Non-contact continuous detection system and method for surface roughness of cable joint

Through contactless optical detection system and machine learning algorithm, the contact damage and detection discontinuity of cable joint surface roughness detection are solved, and the accurate and rapid detection of cable joint surface roughness is achieved, and the detection accuracy and automation are improved.

CN119085541APending Publication Date: 2024-12-06STATE GRID TIANJIN ELECTRIC POWER COMPANY +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411210992.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-30
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The existing cable joint surface roughness detection methods have problems of contact damage and discontinuous detection, resulting in damage to the insulation layer, reducing cable life and safety risks.

Method used

Using contactless optical detection system, the reflectivity and transmittance of the cable surface is measured using light sources, semi-transparent half-mirror and photosensitive sensors, combined with a screw drive motor to achieve continuous measurement along the cable axial direction, and through optical calculations and machine learning algorithms Identify roughness abnormalities.

Benefits of technology

It realizes accurate and fast contact-free detection of the surface roughness of the cable joint, reduces measurement errors and surface wear, avoids detection omissions, and improves detection accuracy and automation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119085541A_ABST
    Figure CN119085541A_ABST
Patent Text Reader

Abstract

The invention relates to a non-contact detection system and method for processing roughness of a cable joint. The non-contact detection system comprises an optical detection unit, a motion unit and a light shield, the optical detection unit comprises a light source, a semi-transparent and semi-reflecting mirror, a first photosensitive sensor and a second photosensitive sensor; the motion unit comprises a base, a lead screw driving motor, a sliding rail, a lead screw and a sliding block. According to the invention, the surface roughness of the cable joint can be accurately and rapidly evaluated.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of power cable joint detection, and relates to a contactless detection system for cable joints, in particular to a contactless continuous detection system and method for the surface roughness of cable joints. Background Art

[0002] In the process of cable manufacturing, processing and handling, the treatment of cable joints is a core and extremely critical step. After the cable is stripped, grinding the surface of the insulation layer is a very important step. The roughness of the surface of the insulation layer after grinding has become an extremely important testing standard.

[0003] At present, the existing method for detecting the roughness of the surface of the polished cable joint is mainly to perform contact measurement using instruments manually. This measurement method mainly relies on the contact of the probe with the surface of the insulation layer of the tested cable, and the probe moves up and down along the concave and convex surface through linear motion. The probe displacement data is then transmitted to the cable joint roughness detection device to obtain the detection result of the cable joint surface roughness.

[0004] However, for the surface of the cable insulation layer, direct contact between the probe and it may scratch the insulation surface and because the radius of the stylus tip is very small, generally made of diamond with a diameter of 0.1 to 10 μm, the pressure on the high-voltage cable insulation layer can reach a considerable value, resulting in scratches when the probe contacts the surface of the cable connector being tested.

[0005] It can be seen that contact testing will damage the surface of the cable insulation layer during testing. If the test is not done properly, the surface of the cable connector will be damaged, which may cause excessive partial discharge, accelerate the aging of the insulation layer, and reduce the service life of the cable.

[0006] Especially for high-voltage cable joints, such defects may lead to reduced insulation performance, electric field distortion, heat generation due to moisture, fire and explosion, frequent cable failures, and even cable breakdown and rupture, which will seriously affect the safety of the power grid. These consequences not only threaten the safe and stable operation of the power system, but also affect the production and life safety of the general public, and may even cause more extensive safety accidents.

[0007] In addition, during the cable processing, the roughness of the entire cable joint surface needs to be evaluated. Since the traditional cable joint surface contact detection method will slide on the surface of the cable joint, it will not only damage the insulation layer, but also the detection process generally targets a single point for detection. When performing multi-point detection, manual operation is required to switch the detection target. The switching process is complicated and time-consuming. The distance between the two detection positions is large. If there is an abnormal roughness point between the two detection points, it is easy to miss it, resulting in an incorrect assessment of the cable surface roughness. Therefore, there is still a lack of effective means to continuously measure the roughness of the entire cable joint.

[0008] Therefore, how to find a non-contact measurement method that does not damage the surface of the cable connector and can continuously measure the roughness of the entire cable connector is a technical difficulty that needs to be solved urgently by technical personnel in this field.

