Loose coil detection device

Abstract

This invention relates to the field of coil breakage detection technology and discloses a Rousselt coil breakage detection device, comprising a high-frequency excitation injection mechanism, an induction detection mechanism, a mechanical scanning mechanism, a signal processing mechanism, and an embedded system mechanism. The high-frequency excitation injection mechanism injects a high-frequency excitation signal into the main circuit of the coil under test, utilizing the skin effect to induce a local impedance change at the breakage point. The mechanical scanning mechanism, equipped with the induction detection mechanism, scans along the coil surface, picking up the spatial leakage magnetic field distortion signal excited by the impedance change, and simultaneously outputs a mechanical displacement pulse. The signal processing mechanism performs digital filtering and synchronous demodulation on the picked-up signal, extracting orthogonal characteristic components. The embedded system mechanism calculates the impedance phase difference value accordingly. When the difference value exceeds a set threshold, the absolute physical coordinates of the breakage point are calculated in conjunction with the mechanical displacement pulse. This invention achieves high-sensitivity, high-precision, non-contact spatial positioning of internal breakage defects in Rousselt coils.

Description

Technical Field

[0001] This invention relates to the field of coil breakage detection technology, specifically a Rotary coil breakage detection device. Background Technology

[0002] As a coreless current sensor, the Root coil is widely used in transient current monitoring and protection devices in power systems due to its wide measurement range, absence of magnetic saturation, and fast response speed. However, during actual manufacturing and long-term operation, the extremely fine enameled wire inside the Root coil is prone to localized strand breakage or hidden wire breakage damage due to the combined effects of winding tension, mechanical vibration, and thermal stress. This microscopic physical structural damage can disrupt the uniformity of current distribution inside the coil, thereby affecting the accuracy and stability of current measurement under actual operating conditions.

[0003] Currently, the main methods for detecting the structural integrity of Roots coils rely on measuring the overall DC resistance or low-frequency AC impedance. Since Roots coils are typically composed of thousands of turns of fine wire, the change in DC resistance caused by a small break in a local conductor is usually at the micro-ohm level. In actual testing, this extremely small change in resistance is easily masked by the thermal drift caused by ambient temperature fluctuations and the background noise of the testing equipment. This makes it difficult for conventional low-frequency testing methods to effectively identify hidden early breakage defects. At the same time, conventional overall electrical parameter testing can only reflect the macroscopic conduction state of the overall coil circuit and cannot obtain the specific spatial physical coordinates of the defect point on the entire coil. This makes it difficult to obtain accurate data support for subsequent targeted repairs and manufacturing process improvements.

[0004] To highlight microscopic defects, some detection schemes attempt to introduce high-frequency signals. However, the physical form of the long stroke of the Rotor coil requires the detection equipment to have spatial scanning capabilities. In complex detection environments, weak local high-frequency leakage magnetic characteristic signals are not only easily submerged by scattered alternating electromagnetic waves in space and conducted common-mode noise in electrical circuits, but are also interfered with by the jitter of the mechanical scanning device itself. Existing devices lack effective physical isolation and digital noise reduction mechanisms in the weak spatial magnetic field pickup and high-frequency characteristic demodulation stages, making it difficult to stably extract distorted phases under strong background noise, thus severely restricting the detection sensitivity and spatial positioning accuracy of wire break defects. Summary of the Invention

[0005] The purpose of this invention is to provide a Root coil breakage detection device, which solves the problem that existing Root coil breakage detection technologies are unable to effectively extract weak local strand distortion signals under strong background electromagnetic noise and mechanical scanning jitter, resulting in low detection sensitivity for internal micro-Eohm level damage and the inability to achieve high-precision breakage point coordinate positioning.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] The roc coil breakage detection device includes:

[0008] The high-frequency excitation injection mechanism 10 is connected to both ends of the main circuit of the rotund coil under test, and is used to inject a high-frequency AC excitation signal into both ends of the main circuit of the rotund coil under test;

[0009] The induction detection mechanism 20 is used to pick up the spatial alternating leakage magnetic field distortion signal caused by the local impedance change outside the Roche coil to be tested, and output the corresponding original induced electromotive force signal.

[0010] The mechanical scanning mechanism 30 is used to carry the sensing and detection mechanism 20 to scan and move along the surface of the rotatable coil to be tested, and output a mechanical displacement pulse signal corresponding to the scanning displacement in real time;

[0011] The signal processing mechanism 40 is connected to the induction detection mechanism 20 and is used to receive the original induced electromotive force signal and perform digital filtering and synchronous demodulation operations to extract in-phase characteristic components and quadrature characteristic components.

[0012] The embedded system mechanism 50 is communicatively connected to the high-frequency excitation injection mechanism 10, the mechanical scanning mechanism 30, and the signal processing mechanism 40, respectively, for receiving the in-phase characteristic component and the quadrature characteristic component, calculating the transient phase angle and the impedance phase difference value; and when the impedance phase difference value exceeds a set threshold, calculating the absolute physical coordinates of the breakpoint based on the number of mechanical displacement pulses generated by the accumulated mechanical displacement pulse signal.

[0013] Preferably, the high-frequency excitation injection mechanism 10 includes a direct digital frequency synthesis unit, a power amplification unit, and an impedance matching network. The direct digital frequency synthesis unit is used to generate an initial high-frequency sinusoidal AC signal according to the control command. The high-frequency sinusoidal AC signal is amplified by the power amplification unit and the impedance is balanced by the impedance matching network before being output as the high-frequency AC excitation signal.

[0014] Preferably, the sensing detection mechanism 20 employs a non-contact differential high-frequency sensing probe, comprising two miniature receiving coils with opposite winding directions, identical geometric dimensions, and arranged parallel to each other along the axial direction of the measured Roche coil. The two miniature receiving coils are connected in reverse series in the electrical circuit to physically cancel common-mode interference signals in the space environment through the differential structure, and output the original induced electromotive force signal containing the distorted phase of the alternating leakage magnetic field in space. The original induced electromotive force signal is the algebraic sum of the transient induced voltages of the two miniature receiving coils.

[0015] Preferably, the mechanical scanning mechanism 30 includes a servo drive motor, a ball screw, a Hall sensor, and an incremental rotary encoder. The Hall sensor, in conjunction with a trigger magnet fixed on the moving part, is used to capture a level flip signal during the system initialization phase to calibrate the mechanical coordinate zero point. The incremental rotary encoder operates synchronously with the rotor of the servo drive motor on the same axis and is used to output the mechanical displacement pulse signal containing displacement increment information in real time during the scanning phase.

[0016] Preferably, the signal processing mechanism 40 includes a programmable gain amplifier and a field-programmable gate array (FPGA). The programmable gain amplifier is used to adaptively adjust the analog amplification gain to amplify and discretize the original induced electromotive force signal. The FPGA is internally configured with a wavelet transform unit. The wavelet transform unit is used to perform a discrete wavelet transform algorithm on the sampled discrete digital signal and to clear the wavelet coefficients corresponding to mechanical jitter and high-frequency white noise by performing hard thresholding on the wavelet coefficients at a specific scale, thereby reconstructing and outputting a denoised high-frequency carrier signal.

[0017] Preferably, the field-programmable gate array is further used to perform digital synchronous demodulation operations, specifically including:

[0018] A local quadrature reference signal with the same frequency and phase as the high-frequency AC excitation signal is generated. This signal is then mixed with the denoised high-frequency carrier signal to generate an in-phase intermediate component and a quadrature intermediate component. A smoothing integral calculation is then performed on the two intermediate components using a digital low-pass filter to extract the in-phase and quadrature characteristic components that have been shifted to the DC frequency band.

