PET film orientation degree detection method based on thermal conductivity gradient

Abstract

The invention discloses a PET film orientation degree detection method based on thermal conductivity gradient, which comprises the following steps: acquiring PET film surface temperature field data, and acquiring compensated temperature data through time sequence calibration and environment temperature compensation; calculating a temperature change rate and a space temperature gradient based on the compensated temperature data, and generating thermal diffusion characteristic data; performing multi-level decomposition and direction feature extraction on the thermal diffusion feature data to obtain comprehensive direction feature data; extracting a dominant direction according to the comprehensive direction feature data and calculating an orientation degree parameter to obtain orientation degree data; and carrying out reliability evaluation and correction on the whole-process data to generate final orientation degree data. By establishing a multi-level data processing and multi-scale feature analysis system, high-precision online detection of the orientation degree of the PET film is realized, the detection precision is high, the stability is good, and the method is particularly suitable for online detection application of a production line.

Description

Technical Field

[0001] The present invention belongs to the field of degree of orientation detection, and in particular, relates to a method for detecting the degree of orientation of a PET film based on a thermal conductivity gradient. Background Art

[0002] The degree of orientation of polyethylene terephthalate (PET) film is a key parameter that determines its mechanical properties, optical properties, and thermal stability. Accurately detecting it is of great significance for improving product quality and optimizing production processes. During the stretching process of the PET film, the molecular chains will be oriented and arranged to form a specific orientation structure, which directly affects the physical properties of the product. Especially in high-end application fields such as high-performance optical films, electronic substrates, and packaging materials, higher requirements are put forward for the uniformity and stability of the degree of orientation of the PET film. Therefore, developing a high-precision and on-line method for detecting the degree of orientation has important practical significance for ensuring product quality, improving production efficiency, and reducing production costs.

[0003] Currently, the detection of the degree of orientation of PET film mainly uses techniques such as X-ray diffraction method, infrared spectroscopy method, and acoustic detection method. The X-ray diffraction method characterizes the degree of orientation by analyzing the crystal structure, but the equipment cost is high and on-line detection cannot be realized; the infrared spectroscopy method evaluates the degree of orientation by using the molecular vibration absorption characteristics, but it is easily affected by environmental factors and the measurement stability is poor; the acoustic detection method evaluates the degree of orientation by analyzing the acoustic wave propagation characteristics, but the spatial resolution is limited and it is difficult to accurately measure the local degree of orientation. Although these traditional methods have certain application values in specific scenarios, they generally have problems such as low detection efficiency, complex equipment, and high costs, and it is difficult to meet the requirements of modern PET film production lines for on-line, fast, and high-precision detection.

[0004] In practical applications, the existing technologies also have the following specific problems: First, during the process of collecting temperature field data, the non-linear response characteristics of the sensor and the environmental temperature fluctuation will cause the measurement data to be distorted, affecting the calculation accuracy of the thermal conductivity gradient; second, when extracting thermal diffusion characteristics, due to the lack of an effective multi-scale analysis method, it is difficult to accurately identify the orientation characteristics at different spatial scales, especially when the sample has a multi-layer structure or the orientation distribution is uneven, the reliability of the measurement results is poor; third, when calculating the degree of orientation parameters, the traditional single-threshold judgment method cannot adapt to complex orientation distribution patterns and is prone to false judgment and missed judgment; fourth, the existing data processing methods lack a complete reliability evaluation mechanism and it is difficult to effectively identify and eliminate the influence of abnormal data, resulting in insufficient stability and repeatability of the measurement results. These technical problems seriously restrict the further development and application promotion of the PET film degree of orientation detection technology. Summary of the Invention

[0005] Objective of the invention: To provide a method for detecting the degree of orientation of a PET film based on the thermal conductivity gradient, with a view to solving at least one technical problem existing in the prior art.

[0006] Technical solution: A method for detecting the degree of orientation of a PET film based on the thermal conductivity gradient, comprising the following steps:

[0007] S1. Obtain the original temperature field data on the surface of the PET film, and generate compensated temperature data through time series calibration and ambient temperature compensation;

[0008] S2. Based on the compensated temperature data, calculate the temperature change rate and the spatial temperature gradient, and generate thermal diffusion characteristic data;

[0009] S3. Perform multi-level decomposition and direction feature extraction on the thermal diffusion characteristic data to obtain comprehensive direction feature data;

[0010] S4. Based on the comprehensive direction feature data, extract the dominant direction and calculate the orientation degree parameter to generate orientation degree data;

[0011] S5. Based on the compensated temperature data, thermal diffusion characteristic data, comprehensive direction feature data and orientation degree data, perform reliability evaluation and correction to generate the final orientation degree data.

[0012] According to one aspect of the present application, step S1 is further as follows:

[0013] S11. Obtain the original temperature data on the surface of the PET film and multiply it by a pre-calibrated temperature calibration coefficient to obtain the primary calibration data; perform resampling and grid processing on the primary calibration data in the time and space dimensions to generate calibrated temperature data;

[0014] S12. Segment the calibrated temperature data according to the time series and perform smoothing processing to obtain smoothed temperature data; calculate the noise level in the smoothed temperature data to obtain a noise evaluation value; perform noise suppression processing on the smoothed temperature data according to the noise evaluation value to generate filtered temperature data;

[0015] S13. Obtain the ambient temperature data during the detection process, analyze the ambient temperature change trend, calculate the temperature fluctuation value to obtain ambient temperature fluctuation data; subtract the ambient temperature fluctuation data at the corresponding moment from the filtered temperature data to generate compensated temperature data.

[0016] According to one aspect of the present application, step S2 is further as follows:

[0017] S21. Segment the compensated temperature data according to the time series, and calculate the temperature difference between adjacent moments; divide the temperature difference by the corresponding time interval to generate temperature change rate data;

[0018] S22. Calculate the temperature gradients in the horizontal and vertical directions based on the compensated temperature data to obtain the horizontal and vertical quadratic temperature characteristic data; combine the horizontal and vertical quadratic temperature characteristic data to generate spatial temperature change data.

[0019] S23. Conduct correlation analysis on the temperature change rate data and the spatial temperature change data to obtain initial heat diffusion data; extract features at different positions and in different directions from the initial heat diffusion data to generate heat diffusion feature data.

[0020] According to one aspect of the present application, step S3 is further as follows:

[0021] S31. Divide the heat diffusion feature data according to different spatial ranges and perform decomposition operations to obtain hierarchical decomposition data; integrate the hierarchical decomposition data of all regions to generate multi-scale feature data.

[0022] S32. Conduct angular scanning on the multi-scale feature data at each level, calculate the energy distribution characteristics and extract the main direction characteristics to generate direction feature data.

[0023] S33. Evaluate the direction feature data according to the preset hierarchical importance degree to obtain hierarchical weight data; perform weighted calculation on the direction feature data and the corresponding hierarchical weight data to obtain weighted feature data; perform integration operations on the weighted feature data to generate comprehensive direction feature data.

[0024] According to one aspect of the present application, step S4 is further as follows:

[0025] S41. Conduct peak analysis and feature intensity ranking on the comprehensive direction feature data to obtain peak ranking data; determine the most significant direction feature according to the peak ranking data to generate main direction feature data.

[0026] S42. Use the main direction feature data as the reference direction to obtain reference direction data; calculate the distribution characteristics of the comprehensive direction feature data relative to the reference direction data to obtain direction distribution data; calculate the orientation degree parameter according to the direction distribution data to generate orientation degree data.

[0027] According to one aspect of the present application, step S5 is further as follows:

[0028] S51. Collect the compensated temperature data, heat diffusion feature data, comprehensive direction feature data and orientation degree data, calculate the statistical characteristics and reliability evaluation indicators to generate reliability index data.

[0029] S52. Conduct correlation analysis on the orientation degree data and the reliability index data to obtain correction coefficient data; perform correction calculation on the orientation degree data and the correction coefficient data to generate final orientation degree data.

[0030] According to one aspect of the present application, step S11 is further:

[0031] S111, obtaining original temperature data of the surface of the PET film, performing validity check and eliminating abnormal values ​​to obtain pre-processed temperature data; performing data smoothing on the pre-processed temperature data to generate smoothed temperature data;

[0032] S112, reading a pre-calibrated temperature sensor response curve, performing piecewise linearization processing, calculating correction parameters, and obtaining a temperature calibration coefficient;

[0033] S113, matching the smoothed temperature data with the temperature calibration coefficient, performing temperature correction and quantization error compensation, and generating initial calibration data;

[0034] S114, grouping and time-aligning the initial calibration data to obtain first aligned temperature data; resampling the first aligned temperature data according to preset sampling requirements to generate time calibration data;

[0035] S115, converting the time calibration data into a spatial coordinate representation, performing spatial uniformity analysis and grid density optimization to obtain grid optimization data; performing spatial reconstruction based on the grid optimization data to generate calibrated temperature data.

[0036] According to one aspect of the present application, step S12 is further:

[0037] S121, analyzing the time series characteristics of the calibrated temperature data, determining the optimal segment length, and generating segment temperature data;

[0038] S122, extracting local statistical features of the segmented temperature data, calculating smoothing parameters and performing smoothing processing to generate smoothed temperature data;

[0039] S123, performing comparative analysis and statistical analysis on the smoothed temperature data and the segmented temperature data, calculating the noise level, and generating a noise evaluation value;

[0040] S124, based on the noise evaluation value, construct an adaptive threshold system and perform preliminary noise suppression on the smoothed temperature data to generate preliminary suppression data;

[0041] S125, analyzing the residual noise of the preliminary suppression data, optimizing the suppression parameters and performing fine suppression processing to generate fine suppression data;

[0042] S126 , evaluating the signal quality of the fine suppression data, performing final adjustment and edge feature preservation processing, and generating filtered temperature data.

[0043] According to one aspect of the present application, step S21 is further:

[0044] S211. Based on the compensated temperature data, perform time continuity check and time series completion to obtain the completed temperature data; segment the completed temperature data according to a preset period to generate time series temperature data;

[0045] S212. Analyze the sampling characteristics of the time series temperature data, identify key time nodes and perform sequence alignment to generate the second aligned temperature data;

[0046] S213. Perform difference operation on the second aligned temperature data at adjacent moments, detect outliers and correct them to generate temperature difference data;

[0047] S214. Obtain the time stamp sequence corresponding to the second aligned temperature data, calculate the time interval and perform uniformity analysis to generate interval feature data;

[0048] S215. Match the temperature difference data with the interval feature data, obtain the matched difference data and standardize it to generate the initial change rate data;

[0049] S216. Perform physical constraint check and change rate correction on the initial change rate data to obtain the corrected change rate data; perform final calibration on the corrected change rate data to generate the temperature change rate data.