[0009] After searching, no public documents of the prior art that are identical or similar to the present invention were found. Summary of the invention

[0010] The purpose of the present invention is to overcome the deficiencies of the prior art and to provide a non-contact continuous detection system and method for the surface roughness of a cable connector, which can achieve accurate and rapid evaluation of the surface roughness of a cable connector.

[0011] The present invention solves the practical problem by adopting the following technical solutions:

[0012] A non-contact detection system for cable joint processing roughness, comprising: an optical detection unit, a motion unit and a light shield;

[0013] The optical detection unit comprises: a light source, a semi-transparent and semi-reflective mirror, a first photosensor, and a second photosensor;

[0014] The optical detection unit is located above the cable connector to be tested; wherein the light source is directed toward the point to be tested on the surface of the cable connector to be tested; the semi-transparent and semi-reflective mirror is located between the light source and the point to be tested, and is composed of two triangular glasses and a semi-transparent and semi-reflective film in the middle; the two photosensors are respectively located on the left and right sides of the semi-transparent and semi-reflective mirror; wherein the first reflected light L1 of the semi-transparent and semi-reflective mirror is directly incident on the first photosensor, and the light L2 reflected by the semi-transparent and semi-reflective mirror is incident on the second photosensor;

[0015] The motion unit comprises: a base, a screw drive motor, a slide rail, a screw and a slider; the screw drive motor and the slide rail are mounted on the base, the screw drive motor drives the screw to rotate, and provides a translational force for the slider on the slide rail, and a bottom plate is mounted on the slider;

[0016] The light source, the semi-transparent and semi-reflective mirror, the first photosensor, and the second photosensor are arranged on a base plate; the light source is located on the upper part of the base plate, a semi-transparent and semi-reflective mirror is arranged below the light source, and the first photosensor and the second photosensor are symmetrically arranged on the left and right sides of the semi-transparent and semi-reflective mirror respectively; a cable connector to be tested is arranged below the semi-transparent and semi-reflective mirror.

[0017] Moreover, the semi-transparent and semi-reflective mirror is installed at the midpoint of the line connecting the first photosensor and the second photosensor, and the plane of the semi-transparent and semi-reflective mirror forms an angle of 45° with the connecting line of the two photosensors.

[0018] Moreover, the light shield is located outside the movement unit and the optical detection unit.

[0019] A non-contact detection method for cable joint processing roughness comprises the following steps:

[0020] Step 1, the light source emits light, and after passing through the semi-transparent and semi-reflective mirror, it is separated into two parts; one part cannot pass through the semi-transparent and semi-reflective mirror after starting from the light source, and is reflected by the reflective surface of the semi-transparent and semi-reflective mirror and emitted to form light L1, which is projected onto the first photosensor; the other part is transmitted through the semi-transparent and semi-reflective mirror to form light L2, which is projected onto the surface of the cable connector to be tested. Due to the concave and convex changes of the cable connector to be tested, different reflected light intensities are generated, and then reflected by the semi-transparent and semi-reflective mirror and emitted into the first photosensor.

[0021] Step 2: Use two photosensors to measure the intensity of the two beams of light respectively, and then measure the reflectivity data of the cable surface at this time, and then calculate the roughness at this location;

[0022] Step 3: At the same time, the driving motor drives the lead screw to rotate, driving the optical detection unit to move along the cable axis, thereby measuring the roughness measurement data of the axis of the cable connector to be tested;

[0023] Step 4: Analyze the frequency domain of the roughness measurement data of the axis of the cable joint to be tested measured in step 3 to obtain the surface roughness anomaly recognition result of the cable joint in the axial direction.

[0024] Step 5: After completing the roughness measurement of the cable axis, rotate the detection instrument along the cable axis through manual operation. After rotation, repeat steps 1 to 4 to obtain the roughness detection condition around the entire cable joint.

[0025] Moreover, the specific steps of step 2 include:

[0026] (1) The roughness is calculated by using the signal division method, with the signal of the first photosensor as the denominator and the signal of the second photosensor as the numerator;

[0027] (2) Quantitatively measure roughness using light intensity, using the Lambertian reflection model, where the intensity of reflected light is proportional to the intensity of incident light and is independent of the observation angle;

[0028] (3) Assume that I0 is the light intensity emitted directly from the light source, I τ and I t are the light intensities after reflection and transmission through the semi-transparent and semi-reflective mirrors, respectively. Assuming the surface roughness of the cable connector is R, the reflected light intensity I τ Related to R;

[0029] I τ It is expressed as the following simplified formula:

[0030] I τ =I0×f(R)#(1)

[0031] Among them, f(R) is a function related to the surface roughness R, and I0 is a constant term. For the above model, f(R) can be expressed as:

[0032] f(R)=ρ×(1-R)#(2)

[0033] Here, ρ is the reflection coefficient, which is usually less than 1.