[0019] Preferably, the embedded system structure 50 internally includes a digital signal processing core and a microcontroller core. The digital signal processing core calls a four-quadrant arctangent function to calculate the transient phase angle of the current position based on the input orthogonal feature components and the in-phase feature components. The digital signal processing core further introduces a transient phase angle based on the input of adjacent sampling periods. Periodic phase dewinding is performed to eliminate numerical winding dead zones and extract the corrected impedance phase difference value.

[0020] Preferably, the microcontroller core uses a temperature sensor to obtain the ambient temperature, calculates a dynamic compensation coefficient based on the linear expansion coefficient of the ball screw, and multiplies the original accumulated mechanical displacement pulse count by the dynamic compensation coefficient to generate a corrected mechanical displacement pulse count. When the fault interruption flag is received, the microcontroller core hardware captures an absolute timestamp, synchronizes it with the corrected mechanical displacement pulse count, and substitutes it into the spatial coordinate mapping equation to solve for the absolute physical coordinates.

[0021] Preferably, it also includes an electromagnetic protection mechanism 60, which is fixedly connected to the mechanical scanning mechanism 30 and covers the outside of the induction detection mechanism 20 and the signal processing mechanism 40. The electromagnetic protection mechanism 60 adopts a composite shielding structure composed of an outer layer of high magnetic permeability material and an inner layer of high electrical conductivity material tightly bonded together. The outer layer of high magnetic permeability material is used to provide a low magnetic resistance bypass channel to close the magnetic field lines for intruding low-frequency magnetic field lines. The inner layer of high electrical conductivity material is used to generate an internal electromagnetic eddy current effect when high-frequency stray electromagnetic waves intrude, so as to dissipate and absorb the energy of high-frequency electromagnetic waves.

[0022] Preferably, the electromagnetic protection mechanism 60 is also connected in series in the power supply network and communication line from the embedded system mechanism 50 to the external interface, and a common-mode choke is connected in series in the specific circuit topology. The common-mode choke is composed of two sets of coils wound in the same direction on the same high-frequency ferrite core. When external high-frequency common-mode noise enters in the same direction along the two conductors of the power supply network or communication line, the two sets of coils generate magnetic flux in the same direction and superimpose it in the core, presenting a high common-mode impedance. The high common-mode impedance is used to divide and block the conducted common-mode noise on the power supply and communication lines.

[0023] In summary, the present invention has at least one of the following beneficial technical effects:

[0024] 1. This invention injects a high-frequency AC signal of a specific frequency into the Röss coil under test through a high-frequency excitation injection mechanism. By utilizing the skin effect of high-frequency current, the weak strand breakage damage inside the conductor is converted into a sudden change in local AC equivalent resistance and parasitic inductance, thereby exciting a spatial alternating leakage magnetic field distortion signal. At the same time, in conjunction with the non-contact symmetrical differential receiving coil structure of the induction detection mechanism, common-mode electromagnetic noise in the environment is canceled in the physical circuit. This design transforms the microscopic physical damage that is difficult to detect in conventional methods into anti-interference electromagnetic parameters, thereby improving the detection sensitivity of the device for small wire breakage defects.

[0025] 2. In the signal processing and feature extraction stages, this invention utilizes a field-programmable gate array (FPGA) to perform discrete wavelet transform and digital phase-locked demodulation, filtering out mechanical jitter and extracting orthogonal feature components. Subsequently, the transient phase angle is calculated by the digital signal processing core and introduced based on... The periodic phase unwinding logic, this data processing mechanism, eliminates the numerical calculation dead zone and spurious step caused by the phase crossing the polarity boundary from the bottom layer, ensuring the continuity and accuracy of the impedance phase difference value calculation, and effectively avoiding false alarms and missed alarms during the scanning process.

[0026] 3. In the breakpoint spatial positioning mechanism, this invention utilizes a microcontroller core to acquire the ambient temperature collected by a temperature sensor in real time, and dynamically compensates the original displacement pulse count output by the rotary encoder based on the linear expansion mechanism of the ball screw metal material. Then, when a fault interruption is captured, the timestamp is hardware-aligned with the compensated pulse count. This synchronous compensation mechanism directly offsets the cumulative stroke error caused by the thermal expansion of the mechanical transmission components due to long stroke at the data stream level, ensuring a strict mapping between electromagnetic characteristics and actual physical position, and improving the positioning accuracy of the absolute three-dimensional coordinates of the breakpoint. Attached Figure Description

[0027] Figure 1 This is an overall structural diagram of the present invention;

[0028] Figure 2 This is a diagram of the internal structure of the electromagnetic protection mechanism of the present invention;

[0029] Figure 3 This is a schematic diagram of the hardware circuit topology and connection of the high-frequency excitation injection mechanism of the present invention;

[0030] Figure 4 This is a schematic diagram of the sensing and detection mechanism of the present invention.

[0031] Figure 5 This is a schematic diagram of the transmission structure and signal synchronization principle of the mechanical scanning mechanism of the present invention;

[0032] Figure 6 This is a diagram showing the hardware architecture and data flow of the signal processing mechanism of the present invention;

[0033] Figure 7 This is a flowchart illustrating the internal architecture and processing flow of the embedded system mechanism of the present invention;

[0034] Figure 8 This is a schematic diagram of the assembly and electrical connection of the electromagnetic protection mechanism of the present invention;

[0035] Figure 9 This is the axial scanning phase difference mapping diagram of the Rotary coil under test according to the present invention.

[0036] Among them, 10 is a high-frequency excitation injection mechanism; 20 is a sensing and detection mechanism; 30 is a mechanical scanning mechanism; 40 is a signal processing mechanism; 50 is an embedded system mechanism; and 60 is an electromagnetic protection mechanism. Detailed Implementation

[0037] The following is in conjunction with the appendix Figure 1 - Appendix Figure 9 The present invention will be further described in detail below.

[0038] This invention provides a rotisserie coil breakage detection device, comprising:

[0039] The high-frequency excitation injection mechanism 10 is used to inject a high-frequency AC excitation signal into the two ends of the main circuit of the ROZ coil under test.

[0040] The induction detection mechanism 20 is used to pick up the leakage magnetic field distortion signal caused by local impedance change outside the ROZZ coil under test.

[0041] The mechanical scanning mechanism 30 is used to carry the induction detection mechanism 20 to perform scanning motion along the surface of the rotatable coil to be tested, and output mechanical displacement pulse signals.

[0042] The signal processing unit 40 is used to receive the raw signal output by the sensing and detection unit 20, and to perform analog gain amplification, digital filtering and synchronous demodulation to extract orthogonal feature components.

[0043] The embedded system mechanism 50 is used to receive orthogonal characteristic components, calculate the transient phase angle and impedance phase difference, and perform timestamp alignment in conjunction with mechanical displacement pulse signals to solve the spatial coordinates of the breakpoint.

[0044] Electromagnetic protection mechanism 60 is used to physically isolate the induction detection mechanism 20 and the signal processing mechanism 40 and suppress conducted interference.

[0045] The excitation output of the high-frequency excitation injection mechanism 10 is directly connected to both ends of the closed main circuit of the rotary coil under test via a low-loss radio frequency cable. The linear drive guide of the mechanical scanning mechanism 30 is mounted parallel to the axial side of the rotary coil under test. The induction detection mechanism 20 is rigidly mounted on the movable slide of the mechanical scanning mechanism 30 by mechanical fasteners, and maintains a set small air gap with the surface of the rotary coil under test. In order to minimize the transmission path of the weak analog signal and suppress parasitic capacitance interference, the signal processing mechanism 40 is arranged close to the induction detection mechanism 20. The signal output of the induction detection mechanism 20 is connected via... A short-distance coaxial cable establishes an electrical connection with the front-end analog input interface of the signal processing mechanism 40. The communication bus of the embedded system mechanism 50 is connected to the frequency tuning terminal of the high-frequency excitation injection mechanism 10, the motor drive and encoder interface of the mechanical scanning mechanism 30, and the high-speed data port of the signal processing mechanism 40. Furthermore, the electromagnetic protection mechanism 60 forms a fully enclosed metal cavity fixed on the mechanical scanning mechanism 30 and completely covers the induction detection mechanism 20 and the adjacent signal processing mechanism 40. It is connected in series on the power supply link and digital bus from the embedded system mechanism 50 to the front-end mechanism via electrical wiring.

