A device and method for determining the process condition of waterproof membrane rolls

Abstract

This invention relates to the field of waterproof material quality testing technology, specifically to a device and method for determining the process status of waterproof membranes. By simultaneously acquiring surface height sequences measured by laser and contact, and determining the surface fabric texture signal and fabric structure compression fluctuation signal, the relative displacement of the melt fabric is first quantified using the delay correlation between the two, thereby directly identifying macroscopic process deviations caused by velocity mismatch. Then, through translation correction and linear gain modeling, the nonlinear morphological residual signal is extracted, and the physical morphology of defects is identified based on its amplitude distribution morphology characteristics, effectively distinguishing between local abrupt changes caused by unplasticized particles and systemic smooth defects caused by excessive casting. Combining the high-frequency texture attenuation rate and linear correlation determined by the frequency domain transfer function, the microscopic reproducibility of the melt is quantified to reflect its rheological characteristics, thus expanding single thickness detection to multi-dimensional feature decoupling diagnosis and improving the accuracy of waterproof membrane process status determination.

Description

Technical Field

[0001] This invention relates to the field of waterproof material quality testing technology, specifically to a device and method for determining the process status of waterproof membrane rolls. Background Technology

[0002] Composite waterproof membranes typically consist of a polymer layer and a fabric reinforcement layer. Their final thickness is a key quality parameter determining their performance and durability. Therefore, current technologies for determining the process condition of waterproof membranes usually employ contact thickness measurement methods, comparing the measured scalar thickness value with the standard process thickness to screen for membranes with abnormal process conditions. However, this method of monitoring a single thickness result suffers from serious information aliasing. The final thickness deviation may be caused by various drastically different process fluctuations: for example, speed mismatch between the extruder and calender rolls can lead to interlayer stretching and thinning; excessively high melt temperature can cause excessive thinning during casting; and unplasticized particles in the raw material can cause localized thickening due to bulging. Since these defects with completely different physical mechanisms all manifest as "thickness value out of tolerance" on contact thickness gauges, operators cannot distinguish the specific source of the fault based solely on the thickness reading. This results in poor accuracy in fault attribution for waterproof membrane process condition determination using current technologies based on a single thickness result. Summary of the Invention

[0003] To address the problem of poor fault attribution accuracy in existing technologies that rely on single thickness results for determining the process condition of waterproof membranes, this application aims to provide a device and method for determining the process condition of waterproof membranes. The specific technical solution adopted is as follows:

[0004] The first aspect of this application provides a method for determining the process condition of waterproof membrane, including:

[0005] The laser-measured surface height sequence and the contact-measured surface height sequence of the waterproof membrane are synchronously acquired at fixed sampling spatial intervals; based on the laser-measured surface height sequence and its relative deviation from the contact-measured surface height sequence, the surface fabric texture signal and the fabric structure compression fluctuation signal are determined.

[0006] The relative displacement of the melt fabric is determined based on the delay correlation between the surface fabric texture signal and the fabric structure compression fluctuation signal; the fabric structure compression fluctuation signal is translated and corrected based on the relative displacement of the melt fabric to determine the corrected compression fluctuation signal; and the fabric texture reproduction gain is determined based on the signal deviation between the surface fabric texture signal and the corrected compression fluctuation signal.

[0007] Based on the deviation between the surface fabric texture signal and the corrected compressed fluctuation signal after gain weighting through the fabric texture reproduction, the nonlinear morphology residual signal is determined; based on the amplitude distribution pattern of the nonlinear morphology residual signal, the residual distribution pattern characteristic value is determined; based on the change of the frequency domain transfer function between the surface fabric texture signal and the corrected compressed fluctuation signal, the corresponding high-frequency texture attenuation rate and texture reproduction linear correlation are determined.

[0008] The process status of the waterproof membrane was determined based on the relative displacement of the melt fabric, the characteristic value of the residual distribution morphology, the high-frequency texture attenuation rate, and the linear correlation of texture reproduction.

[0009] Further, the process of determining the surface fabric texture signal and the fabric structure compression undulation signal based on the laser-measured surface height sequence and its relative deviation from the contact-measured surface height sequence includes:

[0010] The laser-measured surface height sequence is input into a digital bandpass filter with a preset spatial frequency range for bandpass filtering to obtain the surface fabric texture signal.

[0011] Calculate the surface height difference between the laser-measured surface height sequence and the contact-measured surface height sequence at the same spatial location; arrange the surface height differences at all spatial locations in the spatial order of the laser-measured surface height sequence to determine the surface height difference sequence;

[0012] The surface height difference sequence is input into a digital bandpass filter within a preset spatial frequency range for bandpass filtering to obtain the fabric structure compression undulation signal.

[0013] Furthermore, the process of obtaining the relative displacement of the melt fabric includes:

[0014] The maximum number of delayed sampling points is determined by rounding up the ratio between the preset maximum drag displacement and the sampling space interval. The cross-correlation function value of the surface fabric texture signal and the fabric structure compression undulation signal at each spatial delay index value is calculated using the cross-correlation function. The spatial delay index value is the offset of the number of sampling points of the fabric structure compression undulation signal relative to the surface fabric texture signal. The range of the spatial delay index value is all integers between the negative of the maximum number of delayed sampling points and the maximum number of delayed sampling points.

[0015] The relative displacement of the melt fabric is determined by multiplying the spatial delay index value corresponding to the largest cross-correlation function value with the sampling spatial interval.

[0016] Furthermore, the process of obtaining the fabric texture reproduction gain includes:

[0017] Construct an objective function; the objective function value is the sum of squared residuals between the surface fabric texture signal and the corrected compressed fluctuation signal weighted by undetermined coefficients at all signal index values; based on the least squares method, the solution of the undetermined coefficients when the objective function value is minimized is taken as the fabric texture reproduction gain.

[0018] Furthermore, the process of acquiring the nonlinear topographic residual signal includes:

[0019] The modified compression fluctuation signal is weighted by the fabric texture reproduction gain to determine the weighted compression fluctuation signal;

[0020] The nonlinear morphology residual signal is determined by subtracting the weighted compression fluctuation signal from the surface fabric texture signal.

[0021] Furthermore, the process of obtaining the residual distribution morphological feature values ​​includes:

[0022] Calculate the kurtosis of the nonlinear morphological residual signal to determine the residual distribution morphological characteristic value.

[0023] Furthermore, the process of determining the corresponding high-frequency texture attenuation rate and texture reproduction linear correlation based on the change in the frequency domain transfer function between the surface fabric texture signal and the corrected compression fluctuation signal includes:

[0024] The Welch method was used to calculate the auto-power spectral density of the surface fabric texture signal, the auto-power spectral density of the corrected compression fluctuation signal, and the cross-power spectral density between the surface fabric texture signal and the corrected compression fluctuation signal at each frequency.