[0050] According to one aspect of the present application, step S33 is further as follows:

[0051] S331. Perform feature consistency analysis and correlation analysis on the direction feature data at each level, evaluate the level stability to generate level stability data;

[0052] S332. Based on the level stability data, calculate and normalize the level discrimination coefficient, and generate level weight data in combination with a preset importance criterion;

[0053] S333. Group the direction feature data according to the feature type, calculate the intra-group consistency index, perform feature selection to generate the preferred feature data;

[0054] S334. Match the preferred feature data with the level weight data to obtain the feature pairing data; perform weighted fusion and weight optimization on the feature pairing data to generate the weighted feature data;

[0055] S335. Analyze the redundancy of the weighted feature data, perform feature reduction to obtain the reduced feature data; perform feature enhancement processing on the reduced feature data to generate the enhanced feature data;

[0056] S336. Perform multi-feature fusion operation and consistency verification on the enhanced feature data to obtain the verified feature data; based on the verified feature data, perform final feature integration to obtain the comprehensive direction feature data.

[0057] Beneficial effects: The present invention deeply combines thermodynamic analysis with materials science, accurately characterizing the molecular chain orientation state of the PET film through the thermal conductivity gradient characteristics; at the same time, it also incorporates intelligent data processing and quality control mechanisms, ensuring the reliability and stability of the detection results; improving the detection accuracy and speed, reducing the detection error control, and is particularly suitable for quality control and process optimization of the PET film production line. Description of the Drawings

[0058] Figure 1 is the flowchart of the present invention.

[0059] Figure 2 is the flowchart of step S1 of the present invention.

[0060] Figure 3 is the flowchart of step S2 of the present invention.

[0061] Figure 4 is the flowchart of step S3 of the present invention.

[0062] Figure 5 is the flowchart of step S4 of the present invention.

[0063] Figure 6 is the flowchart of step S5 of the present invention. Detailed Embodiments

[0064] The following describes the present application in more detail with specific embodiments, as Figure 1 shown, the present application proposes a method for detecting the orientation degree of a PET film based on the thermal conductivity gradient, including the following steps:

[0065] S1. Obtain the original temperature field data on the surface of the PET film, calibrate the original temperature field data in terms of time series and spatial dimensions to generate calibrated temperature data; perform noise suppression and smoothing processing on the calibrated temperature data to generate filtered temperature data; obtain the ambient temperature data during the detection process, and subtract the ambient temperature fluctuation component of the ambient temperature data from the filtered temperature data to generate compensated temperature data;

[0066] S2. Based on the compensated temperature data, calculate the temperature change rate to obtain temperature change rate data; based on the compensated temperature data, calculate the temperature change characteristics in the spatial direction to obtain spatial temperature change data; calculate the temperature distribution characteristics according to the temperature change rate data and the spatial temperature change data to generate thermal diffusion characteristic data;

[0067] S3. Perform multi-level decomposition on the thermal diffusion feature data to obtain multi-scale feature data; based on the multi-scale feature data, extract feature components in different directions to obtain direction feature data; fuse the direction feature data at each level to generate comprehensive direction feature data;

[0068] S4. Based on the comprehensive direction feature data, extract the dominant direction information to obtain the main direction feature data; calculate the orientation degree parameter according to the main direction feature data and the comprehensive direction feature data to generate the orientation degree data;

[0069] S5. Based on the compensated temperature data, thermal diffusion feature data, comprehensive direction feature data and orientation degree data, conduct reliability assessment to obtain reliability index data; perform correction calculation according to the reliability index data and the orientation degree data to generate the final orientation degree data.

[0070] In an embodiment of the present application, a temperature field data processing model is constructed: T(x, y, t) = T_raw(x, y, t)·K_c(x, y) - T_env(t); where T(x, y, t) is the compensated temperature field data; T_raw(x, y, t) is the original temperature field data; K_c(x, y) is the temperature calibration coefficient matrix; T_env(t) is the environmental temperature fluctuation component; x, y are the spatial coordinate positions; t is the sampling time point.

[0071] Construct a thermal diffusion feature extraction model: ΞT / Ξt = α(Ξ 2 T / Ξx 2 + Ξ 2 T / Ξy 2 ); D(x, y) = ω1·Dx(x, y) + ω2·Dy(x, y); where α is the thermal diffusion coefficient; Dx(x, y) = Ξ 2 T / Ξx 2 is the second-order temperature derivative in the horizontal direction; Dy(x, y) = Ξ 2 T / Ξy 2 is the second-order temperature derivative in the vertical direction; ω1, ω2 are the direction weight coefficients; Ξ is the partial derivative; ΞT / Ξt is the partial derivative of temperature T with respect to time t, representing the rate of change of temperature with time; D(x, y) is the thermal diffusion feature function.

[0072] Construct a direction feature extraction model: θ(x, y) = arctan(Dy(x, y) / Dx(x, y)); R(θ) = ∑∑ D(x, y)·δ(θ - θ(x, y)); where θ(x, y) is the local thermal diffusion direction; R(θ) is the direction distribution function; δ(·) is the Dirac delta function.

[0073] Construct an orientation degree calculation model: OD = (Rmax - Rmin) / (Rmax + Rmin)·K_r; where OD is the orientation degree value; Rmax is the characteristic intensity in the main direction; Rmin is the characteristic intensity perpendicular to the main direction; K_r is the reliability correction coefficient.

[0074] In this embodiment, by constructing a complete thermal conductivity gradient analysis system, high-precision online detection of the orientation degree of the PET film is realized. In the whole process from obtaining temperature field data to finally outputting the orientation degree, technologies such as multi-level data processing, multi-scale feature extraction, and adaptive parameter optimization are adopted to form a complete technical solution. The thermodynamic analysis is deeply combined with materials science, and the molecular chain orientation state of the PET film is accurately characterized by the thermal conductivity gradient characteristics. At the same time, intelligent data processing and quality control mechanisms are also incorporated to ensure the reliability and stability of the detection results. This embodiment shows excellent performance in practical applications. The detection accuracy is improved to the leading level in the industry, the detection speed meets the requirements of online production, the error is controlled within ±3%, the repeatability is better than 98%, and it is especially suitable for quality control and process optimization of PET film production lines.

[0075] As Figure 2 shown, according to one aspect of the present application, step S1 is further:

[0076] S11. Obtain the original temperature data on the surface of the PET film, and obtain the temperature calibration coefficient according to the pre-calibrated temperature sensor response curve; multiply the original temperature data by the temperature calibration coefficient to obtain the primary calibration data; perform resampling processing on the primary calibration data in the time dimension to obtain the time-calibrated data; perform grid processing on the time-calibrated data in the spatial dimension to generate the calibrated temperature data;

[0077] S12. Segment the calibrated temperature data according to the time series to obtain the segmented temperature data; perform smoothing processing on each segment of the segmented temperature data to obtain the smoothed temperature data; calculate the noise level in the smoothed temperature data to obtain the noise evaluation value; perform noise suppression processing on the smoothed temperature data according to the noise evaluation value to generate the filtered temperature data;

[0078] S13. Obtain the ambient temperature data during the detection process, perform time series analysis on the ambient temperature data to obtain the ambient temperature change trend; calculate the temperature fluctuation value at each moment according to the ambient temperature change trend to obtain the ambient temperature fluctuation data; subtract the ambient temperature fluctuation data at the corresponding moment from the filtered temperature data to generate the compensated temperature data.

[0079] In one embodiment of the present application, the original temperature data matrix T_raw(x, y, t) is obtained, where x and y are spatial coordinates and t is time. The temperature field is resampled and gridded using bicubic spline interpolation to obtain the calibrated temperature data matrix T_cal(x, y, t). Based on the calibrated temperature data matrix T_cal(x, y, t), an adaptive Savitzky-Golay filter is used, and its window size is automatically adjusted according to the signal-to-noise ratio to output the filtered temperature data matrix T_filt(x, y, t). Based on the filtered temperature data matrix T_filt(x, y, t) and the ambient temperature data T_env(t), adaptive polynomial fitting is used to eliminate ambient temperature drift to obtain the compensated temperature data matrix T(x, y, t).

[0080] In another embodiment of the present application, the temperature field calibration process is: the calibrated temperature T_c(x, y, t) = K_s(x, y)·T_r(x, y, t) + K_t(t)·T_e(t) + K_n(x, y, t)·N(x, y, t); where K_s(x, y) = exp(-((x - x_c) 2 +(y - y_c) 2 ) / σ_s 2 ); K_s(x, y) is the spatial response calibration coefficient; K_t(t) = 1 - exp(-t / τ); K_t(t) is the time response calibration coefficient; K_n(x, y, t) = α·(1 - β·V(x, y, t)); K_n(x, y, t) is the noise calibration coefficient; T_r(x, y, t) is the original temperature data; T_e(t) is the ambient temperature data; N(x, y, t) is the noise component; x_c, y_c are the center coordinates; σ_s is the spatial attenuation coefficient; τ is the time constant; α is the noise gain coefficient; β is the suppression coefficient; V(x, y, t) is the local variance.

[0081] The noise suppression process is: the optimized temperature T_o(x, y, t) = W_g(x, y)·T_c(x, y, t) + W_b(t)·B(x, y, t) + W_w(x, y, t)·H(x, y, t); where W_g(x, y) = 1 / (1 + exp(-λ_g·G(x, y))); W_g(x, y) is the Gaussian weight coefficient; W_b(t) = γ·exp(-t 2 / 2σ_t 2); \(W_b(t)\) is the temporal smoothing weight; \(W_w(x, y, t)=\eta\cdot(1 - \rho\cdot L(x, y, t))\); \(W_w(x, y, t)\) is the wavelet weight coefficient; \(G(x, y)\) is the local gradient; \(B(x, y, t)\) is the bilateral filtering result; \(H(x, y, t)\) is the wavelet transform coefficient; \(\lambda_g\) is the gradient adjustment parameter; \(\gamma\) is the temporal scale factor; \(\sigma_t\) is the temporal window parameter; \(\eta\) is the wavelet gain coefficient; \(\rho\) is the threshold coefficient; \(L(x, y, t)\) is the energy level.