[0034] Since the intensity of the transmitted light is inversely proportional to the intensity of the reflected light, let T be the transmission coefficient (T = 1-ρ), then:

[0035] I t =T×I0×(1-R)#(3)

[0036] (4) Due to the transmitted light I t The intensity of is inversely proportional to the intensity of the reflected light I, so the roughness can be quantitatively calculated using the following formula:

[0037]

[0038] By applying the above model and simplified formula, the single-point roughness value at the light-irradiated location is calculated using the data sent back by the photosensors on both sides of the semi-transparent and semi-reflective mirror.

[0039] Moreover, the specific method of step 3 is:

[0040] At the same time, the lead screw motor drives the lead screw to rotate, and with the cooperation of the slider rail mechanism, drives the optical detection unit composed of the light source, semi-transparent and semi-reflective mirror, and photosensor on the bottom plate to move along the cable axis; repeat steps 1 and 2 to obtain the roughness measurement data in the cable axis direction.

[0041] Moreover, the specific steps of step 4 include:

[0042] (1) Analyze the frequency domain of the measured roughness measurement data and decompose it into a high-frequency signal (a signal with a drastic change in roughness) and a low-frequency signal (a signal with a gentle change in roughness);

[0043] Finite impulse response filter FIR is directly used for filtering during decomposition;

[0044] For a sequence x(n) of length N, the output y(k) of the FIR filter is as follows:

[0045]

[0046] By adjusting the filter coefficients, the passband cutoff frequency and the stopband cutoff frequency of the FIR filter are changed, thereby obtaining the required bandpass filter, and using the corresponding bandpass filter to process the obtained data to obtain the high-frequency part and the low-frequency part;

[0047] (2) For high-frequency signals, obtain their power spectrum density and use machine learning algorithms to analyze them;

[0048] When obtaining the power spectral density, it is necessary to process the data using Fourier transform;

[0049] Its general expression is:

[0050]

[0051] By accumulating a large amount of data, the identification thresholds corresponding to different cable signals are determined. When the power exceeds the threshold, it is considered that roughness abnormality occurs at that location.

[0052] (3) For low-frequency signals, calculate the fluctuation of their amplitude with distance. Since the amplitude of the optical signal caused by the distance change is small, it is reflected as low-frequency fluctuation in the frequency domain. The change of the cable surface roughness in the axial direction can be inferred through the amplitude fluctuation of the low-frequency signal, thereby obtaining the shape data of the cable surface.

[0053] Advantages and beneficial effects of the present invention:

[0054] 1. The present invention proposes a non-contact continuous detection system and method for the roughness of cable joints, which utilizes optical principles such as interference, diffraction, and refraction of light to achieve non-contact detection of surface roughness without damaging the outer surface of the cable joint, thereby reducing measurement errors and surface wear of objects caused by contact.

[0055] 2. The present invention uses a lead screw drive motor to drive the optical detection unit to move along the axial direction of the cable, thereby realizing continuous measurement of the surface roughness along the axial direction of the cable, and avoiding erroneous quality assessment caused by missing abnormal roughness points that may be generated by single-point measurement during cable defect detection. In addition, non-contact detection also avoids damage or contamination to the surface of the cable connector caused by contact detection, and has a higher degree of automation. Continuous numerical detection can be achieved through optical calculation methods, which increases the amount of data in the detection process, is more friendly to the later analysis using computer algorithms, and is easier to improve detection accuracy and speed from the algorithm level than manual measurement methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of the system structure of the present invention;

[0057] Figure 2 is a process flow chart of the present invention;

[0058] Figure 3 A comparison diagram of roughness change curves obtained by two detection methods in a specific embodiment of the present invention;

[0059] Figure 4 It is a schematic diagram of the roughness of the cable connector surface obtained after rotation in a specific embodiment of the present invention.