[0025] Based on the ratio between the self-power spectral density and the corresponding cross-power spectral density of the surface fabric texture signal at each frequency, the empirical transfer coefficient corresponding to each frequency is determined; the negative of the linear regression slope of the modulus of the empirical transfer coefficients corresponding to all frequencies within the preset spatial frequency range is calculated to determine the corresponding high-frequency texture attenuation rate.

[0026] The signal coherence is calculated based on the auto-power spectral density of the surface fabric texture signal at each frequency, the auto-power spectral density of the modified compression fluctuation signal, and the cross-power spectral density between the surface fabric texture signal and the modified compression fluctuation signal. The local coherence eigenvalues ​​corresponding to each frequency are then determined. The mean of the local coherence eigenvalues ​​corresponding to all frequencies within the preset spatial frequency range is calculated to determine the linear correlation of texture reproduction.

[0027] Furthermore, the process of determining the process status of the waterproof membrane based on the relative displacement of the melt fabric, the fabric texture reproduction gain, the residual distribution morphology characteristic value, the high-frequency texture attenuation rate, and the linear correlation of texture reproduction includes:

[0028] When the residual distribution morphology characteristic value is greater than the preset first distribution threshold, and the absolute value of the relative displacement of the melt fabric is greater than the preset prior relative displacement threshold, the process classification status label of the waterproof membrane is set to local material accumulation caused by dragging.

[0029] When the residual distribution morphology feature value is less than or equal to the preset second distribution threshold, and the high-frequency texture attenuation rate is greater than the preset prior attenuation rate threshold, the process status classification label of the waterproof membrane is set to melt over-leveling.

[0030] When the residual distribution morphological characteristic value is less than or equal to a preset first distribution threshold and greater than a preset second distribution threshold: if the linear correlation of texture reproduction is less than or equal to a preset prior correlation threshold, then the process state classification label of the waterproof membrane is set to unstable composite process; if the linear correlation of texture reproduction is greater than a preset prior correlation threshold, the relative displacement of melt fabric is less than or equal to a preset prior relative displacement threshold, and the high-frequency texture attenuation rate is less than or equal to a preset prior attenuation rate threshold, then the process state classification label of the waterproof membrane is set to ideal process; wherein, the preset first distribution threshold is greater than the preset second distribution threshold, and both the preset first distribution threshold and the preset second distribution threshold are greater than 0.

[0031] Furthermore, the determination of the waterproof membrane process status based on the relative displacement of the melt fabric, the residual distribution morphology characteristic value, the high-frequency texture attenuation rate, and the linear correlation of texture reproduction also includes:

[0032] Waterproof membranes that fail to meet all four conditions—localized material accumulation, excessive melt leveling, unstable composite process, and ideal process—are categorized as having an abnormal process status pending review.

[0033] Secondly, this application provides a device for determining the process condition of waterproof membrane, the device comprising:

[0034] The data acquisition and preprocessing module is used to synchronously acquire the laser-measured surface height sequence and the contact-measured surface height sequence of the waterproof membrane at a fixed sampling spatial interval; based on the laser-measured surface height sequence and its relative deviation from the contact-measured surface height sequence, the surface fabric texture signal and the fabric structure compression fluctuation signal are determined.

[0035] The first determining module is used to determine the relative displacement of the melt fabric based on the delay correlation between the surface fabric texture signal and the fabric structure compression fluctuation signal; to perform translation correction on the fabric structure compression fluctuation signal based on the relative displacement of the melt fabric to determine the corrected compression fluctuation signal; and to determine the fabric texture reproduction gain based on the signal deviation between the surface fabric texture signal and the corrected compression fluctuation signal.

[0036] The second determining module is used to determine the nonlinear morphology residual signal based on the deviation between the surface fabric texture signal and the corrected compressed fluctuation signal after gain weighting through the fabric texture reproduction; determine the residual distribution morphology characteristic value based on the amplitude distribution pattern of the nonlinear morphology residual signal; and determine the corresponding high-frequency texture attenuation rate and texture reproduction linear correlation based on the change of the frequency domain transfer function between the surface fabric texture signal and the corrected compressed fluctuation signal.

[0037] The process status determination module is used to determine the process status of the waterproof membrane based on the relative displacement of the melt fabric, the residual distribution morphology characteristic value, the high-frequency texture attenuation rate, and the linear correlation of texture reproduction.

[0038] Thirdly, this application provides a computer device including a memory and a processor. The memory is used to store computer program code, and the processor is used to call and run the computer program code from the memory to perform the method as described in the first aspect of this application or any embodiment of the first aspect.

[0039] Fourthly, this application provides a computer program product comprising computer program code, which, when executed, performs the method as described in the first aspect of this application or any embodiment thereof.

[0040] Fifthly, this application provides a computer-readable storage medium that stores computer program code, which, when executed, performs the method as described in the first aspect of this application or any embodiment thereof.

[0041] This application has the following beneficial effects:

[0042] This application simultaneously acquires surface height sequences measured by laser and contact, and determines surface fabric texture signals and fabric structure compression fluctuation signals. First, it quantifies the relative displacement of the melt fabric using the delay correlation between the two, thereby directly identifying macroscopic process deviations caused by velocity mismatch. Then, it extracts nonlinear morphology residual signals through translation correction and linear gain modeling, and identifies the physical morphology of defects based on their amplitude distribution morphology characteristics, effectively distinguishing between local abrupt changes caused by unplasticized particles and system smooth defects caused by excessive casting. At the same time, it combines the high-frequency texture attenuation rate and linear correlation determined by the frequency domain transfer function to quantify the microscopic reproducibility of the melt to reflect its rheological characteristics, thereby expanding single thickness detection into multi-dimensional feature decoupling diagnosis, making the fault attribution accuracy based on the determination of waterproof membrane process status higher. Attached Figure Description

[0043] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 This is a flowchart of a method for determining the process status of waterproof membrane according to an embodiment of the present invention;

[0045] Figure 2 This is a structural diagram of a waterproof membrane process condition measuring device provided in one embodiment of the present invention;

[0046] Figure 3 This is a schematic diagram of a computer device structure provided in one embodiment of the present invention. Detailed Implementation

[0047] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a waterproof membrane process state measuring device and method proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment, and specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of that feature.

[0048] 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 invention pertains.

[0049] The following description, in conjunction with the accompanying drawings, details the specific scheme of the waterproof membrane process condition determination device and determination method provided by the present invention.