[0046] The above-mentioned mechanisms constitute a closed-loop detection system in terms of physical space and electrical connection. During the initialization phase of device operation, the embedded system mechanism 50 sends a reset command to the mechanical scanning mechanism 30 via the control bus. The mechanical scanning mechanism 30 completes the zero-point position calibration based on the Hall sensor configured inside. At the same time, the high-frequency excitation injection mechanism 10 is controlled by the parameters of the embedded system mechanism 50 and continuously outputs a sinusoidal AC signal with a set frequency and amplitude to the ROZ coil under test.

[0047] During the scanning and detection phase, the mechanical scanning mechanism 30 drives the transmission component to move the induction detection mechanism 20 along the axial direction of the rotund coil under test at a set speed. During this process, the induction detection mechanism 20 senses the alternating magnetic field in the space in real time. When there is broken strand or broken wire damage inside the rotund coil under test, the skin effect causes a sudden change in the high-frequency parasitic impedance in that area. The induction detection mechanism 20 picks up the local leakage magnetic field distortion and outputs the original induced electromotive force signal to the directly connected signal processing mechanism 40.

[0048] After receiving the original induced electromotive force signal, the signal processing unit 40 performs signal conditioning through an internal programmable gain amplifier. The conditioned signal is then input to a field-programmable gate array (FPGA), which executes a discrete wavelet transform algorithm to filter out low-frequency mechanical jitter and spatial noise. Subsequently, it performs digital phase-locked demodulation and transmits the extracted in-phase and quadrature characteristic components to the embedded system unit 50 via a bus.

[0049] The embedded system mechanism 50 is equipped with a dual-core heterogeneous processing unit. The digital signal processing core receives in-phase and quadrature characteristic components and calculates the transient phase angle at the current position. Based on this angle sequence, the digital signal processing core further calculates the impedance phase difference value of adjacent sampling periods. When the impedance phase difference value exceeds the set threshold, the digital signal processing core generates a fault interruption flag.

[0050] Meanwhile, the microcontroller core in the embedded system mechanism 50 records the number of rotary encoder pulses output by the mechanical scanning mechanism 30 in real time. The microcontroller core uses a temperature sensor to obtain ambient temperature data, reducing the mechanical transmission error caused by it, and directly calculates the dynamic compensation coefficient based on the linear expansion mechanism of the metal material to numerically correct the accumulated mechanical displacement pulses. When a fault interruption flag is received, the microcontroller core triggers the hardware capture function to bind and align the timestamp with the current accumulated temperature compensation pulse count. Finally, the microcontroller core substitutes the aligned data into the spatial mapping equation to calculate and output the absolute physical coordinates of the breakpoint.

[0051] In the above operating process, the electromagnetic protection mechanism 60 blocks the interference of external electromagnetic fields on the induction detection mechanism 20 through a composite shield composed of an outer layer of permalloy and an inner layer of copper foil. In addition, the common mode choke configured in the electromagnetic protection mechanism 60 is connected to the power supply network and communication line to suppress the conducted common mode noise in the circuit, thereby ensuring the high signal-to-noise ratio and transmission stability of the whole machine signal.

[0052] See attached document Figure 3 , Figure 3 This is a schematic diagram of the hardware circuit topology and connection of a high-frequency excitation injection mechanism according to an embodiment of the present invention. In this embodiment, the high-frequency excitation injection mechanism 10 is disposed at both ends of the main circuit of the rotund coil under test, and is used to generate and inject a high-frequency AC excitation signal into the internal conductor of the rotund coil under test. The high-frequency excitation injection mechanism 10 internally includes a direct digital frequency synthesis unit, a power amplification unit, and an impedance matching network. The specific signal injection operation of the high-frequency excitation injection mechanism 10 includes the following steps:

[0053] S101, the direct digital frequency synthesis unit inside the high-frequency excitation injection mechanism 10 generates an initial high-frequency sinusoidal AC signal according to the control command issued by the embedded system mechanism 50. The time-domain expression of this high-frequency sinusoidal AC signal is:

[0054] ;

[0055] In the formula, Representative moment The excitation voltage; This represents the amplitude of the excitation signal; Represents the high-frequency carrier frequency; Represents the initial phase.

[0056] As a preferred method, high-frequency carrier frequency The value range is set to 100kHz to 500kHz, a frequency band that ensures the skin effect in copper or aluminum conductors of standard wire diameters; simultaneously, the excitation signal amplitude... Based on the overall transmission impedance of the ROZ coil under test and the signal-to-noise ratio requirements of the back-end signal processing mechanism 40, closed-loop adaptive adjustment is performed, typically within the range of 5V to 15V. By controlling the direct digital frequency synthesis unit to adjust the high-frequency carrier frequency and the amplitude of the excitation signal, the system can adapt to ROZ coils under test with different wire diameter parameters.

[0057] S102 After generating the initial high-frequency sinusoidal AC signal, the signal is synchronously guided to the power amplifier unit. The power amplifier unit amplifies the input signal through the current stage. For the specific circuit implementation of the power amplifier unit, those skilled in the art can use a conventional push-pull amplifier circuit. The specific component configuration is a well-known technology in the field and will not be described in detail here. The amplified signal is output to the two ends of the main circuit of the ROZ coil under test through the impedance matching network.

[0058] In this embodiment, the impedance matching network is preferably of LC type or A passive matching circuit, consisting of capacitors and inductors, is used to balance the difference between the output impedance of the signal source and the distributed impedance of the ROZU coil under test, thereby reducing the reflection attenuation of high-frequency signals at the interface.

[0059] S103, when the injected high-frequency AC excitation signal propagates within the conductor of the Röss coil under test, it exhibits a skin effect due to the characteristics of the high-frequency electromagnetic field, causing the signal current to concentrate in the surface region of the conductor. The calculation model for the skin depth of the current penetrating the conductor is as follows:

[0060] ;

[0061] In the formula, The skin depth represents the depth to which current penetrates the conductor; This represents the resistivity of the conductor inside the ROZZ coil being tested; The carrier frequency representing the high-frequency AC excitation signal; Representing the absolute permeability of the internal conductor, the system transmits the high-frequency carrier frequency. Locking onto a specific high-frequency band can reduce skin depth. Compressed to a microscale much smaller than the radius of the internal conductor, the alternating current conduction behavior of the conductor is extremely sensitive to its surface morphology and local cross-sectional integrity.

[0062] S104, based on the aforementioned high-frequency skin effect, when there is micro-ohm level strand breakage damage inside the tested Roszc coil, the effective conductive cross-sectional area of ​​the damaged area decreases, and the physical morphology of the wire breaks. At low frequencies, this physical breakage only manifests as a weak change in DC resistance, which is easily drowned out by environmental noise. However, under the action of a high-frequency AC excitation signal, due to the limited skin depth, the surface current conduction path in the strand breakage area is severely distorted, causing a sudden change in the local AC equivalent resistance and local parasitic inductance. The mathematical model of the local high-frequency impedance at the strand breakage point is expressed as:

[0063] ;

[0064] In the formula, Represents the local high-frequency impedance at the point of damage; This represents the AC equivalent resistance caused by the skin effect and the reduction in effective cross-sectional area; This represents the localized parasitic inductance caused by a physical fracture surface; Representing the imaginary unit, parasitic inductance alters the continuous loop of local current due to the presence of physical fracture interfaces. and equivalent resistance All of these produce nonlinear offsets.