[0060] Description of reference numerals:

[0061] 1-screw drive motor; 2-slide rail; 3-screw; 4-slider; 5-base; 6-bottom plate; 7-light source; 8-first photosensor; 9-semi-transparent and semi-reflective mirror; 10-cable connector to be tested; 11-second photosensor; 12-light shield. DETAILED DESCRIPTION

[0062] The embodiments of the present invention are further described in detail below with reference to the accompanying drawings:

[0063] A non-contact detection system for the roughness of cable joint processing, such as Figure 1 As shown, it includes: an optical detection unit, a motion unit and a light shield;

[0064] The optical detection unit comprises: a light source, a semi-transparent and semi-reflective mirror, a first photosensor, and a second photosensor;

[0065] The optical detection unit is located above the cable connector to be tested; wherein the light source is directed toward the point to be tested on the surface of the cable connector to be tested; the semi-transparent and semi-reflective mirror is located between the light source and the point to be tested, and is composed of two triangular glasses and a semi-transparent and semi-reflective film in the middle; the two photosensors are respectively located on the left and right sides of the semi-transparent and semi-reflective mirror; wherein the first reflected light L1 of the semi-transparent and semi-reflective mirror is directly incident on the first photosensor, and the light L2 reflected by the semi-transparent and semi-reflective mirror is incident on the second photosensor;

[0066] In this embodiment, the two beams of light will produce a light intensity difference due to the change in the roughness of the cable surface. By observing and analyzing the numerical value between the light intensity difference and the roughness, their mapping relationship can be inferred, thereby realizing the detection of the cable surface roughness by optical methods.

[0067] The motion unit comprises: a base, a screw drive motor, a slide rail, a screw and a slider; the screw drive motor and the slide rail are mounted on the base, the screw drive motor drives the screw to rotate, and provides a translational force for the slider on the slide rail, and a bottom plate is mounted on the slider;

[0068] The light source, the semi-transparent and semi-reflective mirror, the first photosensor, and the second photosensor are arranged on the bottom plate; the light source is located on the upper part of the bottom plate, a semi-transparent and semi-reflective mirror is arranged below the light source, and the first photosensor and the second photosensor are symmetrically arranged on the left and right sides of the semi-transparent and semi-reflective mirror respectively; a cable connector to be tested is arranged below the semi-transparent and semi-reflective mirror;

[0069] In this embodiment, the semi-transparent and semi-reflective mirror is installed at the midpoint of the line connecting the first photosensor and the second photosensor, and the plane of the semi-transparent and semi-reflective mirror forms a 45° angle with the connecting line of the two photosensors. This ensures that the light emitted by the light source is reflected and vertically enters the photosensor in the shortest path.

[0070] When in use, the motion unit drives the optical detection unit to move horizontally along the axial direction of the cable connector to be tested, and the lead screw driving motor drives the lead screw to rotate, thereby driving the slide block to move left and right under the constraint of the slide rail.

[0071] In this embodiment, the light shield is located outside the moving unit and the optical detection unit to ensure that when the moving unit moves, the light source, the semi-transparent mirror, the two photosensors and the detection point of the cable to be detected can always be shielded.

[0072] The working process of the system of the present invention is:

[0073] 1. Install the cable connector to be tested on the base, and install and adjust the light shield to ensure that it can effectively shield the external light source.

[0074] 2. Turn on the system switch, set the detection parameters, and run the equipment;

[0075] 3. The light source emits light, and the first photosensor and the second photosensor record the collected light intensity data

[0076] 4. The motor drives the lead screw to rotate, thereby driving the base plate and the optical detection unit installed on the base plate to move along the axial direction of the cable connector to be tested.

[0077] 5. When moving along the axial direction, the optical detection unit records the light intensity data collected by the two photosensors in real time;

[0078] 6. After moving to the preset length, the screw drive motor and the optical detection unit stop moving.

[0079] 7. Rotate the device along the axis of the cable connector to be tested and repeat the above process to test again.

[0080] The functions of each component are further described below:

[0081] The driving motor and transmission structure are used to drive the light source to move horizontally along the axial direction of the cable joint to be tested; the light source, semi-transparent and semi-reflective mirror and photosensor are used to detect the surface roughness of the cable; the light shield is used to protect the internal device and block the external light source to avoid affecting the detection result.

[0082] In addition, specific algorithms are needed to analyze and process the results obtained by the sensor in order to optimize the detection results and reduce false identification.