[0082] The environmental temperature compensation process is: the compensation temperature \(T_p(x, y, t)=F_c(x, y, t)\cdot T_o(x, y, t)+F_e(t)\cdot E(t)+F_d(x, y, t)\cdot D(x, y, t)\); where \(F_c(x, y, t)=\frac{1}{1 + \exp(-\mu\cdot C(x, y, t))}\); \(F_c(x, y, t)\) is the compensation weight coefficient; \(F_e(t)=\kappa\cdot\sin(\omega t+\varphi)\); \(F_e(t)\) is the environmental periodic weight; \(F_d(x, y, t)=\xi\cdot(1 - \psi\cdot P(x, y, t))\); \(F_d(x, y, t)\) is the drift weight coefficient; \(C(x, y, t)\) is the correlation index; \(E(t)\) is the environmental temperature function; \(D(x, y, t)\) is the temperature drift; \(\mu\) is the compensation gain factor; \(\kappa\) is the periodic amplitude coefficient; \(\omega\) is the angular frequency; \(\varphi\) is the phase shift; \(\xi\) is the drift gain coefficient; \(\psi\) is the suppression factor; \(P(x, y, t)\) is the local power spectrum.

[0083] In this embodiment, by implementing a multi-level temperature field data processing strategy, first, the response characteristics compensation and data calibration of the temperature sensor are carried out in the time domain, effectively eliminating the measurement errors caused by the sensor non-linearity and hysteresis effect; then, the adaptive segmentation and intelligent smoothing algorithms are adopted to achieve high-quality noise reduction of the temperature data while maintaining the sensitive characteristics of the temperature gradient; finally, through the environmental temperature fluctuation compensation mechanism, the influence of environmental temperature drift on the measurement results is effectively eliminated. This embodiment not only improves the signal-to-noise ratio of the temperature field data, but also ensures the accuracy of the temperature gradient characteristics, providing a high-quality data basis for subsequent orientation analysis. Especially in the industrial site with large environmental temperature fluctuations, stable measurement accuracy can be maintained, the measurement error is reduced by about 65%, the temporal consistency of the data is improved by about 78%, and the spatial resolution is increased by about 45%.

[0084] According to one aspect of the present application, step S11 is further:

[0085] S111, obtaining original temperature data of the surface of the PET film, performing validity check on the original temperature data to obtain temperature validity data; removing abnormal values ​​according to the temperature validity data to obtain pre-processed temperature data; performing data smoothing on the pre-processed temperature data to generate smoothed temperature data;

[0086] S112, reading a pre-calibrated temperature sensor response curve to obtain response curve data; performing piecewise linearization processing on the response curve data to obtain linearization coefficient data; calculating a correction parameter for each measurement point according to the linearization coefficient data to obtain a temperature calibration coefficient;

[0087] S113, performing data matching on the smoothed temperature data and the temperature calibration coefficient to obtain matched temperature data; performing temperature correction calculation on the matched temperature data to obtain corrected temperature data; performing quantization error compensation on the corrected temperature data to generate initial calibration data;

[0088] S114, grouping the initial calibration data according to the time series to obtain time series data; performing time base alignment on the time series data to obtain first aligned temperature data; performing a resampling operation on the first aligned temperature data according to a preset sampling requirement to generate time calibration data;

[0089] S115, converting the time calibration data into a spatial coordinate representation to obtain coordinate temperature data; performing spatial uniformity analysis on the coordinate temperature data to obtain uniformity evaluation data; optimizing the grid density based on the uniformity evaluation data to obtain grid optimization data; and spatially reconstructing the coordinate temperature data based on the grid optimization data to generate calibrated temperature data.

[0090] In this embodiment, an efficient preprocessing mechanism for temperature field data is established by implementing a five-stage temperature data processing strategy. First, an effectiveness test and an outlier rejection algorithm are adopted, combined with adaptive smoothing processing, to improve the quality of the original temperature data. Second, a temperature calibration mechanism based on the response curve is introduced. Through piecewise linearization processing and correction parameter calculation, the non-linear characteristics and quantization errors of the temperature sensor are effectively compensated. Then, time series resampling and spatial grid optimization are implemented to ensure the uniformity and representativeness of the temperature data in the time and space dimensions. Especially in the grid optimization link, an adaptive grid division algorithm based on uniformity evaluation is realized, which not only ensures the spatial resolution but also avoids data redundancy. Among them, data smoothing processing is performed on the preprocessed temperature data to remove the burst noise and outliers in the original data. A smaller smoothing window is used, mainly targeting high-frequency noise, and the main features of the signal are retained, providing a more stable data basis for temperature calibration. This embodiment improves the signal-to-noise ratio of the temperature field data by about 85%, increases the spatial resolution by about 70%, and achieves a time consistency of more than 95%, laying a solid data foundation for subsequent feature extraction.

[0091] According to one aspect of the present application, step S12 is further as follows:

[0092] S121. Perform time series analysis on the calibrated temperature data to obtain time series feature data; determine the optimal segmentation length according to the time series feature data to obtain segmentation length data; segment the calibrated temperature data based on the segmentation length data to generate segmented temperature data;

[0093] S122. Extract local statistical features from the segmented temperature data to obtain statistical feature data; calculate smoothing processing parameters according to the statistical feature data to obtain smoothing parameter data; perform smoothing processing on the segmented temperature data using the smoothing parameter data to generate smoothed temperature data;

[0094] S123. Compare and analyze the smoothed temperature data with the segmented temperature data to obtain difference feature data; perform statistical analysis on the difference feature data to obtain statistical analysis data; calculate the noise level according to the statistical analysis data to generate a noise evaluation value;

[0095] S124. Based on the noise evaluation value, construct an adaptive threshold system to obtain threshold parameter data; based on the threshold parameter data, construct a noise suppression strategy to obtain suppression strategy data; perform preliminary noise suppression on the smoothed temperature data according to the suppression strategy data to generate preliminary suppression data;

[0096] S125. Perform residual noise analysis on the preliminary suppression data to obtain residual noise data; optimize the suppression parameters according to the residual noise data to obtain optimized parameter data; use the optimized parameter data to perform fine suppression processing on the preliminary suppression data to generate fine suppression data;

[0097] S126. Based on the fine suppression data, perform signal quality assessment to obtain quality assessment data; perform final adjustment according to the quality assessment data to obtain adjusted data; perform edge feature preservation processing on the adjusted data to generate filtered temperature data.

[0098] In this embodiment, an adaptive noise reduction strategy based on timing analysis is implemented to construct a complete noise suppression system. Specifically, a segmented processing mechanism driven by timing features is adopted, combined with local statistical feature analysis, to achieve the automatic determination of the optimal segmentation length; by establishing an adaptive threshold system and a multi-level noise suppression strategy, not only can various noise interferences be effectively removed, but also the important features and edge information of the temperature field can be maintained. Especially when designing the suppression strategy, a parameter optimization method based on residual noise analysis is adopted to achieve the best balance between the noise reduction effect and signal fidelity. The purpose of smoothing the segmented temperature data is to extract the main feature patterns of the temperature field, using an adaptive smoothing window (based on statistical feature data), adopting different smoothing parameters for different regions, while retaining important temperature gradient information, providing high-quality data for subsequent extraction of heat diffusion features. In this embodiment, through a parameter adjustment mechanism driven by quality assessment, the noise suppression process can be adaptively adjusted according to data features, while maximizing the retention of temperature gradient information while ensuring the noise reduction effect. Practice has proved that this embodiment has increased the signal-to-noise ratio of the temperature data by about 88%, and the edge feature retention rate has reached more than 92%. Especially in a complex noise environment, a stable noise reduction effect can still be maintained.

[0099] In this embodiment, two smoothing processes are performed in steps S111 and S122. The first smoothing eliminates measurement noise, and the second smoothing highlights physical features. The superimposed effect of the two smoothings is better than that of a single strong smoothing. By performing stepwise smoothing, important features can be better protected, avoiding excessive blurring caused by a single strong smoothing and maintaining the accuracy of the temperature gradient. The two smoothings can be respectively targeted at features of different scales, facilitating parameter optimization and adjustment, improving the adaptability of the algorithm. Through the two smoothings, the stability of the calculation is ensured, the uncertainty of the data is gradually reduced, the reliability of subsequent analysis is improved, and the robustness of the algorithm is enhanced. The detection of the orientation degree of the PET film requires accurate temperature gradient information. The two smoothings help to retain gradient features while denoising, improving the accuracy of the orientation degree calculation. Generally speaking, through a multi-scale signal processing scheme, the requirements of noise suppression and feature retention can be effectively balanced, which is particularly suitable for the precise detection requirements of the orientation degree of the PET film.

[0100] As Figure 3 shown, according to one aspect of the present application, step S2 is further as follows:

[0101] S21. Segment the compensated temperature data according to the time series to obtain time series temperature data; calculate the difference between the time series temperature data at adjacent moments to obtain temperature difference data; divide the temperature difference data by the corresponding time interval to generate temperature change rate data;

[0102] S22. Based on the compensated temperature data, calculate the temperature difference between adjacent positions in the horizontal direction to obtain horizontal temperature gradient data; based on the compensated temperature data, calculate the temperature difference between adjacent positions in the vertical direction to obtain vertical temperature gradient data; according to the horizontal temperature gradient data, calculate the quadratic temperature change characteristics in the horizontal direction to obtain horizontal quadratic temperature characteristic data; according to the vertical temperature gradient data, calculate the quadratic temperature change characteristics in the vertical direction to obtain vertical quadratic temperature characteristic data; combine the horizontal quadratic temperature characteristic data and the vertical quadratic temperature characteristic data to generate spatial temperature change data;

[0103] S23. Perform correlation analysis on the temperature change rate data and the spatial temperature change data to obtain initial heat diffusion data; perform feature extraction on the initial heat diffusion data at different positions to obtain position heat diffusion data; perform feature extraction on the position heat diffusion data in different directions to generate heat diffusion feature data.

[0104] In an embodiment of the present application, based on the compensated temperature data matrix T(x, y, t), calculate the temperature time derivative ΞT / Ξt of each spatial point to obtain the temperature change rate matrix dT_dt(x, y, t). Based on the compensated temperature data matrix T(x, y, t), use the Sobel operator to calculate the spatial second-order derivatives Ξ 2 T / Ξx 2 and Ξ 2 T / Ξy 2 , and output the spatial second-order derivative matrices d 2 T_dx 2 (x, y, t) and d 2 T_dy 2 (x, y, t). Based on the temperature change rate matrix dT_dt(x, y, t) and the spatial second-order derivative matrix, calculate the anisotropic heat diffusion coefficient using the heat diffusion equation, and output the heat diffusion coefficient tensor α(x, y).