[0065] When the high-frequency AC excitation signal passes through the local high-frequency impedance abrupt change point, it excites an alternating leakage magnetic field with phase shift and amplitude change in the external space of the broken strand. Through the above hardware configuration and physical conversion process, the high-frequency excitation injection mechanism 10 converts the micro-ohm level broken strand damage inside the tested Roche coil into a measurable electromagnetic distortion parameter in the external space, thus establishing a stable physical working condition for the non-contact signal pickup of the induction detection mechanism 20.

[0066] See attached document Figure 4 , Figure 4 This is a schematic diagram of the structure of an inductive detection mechanism according to an embodiment of the present invention. In this embodiment, the inductive detection mechanism 20 is mounted on the mechanical scanning mechanism 30 and is disposed near the outer insulation layer of the coil under test. It is used to non-contactly pick up the spatial alternating leakage magnetic field distortion signal excited by internal micro-Euclidean level damage. In order to extract this weak feature from complex background noise, the core component of the inductive detection mechanism 20 is set as a non-contact differential high-frequency inductive probe. The specific signal pickup operation of the inductive detection mechanism 20 includes the following steps:

[0067] S201, the physical configuration of the sensing detection mechanism 20 adopts a symmetrical differential winding structure. Specifically, the non-contact differential high-frequency sensing probe includes two miniature receiving coils with opposite winding directions and identical geometric dimensions.

[0068] As a preferred approach, two miniature receiving coils are arranged parallel to each other along the axis of the rotatable coil under test, with the center-to-center distance between them set between 5 mm and 10 mm. This range is determined based on the spatial attenuation rate of the high-frequency leakage magnetic field. This spacing configuration ensures that the probe accurately covers the local leakage magnetic field gradient change area generated by the weak strand break during the scanning process, while maintaining the miniaturization of the overall probe structure and avoiding a decrease in spatial resolution due to an excessively large detection aperture. As for the specific skeleton material and enameled wire selection of the miniature receiving coils, those skilled in the art can make conventional selections based on the high-frequency working environment. The specific winding process is a well-known technology in the field and will not be described in detail here.

[0069] S202, based on the aforementioned probe configuration, the induction detection mechanism 20 achieves physical cancellation of common-mode interference in the spatial environment through a symmetrical differential structure. In a complex industrial electromagnetic environment, the induced electromotive force generated by the far-field background electromagnetic noise in two closely adjacent miniature receiving coils is in opposite directions and has approximately equal amplitude. When the two miniature receiving coils are connected in reverse series in the electrical circuit, the power frequency interference introduced by the external environment and the common-mode signal generated by the spatially scattered electromagnetic waves in the two coils cancel each other out, and the background noise at the probe output end is effectively suppressed. Unlike the distribution characteristics of far-field noise, the alternating leakage magnetic field caused by the high-frequency impedance change at the local break point of the tested Roche coil exhibits a localized spatial gradient distribution. When the induction detection mechanism 20 sweeps across the damaged area, the two miniature receiving coils are in different magnetic flux density regions of the leakage magnetic field, thereby outputting an effective differential-mode signal containing structural distortion characteristics.

[0070] S203, the differential-mode signal after physical differential cancellation manifests as the original induced electromotive force signal containing distortion information. Its essence is the algebraic sum of the transient induced voltages of the two miniature receiving coils. Combined with the time-domain mathematical model, the expression for the original induced electromotive force signal output by the induction detection mechanism 20 is:

[0071] ;

[0072] In the formula, Representative moment The original induced electromotive force signal output by the induction detection mechanism 20; The current absolute physical coordinates of the induction detection mechanism 20 on the surface of the rotatable coil under test; Represents the probe in coordinates The amplitude of the high-frequency leakage magnetic field induced signal picked up at the location; This represents the high-frequency carrier frequency injected by the high-frequency excitation injection mechanism 10; Represents the probe in coordinates The spatial alternating magnetic field distortion phase picked up at the location; This represents the remaining broadband background noise after cancellation by the differential structure, and the absolute physical coordinates as the mechanical slide advances. Manifested as time The function, therefore and These are all parameters that dynamically reflect the evolution of the internal conductor's spatial morphology.

[0073] S204, the induction detection mechanism 20 continuously outputs the acquired original induced electromotive force signal to the back-end signal processing mechanism 40. In the region where the internal conductor structure of the Roche coil under test is intact, the external leakage magnetic field is uniformly distributed, and the parameters... and Maintaining a relatively constant state, when the sensing and detection mechanism 20 moves to the micro-ohm level broken strand region, the change in local parasitic inductance and AC equivalent resistance caused by the sudden change in internal high-frequency impedance results in a shift in the spatial magnetic field vector, at which point the distorted phase... It will exhibit obvious transient step change.

[0074] Through the above mechanism, the sensing and detection mechanism 20 completes the conversion from macroscopically difficult-to-measure electromagnetic field physical distortion to continuously calculable electrical signal parameters, providing underlying physical data support for subsequent calculation of impedance phase difference and precise location of breakpoint coordinates.

[0075] See attached document Figure 5 , Figure 5 This is a schematic diagram of the transmission structure and signal synchronization principle of a mechanical scanning mechanism according to an embodiment of the present invention. In this embodiment, the mechanical scanning mechanism 30 is used to carry the induction detection mechanism 20 to perform uniform linear scanning along the surface of the rotund coil under test, and converts the continuous physical movement into a synchronous discrete displacement pulse sequence, providing a reference scale for subsequent fault location of the embedded system mechanism 50.

[0076] The specific physical transmission and displacement calibration of the mechanical scanning mechanism 30 includes the following steps:

[0077] S301, the hardware transmission body of the mechanical scanning mechanism 30 adopts a ball screw slide mechanism driven by a servo drive motor. Specifically, the output shaft of the servo drive motor is connected to one end of the ball screw through an elastic coupling. The sensing and detection mechanism 20 is fixedly installed on the nut slide of the ball screw and is constrained by the lateral limit of the parallel linear guide rails. Under the drive of the servo drive motor, the nut slide converts the rotational motion into linear displacement along the axis of the rotatable coil to be measured. The servo drive motor is configured in a closed-loop speed control mode to ensure the smoothness of the scanning process and suppress the parasitic modulation interference caused by mechanical jitter to the detection of weak magnetic fields. At the same time, a spring clamp that can slide up and down is provided on the side of the ball screw slide mechanism. The spring force of the spring clamp automatically clamps the two ends of the rotatable coil to be measured to avoid movement that would lead to inaccurate detection. For the adjustment of the control parameters of the servo motor driver and the specific mechanical assembly process of the ball screw slide, those skilled in the art can make conventional selections based on the actual required effective scanning stroke and load mass. The mechanical selection and assembly are well-known technologies in the field and will not be described in detail here.

[0078] S302, an absolute spatial coordinate system for scanning and detection is established in the above mechanical structure. The mechanical scanning mechanism 30 is equipped with a zero-point calibration circuit based on a Hall sensor. The Hall sensor of this circuit is fixedly installed at the starting end of the linear guide rail, and the corresponding trigger magnet is embedded in the side of the nut slide of the sensing and detection mechanism 20. During the reset phase of the device power-on initialization, the servo drive motor drives the nut slide to retract slowly to the starting end. When the trigger magnet enters the threshold range of the Hall sensor's sensing magnetic field, the Hall sensor output generates a level flip signal. The embedded system mechanism 50 captures the level flip signal through the hardware external interrupt mechanism, sends a hardware braking command to the servo drive motor, and marks the current stopping position as the mechanical coordinate zero point in the logic memory. This physical zero-point calibration mechanism eliminates the mechanical cumulative error caused by system power failure restart or long-term repetitive operation, and establishes the spatial reference for subsequent breakpoint coordinate calculation.

