A plastic pipe wall thickness testing method and device based on terahertz transmission

Abstract

The invention discloses a plastic pipe wall thickness testing method and device based on terahertz transmission, which relates to the field of terahertz transmission technology, including constructing a standard model; setting the frequency of a standard terahertz wave and the width of a beam, and constructing a standard setting array to obtain a plurality of setting arrays to be screened; a terahertz emission source respectively emits terahertz waves under a plurality of setting arrays to be screened, and obtains a plurality of thickness distribution diagrams with different degrees of precision; obtains signal-to-noise ratio influence data and measurement accuracy data when performing a plastic pipe wall thickness test; constructs a thickness distribution diagram resolution evaluation model based on a long-term and short-term neural network; and performs data analysis to obtain a screened setting array, and applies the screening setting array to the plastic pipe wall thickness test; the application solves the problem that the resolution setting of the thickness distribution diagram is not standardized, resulting in the inability to accurately capture subtle thickness changes while ensuring signal strength, thereby affecting the final test effect.

Description

Technical Field

[0001] The present invention relates to the field of terahertz transmission technology, and more specifically, to a method and device for testing the wall thickness of a plastic pipe based on terahertz transmission. Background Art

[0002] The terahertz transmission-based plastic pipe wall thickness test method is a non-contact measurement technology that uses terahertz waves. It is mainly used for online monitoring and quality control of wall thickness during the manufacturing process of plastic pipes. The uniformity of the wall thickness of plastic pipes directly affects their mechanical properties, sealing performance and long-term safety, so high-precision wall thickness measurement is crucial in the production process.

[0003] Terahertz waves are located in the far infrared to microwave region of the electromagnetic spectrum, with wavelengths typically ranging from 30 to 3000 micrometers. This allows them to penetrate most non-metallic materials (such as plastics) without damaging them.

[0004] For example, the invention patent with announcement number: CN112163365B announces an equivalent measurement method and device for the transmission characteristics of terahertz waves in a plasma sheath, including obtaining the electron density and distribution of the target plasma sheath by means of a Langmuir probe; simulating and modeling the target plasma sheath according to the electron density and distribution of the target plasma sheath, and using an equivalent plasma generated by high-voltage discharge to simulate the target plasma sheath; and equivalently measuring the transmission characteristics of terahertz waves in the target plasma sheath according to the transmission characteristics of terahertz waves in the equivalent plasma generated by high-voltage discharge. The present invention solves the technical bottleneck that the current terahertz wave energy is weak and it is impossible to directly obtain experimental data on the transmission characteristics of terahertz waves in the plasma sheath. The purpose of fully carrying out equivalent measurements of the transmission characteristics of terahertz waves in the plasma sheath is achieved.

[0005] The above disclosed technical solutions have at least the following technical problems:

[0006] When existing technologies are applied to plastic pipe wall thickness testing, the resolution of the thickness distribution map is crucial. The resolution determines the level of detail that the system can capture, which directly affects the accuracy of the pipe wall thickness and the reliability of the test.

[0007] If the resolution is higher, it is possible to accurately measure tiny thickness changes on the surface and inner layer of the pipe, making it easier to detect local defects or unevenness on the pipe, thereby providing more accurate quality control data.

[0008] Lower resolution may lead to neglect of slight thickness differences, thus affecting the overall measurement accuracy and even missing some hidden defect areas.

[0009] In the prior art, in the wall thickness test of plastic pipes, the resolution setting of the thickness distribution map is not standardized, which makes it impossible to accurately capture the subtle thickness changes while ensuring the signal strength, thus affecting the final test effect. In view of the above problems, the present invention proposes a solution. Summary of the invention

[0010] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a plastic pipe wall thickness testing method and device based on terahertz transmission, which analyzes different terahertz wave frequencies and beam widths to solve the problem that the resolution setting of the thickness distribution diagram is not standardized, resulting in the inability to accurately capture subtle thickness changes while ensuring signal strength, thereby affecting the final test effect.

[0011] To achieve the above object, the present invention provides the following technical solutions:

[0012] A method for testing the wall thickness of a plastic pipe based on terahertz transmission, comprising the following steps:

[0013] The terahertz emission source and the receiving detector are aligned and installed, the background transmission signal is measured in the absence of plastic pipes, the standard transmission signal is measured in the presence of standard plastic pipes with known thickness and material, and a standard model is constructed; according to historical data, the frequency of the standard terahertz wave and the width of the beam are set, and a standard setting array is constructed, and the standard setting set is subjected to several random number processings to obtain several setting arrays to be screened; the terahertz emission source respectively emits several terahertz waves under the setting arrays to be screened, passes through the plastic pipes to be tested, and the receiving detector receives several terahertz transmission signals for signal processing and reconstruction to obtain several thickness distribution maps with different degrees of precision, and the thickness is reversely calculated based on the standard model; the signal-to-noise ratio influence data and the measurement accuracy data when testing the wall thickness of the plastic pipe are obtained; according to the signal-to-noise ratio influence data and the measurement accuracy data, a thickness distribution map resolution evaluation model is constructed based on a long-term and short-term neural network; according to the output of the thickness distribution map resolution evaluation model and the corresponding setting array to be screened, a resolution-test effect change diagram is constructed based on a Cartesian three-dimensional display model, and data analysis is performed to obtain the screened setting array, which is applied to the plastic pipe wall thickness test.