[0050] This application provides a method for determining the process condition of waterproof membrane. Please refer to [link / reference]. Figure 1 The diagram illustrates a flowchart of a method for determining the process condition of a waterproof membrane according to an embodiment of the present invention. The method includes:

[0051] Step S101: Simultaneously acquire the laser-measured surface height sequence and the contact-measured surface height sequence of the waterproof membrane at a fixed sampling spatial interval; determine the surface fabric texture signal and the fabric structure compression fluctuation signal based on the laser-measured surface height sequence and its relative deviation from the contact-measured surface height sequence.

[0052] First, a dual-modal probe device is provided, integrating a laser displacement sensor and a contact displacement sensor. Both sensors are mounted on the same rigid mechanical structure to ensure high spatial consistency of measurement points during the scanning process. Then, the dual-modal probe device is driven to reciprocate along a guide rail perpendicular to the roll material's forward direction at a preset constant speed, traversing the entire width of the roll material. Simultaneously, after scanning each row of the waterproof roll material, the longitudinal conveying motion of the waterproof roll material on the production line aligns the scanned spatial positions of the next row with those of the previous row, ensuring that the distance between adjacent scanned spatial positions is equal to the sampling interval, resulting in a standard matrix distribution of all scanned spatial positions on the waterproof roll material. Scanning stops after the dual-modal probe's scanning trajectory covers all areas of the roll material surface, yielding the sequence of surface heights obtained from the laser displacement sensor and the contact displacement sensor.

[0053] During the scanning process, two sensors simultaneously record their output signals at a fixed sampling frequency. In this embodiment, the sampling frequency is set to 100Hz, and the preset constant speed is set to 50mm / s, which can be adjusted according to the specific implementation environment. Based on the principle of kinematic discrete sampling, the sampling spatial interval is equal to the ratio between the preset constant speed and the sampling frequency, with the unit being millimeters. That is, after the dual-modal probe device scans the spatial position of each row of waterproof membrane, the length of the longitudinal displacement of the waterproof membrane on the production line is the sampling spatial interval. It should be noted that, to form a more stable spatial position distribution, the dual-modal probe device stops immediately after scanning the spatial position of each row of waterproof membrane. After the longitudinal displacement sampling spatial interval of the waterproof membrane on the production line, the dual-modal probe device continues to reciprocate along a direction perpendicular to the membrane's forward movement until the dual-modal probe scanning trajectory covers all areas of the membrane surface.

[0054] The surface height sequences obtained directly from the sensors, measured by laser and by contact, contain both microscopic fluctuations generated by the internal fabric structure that need to be analyzed, and macroscopic deformations of the roll material caused by process fluctuations during production that need to be filtered out, as well as high-frequency noise from the sensors themselves. Therefore, to reduce the interference of irrelevant signal components on process condition diagnosis, it is necessary to filter out the macroscopic deformation trend terms and high-frequency noise interference terms contained in the laser-measured and contact-measured surface height sequences. Thus, this embodiment of the invention further determines, based on the laser-measured surface height sequence and its relative deviation from the contact-measured surface height sequence, a surface fabric texture signal and a fabric structure compression fluctuation signal containing only microscopic texture feature information.

[0055] Preferably, in some possible implementations of the present invention, the process of determining the surface fabric texture signal and the fabric structure compression undulation signal based on the laser-measured surface height sequence and its relative deviation from the contact-measured surface height sequence includes: inputting the laser-measured surface height sequence into a digital bandpass filter within a preset spatial frequency range for bandpass filtering to obtain the surface fabric texture signal; calculating the surface height difference between the laser-measured surface height sequence and the contact-measured surface height sequence at the same spatial position; arranging the surface height differences at all spatial positions in the spatial position order of the laser-measured surface height sequence to determine the surface height difference sequence; and inputting the surface height difference sequence into a digital bandpass filter within a preset spatial frequency range for bandpass filtering to obtain the fabric structure compression undulation signal.

[0056] In one specific implementation of this invention, the preset spatial frequency range is set to 0.5 cycles / mm to 2 cycles / mm. This is because the typical spacing between warp and weft yarns in a fabric is between 0.5 mm and 2 mm, and the implementer can adjust it according to the specific implementation environment. Based on the principle of a bandpass filter, this filter only allows signal components within the preset spatial frequency range to pass through, thereby effectively filtering out low-frequency components of macroscopic deformation below the lower limit of the range and high-frequency noise components of the sensor above the upper limit of the range. Therefore, the obtained surface fabric texture signal can characterize the morphological features of the waterproof membrane surface that are only related to the microscopic undulations of the fabric texture. The laser-measured surface height sequence is data obtained through non-contact measurement using a laser displacement sensor, which characterizes the surface profile of the waterproof membrane in its natural state. The contact-measured surface height sequence is data obtained through contact measurement using a contact displacement sensor, which characterizes the surface profile of the membrane under pressure. Therefore, based on this characteristic, the obtained fabric structure compression undulation signal actually characterizes the local compressibility changes of the waterproof membrane caused by the difference in the supporting effect of the internal fabric structure, reflecting the spatial distribution information of the internal fabric skeleton. It should be noted that the digital bandpass filter can be either a Butterworth filter or a Chebyshev filter, both of which are well known to those skilled in the art, and will not be further limited or elaborated here.

[0057] Step S102: Determine the relative displacement of the melt fabric based on the delay correlation between the surface fabric texture signal and the fabric structure compression fluctuation signal; perform translation correction on the fabric structure compression fluctuation signal based on the relative displacement of the melt fabric to determine the corrected compression fluctuation signal; determine the fabric texture reproduction gain based on the signal deviation between the surface fabric texture signal and the corrected compression fluctuation signal.

[0058] After obtaining surface fabric texture signals and fabric structure compression undulation signals that can characterize the surface micro-morphology and internal skeleton structure respectively, considering that in actual composite processes, the slight mismatch between the extruder melt supply speed and the calender roll linear speed is the primary macroscopic factor causing process abnormalities, and this mismatch directly causes tangential slippage of the polymer melt layer relative to the fabric substrate layer, resulting in spatial misalignment between the two; therefore, in order to eliminate this macroscopic process error at its source and provide an accurate data basis for subsequent micro-morphology analysis based on position alignment, this step precisely quantifies the relative displacement of the melt fabric by analyzing the spatial delay correlation characteristics between the two signals, thereby effectively identifying speed mismatch faults while eliminating the spatial phase difference between signals, ensuring that subsequent decoupling analysis of nonlinear morphology defects and rheological properties is not disturbed by macroscopic slippage, and improving the hierarchy and accuracy of process fault attribution; the embodiments of the present invention further determine the relative displacement of the melt fabric based on the delay correlation changes between the surface fabric texture signal and the fabric structure compression undulation signal.