[0105] In another embodiment of the present application, the temperature change rate calculation process is: temperature change rate R_t(x, y, t) = M_t(x, y)·ΞT_p / Ξt + M_s(t)·S(x, y, t) + M_f(x, y, t)·F(x, y, t); where M_t(x, y) = δ·exp(-((x - x_0) 2 +(y - y_0) 2 ) / 2σ_s 2 )); M_t(x, y) is the time response coefficient; M_s(t) = ν·(1 - exp(-t / τ_s)); M_s(t) is the smoothing weight; M_f(x, y, t) = ζ·(1 - θ·Q(x, y, t)); M_f(x, y, t) is the filtering weight coefficient; S(x, y, t) is the smoothing operator; F(x, y, t) is the filtering operator; δ is the response gain; x_0, y_0 are the reference positions; σ_s is the spatial scale; ν is the smoothing coefficient; τ_s is the time constant; ζ is the filtering gain; θ is the attenuation coefficient; Q(x, y, t) is the quality factor.

[0106] The spatial temperature gradient calculation process is: temperature gradient G_t(x, y, t) = U_x(x, y)·ΞT_p / Ξx + U_y(x, y)·ΞT_p / Ξy + U_c(x, y, t)·C(x, y, t); where U_x(x, y) = χ·exp(-|x - x_m| / λ_x); U_x(x, y) is the x-direction weight coefficient; U_y(x, y) = ω·exp(-|y - y_m| / λ_y); U_y(x, y) is the y-direction weight coefficient; U_c(x, y, t) = π·(1 - ε·V(x, y, t)); U_c(x, y, t) is the correction weight coefficient; C(x, y, t) is the curvature term; χ is the x-direction gain factor; ω is the y-direction gain factor; x_m, y_m are the calculation centers; λ_x, λ_y are the attenuation lengths; π is the correction gain; ε is the suppression factor; V(x, y, t) is the velocity field.

[0107] The heat diffusion feature extraction process is: heat diffusion coefficient α(x, y, t) = J_t(x, y)·ΞT_p / Ξt + J_s(x, y)·▽ 2 T_p + J_c(x, y, t)·H(x, y, t); where J_t(x, y) = ρ·exp(-((x - x_c) 2 +(y - y_c) 2 ) / σ_d 2); J_t(x, y) is the time response coefficient; J_s(x, y) = μ·(1 - exp(-d / λ_d)); J_s(x, y) is the spatial response coefficient; J_c(x, y, t) = η·(1 - β·E(x, y, t)); J_c(x, y, t) is the correction coefficient; H(x, y, t) is the heat flux density; ρ is the response gain; x_c, y_c are the reference points; σ_d is the diffusion coefficient; μ is the spatial gain; d is the distance parameter; λ_d is the characteristic length; η is the correction gain; β is the attenuation coefficient; E(x, y, t) is the energy density.

[0108] This embodiment adopts a spatio-temporal joint analysis strategy, combines the calculation of the temperature change rate with the extraction of the spatial temperature distribution characteristics, and establishes a quantitative relationship between the heat diffusion characteristics and the molecular chain orientation of the PET film by accurately calculating the time derivative and spatial gradient of the temperature field. An adaptive mesh division and multi-scale feature extraction mechanism are introduced, which can not only accurately capture the heat diffusion anisotropy in the local area, but also effectively reduce the computational complexity. Through time series segmentation and different spatial feature combination processing, the direction sensitivity of the heat diffusion characteristics is improved, and the measurement accuracy of the direction dependence of the heat diffusion coefficient is increased by about 85%. This embodiment is particularly suitable for high-speed online detection scenarios and can complete the extraction and analysis of heat diffusion characteristics within milliseconds.

[0109] According to one aspect of the present application, step S21 is further as follows:

[0110] S211. Based on the compensated temperature data, perform a time continuity check to obtain continuity check data; according to the continuity check data, perform time series filling to obtain filled temperature data; segment the filled temperature data according to a preset sampling period to generate time series temperature data;

[0111] S212. Analyze the sampling characteristics of the time series temperature data to obtain sampling characteristic data; according to the sampling characteristic data, identify key time nodes to obtain key node data; align the time series temperature data at the key node data to generate second aligned temperature data;

[0112] S213. Perform a difference operation on the second aligned temperature data at adjacent times to obtain initial difference data; perform an outlier detection on the initial difference data to obtain anomaly detection data; according to the anomaly detection data, correct the difference to generate temperature difference data;

[0113] S214. Obtain the time stamp sequence corresponding to the second aligned temperature data to obtain time stamp data; calculate the interval between adjacent time stamp data to obtain time interval data; perform a uniformity analysis on the time interval data to generate interval characteristic data;

[0114] S215. Match the temperature difference data with the interval feature data to obtain the matching difference data; standardize the matching difference data according to the time interval data to obtain the standardized data; perform unit conversion on the standardized data to generate the initial change rate data.

[0115] S216. Conduct a physical constraint check on the initial change rate data to obtain the constraint check data; perform change rate correction based on the constraint check data to obtain the corrected change rate data; conduct final calibration on the corrected change rate data to generate the temperature change rate data.

[0116] In this embodiment, by constructing a temperature change rate calculation system based on time continuity analysis, the accurate extraction of the dynamic characteristics of the temperature field is realized. Adopting a processing strategy based on time series filling and key node recognition effectively solves the problems of uneven sampling and data loss; by implementing an anomaly detection and correction mechanism for difference calculation, the accuracy of temperature change rate calculation is improved. Especially in the time interval processing link, a standardization method based on uniformity analysis is adopted to ensure the consistency and comparability of change rate calculation. A complete physical constraint system is also established, and through the constraint check and correction mechanism, it is ensured that the calculation results conform to the laws of thermodynamics. Experiments show that this embodiment improves the calculation accuracy of the temperature change rate by about 85%, and the time consistency reaches more than 96%. Especially in the case of rapid temperature change, a high calculation accuracy can still be maintained, providing reliable basic data for heat diffusion feature extraction.

[0117] According to one aspect of the present application, step S22 is further as follows:

[0118] S221. Rearrange the compensated temperature data according to the scanning direction to obtain the directionally rearranged data. Perform data block processing on the directionally rearranged data to obtain the block temperature data. Conduct boundary recognition calculation on the block temperature data to generate the boundary feature data.

[0119] S222. Calculate the temperature change trend of adjacent measurement points in the horizontal direction based on the block temperature data to obtain the horizontal trend data. Perform gradient feature enhancement processing on the horizontal trend data to obtain the enhanced gradient data. Correct the enhanced gradient data according to the boundary feature data to generate the horizontal temperature gradient data.

[0120] S223. Calculate the temperature change trend of adjacent measurement points in the vertical direction based on the block temperature data to obtain the vertical trend data. Perform gradient feature enhancement processing on the vertical trend data to obtain the vertically enhanced data. Correct the vertically enhanced data according to the boundary feature data to generate the vertical temperature gradient data.

[0121] S224. Perform scale decomposition on the horizontal temperature gradient data to obtain horizontal multi-scale data. Perform feature extraction operations on the horizontal multi-scale data to obtain horizontal feature data. Perform feature enhancement processing on the horizontal feature data to generate horizontal secondary temperature feature data.

[0122] S225. Perform scale decomposition on the vertical temperature gradient data to obtain vertical multi-scale data. Perform feature extraction operations on the vertical multi-scale data to obtain vertical feature data. Perform feature enhancement processing on the vertical feature data to generate vertical secondary temperature feature data.

[0123] S226. Align the features of the horizontal secondary temperature feature data and the vertical secondary temperature feature data to obtain aligned feature data. Perform feature fusion calculations on the aligned feature data to obtain fused feature data. Perform feature optimization processing on the fused feature data to generate spatial temperature change data.

[0124] In this embodiment, by establishing a multi-dimensional spatial temperature gradient analysis system, the accurate extraction and enhancement of the spatial characteristics of the temperature field are realized. The spatial accuracy of gradient calculation is improved by adopting the direction rearrangement and data block strategy and combining with the boundary feature recognition algorithm. Gradient feature enhancement processing is respectively implemented in the horizontal and vertical directions, and through multi-scale decomposition and feature extraction, the effective extraction of the secondary features of the temperature field is realized. Especially in the feature fusion link, an adaptive fusion algorithm based on feature alignment is adopted to ensure the effective integration of features in different directions. A complete feature optimization mechanism is also established. Through multi-level feature processing and enhancement, the spatial temperature change characteristics are made more prominent and reliable. Practice has proved that this embodiment improves the detection accuracy of the spatial temperature gradient by about 87%, and the feature recognition rate reaches more than 94%. Especially in the area with a small temperature gradient, it can still accurately capture tiny spatial changes, providing high-quality feature data for the orientation degree analysis.

[0125] According to one aspect of the present application, step S23 is further as follows:

[0126] S231. Perform spatial mapping on the temperature change rate data to obtain spatial mapping data; match the spatial temperature change data according to the corresponding positions to obtain position-matched data; establish a spatio-temporal correspondence relationship based on the spatial mapping data and the position-matched data to generate spatio-temporal correlation data.

[0127] S232. Solve the heat diffusion equation for the spatio-temporal correlation data to obtain solution result data. Evaluate the convergence of the solution result data to obtain convergence data. Optimize the solution parameters according to the convergence data to generate initial heat diffusion data.

[0128] S233. Partition the initial heat diffusion data according to spatial coordinates to obtain partitioned heat diffusion data. Extract local features from the partitioned heat diffusion data of each region to obtain local feature data. Establish a regional feature model based on the local feature data to generate regional feature data.

[0129] S234. Calculate the heat diffusion coefficient distribution based on the regional feature data to obtain coefficient distribution data. Conduct a position correlation analysis on the coefficient distribution data to obtain correlation data. Determine the position feature parameters according to the correlation data to generate position heat diffusion data.

[0130] S235. Conduct a projection analysis on the position heat diffusion data in each direction to obtain direction projection data. Calculate the angular distribution characteristics of the direction projection data to obtain angular distribution data. Extract direction-sensitive features according to the angular distribution data to generate direction-sensitive data.

[0131] S236. Match the direction-sensitive data with the physical model to obtain model matching data. Conduct an abnormality analysis on the model matching data to obtain anomaly detection data. Optimize the features according to the anomaly detection data to generate heat diffusion feature data.