[0014] In a preferred embodiment, the construction of the standard model is specifically as follows: collecting transmission data at multiple frequency points from the transmission signal of a standard plastic pipe and recording changes in signal intensity; filtering and denoising the collected transmission signal; fitting the standard signal data based on linear regression to obtain a mathematical equation describing the relationship between the pipe thickness and the transmission signal as the standard model.

[0015] In a preferred embodiment, the standard setting array includes a standard terahertz wave frequency, a standard terahertz wave beam width, a beam width range and a frequency range; the setting array to be screened includes a beam width and a terahertz wave frequency; the standard setting set is subjected to several random number processings to obtain several setting arrays to be screened, specifically: according to the frequency range in the standard setting set, a random number generation algorithm is used to generate multiple different terahertz wave frequencies within the range; according to the beam width range in the standard setting set, a random number generation method is used to generate multiple different beam widths; the terahertz wave frequency and beam width random values ​​are combined to obtain several setting arrays to be screened.

[0016] In a preferred embodiment, the signal-to-noise ratio influencing data includes a signal-to-noise ratio influencing coefficient; the specific method for obtaining the signal-to-noise ratio influencing coefficient is as follows: obtaining the signal strength data during the plastic pipe wall thickness test, the signal strength data including the signal strength, the background noise intensity and the target signal strength, and calculating the average value of the signal-to-noise ratio during the test according to the signal strength data; calculating the signal-noise influence factor based on a preset signal-noise influence factor calculation formula according to the average value of the signal-to-noise ratio; calculating the signal stability influence factor based on a preset signal stability influence factor calculation formula combined with the average value of the signal-to-noise ratio; calculating the signal-to-noise ratio influence coefficient according to the ratio of the signal-noise influence factor and the signal stability influence factor during the plastic pipe wall thickness test.

[0017] In a preferred embodiment, the measurement accuracy data includes a detail loss assessment coefficient and a measurement accuracy influence coefficient; the specific method for obtaining the measurement accuracy influence coefficient is as follows: based on historical data, a set of influencing events of the pipe wall thickness test under different setting arrays is obtained, and the set of influencing events of the pipe wall thickness test under different setting arrays includes various influencing events caused by test accuracy, as well as event scores; based on the set of influencing events of the pipe wall thickness test under different setting arrays and the corresponding measurement accuracy, a regression model is constructed to generate the measurement accuracy influence coefficient.

[0018] In a preferred embodiment, the specific method for obtaining the detail loss assessment coefficient is as follows: obtain the resolution within several unit areas when testing the wall thickness of the plastic pipe; calculate the resolution standard deviation and the resolution average according to the resolution within several unit areas when testing the wall thickness of the plastic pipe; calculate the resolution variation coefficient within the unit area according to the resolution standard deviation and the resolution average; calculate the detail loss assessment coefficient based on the resolution variation coefficient within the unit area based on a preset detail loss assessment coefficient calculation formula.

[0019] In a preferred embodiment, the resolution-test effect variation graph is constructed based on the Cartesian three-dimensional display model according to the output of the thickness distribution map resolution evaluation model and the corresponding setting array to be screened, specifically: in the Cartesian three-dimensional display model, the frequency and beam width of the terahertz wave in the setting array to be screened are selected as the X-axis and Y-axis respectively, and the output of the corresponding thickness distribution map resolution evaluation model is used as the Z-axis to construct a three-dimensional coordinate system; the frequency, beam width and corresponding output of the thickness distribution map resolution evaluation model obtained from each setting array to be screened are plotted in the three-dimensional coordinate system, and the blanks between each data point are processed with a smooth surface to obtain a resolution-test effect variation graph.

[0020] The technical effects and advantages of the terahertz transmission-based plastic pipe wall thickness testing method and device of the present invention are as follows:

[0021] 1. The present invention realizes high-precision detection of the wall thickness of plastic pipes by constructing a standard model, optimizing the frequency and beam width, and finely controlling the signal-to-noise ratio and measurement accuracy. This method collects transmission signals at multiple frequency points, establishes a mathematical model in combination with linear regression analysis, and infers the wall thickness of unknown pipes. In addition, by analyzing the signal-to-noise ratio and the measurement accuracy influence coefficient, the relationship between resolution and signal strength is effectively balanced, the accuracy and efficiency in the test process are optimized, and stable and accurate wall thickness data can be obtained in different scenarios. This technology not only improves the detection accuracy, but also can respond flexibly in complex environments, providing a reliable technical guarantee for pipe quality inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The present invention is a schematic flow chart of a method for testing the wall thickness of a plastic pipe based on terahertz transmission. DETAILED DESCRIPTION

[0023] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0024] Embodiment 1, Figure 1 The present invention provides a plastic pipe wall thickness testing method based on terahertz transmission, comprising the following steps:

[0025] S1, align and install the terahertz emission source and the receiving detector, measure the background transmission signal without plastic pipes, measure the standard transmission signal with standard plastic pipes of known thickness and material, and build a standard model;

[0026] The construction of the standard model is specifically as follows:

[0027] Collect transmission data at multiple frequency points from the transmission signal of a standard plastic pipe and record changes in signal intensity;

[0028] Filtering and denoising the collected transmission signal;

[0029] The standard signal data was fitted based on linear regression, and the mathematical equation describing the relationship between the pipe thickness and the transmission signal was obtained as the standard model.