[0079] S303 After establishing the spatial reference, the system needs to digitize and quantize the continuous mechanical displacement. During the scanning operation after zero-point calibration, the incremental rotary encoder at the rear end of the mechanical scanning mechanism 30 operates synchronously with the rotor of the servo drive motor on the same axis, and outputs a pulse signal containing the incremental information of mechanical displacement in real time. Considering that the oscillation during mechanical operation can easily cause pulse miscounting, the incremental rotary encoder outputs A-phase and B-phase orthogonal pulses with a phase difference of 90 degrees. The timer inside the embedded system mechanism 50 is configured in orthogonal decoding mode. By judging the leading or lagging relationship between A-phase and B-phase, the actual moving direction of the nut slide is determined, and then the pulse is incremented or decremented.

[0080] The pulse signals output by the rotary encoder constitute a discrete-time sequence of spatial coordinate mappings. Let the number of calibration pulses output by the rotary encoder per revolution be... The mechanical lead parameter of the ball screw, that is, the physical linear displacement corresponding to one revolution of the nut slide, is... To ensure millimeter-level resolution for locating minute breakpoints, the number of calibration pulses is... The value range is set to 2500 to 10000, mechanical lead parameter The value range is set to 4 mm to 10 mm. In terms of dimensional conversion, the physical displacement equivalent to a single-step pulse is equivalent to... and The ratio at any detection time If the net value of the signed valid position pulses captured by the system is Then the current absolute spatial coordinates of the sensing and detection mechanism 20 The mathematical mapping model is as follows:

[0081] ;

[0082] In the formula, Representative moment The absolute spatial coordinates of the location of the sensing and detection mechanism 20; Represents the zero point of the mechanical coordinates, which is physically calibrated by the Hall sensor; This represents the total number of displacement pulses accumulated from the zero position; Represents the calibration resolution of an incremental rotary encoder; The mechanical lead represents the lead screw of the transmission screw.

[0083] S304, the mechanical scanning mechanism 30 discretizes the continuous spatial displacement of the sensing detection mechanism 20 outside the measured rotund coil into a digital pulse sequence that can be extracted in real time by the embedded system mechanism 50 through the above-mentioned mechanical transmission structure and photoelectric conversion mechanism. Since the original induced electromotive force signal output by the sensing detection mechanism 20 and the mechanical displacement pulse signal output by the mechanical scanning mechanism 30 are strictly synchronized in the time domain, the displacement pulse sequence is directly used as the underlying basis for spatial coordinate transformation to establish the mapping relationship between the electromagnetic field distortion signal and the specific physical position.

[0084] See attached document Figure 6 , Figure 6 This is a hardware architecture and data flow diagram of a signal processing mechanism according to an embodiment of the present invention. In this embodiment, the signal processing mechanism 40 includes a programmable gain amplifier, a field-programmable gate array (FPGA), and a wavelet transform unit. This mechanism is mainly used to receive the raw induced electromotive force signal output by the induction detection mechanism 20, and perform analog gain conditioning, digital denoising, and hardware synchronous demodulation based on lock-in amplification to extract the DC characteristic component representing impedance change. In terms of hardware connection topology, the output terminal of the programmable gain amplifier is connected to the FPGA. The FPGA is internally configured with a wavelet transform unit, which performs wavelet transform to eliminate high-frequency noise. The specific signal processing and feature extraction work of the signal processing mechanism 40 includes the following steps:

[0085] S401, the front-end hardware link of the signal processing mechanism 40 is specifically configured with the aforementioned programmable gain amplifier and high-frequency analog-to-digital converter. Since the original induced electromotive force signal output by the sensing and detection mechanism 20 is extremely weak and has a wide dynamic range, the programmable gain amplifier adaptively adjusts the analog amplification gain according to the real-time peak-to-peak value of the input signal through the internal feedback resistor network.

[0086] As a preferred approach, the amplification gain is set to a range of 10 to 1000 times to ensure that the amplitude of the amplified analog signal can fully cover the optimal dynamic range of the downstream high-frequency analog-to-digital converter, avoiding signal limiting or the introduction of excessive quantization noise. After analog amplification, the output of the programmable gain amplifier transmits the conditioned signal to the high-frequency analog-to-digital converter for discretization sampling. To satisfy the Nyquist sampling theorem and fully preserve the small phase shift details of the high-frequency carrier, the sampling rate of the high-frequency analog-to-digital converter is preferably configured to be 10 to 20 times the carrier frequency injected by the high-frequency excitation injection mechanism 10. The discrete digital signal generated by sampling is directly transmitted to the field-programmable gate array inside the signal processing mechanism 40 through a high-speed serial bus. For the clock tree configuration and serial bus timing control of the high-frequency analog-to-digital converter, those skilled in the art can write conventional logic according to the datasheet of the selected chip. The specific underlying driving timing is a well-known technology in the field and will not be described in detail here.

[0087] S402, upon receiving a discrete digital signal, in order to reconstruct an effective high-frequency carrier signal under low signal-to-noise ratio conditions, calls the internally configured wavelet transform unit of the field-programmable gate array to execute the discrete wavelet transform algorithm to eliminate high-frequency noise and low-frequency mechanical vibration interference in the environment. The continuous mathematical form of the discrete wavelet transform is expressed as:

[0088] ;

[0089] In the formula, These represent the wavelet coefficients after transformation; The scaling factor represents the control frequency scaling. The shift factor represents the control time position; Represents the current amplification gain of the programmable gain amplifier; The original induced electromotive force signal output by the induction detection mechanism 20 represents the signal. The complex conjugate of the mother wavelet function is represented by the integral operation performed in the hardware physical implementation of the wavelet transform unit by a multi-stage cascaded finite impulse response filter bank.

[0090] An effective denoising criterion is established. The wavelet transform unit extracts the background noise energy value of the system in a static, unscanned state as a reference benchmark and calculates the hard threshold judgment parameters. Subsequently, the wavelet transform unit performs hard threshold judgment on the wavelet coefficients at a specific scale, directly clearing the wavelet coefficients corresponding to mechanical jitter and high-frequency white noise, while retaining the high-frequency carrier component representing the effective alternating magnetic field characteristics. Finally, the wavelet transform unit reconstructs and outputs the denoised high-frequency carrier signal through inverse wavelet transform. .

[0091] After wavelet denoising to filter out background interference, S403 uses a field-programmable gate array (FPGA) to extract the subtle phase change features hidden in the carrier wave. The digital synchronous demodulation based on the lock-in amplification principle is performed. In terms of physical principle, this demodulation process uses a mixing operation to shift the measured feature with the same frequency as the reference signal to the DC frequency band, thereby realizing narrowband extraction of weak signals.

[0092] The specific operation is as follows: The digitally controlled oscillator inside the field-programmable gate array (FPGA) generates a local quadrature reference signal in real time, which is in phase and frequency with the high-frequency excitation injection mechanism 10, based on the frequency control word issued by the embedded system mechanism 50. This quadrature reference signal contains in-phase sine and quadrature cosine signals. The FPGA then calls its internal digital signal processing multiplier to slice the high-frequency carrier signal. High-frequency mixing operations are performed with two local quadrature reference signals respectively:

[0093] ;

[0094] ;

[0095] In the formula, This represents the in-phase intermediate component generated by mixing. This represents the orthogonal intermediate components generated by frequency mixing; The reference carrier frequency, representing the local quadrature reference signal, is strictly consistent with the high-frequency carrier frequency injected into the Röss coil under test. Through this multiplication mechanism, the phase and amplitude characteristics of the original high-frequency signal are shifted to the DC frequency band and the second harmonic frequency band in the frequency domain.