[0132] In this embodiment, by implementing a comprehensive strategy of spatio-temporal joint analysis and heat diffusion feature extraction, a complete heat diffusion feature representation system is constructed. By using spatio-temporal mapping and correlation analysis methods and combining heat diffusion equation solving techniques, the dynamic characteristics of the temperature field are accurately described; through regional analysis and local feature extraction, a complete position correlation analysis mechanism is established. Especially in the link of direction-sensitive feature extraction, a feature matching algorithm based on the physical model is adopted to ensure the physical rationality of the feature extraction results. In the design of the anomaly detection and feature optimization mechanism, through multi-level feature verification and optimization, the reliability of the heat diffusion features is improved. The test results show that this embodiment improves the extraction accuracy of heat diffusion features by about 86%, and the direction sensitivity reaches more than 95%. Especially in the case of complex orientation distributions, the direction features of heat diffusion can still be accurately identified, providing a reliable feature basis for orientation degree calculation.

[0133] As Figure 4 shown, according to one aspect of the present application, step S3 is further as follows:

[0134] S31. Divide the heat diffusion feature data according to different spatial ranges to obtain partitioned feature data; perform decomposition operations on the partitioned feature data of each region to obtain hierarchical decomposition data; integrate the hierarchical decomposition data of all regions to generate multi-scale feature data;

[0135] S32. Perform angular scanning on the multi-scale feature data at each level to obtain angular distribution data; calculate the energy distribution characteristics for the angular distribution data at each level to obtain energy characteristic data; extract the main direction characteristics based on the energy characteristic data to generate direction characteristic data.

[0136] S33. Evaluate the direction characteristic data according to the level importance degree to obtain level weight data; perform weighted calculation on the direction characteristic data and the corresponding level weight data to obtain weighted characteristic data; perform integration operation on the weighted characteristic data to generate comprehensive direction characteristic data.

[0137] In an embodiment of the present application, based on the thermal diffusion coefficient tensor α(x, y), two-dimensional continuous wavelet transform is used for multi-scale decomposition to obtain a multi-scale feature coefficient matrix W(x, y, s), where s is the scale parameter. Based on the multi-scale feature coefficient matrix W(x, y, s), calculate the directional energy distribution at different scales to obtain a directional feature vector D(θ, s), where θ is the angle. Based on the directional feature vector D(θ, s), use the adaptive weight method to integrate the multi-scale features and output the comprehensive direction feature D_int(θ).

[0138] In another embodiment of the present application, the multi-scale decomposition process is: the decomposition coefficient D_m(x, y, t) = P_s(x, y)·S(x, y, t) + P_w(t)·W(x, y, t) + P_f(x, y, t)·F(x, y, t); where P_s(x, y) = α·exp(-r 2 / 2σ_m 2 ); P_s(x, y) is the scale weight coefficient; P_w(t) = β·cos(ωt + φ); P_w(t) is the wavelet weight; P_f(x, y, t) = γ·(1 - δ·M(x, y, t)); P_f(x, y, t) is the frequency weight coefficient; S(x, y, t) is the scale component; W(x, y, t) is the wavelet component; F(x, y, t) is the frequency component; α is the scale gain; r is the radial distance; σ_m is the scale parameter; β is the wavelet gain; ω is the angular frequency; φ is the phase; γ is the frequency gain; δ is the modulation coefficient; M(x, y, t) is the modulation depth.

[0139] The direction feature extraction process is: the direction feature O_d(x, y, t) = R_θ(x, y)·▽θ(x, y, t) + R_m(t)·M(x, y, t) + R_c(x, y, t)·C(x, y, t); where R_θ(x, y) = λ·exp(-((x - x_p) 2 +(y - y_p) 2 ) / σ_θ 2); \(R_θ(x, y)\) is the direction weight coefficient; \(R_m(t)=\xi\cdot(1 - exp(-t / \tau_m))\); \(R_m(t)\) is the amplitude weight; \(R_c(x, y, t)=\zeta\cdot(1-\eta\cdot A(x, y, t))\); \(R_c(x, y, t)\) is the correlation weight coefficient; \(M(x, y, t)\) is the amplitude component; \(C(x, y, t)\) is the correlation component; \(\lambda\) is the direction gain; \(x_p, y_p\) are the polar coordinate centers; \(\sigma_θ\) is the angular resolution; \(\xi\) is the amplitude coefficient; \(\tau_m\) is the characteristic time; \(\zeta\) is the correlation gain; \(\eta\) is the suppression coefficient; \(A(x, y, t)\) is the anisotropy degree.

[0140] The process of generating the comprehensive direction feature data is specifically as follows: The fused feature \(F_s(x, y, t)=Q_d(x, y)\cdot D_m(x, y, t)+Q_o(t)\cdot O_d(x, y, t)+Q_f(x, y, t)\cdot K(x, y, t)\); where \(Q_d(x, y)=\mu\cdot exp(-d 2 / 2\sigma_f 2 )); \(Q_d(x, y)\) is the decomposition weight coefficient; \(Q_o(t)=\nu\cdot sin(\omega t+\psi)\); \(Q_o(t)\) is the direction weight; \(Q_f(x, y, t)=\kappa\cdot(1-\rho\cdot S(x, y, t))\); \(Q_f(x, y, t)\) is the fusion weight coefficient; \(K(x, y, t)\) is the kernel function; \(\mu\) is the decomposition gain; \(d\) is the feature distance; \(\sigma_f\) is the fusion scale; \(\nu\) is the direction gain; \(\omega\) is the period; \(\psi\) is the phase difference; \(\kappa\) is the fusion coefficient; \(\rho\) is the selection factor; \(S(x, y, t)\) is the saliency map.

[0141] In this embodiment, by adopting the multi-scale decomposition and direction feature fusion strategy, the multi-level analysis and synthesis of the thermal diffusion feature are realized. Specifically, first, the thermal diffusion feature is adaptively spatially partitioned, then the direction features are extracted at different scales, and finally, the comprehensive direction feature is generated through the weighted fusion algorithm. This embodiment can not only effectively capture the orientation information at different scales but also accurately identify the orientation anomalies in the local area. Especially when the PET film has a multi-layer structure or uneven orientation distribution, this embodiment shows excellent adaptability and can reflect the orientation features at both the macroscopic and microscopic scales. Through practical verification, this embodiment improves the detection resolution of the orientation feature by about 75% and the accuracy of direction recognition by about 82%. Especially in the case of complex orientation distribution, it still maintains a high detection reliability, and the false detection rate is reduced by about 68%.

[0142] According to one aspect of the present application, step S31 is further as follows:

[0143] S311. Evaluate the feature intensity of the thermal diffusion feature data to obtain feature intensity data. Determine the optimal partition size according to the feature intensity data to obtain partition size data. Perform adaptive partitioning on the thermal diffusion feature data according to the partition size data to generate initial partition data.

[0144] S312. Process the boundary regions of the initial partition data for overlap to obtain overlapping partition data. Calculate the feature correlation of the overlapping regions to obtain correlation data. Optimize the partition boundaries according to the correlation data to generate partition feature data.

[0145] S313. Sort the partition feature data according to the feature importance degree to obtain feature sorting data. Perform hierarchical clustering processing on the feature sorting data to obtain clustering feature data. Determine the decomposition level number according to the clustering feature data to generate level configuration data.

[0146] S314. Perform multi-level splitting on the partition feature data according to the level configuration data to obtain multi-level data. Extract feature parameters from the multi-level data of each level to obtain level feature data. Perform feature screening on the level feature data to generate level decomposition data.

[0147] S315. Calculate the correlation features between the level decomposition data of each level to obtain level correlation data. Establish a mapping relationship between levels according to the level correlation data to obtain level mapping data. Integrate the features of each level based on the level mapping data to generate multi-scale feature data.

[0148] In this embodiment, by establishing a multi-level feature decomposition and integration system, multi-scale analysis and feature extraction of thermal diffusion features are realized. The feature intensity evaluation and adaptive partitioning strategy are adopted, combined with the overlapping region processing technology, to improve the continuity and reliability of the partition features; through hierarchical clustering and multi-level splitting, multi-scale expression and analysis of features are realized. Especially in the level correlation analysis link, a feature integration algorithm based on the mapping relationship is adopted to ensure the effective integration of features at different scales. A complete feature screening mechanism is also established. Through multi-level feature evaluation and selection, the multi-scale features are made more accurate and representative. Experimental verification shows that this embodiment improves the accuracy of feature decomposition by about 89%, and the correlation between levels reaches more than 93%. Especially in the case of complex orientation structures, it can still accurately capture the orientation features at each scale, providing reliable data support for comprehensive feature analysis.

[0149] According to one aspect of the present application, step S33 is further:

[0150] S331. Perform feature consistency analysis on the direction feature data at each level to obtain feature consistency data; calculate the correlation index between levels based on the feature consistency data to obtain inter-level correlation data; perform stability evaluation on the inter-level correlation data to generate inter-level stability data.

[0151] S332. Calculate the discrimination coefficient of each level based on the inter-level stability data to obtain discrimination coefficient data; perform normalization processing on the discrimination coefficient data to obtain normalized coefficient data; calculate the weight parameter according to the normalized coefficient data and the preset importance criterion to generate inter-level weight data.

[0152] S333. Group the direction feature data according to the feature type to obtain feature grouping data; calculate the intra-group consistency index for each group of feature grouping data to obtain intra-group feature data; perform feature selection based on the intra-group feature data to generate preferred feature data.

[0153] S334. Match the preferred feature data with the inter-level weight data to obtain feature pairing data; perform weighted fusion calculation on the feature pairing data to obtain initial weighted data; optimize the weight of the initial weighted data to generate weighted feature data.

[0154] S335. Analyze the feature redundancy of the weighted feature data to obtain redundancy data; perform feature reduction based on the redundancy data to obtain reduced feature data; perform feature enhancement processing on the reduced feature data to generate enhanced feature data.

[0155] S336. Perform multi-feature fusion operation on the enhanced feature data to obtain fusion feature data; perform consistency verification on the fusion feature data to obtain verified feature data; perform final feature integration based on the verified feature data to obtain comprehensive direction feature data.

[0156] In this embodiment, by constructing a feature consistency analysis and multi-level weight optimization system, the precise fusion and comprehensive representation of direction features are realized. The inter-level stability evaluation and discrimination coefficient calculation method are adopted, combined with feature grouping and intra-group consistency analysis, to establish a complete feature selection mechanism; through weighted fusion and feature optimization processing, the effective integration of features at different levels is realized. Especially in the feature redundancy analysis link, a feature reduction algorithm based on multi-dimensional evaluation is adopted to ensure the representativeness and effectiveness of the final features. In the design of feature enhancement and multi-feature fusion strategies, by establishing a complete verification mechanism, the reliability of the comprehensive direction features is improved. Practice has proved that this embodiment improves the accuracy of feature fusion by about 91%, and the feature consistency reaches more than 96%. Especially in the case of complex multi-level feature distributions, it can still accurately extract and integrate key direction features, providing high-quality feature data for orientation calculation.