[0030] It should be noted that when collecting the transmission signal of standard plastic pipes, multiple frequency points are selected for collection in order to cover the transmission characteristics of terahertz waves in pipes of different frequency bands, ensuring that the physical properties of the pipes can be fully captured. Transmission data at different frequency points can help improve the accuracy of the model, especially for pipes of different thicknesses and materials, the transmittance of terahertz waves at different frequencies will vary.

[0031] When fitting the signal data based on linear regression, it is assumed that there is a certain linear relationship between the pipe thickness and the transmission signal. The fitting equation obtained by calculation can reflect the mathematical relationship between the thickness of the standard pipe and the corresponding transmission signal intensity. After the standard model is established, the thickness of the unknown pipe can be inferred through the model.

[0032] This standard model not only provides a theoretical basis for subsequent thickness inverse calculations, but also provides a prediction reference for a transmission signal for pipes of different thicknesses, so that the actual thickness of the pipe can be determined more accurately based on the transmission signal in subsequent tests.

[0033] It should be noted that the construction of the standard model generally includes collecting transmission signal data of pipes of different thicknesses and materials at specific frequency points through experiments, processing the raw data using filtering and denoising techniques, and then establishing a mathematical relationship between the signal and thickness through regression analysis (such as linear regression or nonlinear regression). After verification, the model can be used as a reference to estimate the wall thickness of unknown pipes. Due to the mature signal processing and regression analysis algorithms, this technology has been widely used in terahertz transmission testing and will not be repeated here.

[0034] S2, according to historical data, set the frequency of the standard terahertz wave and the width of the beam, and construct a standard setting array, perform random number processing on the standard setting set several times, and obtain several setting arrays to be screened;

[0035] The standard setting array includes a standard terahertz wave frequency, a standard terahertz wave beam width, a beam width range, and a frequency range;

[0036] The array of settings to be screened includes a beam width and a terahertz wave frequency;

[0037] The standard setting set is subjected to several random number processings to obtain several arrays of settings to be screened, specifically:

[0038] According to the standard setting of a centralized frequency range, a random number generation algorithm is used to generate a plurality of different terahertz wave frequencies within the range;

[0039] According to the standard setting of the concentrated beam width range, a plurality of different beam widths are generated using a random number generation method;

[0040] The frequency of the terahertz wave and the random values ​​of the width of the beam are combined to obtain several arrays of settings to be screened.

[0041] S3, the terahertz emission source emits several terahertz waves under the array to be screened, passes through the plastic pipe to be tested, and the receiving detector receives several terahertz transmission signals for signal processing and reconstruction, obtains several thickness distribution maps with different degrees of precision, and performs reverse calculation of thickness based on the standard model;

[0042] The receiving detector receives a plurality of terahertz transmission signals for signal processing and reconstruction, and obtains a plurality of thickness distribution maps with different degrees of precision, specifically:

[0043] Based on the wall thickness data inferred from the standard model, the local thickness data is converted into a thickness distribution map through an algorithm. For different frequency and beam width settings, thickness distribution maps with different resolutions are obtained.

[0044] It should be noted that when using terahertz waves of different frequencies for measurement, higher frequency terahertz waves have higher spatial resolution, because high frequency waves have shorter wavelengths and can carefully detect tiny structural changes on the surface and inside of the pipe. This makes the generated thickness distribution map more precise and can reflect the microscopic thickness differences of the pipe. Lower frequency waves have longer wavelengths and can penetrate thicker materials, but their spatial resolution is lower, and the generated distribution map will be relatively rough, making it difficult to accurately locate tiny thickness changes.

[0045] When the beam width is narrow, the focusing ability of the terahertz wave is enhanced, and a small area on the surface of the tube can be detected more accurately, producing a high-resolution thickness distribution map. This means that the thickness changes in the map will be more obvious, and the wall thickness differences in local areas can be accurately revealed. When the beam width is wide, the area covered by the beam increases, resulting in a decrease in spatial resolution, a decrease in the fineness of the thickness distribution map, and the possibility of blurring out some small-area thickness changes.

[0046] When high frequency and narrow beam are used together, it can provide higher spatial resolution while more accurately measuring thickness changes in local areas. This combination is suitable for applications that require high-precision and refined measurements, especially when detecting tiny defects or uneven thickness, and can provide accurate data support.