[0096] S404, in order to separate the effective features carrying local impedance distortion information and eliminate the second harmonic component, uses a field-programmable gate array to separate the in-phase intermediate component. Orthogonal intermediate components Two parallel digital low-pass filters are fed into the filter. The filtering operation of the digital low-pass filters is equivalent in the time domain to an integration period of a specific time. Perform smooth integration calculation on the signal within:

[0097] ;

[0098] ;

[0099] In the formula, This represents the extracted in-phase feature components; Represents the extracted orthogonal feature components; The equivalent integration period of the digital low-pass filter; Representing the integration time variable, as a preferred method, to ensure that the second harmonic and residual high-frequency clutter are sufficiently attenuated, the equivalent integration period... The value range is set to 100 to 500 times the high-frequency carrier period. After processing by a digital low-pass filter, the dynamically alternating high-frequency magnetic field signal is converted into two slowly changing DC level signals. The signal processing unit 40 then processes these two in-phase characteristic components that reflect the physical fracture interface. Orthogonal eigencomponents The data is continuously transmitted to the embedded system mechanism 50 via an internal high-speed parallel data bus, providing a digital reference for subsequent impedance phase difference calculations.

[0100] See attached document Figure 7 , Figure 7 This is a flowchart illustrating the internal architecture and processing flow of an embedded system mechanism according to an embodiment of the present invention. In this embodiment, the embedded system mechanism 50 is configured with a dual-core heterogeneous processing unit, which specifically includes an ARM Cortex-M7 core and a DSP core.

[0101] The ARM Cortex-M7 core handles human-computer interaction and data storage, while the DSP core runs fault feature extraction algorithms, achieving efficient decoupling between low-level real-time computing and high-level integrated management. The embedded system mechanism 50 is responsible for receiving feature data extracted by the front-end hardware and calculating the absolute physical coordinates of the breakpoints by combining spatial calibration information. The specific data calculation and control work of the embedded system mechanism 50 includes the following steps:

[0102] S501, the DSP core receives the in-phase and quadrature characteristic components transmitted in real time by the signal processing mechanism 40 through its internal high-speed memory interface. After acquiring the quadrature characteristic data, the DSP core performs transient phase angle calculation. To avoid algorithm dead zones caused by division-by-zero overflow when the in-phase characteristic component crosses zero, the DSP core calls the four-quadrant arctangent function to calculate the transient phase angle at the current position. The mathematical model of this calculation process is expressed as follows:

[0103] ;

[0104] In the formula, Representative moment The extracted transient phase angle of the alternating leakage magnetic field; Representative moment Orthogonal feature components input to the DSP core; Representative moment The in-phase characteristic components input to the DSP core; This represents the four-quadrant arctangent function containing quadrant judgment logic. Through this function, the DSP core converts the orthogonal voltage amplitude in a two-dimensional Cartesian coordinate system into a continuously distributed phase physical quantity in a polar coordinate system. Since the signal processing mechanism 40 performs synchronous phase-locked demodulation at the front end, the transient phase angle calculated by the DSP core here... Mathematically, this is strictly equivalent to the sensing detection mechanism 20 in the current absolute physical coordinates. Spatial alternating magnetic field distortion phase picked up at the location Thus, a unique equivalent mapping relationship was successfully established between time-domain phase data and spatial-domain structural distortion characteristics.

[0105] S502, after calculating the continuous transient phase angle sequence, the DSP core further extracts dynamic features characterizing the spatial impedance abrupt change. Since the strand breakage damage of the tested Roche coil manifests as a high-frequency impedance abrupt change within a local spatial range, this physical structural abrupt change is mapped to a drastic step change in phase angle between adjacent sampling points. Considering that the range of the arctangent function is limited to... arrive When the phase normally crosses the polarity boundary, numerical wrapping occurs, and direct subtraction will cause spurious numerical abrupt changes. Therefore, the DSP core introduces a method based on... After phase dewinding of the period, the formula for calculating the impedance phase difference between adjacent sampling periods after correction is as follows:

[0106] ;

[0107] In the formula, Representative moment Impedance phase difference after unwinding; Represents the discrete sampling period set by the system; Represents floating-point modulo operation; the absolute value sign is used to extract the true amplitude of the phase step without distinguishing polarity. After the calculation is completed, the DSP core will generate the impedance phase difference value in real time. Compared with the preset fault safety threshold Compare them.

[0108] As a preferred approach, this fail-safe threshold The phase noise floor is extracted from the intact section of the Röss coil under test and set to a range of 3 to 5 times the standard deviation of this noise floor. This filters out minute electromagnetic fluctuations caused by conventional manufacturing tolerances. Greater than or equal to When the probe is scanning the internal broken area, the DSP core immediately generates and sends a fault interrupt flag to the internal interrupt controller.

[0109] In the S503, while the DSP core performs the aforementioned feature extraction, the ARM Cortex-M7 core is responsible for maintaining the coordinate accuracy of the mechanical transmission system. Long-stroke reciprocating scanning causes the metal lead screw to expand due to heat, resulting in a slight drift in the physical mechanical lead. The ARM Cortex-M7 core uses temperature sensors placed on the surface of the transmission components to collect the ambient temperature in real time and generates dynamic compensation coefficients based on the linear expansion mechanism of the lead screw material. This effectively corrects the original cumulative pulse count acquired by the rotary encoder. The real-time effective pulse count correction model after dynamic compensation is as follows:

[0110] ;

[0111] In the formula, Representative moment The number of effective displacement pulses after temperature compensation; This represents the raw, cumulative number of pulses obtained from the rotary encoder; The equivalent linear expansion coefficient representing the metal material of the ball screw; This represents the temperature rise difference between the current real-time temperature and the initial temperature at the system's zero-point calibration time. This mechanism effectively calculates the stroke elongation caused by thermal deformation from the underlying data level, fundamentally eliminating coordinate lag error.

[0112] In S504, when a fault interrupt flag triggered by the DSP core is received, the ARM Cortex-M7 core immediately responds to the inter-core interrupt. In the interrupt service routine, the ARM Cortex-M7 core synchronously reads the absolute timestamp register of the current system timer and the cumulative pulse register of the encoder interface, and then compares the timestamp with the effective displacement pulse count calculated in S503. Strict frame binding is performed in the dual-core shared memory area. The ARM Cortex-M7 core extracts the alignment data, substitutes it into the spatial absolute coordinate mapping equation, and calculates the precise three-dimensional physical coordinates of the breakpoint that causes the impedance change. Subsequently, the system saves the coordinates, transient phase angle and timestamp to a non-volatile storage medium and presents the detection results through the display peripheral, thus forming a data closed loop from leakage magnetic feature picking to breakpoint spatial positioning.

[0113] See attached document Figure 8 , Figure 8 This is a schematic diagram of the assembly and electrical connection of an electromagnetic protection mechanism according to an embodiment of the present invention. In this embodiment, the electromagnetic protection mechanism 60 covers the outside of the induction detection mechanism 20 and the signal processing mechanism 40, and is connected in series in the power supply and communication link of the system. This mechanism is used to suppress the interference of low-frequency alternating magnetic fields, high-frequency stray electric field radiation, and conducted common-mode noise on weak leakage magnetic distortion signals in harsh industrial environments. The specific structural configuration and noise suppression operation of the electromagnetic protection mechanism 60 include the following steps:

[0114] S601, the electromagnetic protection mechanism 60 adopts a three-dimensional fully enclosed composite shielding structure in terms of spatial physical form. The composite shielding is composed of an outer layer of high magnetic permeability material and an inner layer of high electrical conductivity material tightly bonded together.

[0115] As a preferred approach, the outer layer material is permalloy, with a thickness ranging from 1 mm to 3 mm. This thickness range provides sufficient magnetic cross-sectional area to prevent magnetic saturation under strong external magnetic fields without increasing the overall mass of the probe. The inner layer material is copper foil, with a thickness ranging from 0.1 mm to 0.5 mm. This thickness is much greater than the skin depth of conventional megahertz-level high-frequency interference, ensuring the formation of a continuous electromagnetic eddy current barrier. This double-layer composite structure can meet the attenuation requirements of electromagnetic fields of different frequency bands and properties. For the specific sheet metal forming and interlayer insulation bonding process of this composite shield, those skilled in the art can perform conventional processing according to the actual mechanical shape of the probe. The material stamping and bonding processes are well-known technologies in the field and will not be described in detail here.