[0083] A non-contact detection method for the roughness of cable joint processing, such as Figure 2 As shown, the following steps are included:

[0084] Step 1, the light source emits light, and after passing through the semi-transparent and semi-reflective mirror, it is separated into two parts; one part cannot pass through the semi-transparent and semi-reflective mirror after starting from the light source, and is reflected by the reflective surface of the semi-transparent and semi-reflective mirror and emitted to form light L1, which is projected onto the first photosensor; the other part is transmitted through the semi-transparent and semi-reflective mirror to form light L2, which is projected onto the surface of the cable connector to be tested. Due to the concave and convex changes of the cable connector to be tested, different reflected light intensities are generated, and then reflected by the semi-transparent and semi-reflective mirror and emitted into the first photosensor.

[0085] Step 2: Use two photosensors to measure the intensity of the two beams of light respectively, and then measure the reflectivity data of the cable surface at this time, and then calculate the roughness at this location;

[0086] The two photosensors are used to receive the light reflected by the semi-transparent and semi-reflective mirrors, measure the intensity of the light reflected by the semi-transparent and semi-reflective mirrors, generate voltage signals, transmit them to the host computer for comparison, and calculate the roughness.

[0087] The specific steps of step 2 include:

[0088] (1) The roughness is measured by using the signal division method, with the signal of the first photosensor as the denominator and the signal of the second photosensor as the numerator, so as to shield the influence of light on the current.

[0089] (2) Quantitatively measure roughness using light intensity, using the Lambertian reflection model, where the intensity of reflected light is proportional to the intensity of incident light and is independent of the observation angle;

[0090] (2) Assume that I0 is the light intensity emitted directly from the light source, I τ and I t are the light intensities after reflection and transmission through the semi-transparent and semi-reflective mirrors, respectively. Assuming the surface roughness of the cable connector is R, the reflected light intensity I τ Related to R;

[0091] I τ It can be expressed as the following simplified formula:

[0092] I τ =I0×f(R)#(1)

[0093] Among them, f(R) is a function related to the surface roughness R, and I0 is a constant term. For the above model, f(R) can be expressed as:

[0094] f(R)=ρ×(1-R)#(2)

[0095] Here, ρ is the reflection coefficient, which is usually less than 1.

[0096] Since the intensity of the transmitted light is inversely proportional to the intensity of the reflected light, let T be the transmission coefficient (T = 1-ρ), then:

[0097] I t =T×I0×(1-R)#(3)

[0098] (4) Due to the transmitted light I t The intensity of the reflected light I τ The strength is inversely proportional to the roughness, so the following formula can be used to quantitatively calculate the roughness:

[0099]

[0100] By applying the above model and simplified formula, the roughness value of a single point where light is irradiated can be calculated using the data sent back by the photosensors on both sides of the semi-transparent and semi-reflective mirror.

[0101] Step 3: At the same time, the driving motor drives the lead screw to rotate, driving the optical detection unit to move along the cable axis, thereby measuring the roughness measurement data of the axis of the cable connector to be tested;

[0102] The specific method of step 3 is:

[0103] At the same time, the lead screw motor drives the lead screw to rotate, and with the cooperation of the slider rail mechanism, drives the optical detection unit composed of the light source, semi-transparent and semi-reflective mirror, and photosensor on the bottom plate to move along the cable axis; repeat steps 1 and 2 to obtain the roughness measurement data in the cable axis direction.

[0104] Step 4: Analyze the frequency domain of the roughness measurement data of the axis of the cable joint to be tested measured in step 3 to obtain the surface roughness abnormality recognition result in the axial direction of the cable joint;

[0105] The specific method of step 4 is:

[0106] The frequency domain of the measured roughness measurement data is analyzed and decomposed into high-frequency signals (signals with drastic roughness changes) and low-frequency signals (signals with gentle roughness changes);

[0107] In this embodiment, the frequency domain of the roughness measurement data of the axis of the cable connector to be tested is related to the axial movement speed of the detection device. Taking 1cm / s as an example, a signal higher than 10Hz is considered a high-frequency signal;

[0108] Finite Impulse Response (FIR) filters are used directly for filtering during decomposition. For a sequence x(n) of length N, the output y(k) of the FIR filter is as follows:

[0109]

[0110] By adjusting the filter coefficients, the passband cutoff frequency and stopband cutoff frequency of the FIR filter can be changed to obtain the required bandpass filter. The high-frequency part and the low-frequency part can be obtained by processing the obtained data with the corresponding bandpass filter.