[0047] Combination of low frequency and wide beam: When low frequency and wide beam are used together, although the measurement resolution is low, it can better penetrate thicker pipes and effectively measure a wider range of pipe thickness. This combination is suitable for more uniform pipe inspection, especially when the thickness does not vary much or the structure is relatively simple, which can reduce measurement time and calculation complexity.

[0048] There is already mature existing technology for the process of performing reverse calculation of thickness based on the standard model, receiving a plurality of terahertz transmission signals by the receiving detector for signal processing and reconstruction, and obtaining a plurality of thickness distribution maps with different degrees of precision, which will not be described in detail here.

[0049] S4, obtaining signal-to-noise ratio influence data and measurement accuracy data when testing the wall thickness of plastic pipes;

[0050] The signal-to-noise ratio impact data includes a signal-to-noise ratio impact coefficient; the measurement accuracy data includes a detail loss assessment coefficient and a measurement accuracy impact coefficient.

[0051] The signal-to-noise ratio impact factor is used to quantify the impact of changes in the signal-to-noise ratio during the measurement process at different resolutions on the test results. In the case of high resolution, due to the smaller area of ​​data collected, the signal may be weaker and the noise interference is more significant, thus affecting the accuracy and reliability of the measurement results. Low-resolution measurements usually collect signals from a larger area. Relatively speaking, the signal-to-noise ratio is higher, the noise interference is smaller, and the data is more stable, which is suitable for use in scenarios where accuracy requirements are not extreme. By evaluating this balance, the signal-to-noise ratio impact factor helps optimize the resolution selection to ensure that good measurement accuracy and stability are maintained while ensuring signal strength.

[0052] Analyzing the signal-to-noise ratio influence coefficient has the following advantages for evaluating the test results under different resolutions of thickness distribution maps and solving the problem that the resolution setting of thickness distribution maps is not standardized, resulting in the inability to accurately capture subtle thickness changes while ensuring signal strength, thus affecting the final test results:

[0053] Balance signal strength and resolution: By evaluating the signal-to-noise ratio, the choice of test resolution can be optimized to ensure that subtle thickness changes can be accurately captured while ensuring signal strength, thereby avoiding the loss of measurement accuracy due to weak signals.

[0054] Improve test reliability: Under high resolution, although a more detailed thickness distribution map can be obtained, the reliability of the measurement results may be affected due to the large noise interference. The signal-to-noise ratio influence factor can help determine the appropriate resolution, reduce the interference of noise on the measurement results, and improve the accuracy of the test results.

[0055] Optimize test strategy: This coefficient can provide a basis for the resolution selection during the test process. By setting the resolution reasonably, the test can meet the accuracy requirements without over-relying on the additional noise interference brought by higher resolution, thereby improving the overall measurement accuracy and stability.

[0056] Flexible adaptation to different application scenarios: The signal-to-noise ratio influence coefficient allows you to select the appropriate resolution based on actual conditions under different application requirements, avoiding performance loss caused by too high or too low resolution, and ensuring that the test process is efficient and accurate.

[0057] The specific method for obtaining the signal-to-noise ratio influence coefficient is as follows:

[0058] Acquire signal strength data during the plastic pipe wall thickness test, wherein the signal strength data includes signal strength, background noise strength, and target signal strength, and calculate an average value of the signal-to-noise ratio during the test according to the signal strength data;

[0059] Calculate the signal-noise impact factor based on the average value of the signal-to-noise ratio based on a preset signal-noise impact factor calculation formula;

[0060] The signal stability influence factor is calculated according to the preset signal stability influence factor calculation formula combined with the average value of the signal-to-noise ratio;

[0061] The signal-to-noise ratio influence coefficient is calculated according to the ratio of the signal noise influence factor and the signal stability influence factor in the process of plastic pipe wall thickness testing.

[0062] The signal noise impact factor calculation formula is specifically:

[0063]

[0064] The calculation formula of the signal stability influencing factor is specifically:

[0065]

[0066] The specific calculation formula of the signal-to-noise ratio influence coefficient is as follows:

[0067]

[0068] In the formula, is the signal noise impact factor, is the signal stability influencing factor, is the average value of the signal-to-noise ratio, is the total number of measurement points during the i-th wall thickness measurement, is the signal-to-noise ratio influence coefficient, is the starting measurement value of the i-th wall thickness test, is the target measurement value of the i-th wall thickness test, is the measurement depth of the i-th test device, is the time required for the i-th test.

[0069] The measurement accuracy impact coefficient is used to reflect the balance between measurement speed and data processing complexity. Since the low-resolution thickness distribution map collects a larger area, it can be measured in a shorter time, thereby improving the measurement accuracy. It is particularly suitable for large-scale production lines or rapid inspection scenarios. However, low resolution also means that the data is relatively rough and may not reflect small changes or local defects. Relatively speaking, although high-resolution measurements can provide more detailed thickness information, the measurement accuracy is reduced because more detailed data needs to be processed, the data volume increases, the computing resource requirements increase, and the analysis process becomes more complicated. The measurement accuracy impact coefficient helps quantify this trade-off between speed and accuracy and provides a basis for selecting the appropriate resolution.