[0116] S602, based on the physical properties of the aforementioned composite material, the electromagnetic protection mechanism 60 forms two independent attenuation paths for external space radiation interference. When a low-frequency alternating magnetic field intrudes from the external environment, the outer permalloy layer, with its extremely high initial relative permeability, provides a low-resistance bypass channel for the low-frequency magnetic field lines, guiding them to close within the shielding wall, thereby reducing the magnetic flux density penetrating to the internal induction detection mechanism 20. When a high-frequency stray electric field or high-frequency electromagnetic wave intrudes, the inner copper foil generates a strong eddy current effect. From the perspective of the physical mechanism of energy conversion, the eddy currents generated internally will generate a reverse alternating magnetic field to cancel the incident wave, while simultaneously converting electromagnetic energy into the material's thermal energy. According to the theory of electromagnetic field reflection and absorption, the mathematical model of the absorption loss of high-frequency electromagnetic waves by the composite shielding body is expressed as follows:

[0117] ;

[0118] In the formula, The value in decibels represents the absorption loss of the inner copper foil against high-frequency stray electromagnetic waves. Represents the physical thickness of the copper foil; Represents the frequency of intrusive high-frequency spurious interference; Represents the absolute permeability of the copper foil; The value represents the electrical conductivity of the copper foil. This model shows that by setting a specific thickness for the inner copper foil, the system can efficiently convert the high-frequency electric field energy transmitted into the interior into heat dissipation.

[0119] S603, in order to avoid secondary radiation caused by the fluctuation of its own potential after the shield absorbs high-frequency electric field energy, the inner copper foil and the outer permalloy are connected to the low impedance grounding point of the system through a braided copper strip. The system adopts a strict single-point grounding process to prevent ground loop interference current from being generated between different ground potentials, so that the induced charge on the composite shield is discharged to the ground of the main equipment along a single path, ensuring that the reference ground potential of the internal signal processing mechanism 40 remains absolutely stable.

[0120] S604, in addition to space radiation interference, the electromagnetic protection mechanism 60 connects a common-mode choke in series in the device's power supply network and communication lines to suppress common-mode noise conducted along the electrical circuit. In the specific circuit topology, the common-mode choke consists of two sets of coils wound in the same direction on the same high-frequency ferrite core. For the positive and negative buses providing DC power or differential communication twisted-pair cables, when the normal operating current flows through the two sets of coils, it generates magnetic fluxes of equal magnitude and opposite direction. The magnetic fluxes inside the core cancel each other out, and the common-mode choke does not attenuate the effective differential-mode signal. When external high-frequency common-mode noise enters along the two wires in the same direction, the two sets of coils generate magnetic fluxes in the same direction and superimpose them in the core, exhibiting extremely high common-mode impedance. The mathematical model of this common-mode impedance is:

[0121] ;

[0122] In the formula, This represents the equivalent common-mode impedance presented by the common-mode choke to conducted noise; The frequency representing common-mode noise; This represents the equivalent inductance of the common-mode choke. Represents the equivalent loss resistance of the magnetic core at high frequencies; Represents the imaginary unit. Through this series connection method, the electromagnetic protection mechanism 60 utilizes the equivalent common-mode impedance. Effective voltage division and blocking of high-frequency surges and conducted interference on power supply and communication cables ensure the reliability of data transmission of the embedded system mechanism 50 under complex working conditions.

[0123] Specific application examples:

[0124] In this embodiment, an industrial flexible rotisserie coil with a rated current of 1000A, an internal conductor wire diameter of 8mm, and a total length of 2000mm was selected as the test object. To verify the effectiveness of the wire breakage detection device, three physical damages of different degrees were set in the internal conductor of the rotisserie coil under test. The specific locations and states are as follows: the first location is located at an axial distance of 500.53mm, the damage state is a single strand of copper wire broken, and the cross-sectional area reduction rate is about 5%; the second location is located at an axial distance of 1203.27mm, the damage state is multiple strands of copper wire broken, and the cross-sectional area reduction rate is about 30%; the third location is located at an axial distance of 1850.85mm, the damage state is a deep wire break, and the cross-sectional area reduction rate is about 90%.

[0125] The device parameters are configured as follows: the carrier frequency output of the high-frequency excitation injection mechanism is set to 350kHz, the excitation voltage amplitude is set to 10V, the ball screw lead parameter of the mechanical scanning mechanism is 5mm, the rotary encoder resolution is set to 5000 pulses / revolution, the scanning line speed driven by the servo motor is constant at 20mm / s, and in the signal processing mechanism, the initial gain of the programmable gain amplifier is set to 100 times, the sampling rate of the analog-to-digital converter is set to 5MS / s, and the equivalent integration period of the phase-locked loop digital low-pass filter is set to 200 times the high-frequency carrier period.

[0126] During the test run, the embedded system sends a reset command to the mechanical scanning mechanism. The slide retracts, triggering the Hall sensor to complete the zero-point coordinate calibration. The high-frequency excitation injection mechanism continuously injects AC signals into the closed main circuit. The mechanical scanning mechanism drives the induction detection mechanism to move along the coil surface. The induction detection mechanism picks up the leakage magnetic field distortion signal and outputs it to the signal processing mechanism. The signal processing mechanism performs discrete wavelet transform to remove background noise, generates a local orthogonal reference signal through digital control oscillator to complete mixing and filtering, and outputs in-phase and orthogonal characteristic components. The DSP core of the embedded system calculates the transient phase angle through the four-quadrant arctangent function, performs unwinding processing, and calculates the impedance phase difference value of adjacent sampling periods. The ARM Cortex-M7 core updates the linear expansion compensation coefficient in conjunction with the real-time temperature data of the temperature sensor and corrects the displacement pulse. When the impedance phase difference value exceeds the preset 0.15rad safety threshold, an interrupt is triggered, the timestamp is recorded, and the physical coordinates of the breakpoint are output.

[0127] To evaluate the positioning accuracy and detection sensitivity of this device, a traditional low-frequency inductance detection method was introduced as a control group. The control group was given a 1kHz sinusoidal excitation signal, and the fault characteristics were extracted by measuring the change in the external inductance. Both systems performed scanning detection on the same Röss coil under test and recorded the absolute physical coordinates when the judgment threshold was triggered. The data records are shown in Table 1.

[0128] Table 1 Comparison of Fault Location Data between the Device of the Present Invention and the Low-Frequency Inductance Detection Method

[0129] Damage number Damage type Actual physical coordinates (mm) The present invention detects coordinates (mm). Positioning error (mm) of this invention Control group detection coordinates (mm) Positioning error of the control group (mm) 1 Single strand fracture 500.53 500.81 +0.28 No response - 2 Multiple strands fractured 1203.27 1202.94 -0.33 1210.56 +7.29 3 Deep disconnection 1850.85 1851.12 +0.27 1845.33 -5.52

[0130] Based on the data in Table 1 and... Figure 9 The proposed Röss coil breakage detection device is effective in locating micro-ohmic level strand breakage damage. In the test of damage number 1, the actual damage was a single strand breakage leading to a slight reduction in cross-sectional area. The low-frequency inductance method did not output an effective response, indicating that conventional low-frequency methods are difficult to distinguish slight changes in DC resistance or low-frequency impedance. However, the device of this invention outputs a positioning coordinate of 500.81 mm with an error of 0.28 mm. This result verifies the ability of the high-frequency skin effect model to convert weak structural deformations. The high-frequency carrier compresses the current penetration depth, enabling the inductive detection mechanism to obtain local high-frequency leakage magnetic field distortion.

