. The moisture characteristics are fitted to the calibrated moisture content to establish a mapping relationship between the thickness-compensated moisture characteristics and the actual moisture content. During online operation, the extracted moisture characteristics are input and the corresponding membrane moisture content is directly output.

7. The method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission according to claim 1, characterized in that: The step of outputting the moisture content also includes: Using dry film samples of different thicknesses, linear fitting was performed with absorbance in the strong absorption band of water as the horizontal axis and absorbance in the weak absorption reference band of water as the vertical axis to obtain the ratio of the absorption coefficients of the non-aqueous components of the material in the two bands. The ratio of absorption coefficients is multiplied by the absorbance in the strong absorption band to obtain the contribution of the absorbance in the reference band to the absorbance in the strong absorption band. The contribution value is subtracted from the absorbance of the reference band to obtain the net absorbance of the corresponding moisture absorption. The net absorbed light intensity is mapped to the water surface density through a calibration coefficient, and the water content is obtained by combining the film thickness value with the material density.

8. The method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission according to claim 1, characterized in that: The output process of the film thickness value and moisture content also includes applicable range determination and iterative decoupling. By setting a preset upper limit value for moisture content, when the online measured moisture content exceeds the upper limit value, an alarm is output and the iterative decoupling process is activated, specifically including: The initial thickness value is obtained from the absorbance of the reference wavelength band; The surface density of water was obtained from the net absorbed light intensity and then converted to the initial water content. The corrected absorbance is obtained by subtracting the moisture contribution from the absorbance in the reference band, and the film thickness value is updated accordingly. Repeat the process of recalculating the moisture contribution based on the corrected and updated film thickness value, updating the moisture areal density and water content results, until the film thickness value and water content converge, and output the final film thickness value and water content.

9. The method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission according to claim 1, characterized in that: The method of identifying and marking abnormal operating conditions by combining the consistency and fluctuation status of dual-channel signals specifically includes the following criteria: The transmittance of any band exceeds the preset saturation or underrange range. The short-time variance and jump amplitude of transmittance exceed the preset threshold. The ratio or net absorbance of the two channels deviates from historical statistical characteristics; An alarm was triggered when multiple sampling points showed abnormalities. Data deemed abnormal is subject to deweighting, removal, or invalidation, and the events are logged for diagnostic maintenance.

10. A thin film thickness and moisture content detection system based on dual narrowband near-infrared transmission, characterized in that: The method for detecting film thickness and moisture content based on dual narrowband near-infrared transmission according to any one of claims 1 to 9 includes: The signal classification module is used to simultaneously acquire two near-infrared narrowband transmission signals with different center wavelengths in the film's transport direction. One band is a weak moisture absorption reference band, and the other band is a strong moisture absorption band. The feature extraction module is used to perform dark current subtraction and no-load normalization on the transmission signals of the weak absorption reference band and the strong absorption band of water, respectively, to obtain the transmittance, and to obtain the absorbance feature based on the transmittance conversion. The thickness output module is used to extract the absorbance characteristics of the weak absorption reference band of moisture as a thickness-related parameter, and to establish the mapping relationship between the thickness-related parameter and the geometric thickness of the film through dry film calibration, and output the film thickness value. The moisture content output module is used to extract the absorbance characteristics of the strong absorption band of moisture and introduce thickness-related parameters as compensation variables into the moisture calculation. By eliminating the coupling effect of material thickness variation and scattering interference on moisture absorption characteristics through dual-wavelength difference, the moisture content of the thin film is decoupled and output. The operating condition identification module is used to output the film thickness and moisture content values ​​obtained by synchronous detection, and to identify and mark abnormal operating conditions by combining the consistency and fluctuation status of the dual-channel signals.

Images