[0070] Analyzing the measurement accuracy influence coefficient has the following advantages for evaluating the test results under different resolutions of the thickness distribution map and solving the problem that the resolution setting of the thickness distribution map is not standardized, resulting in the inability to accurately capture subtle thickness changes while ensuring signal strength, thus affecting the final test results:

[0071] Optimize the balance between speed and accuracy: The measurement accuracy influence coefficient helps find the best balance between high resolution and low resolution. It ensures that the thickness distribution information can be quickly obtained while meeting the production or test requirements, and the computing resources are reasonably allocated when higher accuracy is required. By controlling the resolution setting, while ensuring the signal strength, it can also accurately capture the slight thickness changes, avoiding unnecessary complexity and long processing time caused by too high resolution.

[0072] Improve the efficiency of large-scale inspection: In large-scale production or batch inspection, low-resolution measurement can significantly improve the test accuracy, especially when the thickness distribution in a large area is quickly scanned during the inspection process. The measurement accuracy influence coefficient can effectively improve the overall production accuracy, reduce the test time, and provide efficient real-time data support for the inspection system.

[0073] Reduce data processing pressure: Although high resolution provides detailed thickness data, it also increases the burden of data processing and storage. The measurement accuracy impact coefficient helps select the appropriate resolution by evaluating the data processing complexity at different resolutions to avoid excessive consumption of computing resources and long processing time, making the system more operational in practical applications.

[0074] Improve operational adaptability: In different application scenarios, the measurement accuracy impact coefficient can flexibly adjust the resolution according to the measurement requirements. For scenarios with rapid production inspection or limited inspection equipment resources, lower resolution can effectively improve work accuracy. For scenarios that require high-precision inspection, the measurement accuracy impact coefficient can guide how to balance accuracy and efficiency through reasonable resolution configuration and data processing strategies.

[0075] By comprehensively analyzing the coefficients affecting measurement accuracy, we can effectively avoid test errors caused by unreasonable resolution settings that lead to the inability to simultaneously ensure signal strength and capture subtle changes, thereby improving the accuracy and operability of the overall test results.

[0076] The specific method for obtaining the measurement accuracy influence coefficient is as follows:

[0077] According to the historical data, a set of influencing events of the pipe wall thickness test under different setting arrays is obtained, wherein the set of influencing events of the pipe wall thickness test under different setting arrays includes various influencing events caused by the test accuracy and event scores;

[0078] Based on the influencing event set of pipe wall thickness test under different setting arrays and the corresponding measurement accuracy, a regression model is constructed to generate the measurement accuracy influence coefficient.

[0079] The specific calculation formula of the measurement accuracy influence coefficient is as follows:

[0080]

[0081] In the formula, is the measurement accuracy influence coefficient, … The preset measurement accuracy corresponding to the influencing events of the pipe wall thickness test under different setting arrays, … is the proportion of the influencing events of the pipe wall thickness test under different setting arrays among all the influencing events of the pipe wall thickness test under different setting arrays. is the accuracy of the influencing event of the pipe wall thickness test under different setting arrays, n is the number of the influencing event of the pipe wall thickness test under different setting arrays, and e is a natural number.

[0082] The detail loss assessment coefficient is used to quantify the degree of detail loss caused by the inability to accurately capture small changes in pipe wall thickness under low resolution conditions, especially when there are local defects on the surface or inside the pipe. Low resolution may not accurately reflect the true structure of the pipe, miss small changes or defects, and thus affect the accuracy of the measurement results and the reliability of quality assessment. This coefficient helps evaluate the test effect of the thickness distribution map at different resolutions, ensuring that key details are not lost when optimizing the resolution, thereby ensuring the accuracy of the pipe wall thickness test.

[0083] Analyzing the detail loss evaluation coefficient has the following advantages for evaluating the test results under different resolutions of thickness distribution maps and solving the problem that the resolution setting of thickness distribution maps is not standardized, resulting in the inability to accurately capture subtle thickness changes while ensuring signal strength, thus affecting the final test results:

[0084] Quantifying Loss of Detail: The Loss of Detail Assessment Factor can quantify the loss of detail caused by low resolution, especially when identifying small changes in pipe wall thickness, helping to identify potential problems with resolution settings.

[0085] Identify the problem of insufficient resolution: Low resolution often cannot capture local defects and tiny unevenness on the surface or inside of the pipe. The detail loss assessment coefficient can clarify this problem and provide a basis for resolution optimization.

[0086] Optimize test accuracy: Through quantitative analysis of detail loss, the detail loss evaluation factor can help determine the most suitable resolution setting, thereby ensuring that the test can accurately capture the slightest changes in pipe wall thickness.

[0087] Improve test accuracy and reliability: The detail loss assessment factor helps to obtain more accurate test results during the test process and improve the overall quality of the test by accurately evaluating the impact of low resolution.

[0088] Avoiding loss of precision: The detail loss assessment factor prevents loss of precision due to improper resolution settings, ensuring that small changes in wall thickness are accurately captured while maintaining signal strength, avoiding erroneous assessments due to low resolution.