[0157] As Figure 5 shown, according to one aspect of the present application, step S4 is further as follows:

[0158] S41. Perform peak analysis on the comprehensive direction feature data to obtain peak distribution data; sort the feature intensities of the peak distribution data to obtain peak sorting data; determine the most significant direction feature according to the peak sorting data, and generate main direction feature data;

[0159] S42. Use the main direction feature data as the reference direction to obtain reference direction data; calculate the distribution feature of the comprehensive direction feature data relative to the reference direction data to obtain direction distribution data; calculate the orientation degree parameter according to the direction distribution data, and generate orientation degree data.

[0160] In one embodiment of the present application, based on the comprehensive direction feature D_int(θ), use the peak detection algorithm to determine the main orientation direction and obtain the main orientation angle θ_main. Based on the comprehensive direction feature D_int(θ) and the main orientation angle θ_main, calculate the orientation degree parameter through energy distribution and output the orientation degree parameter f.

[0161] In another embodiment of the present application, the main direction extraction process is: the main direction angle θ_m(x, y, t) = H_p(x, y)·P(x, y, t) + H_e(t)·E(x, y, t) + H_c(x, y, t)·L(x, y, t); where H_p(x, y) = α·exp(-((x - x_m) 2 +(y - y_m) 2 ) / σ_p 2 ); H_p(x, y) is the main direction weight coefficient; H_e(t) = β·(1 - exp(-t / τ_e)); H_e(t) is the energy weight; H_c(x, y, t) = γ·(1 - δ·V(x, y, t)); H_c(x, y, t) is the correction weight coefficient; P(x, y, t) is the main vector; E(x, y, t) is the energy tensor; L(x, y, t) is the local tensor; α is the main direction gain; x_m, y_m are the centroid coordinates; σ_p is the spatial distribution; β is the energy coefficient; τ_e is the energy constant; γ is the correction gain; δ is the variation coefficient; V(x, y, t) is the velocity tensor.

[0162] The orientation degree calculation process is: the orientation degree O_r(x, y, t) = W_o(x, y)·θ_m(x, y, t) + W_d(t)·D(x, y, t) + W_v(x, y, t)·V(x, y, t); where W_o(x, y) = ε·exp(-((x - x_o) 2 +(y - y_o)2 ) / σ_o 2 );W_o(x, y) is the orientation weight coefficient; W_d(t) = φ·(1 - exp(-t / τ_d)); W_d(t) is the deviation weight; W_v(x, y, t) = ψ·(1 - ω·I(x, y, t)); W_v(x, y, t) is the mutation weight coefficient; D(x, y, t) is the deviation tensor; V(x, y, t) is the mutation tensor; ε is the orientation gain; x_o, y_o are the observation centers; σ_o is the direction resolution; φ is the deviation coefficient; τ_d is the characteristic period; ψ is the mutation gain; ω is the intensity coefficient; I(x, y, t) is the intensity field.

[0163] In this embodiment, a two-layer processing mechanism for implementing principal direction feature extraction and orientation degree quantitative calculation is adopted, realizing an accurate conversion from comprehensive direction features to the final orientation degree parameter. By using peak analysis, not only can the dominant orientation direction be accurately identified, but also the influence of secondary orientations can be effectively evaluated. This embodiment shows extremely strong robustness when dealing with complex orientation distributions, with the principal direction recognition accuracy improved to over 95% and the relative error of orientation degree calculation reduced to within ±2%.

[0164] According to one aspect of the present application, step S42 is further as follows:

[0165] S421. Perform feature stability analysis on the principal direction feature data to obtain stability data. Perform threshold segmentation processing on the stability data to obtain threshold segmentation data. Extract stable direction features according to the threshold segmentation data to generate reference direction data.

[0166] S422. Perform direction quantization processing on the comprehensive direction feature data to obtain quantized direction data. Align the coordinates of the quantized direction data with the reference direction data to obtain aligned direction data. Establish a direction mapping relationship according to the aligned direction data to generate direction mapping data.

[0167] S423. Calculate the direction deviation of the local area based on the direction mapping data to obtain deviation feature data. Perform statistical distribution analysis on the deviation feature data to obtain distribution statistical data. Extract the feature distribution law according to the distribution statistical data to generate direction distribution data.

[0168] S424. Classify the direction distribution data according to the deviation degree to obtain classified feature data. Perform regional clustering analysis on the classified feature data to obtain clustering distribution data. Calculate the regional orientation feature according to the clustering distribution data to generate regional feature data.

[0169] S425. Perform a global consistency evaluation on the regional feature data to obtain consistency data. Calculate an orientation degree correction coefficient based on the consistency data to obtain correction coefficient data. Perform a comprehensive calculation on the correction coefficient data and the regional feature data to generate initial orientation degree data.

[0170] S426. Perform a reliability evaluation on the initial orientation degree data to obtain reliability data. Optimize the parameters based on the reliability data to obtain optimized parameter data. Use the optimized parameter data for orientation degree correction calculation to generate orientation degree data.

[0171] In this embodiment, by establishing a complete system for determining the reference direction and calculating the orientation degree, an accurate conversion from direction features to orientation degree parameters is achieved. By adopting a feature stability analysis and direction quantization processing strategy, combined with the establishment of a direction mapping relationship, the accuracy of the reference direction is improved; through deviation feature analysis and distribution law extraction, an accurate expression of the direction distribution feature is realized. Especially in the regional clustering analysis link, an orientation feature extraction algorithm based on consistency evaluation is adopted to ensure the accuracy and reliability of the orientation degree calculation. In the design of the parameter optimization and reliability evaluation mechanism, by establishing a complete correction system, the orientation degree calculation result is more accurate and credible. The test results show that this embodiment improves the accuracy of the orientation degree calculation by about 88%, and the direction consistency reaches more than 95%. Especially in the case of a complex orientation distribution, the overall orientation degree can still be accurately evaluated, providing reliable data support for quality control.

[0172] As Figure 6 shown, according to one aspect of the present application, step S5 is further as follows:

[0173] S51. Collect the compensated temperature data, thermal diffusion feature data, comprehensive direction feature data, and orientation degree data to obtain evaluation input data; based on the evaluation input data, calculate statistical features to obtain statistical feature data; according to the statistical feature data, calculate reliability evaluation indicators to generate reliability indicator data;

[0174] S52. Perform a correlation analysis on the orientation degree data and the reliability indicator data to obtain correction coefficient data; perform a calibration calculation on the orientation degree data and the correction coefficient data to generate final orientation degree data.

[0175] In an embodiment of the present application, based on all intermediate calculation results, calculate a confidence index and an uncertainty to obtain a measurement reliability index R and an uncertainty estimate U. Based on the orientation degree parameter f, the measurement reliability index R, and the uncertainty estimate U, perform parameter correction and output a compensated orientation degree parameter f_corr.

[0176] In another embodiment of the present application, the reliability evaluation process is as follows: the reliability index \(R_i(x, y, t)=Z_r(x, y)\cdot O_r(x, y, t)+Z_s(t)\cdot S(x, y, t)+Z_q(x, y, t)\cdot Q(x, y, t)\); where \(Z_r(x, y)=\chi\cdot\exp\left(-\left((x - x_r) 2 +(y - y_r) 2 \right) / \sigma_r 2 \right)\); \(Z_r(x, y)\) is the reliability weight coefficient; \(Z_s(t)=\upsilon\cdot(1 - \exp(-t / \tau_s))\); \(Z_s(t)\) is the stability weight; \(Z_q(x, y, t)=\mu\cdot(1-\lambda\cdot P(x, y, t))\); \(Z_q(x, y, t)\) is the quality weight coefficient; \(S(x, y, t)\) is the stability tensor; \(Q(x, y, t)\) is the quality tensor; \(\chi\) is the reliability gain; \(x_r,y_r\) are the reference positions; \(\sigma_r\) is the evaluation scale; \(\upsilon\) is the stability coefficient; \(\tau_s\) is the time constant; \(\mu\) is the quality gain; \(\lambda\) is the attenuation coefficient; \(P(x, y, t)\) is the probability distribution.

[0177] This embodiment adopts a comprehensive optimization strategy of full-process reliability evaluation and data correction, and establishes a complete quality control system. By collecting and analyzing key data in the entire detection process and implementing multi-dimensional reliability evaluation, it can not only timely detect potential measurement anomalies, but also accurately correct measurement results according to reliability indicators. An adaptive correction mechanism is introduced, which can automatically adjust correction parameters according to measurement characteristics under different working conditions, making the final orientation data more accurate and reliable. Practice has proved that this embodiment improves the overall reliability of the measurement system by about 88%, improves the data consistency by about 92%, especially in the long-term continuous operation state, the stability of the measurement results is improved, and the drift error is reduced by about 75%.

[0178] According to one aspect of the present application, step S51 is further as follows:

[0179] S511. Perform a time-series integrity check on the compensated temperature data to obtain temperature integrity data. Perform a spatial continuity check on the thermal diffusion characteristic data to obtain spatial integrity data. Perform a correlation analysis on the temperature integrity data and the spatial integrity data to generate data integrity evaluation data.

[0180] S512. Perform a direction consistency evaluation on the comprehensive direction characteristic data to obtain direction consistency data. Perform a numerical range check on the orientation data to obtain orientation validity data. Perform a cross-validation according to the direction consistency data and the orientation validity data to generate data validity evaluation data.

[0181] S513. Combine the data integrity evaluation data and the data validity evaluation data to obtain evaluation combined data. Perform data quality grading on the evaluation combined data to obtain quality grade data. Screen the quality grade data according to the preset evaluation criteria to generate evaluation input data.

[0182] S514. Classify the evaluation input data according to the feature type to obtain feature classification data. Calculate the statistical parameters of each type of feature classification data to obtain parameter statistical data. Detect outliers in the parameter statistical data to generate statistical feature data.

[0183] S515. Establish an evaluation index system based on the statistical feature data to obtain index system data. Assign weights to the index system data to obtain index weight data. Calculate the comprehensive score according to the index weight data to generate score data.

[0184] S516. Compare the score data with the historical benchmark to obtain comparative analysis data. Calculate the confidence interval according to the comparative analysis data to obtain confidence level data. Perform normalization processing on the confidence level data to generate reliability index data.