[0111] For high-frequency signals, obtain their power spectrum density and use machine learning algorithms to analyze them;

[0112] When obtaining the power spectrum density, it is necessary to use Fourier transform to process the data. Its general expression is:

[0113]

[0114] Through the accumulation of a large amount of data, the identification threshold corresponding to different cable signals is determined. When the power exceeds the threshold, it is considered that the roughness is abnormal.

[0115] In this embodiment, the threshold is determined by the average value obtained by multiple measurements of the standard cable joint. For the cable joint to be tested of the 110kV high-voltage cable, it is generally 8μm. After obtaining the power spectrum density, the spectral power at this position is greater than 0.6, which is considered to be abnormal roughness, indicating the presence of pits or protrusions.

[0116] For low-frequency signals, the fluctuation of its amplitude with distance is calculated. Since the change of the optical signal caused by the distance change is small, it is reflected as low-frequency fluctuation in the frequency domain. The change of the cable surface roughness in the axial direction can be inferred through the amplitude fluctuation of the low-frequency signal, thus obtaining the appearance data of the cable surface.

[0117] The working principle of step 4 is:

[0118] Since the cable cannot be strictly straight during production, and the cable surface will be uneven during stripping, which will cause the distance between the cable surface and the detection device to change. This distance change will cause the measured data to change. Therefore, it is necessary to distinguish the impact caused by the distance change from the impact caused by the roughness change.

[0119] Unqualified roughness will cause the measured signal to change in frequency at a faster rate, while the signal frequency caused by the distance change is slower. To solve this problem, the measured signal is analyzed in the frequency domain using the above formula (8). Therefore, the measured signal is decomposed into a group of high-frequency signals and low-frequency signals for subsequent analysis. Among them, the high- and low-frequency signal thresholds are related to the axial movement speed of the detection equipment. Taking 1cm / s as an example, a signal higher than 10Hz is considered a high-frequency signal.

[0120] Step 5: After completing the roughness measurement of the cable axis, rotate the detection instrument along the cable axis through manual operation. After rotation, repeat steps 1 to 4 to obtain the roughness detection condition around the entire cable joint.

[0121] The present invention is further described below by means of specific examples:

[0122] The following experiment demonstrates the difference between the present invention and the traditional detection method:

[0123] A cable connector is selected, the length of which is greater than 100 mm. After cleaning, a 100 mm portion is selected for testing. The present invention is compared with a conventional contact roughness testing instrument.

[0124] The parameters of the traditional roughness tester are as follows:

[0125] Table 1 Parameters of commonly used roughness testers

[0126] Measuring range 50μm Measuring stroke 6mm Guide head pressure 0.5N Sampling length 0.25mm Measurement error ≤10%

[0127] Figure 3 The roughness change curves obtained by the two detection methods are shown, where the two points with asterisks are the locations where the roughness data is abnormal. As can be seen from the figure, the detection accuracy and resolution of this device are higher. By realizing automated roughness detection along the cable axis, a higher number of sampling points is achieved compared to manual detection. The traditional contact detection method failed to detect the two abnormal points due to insufficient resolution.

[0128] After completing the measurement of the axial data, the detection instrument is rotated along the cable axis through manual operation. After rotation, steps 1 to 4 are repeated to obtain the roughness detection condition around the entire cable joint.

[0129] Since the cable cannot be strictly straight during production, and the cable surface will be uneven during stripping, which will cause the distance between the cable surface and the detection device to change. This distance change will cause the measured data to change. Therefore, it is necessary to distinguish the impact caused by the distance change from the impact caused by the roughness change.

[0130] To solve this problem, the measured signal is analyzed in the frequency domain. Unqualified roughness will cause the measured signal to have a faster frequency change, while the signal frequency caused by the distance change is slower. Therefore, the measured signal is decomposed into a group of high-frequency signals and low-frequency signals for subsequent analysis.

[0131] For high-frequency signals, the power spectrum density is calculated and analyzed using a machine learning algorithm. By accumulating a large amount of data, the identification thresholds corresponding to different cable signals are determined. When the power exceeds the threshold, it is considered that the roughness is abnormal.