[0089] The specific method for obtaining the detail loss assessment coefficient is as follows:

[0090] Obtain the resolution within a certain unit area when testing the wall thickness of plastic pipes;

[0091] According to the resolution within a certain unit area during the plastic pipe wall thickness test, the resolution standard deviation and the resolution average are calculated;

[0092] The resolution coefficient of variation within a unit area is calculated based on the resolution standard deviation and the resolution mean;

[0093] The resolution variation coefficient within a unit area is used to calculate the detail loss assessment coefficient based on a preset detail loss assessment coefficient calculation formula.

[0094] The specific calculation formula of the resolution variation coefficient within the unit area is as follows:

[0095]

[0096] The specific calculation formula of the detail loss evaluation coefficient is as follows:

[0097]

[0098] In the formula, is the coefficient of variation of resolution per unit area, is the resolution within the jth unit area, The total number of unit resolutions collected during the plastic pipe wall thickness test; is the detail loss assessment coefficient.

[0099] This embodiment achieves high-precision detection of the wall thickness of plastic pipes by constructing a standard model, optimizing the frequency and beam width, and finely controlling the signal-to-noise ratio and measurement accuracy. This method collects transmission signals at multiple frequency points, establishes a mathematical model in combination with linear regression analysis, and infers the wall thickness of unknown pipes. In addition, by analyzing the signal-to-noise ratio and the measurement accuracy influence coefficient, the relationship between resolution and signal strength is effectively balanced, the accuracy and efficiency in the test process are optimized, and stable and accurate wall thickness data can be obtained in different scenarios. This technology not only improves the detection accuracy, but also can respond flexibly in complex environments, providing a reliable technical guarantee for pipe quality inspection.

[0100] Embodiment 2, S5, constructing a thickness distribution map resolution evaluation model based on a long-term and short-term neural network according to the signal-to-noise ratio impact data and the measurement accuracy data;

[0101] The thickness distribution map resolution evaluation model is constructed based on the long-term and short-term neural network according to the signal-to-noise ratio impact data and the measurement accuracy data, specifically:

[0102] The obtained signal-to-noise ratio influence coefficient, measurement accuracy influence coefficient and detail loss evaluation coefficient are used to construct a thickness distribution map resolution evaluation model, and generate a thickness distribution map resolution evaluation coefficient;

[0103] The specific calculation formula of the thickness distribution map resolution evaluation coefficient is as follows:

[0104]

[0105] In the formula, is the thickness distribution map resolution evaluation coefficient, is the preset proportional coefficient of the signal-to-noise ratio influence coefficient, It is the preset proportionality coefficient of the measurement accuracy influence coefficient. is the preset scaling factor for the detail loss evaluation factor, is the signal-to-noise ratio influence coefficient, is the measurement accuracy influence coefficient, is the detail loss assessment coefficient.

[0106] S6, constructing a resolution-test effect variation diagram based on the Cartesian three-dimensional display model according to the output of the thickness distribution map resolution evaluation model and the corresponding setting array to be screened, and performing data analysis to obtain a screened setting array, which is applied to the plastic pipe wall thickness test.

[0107] The resolution-test effect change graph is constructed based on the Cartesian three-dimensional display model according to the output of the thickness distribution graph resolution evaluation model and the corresponding array of settings to be screened, specifically:

[0108] In the Cartesian three-dimensional display model, the frequency of the terahertz wave and the width of the beam in the array to be screened are selected as the X-axis and Y-axis respectively, and the output of the corresponding thickness distribution map resolution evaluation model is used as the Z-axis to construct a three-dimensional coordinate system;

[0109] The output of the resolution evaluation model of the frequency, beam width and corresponding thickness distribution map obtained from each array of settings to be screened is plotted in a three-dimensional coordinate system, and the blanks between each data point are processed with a smooth surface to obtain a resolution-test effect change graph.

[0110] The data analysis is performed to obtain a screened setting array, which is applied to the plastic pipe wall thickness test, specifically:

[0111] Acquire multiple peaks of the three-dimensional surface in the resolution-test effect variation diagram, construct a set of local optimal points, compare the stability between multiple peaks based on the rolling window averaging method, and use the setting array corresponding to the stable local optimal point as the screened setting array and apply it to the plastic pipe wall thickness test.

[0112] This embodiment achieves high-precision detection of the wall thickness of plastic pipes by constructing a standard model, optimizing the frequency and beam width, and finely controlling the signal-to-noise ratio and measurement accuracy. This method collects transmission signals at multiple frequency points, establishes a mathematical model in combination with linear regression analysis, and infers the wall thickness of unknown pipes. In addition, by analyzing the signal-to-noise ratio and the measurement accuracy influence coefficient, the relationship between resolution and signal strength is effectively balanced, the accuracy and efficiency in the test process are optimized, and stable and accurate wall thickness data can be obtained in different scenarios. This technology not only improves the detection accuracy, but also can respond flexibly in complex environments, providing a reliable technical guarantee for pipe quality inspection.