[0131] In tests with damage numbers 2 and 3, the positioning errors of the control group were 7.29 mm and 5.52 mm, respectively, while the positioning errors of the device of this invention were 0.33 mm and 0.27 mm, respectively. The error distribution of this scheme is within the range of ±0.35 mm. This error control level demonstrates the corrective effect of the dual-core heterogeneous processing mechanism and dynamic temperature compensation model in long-stroke coordinate mapping. The ARM Cortex-M7 core utilizes pulse compensation performed by temperature sensor data and the metal linear expansion mechanism to reduce the cumulative positional shift caused by mechanical lead thermal drift.

[0132] Meanwhile, the signal processing mechanism combines discrete wavelet transform with digital phase-locked demodulation algorithm to suppress low-frequency jitter noise caused by the operation of the mechanical slide and extract DC characteristic components; the embedded system mechanism eliminates numerical jump dead zones through phase dewinding processing and establishes the correspondence between impedance phase difference and physical fracture surface; the electromagnetic protection mechanism uses a permalloy and copper foil composite shield with a common-mode choke design to ensure that the system maintains effective differential-mode signal output under strong background noise. The comprehensive application of the above structures and algorithms forms a closed loop for data flow, transforming minute internal damage into coordinate output with millimeter-level positioning accuracy.

[0133] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A rotund coil breakage detection device, characterized in that, include: The high-frequency excitation injection mechanism 10 is connected to both ends of the main circuit of the rotund coil under test, and is used to inject a high-frequency AC excitation signal into both ends of the main circuit of the rotund coil under test; The induction detection mechanism 20 is used to pick up the spatial alternating leakage magnetic field distortion signal caused by the local impedance change outside the Roche coil to be tested, and output the corresponding original induced electromotive force signal. The mechanical scanning mechanism 30 is used to carry the sensing and detection mechanism 20 to scan and move along the surface of the rotatable coil to be tested, and output a mechanical displacement pulse signal corresponding to the scanning displacement in real time; The signal processing mechanism 40 is connected to the induction detection mechanism 20 and is used to receive the original induced electromotive force signal and perform digital filtering and synchronous demodulation operations to extract in-phase characteristic components and quadrature characteristic components. The embedded system mechanism 50 is communicatively connected to the high-frequency excitation injection mechanism 10, the mechanical scanning mechanism 30, and the signal processing mechanism 40, respectively. It is used to receive the in-phase characteristic component and the quadrature characteristic component, calculate the transient phase angle and the impedance phase difference value, and when the impedance phase difference value exceeds the set threshold, calculate the absolute physical coordinates of the breakpoint based on the number of mechanical displacement pulses generated by the accumulated mechanical displacement pulse signal.

2. The ROJ coil breakage detection device according to claim 1, characterized in that, The high-frequency excitation injection mechanism 10 includes a direct digital frequency synthesis unit, a power amplification unit, and an impedance matching network. The direct digital frequency synthesis unit generates an initial high-frequency sinusoidal AC signal according to the control command. The high-frequency sinusoidal AC signal is amplified by the power amplification unit and the impedance is balanced by the impedance matching network before being output as the high-frequency AC excitation signal.

3. The ROJ coil breakage detection device according to claim 1, characterized in that, The sensing detection mechanism 20 employs a non-contact differential high-frequency sensing probe, comprising two miniature receiving coils with opposite winding directions, identical geometric dimensions, and parallel arrangement along the axial direction of the measured Roche coil. The two miniature receiving coils are connected in reverse series in the electrical circuit to physically cancel common-mode interference signals in the space environment through the differential structure, and output the original induced electromotive force signal containing the distorted phase of the alternating leakage magnetic field in space. The original induced electromotive force signal is the algebraic sum of the transient induced voltages of the two miniature receiving coils.

4. The ROJ coil breakage detection device according to claim 1, characterized in that, The mechanical scanning mechanism 30 includes a servo drive motor, a ball screw, a Hall sensor, and an incremental rotary encoder. The Hall sensor, in conjunction with a trigger magnet fixed on the moving part, is used to capture level flip signals during the system initialization phase to calibrate the mechanical coordinate zero point. The incremental rotary encoder operates synchronously with the rotor of the servo drive motor and is used to output the mechanical displacement pulse signal containing displacement increment information in real time during the scanning phase.

5. The ROJ coil breakage detection device according to claim 1, characterized in that, The signal processing mechanism 40 includes a programmable gain amplifier and a field-programmable gate array (FPGA). The programmable gain amplifier is used to adaptively adjust the analog amplification gain to amplify and discretize the original induced electromotive force signal. The FPGA is internally configured with a wavelet transform unit. The wavelet transform unit is used to perform a discrete wavelet transform algorithm on the sampled discrete digital signal and to clear the wavelet coefficients corresponding to mechanical jitter and high-frequency white noise by performing hard thresholding on the wavelet coefficients at a specific scale, thereby reconstructing and outputting a denoised high-frequency carrier signal.

6. The ROJ coil breakage detection device according to claim 5, characterized in that, The field-programmable gate array is also used to perform digital synchronous demodulation operations, specifically including: A local quadrature reference signal with the same frequency and phase as the high-frequency AC excitation signal is generated. This signal is then mixed with the denoised high-frequency carrier signal to generate an in-phase intermediate component and a quadrature intermediate component. A smoothing integral calculation is then performed on the two intermediate components using a digital low-pass filter to extract the in-phase and quadrature characteristic components that have been shifted to the DC frequency band.

7. The ROJ coil breakage detection device according to claim 1, characterized in that, The embedded system 50 internally includes a digital signal processing core and a microcontroller core. The digital signal processing core calls a four-quadrant arctangent function to calculate the transient phase angle of the current position based on the input orthogonal and in-phase feature components. The digital signal processing core further introduces a transient phase angle based on the input of adjacent sampling periods. Periodic phase dewinding is performed to eliminate numerical winding dead zones and extract the corrected impedance phase difference value.

8. The ROJ coil breakage detection device according to claim 7, characterized in that, The microcontroller core uses a temperature sensor to obtain the ambient temperature, calculates a dynamic compensation coefficient based on the linear expansion coefficient of the ball screw, and multiplies the original accumulated mechanical displacement pulse count by the dynamic compensation coefficient to generate a corrected mechanical displacement pulse count. When the fault interruption flag is received, the microcontroller core hardware captures an absolute timestamp, synchronizes it with the corrected mechanical displacement pulse count, and substitutes it into the spatial coordinate mapping equation to solve for the absolute physical coordinates.

9. The ROJ coil breakage detection device according to claim 1, characterized in that, It also includes an electromagnetic protection mechanism 60, which is fixedly connected to the mechanical scanning mechanism 30 and covers the outside of the induction detection mechanism 20 and the signal processing mechanism 40. The electromagnetic protection mechanism 60 adopts a composite shielding structure composed of an outer layer of high magnetic permeability material and an inner layer of high electrical conductivity material. The outer layer of high magnetic permeability material is used to provide a low magnetic resistance bypass channel to close the magnetic field lines for intruding low-frequency magnetic field lines. The inner layer of high electrical conductivity material is used to generate an internal electromagnetic eddy current effect when high-frequency stray electromagnetic waves intrude, so as to dissipate and absorb the energy of high-frequency electromagnetic waves.

10. The Root coil breakage detection device according to claim 9, characterized in that, The electromagnetic protection mechanism 60 is also connected in series in the power supply network and communication line of the embedded system mechanism 50 to the external interface, and a common-mode choke is connected in series in the specific circuit topology. The common-mode choke is composed of two sets of coils wound in the same direction on the same high-frequency ferrite core. When external high-frequency common-mode noise enters in the same direction along the two conductors of the power supply network or communication line, the two sets of coils generate magnetic flux in the same direction and superimpose in the core, presenting a high common-mode impedance. The high common-mode impedance is used to divide and block the conducted common-mode noise on the power supply and communication lines.

Images