[0059] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the relative displacement of the melt fabric includes:

[0060] The maximum number of delayed sampling points is determined by rounding up the ratio between the preset maximum drag displacement and the sampling space interval. The cross-correlation function value of the surface fabric texture signal and the fabric structure compression undulation signal at each spatial delay index value is calculated using a cross-correlation function. The spatial delay index value is the offset of the number of sampling points of the fabric structure compression undulation signal relative to the surface fabric texture signal, and the range of the spatial delay index value is all integers between the negative of the maximum number of delayed sampling points and the maximum number of delayed sampling points. The relative displacement of the melt fabric is determined by multiplying the spatial delay index value corresponding to the maximum cross-correlation function value with the sampling space interval. It should be noted that in this embodiment of the invention, the spatial delay index value is defined as the spatial lag of the fabric structure compression undulation signal relative to the surface fabric texture signal, i.e., the number of lag sampling space positions.

[0061] In one specific implementation of this invention, the maximum drag displacement is preset to 5 mm, which can be adjusted according to the specific implementation environment. The larger the cross-correlation function value, the higher the waveform matching degree between the surface fabric texture signal and the fabric structure compression undulation signal under the spatial delay, that is, the best spatial overlap between the surface microtexture and the internal fabric skeleton. Therefore, the spatial delay index value corresponding to the maximum cross-correlation function value corresponds to the spatial offset required for the two signals to reach the best alignment state. The obtained relative displacement of the melt fabric characterizes the actual macroscopic tangential slip distance of the polymer melt layer relative to the fabric substrate layer during the production process.

[0062] After quantifying the macroscopic slip distance of the polymer melt layer relative to the fabric substrate layer through cross-correlation analysis, given that subsequent microscopic defect analysis (nonlinear morphology residual signal) and rheological property evaluation (high-frequency texture attenuation rate and texture reproduction linear correlation) both strictly depend on the precise point-to-point correspondence between the surface signal and the reference signal in space, if the positional misalignment caused by this slip is not eliminated, the surface texture and the fabric skeleton will not be aligned in spatial coordinates, which will lead to deviations or even complete failure in the linear reproduction gain calculation. Therefore, in order to eliminate the interference of macroscopic velocity mismatch on microscopic morphology analysis and construct a spatially aligned reference signal, this embodiment of the invention performs translation correction on the fabric structure compression fluctuation signal based on the relative displacement of the melt fabric, determines the corrected compression fluctuation signal, and corrects its spatial phase while retaining the amplitude information of the internal skeleton structure, ensuring that subsequent steps can accurately strip nonlinear defects and quantify the linear reproduction capability of the melt in a spatially aligned coordinate system.

[0063] In one specific implementation of this invention, the process of obtaining the modified compression fluctuation signal includes: when the relative displacement of the melt fabric is positive, the fabric structure compression fluctuation signal needs to be translated in the negative direction of the horizontal axis of the coordinate system corresponding to the signal; when the relative displacement of the melt fabric is negative, the fabric structure compression fluctuation signal needs to be translated in the positive direction of the horizontal axis of the coordinate system corresponding to the signal, and the translation distance is the absolute value of the relative displacement of the melt fabric.

[0064] After completing the spatial translation correction of the fabric structure compression fluctuation signal and strictly aligning the surface microtexture with the internal fabric skeleton in space, considering that in an ideal composite process, the surface morphology should be a linear mapping of the internal skeleton structure under the melt coverage, and its amplitude relationship reflects the melt's ability to fill and reproduce the skeleton; however, in actual processes, the rheological properties of the melt, such as viscosity differences, can cause an overall change in this reproduction degree. For example, excessive casting can cause the surface undulations to be smaller than the overall internal skeleton. Therefore, in order to quantify the overall response intensity of the melt layer to the internal fabric structure and establish a linear benchmark model for subsequent separation of nonlinear defects, this embodiment of the invention further uses the principle of linear regression model to solve for the optimal linear proportionality coefficient, i.e., the fabric texture reproduction gain, based on the signal deviation between the surface fabric texture signal and the corrected compression fluctuation signal. This decouples the complex morphology analysis into two dimensions: "linear reproduction degree" and "nonlinear residual," providing a key parameter basis for accurately diagnosing the melt rheological state and separating local abnormal defects.

[0065] Preferably, in some possible implementations of the embodiments of the present invention, the process of obtaining the fabric texture reproduction gain includes:

[0066] An objective function is constructed; the objective function value is the sum of squared residuals at all signal index values ​​between the surface fabric texture signal and the corrected compressed fluctuation signal weighted by undetermined coefficients. The solution of the undetermined coefficients that minimizes the objective function value is used as the fabric texture reproduction gain. Minimizing the objective function value indicates that the linear model's fit to the surface fabric texture signal is optimal; that is, the undetermined coefficients can explain the components linearly transmitted from the internal fabric structure in the surface texture to the greatest extent. Therefore, the obtained fabric texture reproduction gain corresponds to the overall linear transmission rate of the polymer melt layer to the internal fabric texture fluctuations. Its magnitude directly reflects the melt's ability to adhere to and reproduce the substrate skeleton under current process conditions, and is a key quantitative indicator for judging whether the melt has undergone excessive leveling or insufficient filling.

[0067] Step S103: Determine the nonlinear morphology residual signal based on the deviation between the surface fabric texture signal and the corrected compressed fluctuation signal after gain weighting through fabric texture reproduction; determine the residual distribution morphology characteristic value based on the amplitude distribution pattern of the nonlinear morphology residual signal; determine the corresponding high-frequency texture attenuation rate and texture reproduction linear correlation based on the change of the frequency domain transfer function between the surface fabric texture signal and the corrected compressed fluctuation signal.

[0068] After determining the fabric texture reproduction gain, which characterizes the overall reproducibility of the melt, in order to further separate abnormal fluctuations from the mixed surface morphology that cannot be explained by the normal fabric structure reproduction, i.e., to identify nonlinear defects generated during the production process, this embodiment of the invention constructs a benchmark prediction signal containing only ideal linear reproduction components using the fabric texture reproduction gain and the corrected compression fluctuation signal, and separates it from the real surface fabric texture signal. The nonlinear morphology residual signal obtained by this process eliminates reasonable fluctuations caused by the support of the normal fabric skeleton, thereby isolating nonlinear deviations containing process anomalies such as particle lifting, bubbles, or leveling cutoff. This provides a pure analytical object for subsequent accurate identification of the physical type of defects through morphological analysis, significantly improving the accuracy of measuring micro-process defects.