[0113] Embodiment 3, a device using a terahertz transmission-based plastic pipe wall thickness testing method, comprising: a standard model building module, a to-be-screened setting array acquisition module, a thickness distribution map acquisition module, a data analysis module, a thickness distribution map resolution evaluation module, and a setting array screening module:

[0114] A standard model building module, used to align and install the terahertz emission source and the receiving detector, measure the background transmission signal in the absence of plastic pipes, measure the standard transmission signal in the presence of standard plastic pipes with known thickness and material, and build a standard model;

[0115] The module for obtaining the array of settings to be screened is used to set the frequency and beam width of the standard terahertz wave according to historical data, and to construct a standard setting array, and to perform random number processing on the standard setting set several times to obtain several arrays of settings to be screened;

[0116] The thickness distribution map acquisition module is used for the terahertz emission source to respectively emit several terahertz waves under the array to be screened, pass through the plastic pipe to be tested, and the receiving detector receives several terahertz transmission signals for signal processing and reconstruction, so as to obtain several thickness distribution maps with different degrees of precision, and perform reverse calculation of thickness based on the standard model;

[0117] Data analysis module, used to obtain signal-to-noise ratio impact data and measurement accuracy data when testing the wall thickness of plastic pipes;

[0118] A thickness distribution map resolution evaluation module is used to construct a thickness distribution map resolution evaluation model based on a long-term and short-term neural network according to signal-to-noise ratio impact data and measurement accuracy data;

[0119] The setting array screening module is used to construct a resolution-test effect change diagram based on the Cartesian three-dimensional display model according to the output of the thickness distribution map resolution evaluation model and the corresponding setting array to be screened, and perform data analysis to obtain the screened setting array and apply it to the plastic pipe wall thickness test.

[0120] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0121] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0122] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0123] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0124] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0125] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A plastic pipe wall thickness testing method based on terahertz transmission, characterized in that: The steps include: Align and install the terahertz emission source and the receiving detector, measure the background transmission signal in the absence of plastic pipes, measure the standard transmission signal in the presence of standard plastic pipes with known thickness and material, and build a standard model; According to historical data, the frequency of the standard terahertz wave and the width of the beam are set, and a standard setting array is constructed. The standard setting set is subjected to several random number processings to obtain several setting arrays to be screened. The terahertz emission source emits several terahertz waves under the array to be screened, which pass through the plastic pipe to be tested. The receiving detector receives several terahertz transmission signals for signal processing and reconstruction, and obtains several thickness distribution maps with different degrees of precision, and performs reverse calculation of thickness based on the standard model. Obtain data on the impact of signal-to-noise ratio and measurement accuracy when testing the wall thickness of plastic pipes; According to the signal-to-noise ratio impact data and measurement accuracy data, a thickness distribution map resolution evaluation model is constructed based on long-term and short-term neural networks; According to the output of the thickness distribution map resolution evaluation model and the corresponding setting array to be screened, a resolution-test effect change graph is constructed based on the Cartesian three-dimensional display model, and data analysis is performed to obtain the screened setting array, which is then applied to the plastic pipe wall thickness test.

2. A plastic pipe wall thickness testing method based on terahertz transmission according to claim 1, characterized in that: The construction of the standard model is specifically as follows: Collect transmission data at multiple frequency points from the transmission signal of a standard plastic pipe and record changes in signal intensity; Filtering and denoising the collected transmission signal; The standard signal data was fitted based on linear regression, and the mathematical equation describing the relationship between the pipe thickness and the transmission signal was obtained as the standard model.

3. A plastic pipe wall thickness testing method based on terahertz transmission according to claim 2, characterized in that: The standard setting array includes a standard terahertz wave frequency, a standard terahertz wave beam width, a beam width range, and a frequency range; The array of settings to be screened includes a beam width and a terahertz wave frequency; The standard setting set is subjected to several random number processings to obtain several arrays of settings to be screened, specifically: According to the standard setting of a centralized frequency range, a random number generation algorithm is used to generate a plurality of different terahertz wave frequencies within the range; According to the standard setting of the concentrated beam width range, a plurality of different beam widths are generated using a random number generation method; The frequency of the terahertz wave and the random values ​​of the width of the beam are combined to obtain several arrays of settings to be screened.

4. The method for testing the wall thickness of a plastic pipe based on terahertz transmission according to claim 3, characterized in that: The signal-to-noise ratio impact data includes a signal-to-noise ratio impact coefficient; a specific method for obtaining the signal-to-noise ratio impact coefficient is as follows: Acquire signal strength data during the plastic pipe wall thickness test, wherein the signal strength data includes signal strength, background noise strength, and target signal strength, and calculate an average value of the signal-to-noise ratio during the test according to the signal strength data; Calculate the signal-noise impact factor based on the average value of the signal-to-noise ratio based on a preset signal-noise impact factor calculation formula; The signal stability influence factor is calculated according to the preset signal stability influence factor calculation formula combined with the average value of the signal-to-noise ratio; The signal-to-noise ratio influence coefficient is calculated according to the ratio of the signal noise influence factor and the signal stability influence factor in the process of plastic pipe wall thickness testing.