[0132] For low-frequency signals, the light measurement intensity will fluctuate slightly in the cable axis as the automatic measurement proceeds. The amplitude fluctuation can represent the change in roughness. By continuously obtaining the amplitude data, the continuously changing value of the roughness can be obtained, and then a graph of the overall roughness change of the cable connector surface can be drawn.

[0133] Figure 4 The visualization of the overall roughness change diagram formed by continuous testing of the cable joint to be tested through this example is shown, where the horizontal axis "axial position" represents the change in the axial direction of the cable joint to be tested as the motor moves, the vertical axis "rotation angle" represents the roughness change on different axes after manual rotation at different angles, and the vertical axis "surface roughness" represents the roughness of the cable joint to be tested calculated from the light intensity change. Through the overall change diagram, the abnormal roughness points of the cable joint to be tested can be clearly found, which is more conducive to the rapid evaluation and correction of roughness during the cable joint manufacturing process.

[0134] The working principle of the present invention is:

[0135] The light source emits light, which passes through a semi-transparent and semi-reflective mirror to form two beams of light, one of which is directly reflected and the other is projected onto the cable surface and then reflected again. Two photosensitive sensors measure the intensity of the two beams of light respectively, calculate the reflectivity data of the cable surface, perform frequency domain analysis on the measured signals, distinguish the signal changes caused by roughness and distance changes, and quantitatively calculate the roughness of the cable surface through the light intensity data.

[0136] It should be emphasized that the embodiments of the present invention are illustrative rather than restrictive. Therefore, the present invention includes but is not limited to the embodiments described in the specific implementation modes. Any other implementation modes derived by those skilled in the art based on the technical solutions of the present invention also fall within the scope of protection of the present invention.

Claims

1. A non-contact detection system for cable joint processing roughness, characterized by: include: Optical detection unit, motion unit and light shield; The optical detection unit comprises: a light source, a semi-transparent and semi-reflective mirror, a first photosensor, and a second photosensor; The optical detection unit is located above the cable connector to be tested; wherein the light source is directed toward the point to be tested on the surface of the cable connector to be tested; the semi-transparent and semi-reflective mirror is located between the light source and the point to be tested, and is composed of two triangular glasses and a semi-transparent and semi-reflective film in the middle; the two photosensors are respectively located on the left and right sides of the semi-transparent and semi-reflective mirror; wherein the first reflected light L1 of the semi-transparent and semi-reflective mirror is directly incident on the first photosensor, and the light L2 reflected by the semi-transparent and semi-reflective mirror is incident on the second photosensor; The motion unit comprises: a base, a screw drive motor, a slide rail, a screw and a slider; the screw drive motor and the slide rail are mounted on the base, the screw drive motor drives the screw to rotate, and provides a translational force for the slider on the slide rail, and a bottom plate is mounted on the slider; The light source, the semi-transparent and semi-reflective mirror, the first photosensor, and the second photosensor are arranged on a base plate; the light source is located on the upper part of the base plate, a semi-transparent and semi-reflective mirror is arranged below the light source, and the first photosensor and the second photosensor are symmetrically arranged on the left and right sides of the semi-transparent and semi-reflective mirror respectively; a cable connector to be tested is arranged below the semi-transparent and semi-reflective mirror.

2. A non-contact detection system for cable joint processing roughness according to claim 1, characterized in that: The semi-transparent and semi-reflective mirror is installed at the midpoint of the line connecting the first photosensor and the second photosensor, and the plane of the semi-transparent and semi-reflective mirror forms an angle of 45° with the connecting line of the two photosensors.

3. A non-contact detection system for cable joint processing roughness according to claim 1, characterized in that: The light shield is located outside the movement unit and the optical detection unit.