[0069] In this embodiment of the invention, the nonlinear morphology residual signal is determined based on the deviation between the surface fabric texture signal and the corrected compression fluctuation signal after weighting by the fabric texture reproduction gain. Specifically, the corrected compression fluctuation signal is weighted by the fabric texture reproduction gain to determine the weighted compression fluctuation signal; the weighted compression fluctuation signal is subtracted from the surface fabric texture signal to determine the nonlinear morphology residual signal.

[0070] After successfully extracting the nonlinear morphological residual signal containing only nonlinear process deviations, in order to further identify the specific physical mechanisms causing these deviations, that is, to distinguish whether the deviations are dominated by local spikes caused by isolated hard events such as unplasticized particles and bubbles, or by systematic rheological events such as top truncation or smoothing caused by excessively low melt viscosity, this step introduces statistical morphological analysis methods. By analyzing the amplitude distribution pattern of the residual signal, the characteristic values ​​of the residual distribution pattern are determined, which transforms the thickness fluctuations that are difficult to distinguish intuitively into clear distribution pattern indicators. This enables accurate classification of defects with different physical causes and provides core qualitative basis for subsequent process state logic diagnosis.

[0071] In one specific implementation of this invention, the kurtosis of the nonlinear topographic residual signal is calculated to determine the morphological characteristic value of the residual distribution. It should be noted that the calculation of kurtosis is a technique well-known to those skilled in the art, and will not be further limited or elaborated upon here. According to the definition of kurtosis, the greater the kurtosis of the nonlinear morphological residual signal, that is, the larger the residual distribution morphological characteristic value, the sharper the peak and the thicker the tail of the probability density distribution curve of the nonlinear morphological residual signal. This is more consistent with the nonlinear defect mode dominated by a small number of large-amplitude abnormal mutations, such as hard particles. The smaller the kurtosis of the nonlinear morphological residual signal, that is, the smaller the residual distribution morphological characteristic value, the flatter the probability density distribution curve of the nonlinear morphological residual signal. This is more consistent with the nonlinear defect mode dominated by a wide range of truncated or smoothed fluctuations, such as texture clipping caused by excessive casting. Therefore, based on the magnitude of the obtained nonlinear morphological residual signal kurtosis value, it is possible to directly determine whether the current process defect originates from local foreign matter or a systemic rheological anomaly, thus providing a decisive morphological criterion for the accurate location of the fault source.

[0072] After completing the defect analysis in the temporal and morphological dimensions, in order to further explore the differences in the response of melt rheological properties at different spatial scales, especially to quantify the melt's ability to reproduce fine textures, i.e., high-frequency components, and the stability of its linear transfer, this invention further decouples two key frequency domain features—high-frequency texture attenuation rate and texture reproduction linear correlation—by analyzing the frequency domain transfer function between the surface fabric texture signal and the corrected compression fluctuation signal. The high-frequency texture attenuation rate directly reflects the smoothing effect of melt viscosity on microstructures, while the linear correlation evaluates the explanatory power of the linear model across the entire frequency band. These two features supplement the frequency correlation information that is difficult to capture in the temporal domain analysis, thereby enabling a more sensitive and comprehensive diagnosis of melt rheological states such as viscosity decrease caused by excessively high temperatures. Therefore, this embodiment of the invention further determines the corresponding high-frequency texture attenuation rate and texture reproduction linear correlation based on the changes in the frequency domain transfer function between the surface fabric texture signal and the corrected compression fluctuation signal.

[0073] Preferably, in some possible implementations of the present invention, the process of determining the corresponding high-frequency texture attenuation rate and texture reproduction linear correlation based on the change in the frequency domain transfer function between the surface fabric texture signal and the corrected compression fluctuation signal includes:

[0074] The Welch method is used to calculate the auto-power spectral density of the surface fabric texture signal, the auto-power spectral density of the corrected compression fluctuation signal, and the cross-power spectral density between the surface fabric texture signal and the corrected compression fluctuation signal at each frequency. Based on the ratio between the auto-power spectral density of the surface fabric texture signal and the corresponding cross-power spectral density at each frequency, the empirical transfer coefficient corresponding to each frequency is determined. The negative of the linear regression slope of the modulus of the empirical transfer coefficients corresponding to all frequencies within the preset spatial frequency range is calculated to determine the corresponding high-frequency texture attenuation rate.

[0075] Among them, the empirical transfer coefficient characterizes the gain characteristics of the system at different frequencies, and the corresponding linear regression slope, i.e. the high-frequency texture attenuation rate, quantifies the trend of the gain decreasing with increasing frequency. The larger the negative value of the slope, i.e. the larger the high-frequency texture attenuation rate, the stronger the smoothing effect of the melt on the high-frequency fine texture, which corresponds to the reduction of melt viscosity or excessive casting.

[0076] The signal coherence is calculated based on the auto-power spectral density of the surface fabric texture signal at each frequency, the auto-power spectral density of the modified compression fluctuation signal, and the cross-power spectral density between the surface fabric texture signal and the modified compression fluctuation signal. The local coherence eigenvalues ​​corresponding to each frequency are then determined. The mean of the local coherence eigenvalues ​​corresponding to all frequencies within the preset spatial frequency range is calculated to determine the linear correlation of texture reproduction.

[0077] Signal coherence characterizes the degree of linear causal correlation between input and output signals at a specific frequency. The larger the mean of the local coherence feature values ​​obtained based on signal coherence, i.e., the closer the linear correlation of texture reproduction is to 1, the more the surface texture can be linearly explained by the internal skeleton structure. Conversely, it means that there is significant nonlinear distortion or noise interference, thus providing a quantitative basis in the frequency domain for judging the stability of the process. It should be noted that the calculation of self-power spectral density, cross-power spectral density, and signal coherence using the Welch method are all techniques well known to those skilled in the art, and will not be further limited or elaborated here.

[0078] Step S104: Determine the process status of the waterproof membrane based on the relative displacement of the melt fabric, the characteristic value of the residual distribution morphology, the high-frequency texture attenuation rate, and the linear correlation of texture reproduction.