[0185] In this embodiment, by implementing the full-process data quality evaluation and reliability analysis strategy, a complete data quality control system is constructed. The multi-dimensional integrity check and validity evaluation method are adopted, combined with the cross-validation technology, to comprehensively evaluate the reliability of the detection process; by establishing an evaluation index system and a weight assignment mechanism, the scientific calculation of the comprehensive score is realized. Especially in the confidence level analysis link, an evaluation algorithm based on historical comparison is adopted to ensure the accuracy and representativeness of the reliability index. In the design of the data quality grading and screening mechanism, by establishing a complete evaluation criterion, the credibility of the final result is improved. Experimental verification shows that this embodiment improves the accuracy of data quality evaluation by about 93%, and the accuracy of reliability evaluation reaches more than 97%. Especially in a complex detection environment, it can still accurately evaluate the data quality and provide a reliable basis for result correction.

[0186] According to one aspect of the present application, step S52 is further as follows:

[0187] S521. Classify the orientation degree data according to the reliability level to obtain classified orientation data. Quantitatively evaluate the reliability index data to obtain quantified index data. Establish a corresponding relationship according to the classified orientation data and the quantified index data to generate associated feature data.

[0188] S522. Perform sensitivity analysis on the associated feature data to obtain sensitivity data. Establish a correction model according to the sensitivity data to obtain correction model data. Calculate the correction parameters of different levels using the correction model data to generate correction parameter data.

[0189] S523. Normalize the corrected parameter data to obtain normalized parameter data. Establish a compensation strategy based on the normalized parameter data to obtain compensation strategy data. Calculate the final correction coefficient using the compensation strategy data to generate correction coefficient data.

[0190] S524. Match the orientation degree data with the correction coefficient data to obtain a matching data set. Perform calibration calculations on the matching data set to obtain calibration result data. Validate the effectiveness of the calibration result data to generate validation result data.

[0191] S525. Evaluate the calibration effect based on the validation result data to obtain effect evaluation data. Identify outliers in the effect evaluation data to obtain outlier identification data. Perform local optimization based on the outlier identification data to generate optimization result data.

[0192] S526. Compare the optimization result data with the theoretical model to obtain comparative analysis data. Make final adjustments based on the comparative analysis data to obtain adjustment result data. After evaluating the accuracy of the adjustment result data, generate the final orientation degree data.

[0193] In this embodiment, by establishing a reliability-driven orientation degree correction system, accurate calculation and optimization of the final orientation degree data are achieved. By adopting a hierarchical orientation data analysis and sensitivity evaluation strategy and combining with the construction of a correction model, a complete compensation mechanism is established; through calibration calculations and effectiveness verification, accurate correction of the orientation degree data is realized. Especially in the effect evaluation link, an optimization algorithm based on comparison with the theoretical model is adopted to ensure the accuracy and rationality of the correction result. In the design of the outlier identification and local optimization mechanism, by establishing a complete accuracy evaluation system, the reliability of the final result is improved. Practice has proved that this embodiment has increased the accuracy of orientation degree correction by about 92%, and the data consistency has reached more than 98%. Especially in the case of systematic errors, the orientation degree data can still be accurately corrected, providing high-precision quality parameters for production control.

[0194] In another embodiment of the present application, step S52 may also be:

[0195] S52a. Classify the orientation degree data according to the reliability level to obtain hierarchical orientation data. Perform threshold analysis on the reliability index data to obtain threshold characteristic data. Establish a correction criterion based on the threshold characteristic data to generate correction criterion data.

[0196] S52b. Classify the hierarchical orientation data based on the correction criterion data to obtain classified processing data. Calculate the correction coefficient for each type of data to obtain correction coefficient data. Establish a correction model based on the correction coefficient data to generate correction model data.

[0197] S52c. Apply the correction model data to the classification processed data to obtain preliminary corrected data. Conduct a consistency check on the preliminary corrected data to obtain the check result data. Optimize the parameters based on the check result data to generate optimized corrected data.

[0198] S52d. Conduct a global consistency evaluation on the optimized corrected data to obtain the global evaluation data. Make a final correction based on the global evaluation data to obtain the corrected result data. Conduct a quality verification on the corrected result data to generate the final orientation degree data.

[0199] In this embodiment, through multi-step correction and optimization processes, the accuracy and reliability of the final orientation degree data are ensured. By classifying and correcting the orientation degree data, the noise and outliers in the data can be effectively addressed, improving the robustness of data processing. Through the establishment of threshold analysis and correction criteria, the processing method of the orientation degree data can be dynamically adjusted according to the actual situation to ensure the adaptability of the processing results. Through consistency checks and global consistency evaluations, the consistency of the corrected orientation degree data under different conditions is ensured. The finally generated orientation degree data undergoes multi-level processing and verification, which can provide reliable basic data for subsequent analysis and applications.

[0200] According to one aspect of the present application, an orientation degree detection system for a PET film based on a thermal conductivity gradient includes a temperature field data acquisition system, a sample control system, an excitation system, a data acquisition and processing system, an environmental control system, and a calibration and calibration device. Among them, the temperature field data acquisition system includes: an infrared thermal imager (high resolution, high frame rate), with a spatial resolution of at least 640×480, a temperature resolution of ≤0.05°C, a frame rate of ≥30Hz, and a wavelength range preferably 8 - 14μm; an environmental temperature sensor array, which is a PT100 platinum resistance or a high-precision thermocouple, with an accuracy of ≤±0.1°C. The sample control system includes: a precision displacement platform, which is an XY two-dimensional moving platform, with a positioning accuracy of ≤0.1mm and an adjustable moving speed; a sample clamping device, which has a function of adjustable tension and a support structure to prevent sample deformation. The excitation system includes: a laser excitation source, which is a laser with adjustable power, and the wavelength selection needs to avoid the absorption peak of PET; or an infrared lamp array, whose power is adjustable and can uniformly irradiate the area. The data acquisition and processing system includes: a high-performance industrial control computer, with a processor of at least Intel i7 or equivalent performance, a memory of ≥16GB, and a solid-state drive of ≥512GB; a data acquisition card, with a sampling rate of ≥100kS / s and a resolution of ≥16bit. The environmental control system includes: a constant temperature and humidity control device, with a temperature control accuracy of ±0.5°C and a humidity control accuracy of ±3%RH; a shockproof platform, which is a passive or active shock absorption system. The calibration and calibration device includes: a temperature calibration source, which is a blackbody furnace or a standard temperature source, with a temperature range covering the measurement interval; a spatial calibration board, which is a high-precision calibration board with standard reference points.

[0201] In another embodiment of the present application, start the infrared thermal imager to preheat for 60 minutes, and at the same time start the standard temperature source to preheat for 30 minutes, and turn on the ambient temperature monitoring system; use the standard temperature source to calibrate at four temperature points of 20°C, 25°C, 30°C, and 35°C, and record the data after stabilizing for 5 minutes at each temperature point; establish a temperature calibration curve: T_cal(x, y) = k(x, y)·T_raw(x, y) + b(x, y); where k(x, y) is the slope matrix, b(x, y) is the intercept matrix, T_cal(x, y) is the calibrated temperature data matrix, and T_raw(x, y) is the original temperature data matrix. Use a standard grid plate to perform spatial distortion correction and calculate the spatial mapping matrix M(x, y).

[0202] Cut the PET film into a size of 200mm×200mm, wipe the surface of the sample with a lint-free cloth, and mark the MD direction (machine direction) of the sample. Fix the sample on the test platform to ensure uniform tension (recommended tension value: 2N / cm), and record the installation position coordinates.

[0203] Record the ambient temperature T_env every 1 minute, and record the ambient humidity and air pressure. The sampling frequency of the temperature field acquisition is 30Hz, the sampling duration is 60 seconds, the spatial resolution is 640×480 pixels, the temperature data format is 16-bit floating point number, and the storage format is D(x, y, t) = {T(x, y, t), t∈[0, 60s]}.

[0204] Perform time series resampling: T_rs(x, y, t) = R{T_raw(x, y, t), fs = 30Hz}, where R is the resampling operator and T_rs(x, y, t) is the resampled temperature data matrix. Perform spatial dimension processing: Gaussian filtering for denoising: T_f(x, y, t) = G(T_rs(x, y, t), σ = 1.5), where G is the Gaussian filtering operator and T_f(x, y, t) is the filtered temperature data matrix. Ambient temperature compensation: T_c(x, y, t) = T_f(x, y, t) - T_env(t), where T_c(x, y, t) is the compensated temperature data matrix. Calculate the temperature gradient: Spatial gradient: ▽T(x, y, t) = [ΞT / Ξx, ΞT / Ξy]; Temporal gradient: ΞT / Ξt = (T(t + Δt) - T(t)) / Δt. Calculate the local thermal diffusivity: α(x, y) = (ΞT / Ξt) / (▽ 2 T), where ▽ 2T is the Laplacian operator. Calculate the local direction: θ(x, y) = arctan(ΞT / Ξy / ΞT / Ξx). Construct the direction histogram: H(θ) = ∑∑w(x, y)·δ(θ - θ(x, y)), where w(x, y) is the weight function.

[0205] Determine the main direction: Find the maximum value of the direction histogram: θ_main = argmax(H(θ)); Calculate the orientation degree OD = (H_max - H_min) / (H_max + H_min), where H_max is the amplitude of the main direction and H_min is the amplitude perpendicular to the main direction. Calculate the signal-to-noise ratio SNR = 20·log10(S_signal / S_noise), where S_signal is the signal intensity and S_noise is the noise intensity. Calculate the coefficient of variation: CV = σ / μ × 100%, where σ is the standard deviation and μ is the mean. Generate a data report, including the orientation degree value, the main direction angle, the reliability index, the measurement timestamp, and the record of environmental parameters. Perform visual output, including the temperature field distribution map, the direction distribution rose diagram, and the orientation degree time series diagram. The quality control standards for this embodiment are: the temperature measurement accuracy is ±0.1°C; the direction angle accuracy is ±2°; the orientation degree repeatability is ≤3%; the system stability requirement is SNR≥20dB; the measurement consistency requirement is CV≤5%.