5. The method for testing the wall thickness of a plastic pipe based on terahertz transmission according to claim 4, characterized in that: The measurement accuracy data includes a detail loss assessment coefficient and a measurement accuracy influence coefficient; the specific method for obtaining the measurement accuracy influence coefficient is as follows: According to the historical data, a set of influencing events of the pipe wall thickness test under different setting arrays is obtained, wherein the set of influencing events of the pipe wall thickness test under different setting arrays includes various influencing events caused by the test accuracy and event scores; Based on the influencing event set of pipe wall thickness test under different setting arrays and the corresponding measurement accuracy, a regression model is constructed to generate the measurement accuracy influence coefficient.

6. A plastic pipe wall thickness testing method based on terahertz transmission according to claim 5, characterized in that: The specific method for obtaining the detail loss assessment coefficient is as follows: Obtain the resolution within a certain unit area when testing the wall thickness of plastic pipes; According to the resolution within a certain unit area during the plastic pipe wall thickness test, the resolution standard deviation and the resolution average are calculated; The resolution coefficient of variation within a unit area is calculated based on the resolution standard deviation and the resolution mean; The resolution variation coefficient within a unit area is used to calculate the detail loss assessment coefficient based on a preset detail loss assessment coefficient calculation formula.

7. The method for testing the wall thickness of a plastic pipe based on terahertz transmission according to claim 6, characterized in that: The resolution-test effect change graph is constructed based on the Cartesian three-dimensional display model according to the output of the thickness distribution graph resolution evaluation model and the corresponding array of settings to be screened, specifically: In the Cartesian three-dimensional display model, the frequency of the terahertz wave and the width of the beam in the array to be screened are selected as the X-axis and Y-axis respectively, and the output of the corresponding thickness distribution map resolution evaluation model is used as the Z-axis to construct a three-dimensional coordinate system; The output of the resolution evaluation model of the frequency, beam width and corresponding thickness distribution map obtained from each array of settings to be screened is plotted in a three-dimensional coordinate system, and the blanks between each data point are processed with a smooth surface to obtain a resolution-test effect change graph.

8. The method for testing the wall thickness of a plastic pipe based on terahertz transmission according to claim 7, characterized in that: The data analysis is performed to obtain a screened setting array, which is applied to the plastic pipe wall thickness test, specifically: Acquire multiple peaks of the three-dimensional surface in the resolution-test effect variation diagram, construct a set of local optimal points, compare the stability between multiple peaks based on the rolling window averaging method, and use the setting array corresponding to the stable local optimal point as the screened setting array and apply it to the plastic pipe wall thickness test.

9. The method for testing the wall thickness of a plastic pipe based on terahertz transmission according to claim 8, characterized in that: The signal noise impact factor calculation formula is specifically: The calculation formula of the signal stability influencing factor is specifically: The specific calculation formula of the signal-to-noise ratio influence coefficient is as follows: In the formula, is the signal noise impact factor, is the signal stability influencing factor, is the average value of the signal-to-noise ratio, is the total number of measurement points during the i-th wall thickness measurement, is the signal-to-noise ratio influence coefficient, is the starting measurement value of the i-th wall thickness test, is the target measurement value of the i-th wall thickness test, is the measurement depth of the i-th test device, is the time required for the i-th test.

10. A device using the terahertz transmission-based plastic pipe wall thickness testing method as described in any one of claims 1 to 9, characterized in that: include: Standard model building module, to-be-screened setting array acquisition module, thickness distribution map acquisition module, data analysis module, thickness distribution map resolution evaluation module, and setting array screening module: A standard model building module, used to align and install the terahertz emission source and the receiving detector, measure the background transmission signal in the absence of plastic pipes, measure the standard transmission signal in the presence of standard plastic pipes with known thickness and material, and build a standard model; The module for obtaining the array of settings to be screened is used to set the frequency and beam width of the standard terahertz wave according to historical data, and to construct a standard setting array, and to perform random number processing on the standard setting set several times to obtain several arrays of settings to be screened; The thickness distribution map acquisition module is used for the terahertz emission source to respectively emit several terahertz waves under the array to be screened, pass through the plastic pipe to be tested, and the receiving detector receives several terahertz transmission signals for signal processing and reconstruction, so as to obtain several thickness distribution maps with different degrees of precision, and perform reverse calculation of thickness based on the standard model; Data analysis module, used to obtain signal-to-noise ratio impact data and measurement accuracy data when testing the wall thickness of plastic pipes; A thickness distribution map resolution evaluation module is used to construct a thickness distribution map resolution evaluation model based on a long-term and short-term neural network according to signal-to-noise ratio impact data and measurement accuracy data; The setting array screening module is used to construct a resolution-test effect change diagram based on the Cartesian three-dimensional display model according to the output of the thickness distribution map resolution evaluation model and the corresponding setting array to be screened, and perform data analysis to obtain the screened setting array and apply it to the plastic pipe wall thickness test.

Images