[0079] After systematically decoupling the relative displacement of the melt fabric characterizing macroscopic slippage, the residual distribution morphology feature value identifying the physical form of defects, and the high-frequency texture attenuation rate and texture reproduction linear correlation degree quantifying rheological properties and reproduction quality, a multi-level logical diagnostic model was further constructed based on the relative displacement of the melt fabric, the residual distribution morphology feature value, the high-frequency texture attenuation rate, and the texture reproduction linear correlation degree to transform these independent and physically meaningful multi-dimensional features into process state conclusions that can directly guide production adjustments. This model comprehensively utilizes the complementary information of the above feature parameters to conduct a comprehensive qualitative and quantitative determination of the production process state of waterproof membranes. This process no longer relies on a single thickness deviation judgment, but accurately distinguishes specific states such as dragging and accumulation, excessive melt leveling, unstable composite process, and ideal process through feature combination logic. This achieves a leap from "discovering problems" to "locating the root cause of problems," significantly improving the accuracy of fault attribution and the response efficiency of production control.

[0080] Preferably, in some possible implementations of the embodiments of the present invention, the process of determining the process status of waterproof membrane based on the relative displacement of the melt fabric, the fabric texture reproduction gain, the residual distribution morphology characteristic value, the high-frequency texture attenuation rate, and the linear correlation of texture reproduction includes:

[0081] When the residual distribution morphology characteristic value is greater than the preset first distribution threshold, and the absolute value of the relative displacement of the melt fabric is greater than the preset prior relative displacement threshold, the process classification status label of the waterproof membrane is set to local material accumulation caused by dragging. A residual distribution morphology characteristic value greater than the preset first distribution threshold indicates a large residual distribution morphology characteristic value, suggesting a significant spike abrupt change in the nonlinear morphology residual signal, implying the existence of local high-amplitude anomalies. Conversely, an absolute value of the relative displacement of the melt fabric greater than the preset prior relative displacement threshold indicates a large relative displacement of the melt fabric, corresponding to significant macroscopic slippage between the melt and the fabric. These two logical sets exclude cases of simple raw material particles (no slippage) or simple velocity mismatch (no spikes), directly pointing to spike-shaped defects formed by local melt accumulation or pulling due to velocity mismatch. Therefore, its process status classification label is set to local material accumulation caused by dragging.

[0082] When the residual distribution morphology characteristic value is less than or equal to the preset second distribution threshold, and the high-frequency texture attenuation rate is greater than the preset prior attenuation rate threshold, the process status classification label of the waterproof membrane is set to melt over-leveling. A residual distribution morphology characteristic value less than the preset second distribution threshold indicates a small residual distribution morphology characteristic value, suggesting that the distribution morphology of the nonlinear residual exhibits a flat-top characteristic, meaning that the originally expected texture peaks are truncated or smoothed out, rather than additional sharp peaks appearing. A high-frequency texture attenuation rate greater than the preset prior attenuation rate threshold indicates a large high-frequency texture attenuation rate, meaning that the gain of the frequency domain transfer function decreases sharply with increasing frequency, reflecting a significant decrease in the ability to reproduce high-frequency fine textures. The combination of these two characteristics indicates that excessive melt fluidity, such as excessively high temperature or excessively low viscosity, causes excessive leveling under the influence of gravity and surface tension, making it unable to maintain the fine texture of the fabric skeleton. Therefore, its process status classification label is set to melt over-leveling.

[0083] When the residual distribution morphological characteristic value is less than or equal to the preset first distribution threshold and greater than the preset second distribution threshold: if the linear correlation of texture reproduction is less than or equal to the preset prior correlation threshold, then the process status classification label of the waterproof membrane is set as unstable composite process.

[0084] If the residual distribution morphology characteristic value is less than or equal to the preset first distribution threshold and greater than the preset second distribution threshold, it indicates that the residual distribution morphology characteristic value is in the middle region. This suggests that there are neither significant sharp peaks (i.e., non-hard particles) nor extreme flat-top truncations (i.e., non-excessive leveling), meaning that the defect morphology does not have typical physical orientation. If the linear correlation of texture reproduction is low at this time, it indicates that there is still a good linear correspondence between the surface texture and the internal skeleton. This suggests that the current process deviation is not due to a fundamental change in material properties or the introduction of foreign matter, but rather a general, non-specific random fluctuation. Therefore, the corresponding process state classification label is set to composite process instability, indicating that the operator's process parameters may be in a critical state or there may be external disturbances. General stability adjustments are required rather than targeted troubleshooting.

[0085] Under the condition that the residual distribution morphological characteristic value is less than or equal to the preset first distribution threshold and greater than the preset second distribution threshold, if the linear correlation of texture reproduction is greater than the preset prior correlation threshold, the relative displacement of melt fabric is less than or equal to the preset prior relative displacement threshold, and the high-frequency texture attenuation rate is less than or equal to the preset prior attenuation rate threshold, then the process status classification label of the waterproof membrane is set as ideal process; wherein, the preset first distribution threshold is greater than the preset second distribution threshold, and both the preset first distribution threshold and the preset second distribution threshold are greater than 0.

[0086] If the relative displacement of the melt fabric is less than or equal to the preset a priori relative displacement threshold, it indicates that the relative displacement of the melt fabric is small, indicating that the extrusion and calendering speeds are well matched and there is no macroscopic slippage. If the high-frequency texture attenuation rate is less than or equal to the preset a priori attenuation rate threshold, it indicates that the high-frequency texture attenuation rate is small, the melt has a strong ability to reproduce fine textures, and the rheological state is suitable. On this basis, if the value of the linear correlation of texture reproduction, which characterizes the linear consistency between surface texture and internal skeleton, is large, it indicates that the process meets the predetermined standards in the three dimensions of macroscopic synchronization, microscopic rheology, and system stability, and there are no significant sharp peaks or extreme flat-top truncations. Therefore, it is judged as an ideal process state.

[0087] In one specific implementation of this invention, the preset first distribution threshold is set to 5, and the preset second distribution threshold is set to 2, which can be adjusted according to the specific implementation environment. The preset prior correlation threshold, the preset prior attenuation rate threshold, and the preset prior relative displacement threshold need to be determined based on the statistical characteristics of historical normal production batch data. For example, for the preset prior correlation threshold, by collecting a large amount of production data of qualified waterproof membrane products, the corresponding texture reproduction linear correlation is calculated, and based on statistical principles, the mean of the texture reproduction linear correlation of all qualified waterproof membrane products plus or minus three times the standard deviation is used to set the normal fluctuation range boundary of the texture reproduction linear correlation, and the lower boundary of this normal fluctuation range boundary is taken as the preset prior correlation threshold. For the preset prior attenuation rate threshold, the upper boundary of the normal fluctuation range boundary of the high-frequency texture attenuation rate calculated from qualified waterproof membrane products is taken as the preset prior attenuation rate threshold. For the preset prior relative displacement threshold, the upper boundary of the normal fluctuation range boundary of the absolute value of the relative displacement of the melt fabric calculated from qualified waterproof membrane products is taken as the preset prior relative displacement threshold, which will not be further elaborated here.