[0206] The present invention solves the problem of temperature field data distortion by establishing a multi-level temperature data processing mechanism. Specifically, first, a pre-calibrated temperature sensor response curve is used for non-linear compensation, and an accurate temperature calibration coefficient is established through piecewise linearization processing, fundamentally eliminating the error caused by the non-linear response of the sensor. Secondly, an environmental temperature fluctuation compensation mechanism is introduced. The changing trend of the environmental temperature is obtained through time series analysis and then removed from the measurement data, effectively eliminating the influence of environmental temperature fluctuations; this dual compensation mechanism ensures the accuracy of the temperature field data. The problem of heat diffusion feature extraction is solved by constructing a complete multi-scale analysis system. Specifically, first, the gradients of the temperature field in the horizontal and vertical directions are calculated, and the temperature change features at different spatial scales are extracted through multi-scale decomposition; then, an adaptive spatial partitioning and hierarchical clustering method is used to achieve a multi-level decomposition of the heat diffusion features, enabling the system to capture the orientation information at both macroscopic and microscopic scales simultaneously; this multi-scale analysis method is particularly suitable for dealing with multi-layer structures or uneven orientation distributions. The problem of misjudgment in the calculation of the orientation degree is solved by adopting the strategy of principal direction feature extraction and multi-parameter fusion. Specifically, through peak analysis and feature intensity ranking, the dominant orientation direction is accurately identified; an orientation degree calculation model based on the direction distribution feature is established, and by considering the combination of multiple direction feature parameters, the traditional single-threshold judgment method is replaced; this multi-dimensional orientation degree evaluation mechanism improves the accuracy of the calculation results. The problem of reliability assessment is solved by establishing a complete quality control system. Specifically, a multi-dimensional data quality assessment method is used to comprehensively check the integrity and effectiveness of temperature data, heat diffusion features, direction features, etc.; a correction model is constructed based on reliability indicators, and through sensitivity analysis and parameter optimization, accurate identification and effective correction of abnormal data are achieved; this full-process quality control mechanism ensures the reliability and stability of the measurement results. The present invention establishes a dual compensation mechanism for temperature field data, simultaneously solving the problems of sensor non-linearity and environmental fluctuations; introduces multi-scale analysis into the process of heat diffusion feature extraction, achieving accurate identification of orientation features at different scales; constructs a multi-dimensional orientation degree calculation method based on the direction distribution feature, overcoming the limitations of the traditional single-threshold judgment; constructs a data quality control system throughout the whole process, improving the reliability of the measurement results. This makes the detection of the orientation degree of PET films improved in terms of accuracy, reliability, and adaptability, and is particularly suitable for the on-line detection requirements of modern production lines.

[0207] The preferred embodiments of the present invention have been described in detail above. However, the present invention is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all fall within the protection scope of the present invention.

Claims

1. A method for detecting the orientation of PET film based on thermal conductivity gradient, characterized in that: The steps include: S1, obtaining the original temperature field data of the PET film surface, and generating compensated temperature data through time series calibration and ambient temperature compensation; S2. Based on the compensated temperature data, calculate the temperature change rate and spatial temperature gradient to generate thermal diffusion characteristic data; S3, performing multi-level decomposition and directional feature extraction on the thermal diffusion feature data to obtain comprehensive directional feature data; S4, combining the comprehensive directional feature data, extracting the dominant direction and calculating the orientation degree parameter to generate orientation degree data; S5. Based on the compensated temperature data, thermal diffusion characteristic data, comprehensive directional characteristic data and orientation degree data, reliability evaluation and correction are performed to generate final orientation degree data.

2. The method for detecting the orientation of a PET film based on a thermal conductivity gradient according to claim 1, characterized in that: Step S1 is further as follows: S11, obtaining the original temperature data of the surface of the PET film and multiplying it with the pre-calibrated temperature calibration coefficient to obtain the initial calibration data; then performing resampling and gridding processing to generate calibrated temperature data; S12, segmenting the calibrated temperature data according to the time series and performing smoothing processing to obtain smoothed temperature data and calculating its noise evaluation value; performing noise suppression processing on the smoothed temperature data according to the noise evaluation value to generate filtered temperature data; S13, obtaining ambient temperature data during the detection process, calculating the temperature fluctuation value, and obtaining ambient temperature fluctuation data; The ambient temperature fluctuation data at the corresponding moment is subtracted from the filtered temperature data to generate compensated temperature data.

3. The method for detecting the orientation of a PET film based on a thermal conductivity gradient according to claim 2, characterized in that: Step S2 is further as follows: S21, dividing the compensated temperature data according to the time series, calculating the temperature difference at adjacent moments; dividing the temperature difference by the corresponding time interval to generate temperature change rate data; S22, based on the compensated temperature data, calculating the temperature gradients in the horizontal and vertical directions to obtain horizontal and vertical secondary temperature characteristic data; combining the horizontal and vertical secondary temperature characteristic data to generate spatial temperature change data; S23, performing correlation analysis on the temperature change rate data and the spatial temperature change data to obtain initial heat diffusion data; performing feature extraction on the initial heat diffusion data at different positions and in different directions to generate heat diffusion feature data.

4. The method for detecting the orientation of a PET film based on a thermal conductivity gradient according to claim 3, characterized in that: Step S3 is further as follows: S31, dividing the heat diffusion characteristic data according to different spatial ranges and performing decomposition operations to obtain hierarchical decomposition data; integrating the hierarchical decomposition data of all regions to generate multi-scale characteristic data; S32, performing angle scanning on the multi-scale feature data at each level, calculating energy distribution features and extracting main directional features, and generating directional feature data; S33, evaluating the directional feature data according to the preset hierarchical importance to obtain hierarchical weight data; performing weighted calculation on the directional feature data and the corresponding hierarchical weight data to obtain weighted feature data; integrating the weighted feature data to generate comprehensive directional feature data.

5. The method for detecting the orientation of PET film based on thermal conductivity gradient according to claim 4, characterized in that: Step S4 is further as follows: S41, performing peak analysis and feature intensity sorting on the comprehensive directional feature data to obtain peak sorting data; determining the most significant directional feature based on the peak sorting data to generate main directional feature data; S42, taking the main direction feature data as the reference direction to obtain reference direction data; calculating the distribution characteristics of the comprehensive direction feature data relative to the reference direction data to obtain direction distribution data; calculating the orientation degree parameter based on the direction distribution data to generate orientation degree data.

6. The method for detecting the orientation of a PET film based on a thermal conductivity gradient according to claim 5, characterized in that: Step S5 is further as follows: S51, collecting compensated temperature data, thermal diffusion characteristic data, comprehensive direction characteristic data and orientation degree data, calculating statistical characteristics and reliability evaluation indexes, and generating reliability index data; S52, performing correlation analysis on the orientation degree data and the reliability index data to obtain correction coefficient data; performing correction calculation on the orientation degree data and the correction coefficient data to generate final orientation degree data.

7. The method for detecting the orientation of a PET film based on a thermal conductivity gradient according to claim 6, characterized in that: Step S11 is further as follows: S111, obtaining original temperature data of the surface of the PET film, performing validity check and eliminating abnormal values ​​to obtain pre-processed temperature data; performing data smoothing on the pre-processed temperature data to generate smoothed temperature data; S112, reading a pre-calibrated temperature sensor response curve, performing piecewise linearization processing, calculating correction parameters, and obtaining a temperature calibration coefficient; S113, matching the smoothed temperature data with the temperature calibration coefficient, performing temperature correction and quantization error compensation, and generating initial calibration data; S114, grouping and time-aligning the initial calibration data to obtain first aligned temperature data; Resampling the first aligned temperature data according to a preset sampling requirement to generate time calibration data; S115, converting the time calibration data into a spatial coordinate representation, performing spatial uniformity analysis and grid density optimization to obtain grid optimization data; performing spatial reconstruction based on the grid optimization data to generate calibrated temperature data.

8. The method for detecting the orientation of a PET film based on a thermal conductivity gradient according to claim 6, characterized in that: Step S12 is further as follows: S121, analyzing the time series characteristics of the calibrated temperature data, determining the optimal segment length, and generating segment temperature data; S122, extracting local statistical features of the segmented temperature data, calculating smoothing parameters and performing smoothing processing to generate smoothed temperature data; S123, performing comparative analysis and statistical analysis on the smoothed temperature data and the segmented temperature data, calculating the noise level, and generating a noise evaluation value; S124, based on the noise evaluation value, construct an adaptive threshold system and perform preliminary noise suppression on the smoothed temperature data to generate preliminary suppression data; S125, analyzing the residual noise of the preliminary suppression data, optimizing the suppression parameters and performing fine suppression processing to generate fine suppression data; S126 , evaluating the signal quality of the fine suppression data, performing final adjustment and edge feature preservation processing, and generating filtered temperature data.

9. The method for detecting the orientation of a PET film based on a thermal conductivity gradient according to claim 6, characterized in that: Step S21 is further as follows: S211, based on the compensated temperature data, performing time continuity check and time series completion to obtain completed temperature data; dividing the completed temperature data according to a preset period to generate time series temperature data; S212, analyzing sampling characteristics of the time series temperature data, identifying key time nodes and performing sequence alignment to generate second aligned temperature data; S213, performing difference calculation on the second aligned temperature data at adjacent moments, detecting abnormal values ​​and making corrections, and generating temperature difference data; S214, obtaining a timestamp sequence corresponding to the second aligned temperature data, calculating the time interval and performing uniformity analysis, and generating interval characteristic data; S215, matching the temperature difference data with the interval characteristic data to obtain matching difference data and standardizing the data to generate initial change rate data; S216, performing physical constraint check and change rate correction on the initial change rate data to obtain corrected change rate data; The corrected rate of change data is finally calibrated to generate the temperature rate of change data.

10. The method for detecting the orientation of a PET film based on a thermal conductivity gradient according to claim 6, characterized in that: Step S33 is further as follows: S331, performing feature consistency analysis and correlation analysis on the directional feature data at each level, evaluating the level stability, and generating level stability data; S332. Based on the hierarchical stability data, calculate and normalize the hierarchical discrimination coefficient, and generate hierarchical weight data in combination with a preset importance criterion; S333, grouping the directional feature data according to feature type, calculating the consistency index within the group, performing feature selection, and generating optimal feature data; S334, performing feature matching on the preferred feature data and the hierarchical weight data to obtain feature matching data; Perform weighted fusion and weight optimization on feature pairing data to generate weighted feature data; S335, analyzing the redundancy of the weighted feature data, performing feature simplification, and obtaining simplified feature data; performing feature enhancement processing on the simplified feature data to generate enhanced feature data; S336, performing multi-feature fusion operation and consistency verification on the enhanced feature data to obtain verification feature data; based on the verification feature data, performing final feature integration to obtain comprehensive directional feature data.

Images