4. The contactless detection method of the contactless detection system for cable joint processing roughness according to claims 1-3, characterized in that: The following steps are involved: Step 1, the light source emits light, and after passing through the semi-transparent and semi-reflective mirror, it is separated into two parts; one part cannot pass through the semi-transparent and semi-reflective mirror after starting from the light source, and is reflected by the reflective surface of the semi-transparent and semi-reflective mirror and emitted to form light L1, which is projected to the first photosensor, and the other part is transmitted through the semi-transparent and semi-reflective mirror to form light L2, which is projected to the surface of the cable connector to be tested. Due to the concave-convex changes of the cable connector to be tested, different reflected light intensities are generated, and then reflected by the semi-transparent and semi-reflective mirror and emitted into the first photosensor; Step 2: Use two photosensors to measure the intensity of the two beams of light respectively, and then measure the reflectivity data of the cable surface at this time, and then calculate the roughness at this location; Step 3: At the same time, the driving motor drives the lead screw to rotate, driving the optical detection unit to move along the cable axis, thereby measuring the roughness measurement data of the axis of the cable connector to be tested; Step 4: Analyze the frequency domain of the roughness measurement data of the axis of the cable joint to be tested measured in step 3 to obtain the surface roughness abnormality recognition result in the axial direction of the cable joint; Step 5: After completing the roughness measurement of the cable axis, rotate the detection instrument along the cable axis through manual operation. After rotation, repeat steps 1 to 4 to obtain the roughness detection condition around the entire cable joint.

5. The contactless detection method of the contactless detection system for cable joint processing roughness according to claim 4, characterized in that: The specific steps of step 2 include: (1) The roughness is calculated by using the signal division method, with the signal of the first photosensor as the denominator and the signal of the second photosensor as the numerator; (2) Quantitatively measure roughness using light intensity, using the Lambertian reflection model, where the intensity of reflected light is proportional to the intensity of incident light and is independent of the observation angle; (3) Assume that I0 is the light intensity emitted directly from the light source, I τ and I t are the light intensities after reflection and transmission through the semi-transparent and semi-reflective mirrors, respectively. Assuming the surface roughness of the cable connector is R, the reflected light intensity I τ Related to R; I τ It is expressed as the following simplified formula: I τ =I0×f(R)#(1) Among them, f(R) is a function related to the surface roughness R, and I0 is a constant term; for the above model, f(R) can be expressed as: f(R)=ρ×(1-R)#(2) Where ρ is the reflection coefficient, which is usually less than 1; Since the intensity of the transmitted light is inversely proportional to the intensity of the reflected light, let T be the transmission coefficient (T = 1-ρ), then: I t =T×I0×(1-R)#(3) (4) Due to the transmitted light I t The intensity of the reflected light I τ The strength is inversely proportional to the roughness, so the following formula can be used to quantitatively calculate the roughness: By applying the above model and simplified formula, the single-point roughness value at the light-irradiated location is calculated using the data sent back by the photosensors on both sides of the semi-transparent and semi-reflective mirror.

6. The non-contact detection method of the non-contact detection system for cable joint processing roughness according to claim 4, characterized in that: The specific method of step 3 is: At the same time, the lead screw motor drives the lead screw to rotate, and with the cooperation of the slider rail mechanism, drives the optical detection unit composed of the light source, semi-transparent and semi-reflective mirror, and photosensor on the bottom plate to move along the cable axis; repeat steps 1 and 2 to obtain the roughness measurement data in the cable axis direction.

7. The non-contact detection method of the non-contact detection system for cable joint processing roughness according to claim 4, characterized in that: The specific steps of step 4 include: (1) Analyze the frequency domain of the measured roughness measurement data and decompose it into a high-frequency signal (a signal with a drastic change in roughness) and a low-frequency signal (a signal with a gentle change in roughness); Finite impulse response filter FIR is directly used for filtering during decomposition; For a sequence x(n) of length N, the output y(k) of the FIR filter is as follows: By adjusting the filter coefficients, the passband cutoff frequency and the stopband cutoff frequency of the FIR filter are changed, thereby obtaining the required bandpass filter, and using the corresponding bandpass filter to process the obtained data to obtain the high-frequency part and the low-frequency part; (2) For high-frequency signals, obtain their power spectrum density and use machine learning algorithms to analyze them; When obtaining the power spectral density, it is necessary to process the data using Fourier transform; Its general expression is: By accumulating a large amount of data, the identification thresholds corresponding to different cable signals are determined. When the power exceeds the threshold, it is considered that roughness abnormality occurs at that location. (3) For low-frequency signals, calculate the fluctuation of their amplitude with distance. Since the amplitude of the optical signal caused by the distance change is small, it is reflected as low-frequency fluctuation in the frequency domain. The change of the cable surface roughness in the axial direction can be inferred through the amplitude fluctuation of the low-frequency signal, thereby obtaining the shape data of the cable surface.