[0088] For waterproof membranes without labels, in another specific implementation of this invention, the determination of the waterproof membrane's process status based on the relative displacement of the melt fabric, the characteristic value of the residual distribution morphology, the high-frequency texture attenuation rate, and the linear correlation of texture reproduction further includes: setting the process status classification label of waterproof membranes that do not meet the four conditions of local material accumulation, excessive melt leveling, unstable composite process, and ideal process as anomalies pending review. For waterproof membranes, if all four conditions of local material accumulation, excessive melt leveling, unstable composite process, and ideal process are not met, it indicates that a clear single fault mode cannot be matched in the preset process status logic tree and it does not belong to the ideal process status, so further manual anomaly review is required.

[0089] In summary, a method for determining the process status of waterproof membranes involves simultaneously acquiring surface height sequences measured by laser and contact, and determining surface fabric texture signals and fabric structure compression fluctuation signals. First, the relative displacement of the molten fabric is quantified using the delay correlation between the two, thereby directly identifying macroscopic process deviations caused by velocity mismatch. Then, through translation correction and linear gain modeling, the nonlinear morphological residual signal is extracted, and the physical morphology of defects is identified based on its amplitude distribution morphology characteristics, effectively distinguishing between local abrupt changes caused by unplasticized particles and systemic smooth defects caused by excessive casting. Simultaneously, by combining the high-frequency texture attenuation rate and linear correlation determined by the frequency domain transfer function, the microscopic reproducibility of the melt is quantified to reflect its rheological characteristics, thus expanding single-thickness detection into multi-dimensional feature decoupling diagnosis, resulting in higher accuracy in fault attribution based on the determination of the process status of waterproof membranes.

[0090] This application also provides a device for determining the process condition of waterproof membranes. Please refer to [link / reference]. Figure 2 The diagram shows a structural diagram of a waterproof membrane process condition determination device according to an embodiment of the present invention. The device includes: a data acquisition and preprocessing module 201, a first determination module 202, a second determination module 203, and a process condition determination module 204.

[0091] The data acquisition and preprocessing module 201 is used to synchronously acquire the laser-measured surface height sequence and the contact-measured surface height sequence of the waterproof membrane at a fixed sampling spatial interval; and to determine the surface fabric texture signal and the fabric structure compression fluctuation signal based on the laser-measured surface height sequence and its relative deviation from the contact-measured surface height sequence.

[0092] The first determining module 202 is used to determine the relative displacement of the melt fabric based on the delay correlation between the surface fabric texture signal and the fabric structure compression fluctuation signal; to perform translation correction on the fabric structure compression fluctuation signal based on the relative displacement of the melt fabric to determine the corrected compression fluctuation signal; and to determine the fabric texture reproduction gain based on the signal deviation between the surface fabric texture signal and the corrected compression fluctuation signal.

[0093] The second determining module 203 is used to determine the nonlinear morphology residual signal based on the deviation between the surface fabric texture signal and the corrected compression fluctuation signal after gain weighting through fabric texture reproduction; determine the residual distribution morphology characteristic value based on the amplitude distribution pattern of the nonlinear morphology residual signal; and determine the corresponding high-frequency texture attenuation rate and texture reproduction linear correlation based on the change of the frequency domain transfer function between the surface fabric texture signal and the corrected compression fluctuation signal.

[0094] The process condition determination module 204 is used to determine the process condition of waterproof membrane based on the relative displacement of melt fabric, residual distribution morphology characteristic value, high frequency texture attenuation rate and texture reproduction linear correlation.

[0095] It should be noted that the apparatus provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the waterproof membrane process state determination device and the waterproof membrane process state determination method embodiment provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiment, which will not be repeated here.

[0096] This application also provides a computer device; please refer to [link / reference]. Figure 3 The diagram illustrates a computer device structure according to an embodiment of the present invention. The computer device includes a memory 301, a processor 302, and a computer program 303 stored in the memory 301 and running on the processor 302. When the processor 302 executes the computer program 303, the computer device can execute any of the aforementioned methods for determining the process status of waterproof membrane.

[0097] This application also provides a computer program product that, when run on a computer device, enables the computer device to execute any of the aforementioned methods for determining the process status of waterproof membranes.

[0098] This application also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer device, the computer device can execute any of the aforementioned methods for determining the process status of waterproof membranes.

[0099] In the embodiments provided in this application, it should be understood that the computer device, computer program product and computer-readable storage medium provided are all used to perform the corresponding methods provided above, and therefore the beneficial effects they can achieve can be referred to the beneficial effects of the methods provided above, which will not be repeated here.

[0100] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0101] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method of determining the process condition of a waterproofing membrane, characterized in that, The method includes: The laser-measured surface height sequence and the contact-measured surface height sequence of the waterproof membrane are synchronously acquired at fixed sampling spatial intervals; based on the laser-measured surface height sequence and its relative deviation from the contact-measured surface height sequence, the surface fabric texture signal and the fabric structure compression fluctuation signal are determined. The relative displacement of the melt fabric is determined based on the delay correlation between the surface fabric texture signal and the fabric structure compression fluctuation signal; the fabric structure compression fluctuation signal is translated and corrected based on the relative displacement of the melt fabric to determine the corrected compression fluctuation signal; and the fabric texture reproduction gain is determined based on the signal deviation between the surface fabric texture signal and the corrected compression fluctuation signal. Based on the deviation between the surface fabric texture signal and the corrected compressed fluctuation signal after gain weighting through the fabric texture reproduction, the nonlinear morphology residual signal is determined; based on the amplitude distribution pattern of the nonlinear morphology residual signal, the residual distribution pattern characteristic value is determined; based on the change of the frequency domain transfer function between the surface fabric texture signal and the corrected compressed fluctuation signal, the corresponding high-frequency texture attenuation rate and texture reproduction linear correlation are determined. The process status of the waterproof membrane was determined based on the relative displacement of the melt fabric, the characteristic value of the residual distribution morphology, the high-frequency texture attenuation rate, and the linear correlation of texture reproduction. The process of determining the surface fabric texture signal and the fabric structure compression undulation signal based on the surface height sequence measured by laser and its relative deviation from the surface height sequence measured by contact includes: The laser-measured surface height sequence is input into a digital bandpass filter with a preset spatial frequency range for bandpass filtering to obtain the surface fabric texture signal. Calculate the surface height difference between the laser-measured surface height sequence and the contact-measured surface height sequence at the same spatial location; arrange the surface height differences at all spatial locations in the spatial order of the laser-measured surface height sequence to determine the surface height difference sequence; The surface height difference sequence is input into a digital bandpass filter with a preset spatial frequency range for bandpass filtering to obtain the fabric structure compression undulation signal. The process of obtaining the relative displacement of the melt fabric includes: The maximum number of delayed sampling points is determined by rounding up the ratio between the preset maximum drag displacement and the sampling space interval. The cross-correlation function value of the surface fabric texture signal and the fabric structure compression undulation signal at each spatial delay index value is calculated using the cross-correlation function. The spatial delay index value is the spatial lag of the fabric structure compression undulation signal relative to the surface fabric texture signal, and the range of the spatial delay index value is all integers between the negative of the maximum number of delayed sampling points and the maximum number of delayed sampling points. The relative displacement of the melt fabric is determined by multiplying the spatial delay index value corresponding to the largest cross-correlation function value with the sampling spatial interval. The process of obtaining the fabric texture reproduction gain includes: Construct an objective function; the objective function value is the sum of squared residuals between the surface fabric texture signal and the corrected compressed fluctuation signal weighted by undetermined coefficients at all signal index values; based on the least squares method, the solution of the undetermined coefficients when the objective function value is minimized is taken as the fabric texture reproduction gain; The process of determining the corresponding high-frequency texture attenuation rate and texture reproduction linear correlation based on the change in the frequency domain transfer function between the surface fabric texture signal and the corrected compression fluctuation signal includes: The Welch method was used to calculate the auto-power spectral density of the surface fabric texture signal, the auto-power spectral density of the corrected compression fluctuation signal, and the cross-power spectral density between the surface fabric texture signal and the corrected compression fluctuation signal at each frequency. Based on the ratio between the self-power spectral density and the corresponding cross-power spectral density of the surface fabric texture signal at each frequency, the empirical transfer coefficient corresponding to each frequency is determined; the negative of the linear regression slope of the modulus of the empirical transfer coefficients corresponding to all frequencies within the preset spatial frequency range is calculated to determine the corresponding high-frequency texture attenuation rate. The signal coherence is calculated based on the auto-power spectral density of the surface fabric texture signal at each frequency, the auto-power spectral density of the modified compression fluctuation signal, and the cross-power spectral density between the surface fabric texture signal and the modified compression fluctuation signal. The local coherence eigenvalues ​​corresponding to each frequency are then determined. The mean of the local coherence eigenvalues ​​corresponding to all frequencies within the preset spatial frequency range is calculated to determine the linear correlation of texture reproduction.

2. A method of determining the process state of a waterproofing membrane as defined in claim 1, wherein The process of acquiring the nonlinear topographic residual signal includes: The modified compression fluctuation signal is weighted by the fabric texture reproduction gain to determine the weighted compression fluctuation signal; The nonlinear morphology residual signal is determined by subtracting the weighted compression fluctuation signal from the surface fabric texture signal.

3. A method of determining the process state of a waterproofing membrane as defined in claim 1, wherein The process of obtaining the residual distribution morphological feature values ​​includes: Calculate the kurtosis of the nonlinear morphological residual signal to determine the residual distribution morphological characteristic value.

4. A method of determining the process state of a waterproofing membrane as defined in claim 1, wherein The process of determining the process status of waterproof membrane based on the relative displacement of the melt fabric, the characteristic value of residual distribution morphology, the high-frequency texture attenuation rate, and the linear correlation of texture reproduction includes: When the residual distribution morphology characteristic value is greater than the preset first distribution threshold, and the absolute value of the relative displacement of the melt fabric is greater than the preset prior relative displacement threshold, the process classification status label of the waterproof membrane is set to local material accumulation caused by dragging. When the residual distribution morphology feature value is less than or equal to the preset second distribution threshold, and the high-frequency texture attenuation rate is greater than the preset prior attenuation rate threshold, the process status classification label of the waterproof membrane is set to melt over-leveling. When the residual distribution morphological characteristic value is less than or equal to a preset first distribution threshold and greater than a preset second distribution threshold: if the linear correlation of texture reproduction is less than or equal to a preset prior correlation threshold, then the process state classification label of the waterproof membrane is set to unstable composite process; if the linear correlation of texture reproduction is greater than a preset prior correlation threshold, the relative displacement of melt fabric is less than or equal to a preset prior relative displacement threshold, and the high-frequency texture attenuation rate is less than or equal to a preset prior attenuation rate threshold, then the process state classification label of the waterproof membrane is set to ideal process; wherein, the preset first distribution threshold is greater than the preset second distribution threshold, and both the preset first distribution threshold and the preset second distribution threshold are greater than 0.

5. A method of determining the process state of a waterproofing membrane as defined in claim 4, wherein The determination of the waterproof membrane process status based on the relative displacement of the melt fabric, the residual distribution morphology characteristic value, the high-frequency texture attenuation rate, and the linear correlation of texture reproduction also includes: Waterproof membranes that fail to meet all four conditions—localized material accumulation, excessive melt leveling, unstable composite process, and ideal process—are categorized as having an abnormal process status pending review.

6. A waterproofing membrane process condition measuring device characterized by, The apparatus is used to implement the method for determining the process status of waterproof membrane as described in any one of claims 1 to 5; the apparatus includes: The data acquisition and preprocessing module is used to synchronously acquire the laser-measured surface height sequence and the contact-measured surface height sequence of the waterproof membrane at a fixed sampling spatial interval; based on the laser-measured surface height sequence and its relative deviation from the contact-measured surface height sequence, the surface fabric texture signal and the fabric structure compression fluctuation signal are determined. The first determining module is used to determine the relative displacement of the melt fabric based on the delay correlation between the surface fabric texture signal and the fabric structure compression fluctuation signal; to perform translation correction on the fabric structure compression fluctuation signal based on the relative displacement of the melt fabric to determine the corrected compression fluctuation signal; and to determine the fabric texture reproduction gain based on the signal deviation between the surface fabric texture signal and the corrected compression fluctuation signal. The second determining module is used to determine the nonlinear morphology residual signal based on the deviation between the surface fabric texture signal and the corrected compressed fluctuation signal after gain weighting through the fabric texture reproduction; determine the residual distribution morphology characteristic value based on the amplitude distribution pattern of the nonlinear morphology residual signal; and determine the corresponding high-frequency texture attenuation rate and texture reproduction linear correlation based on the change of the frequency domain transfer function between the surface fabric texture signal and the corrected compressed fluctuation signal. The process status determination module is used to determine the process status of the waterproof membrane based on the relative displacement of the melt fabric, the residual distribution morphology characteristic value, the high-frequency texture attenuation rate, and the linear correlation of texture reproduction.

Images