A high-frequency dynamic response type LVDT displacement sensor and a displacement measurement method thereof

Abstract

Provided are a high-frequency dynamic response type LVDT displacement sensor and a displacement measurement method thereof, and belong to the field of precision displacement measurement. The sensor comprises a coil framework, a primary coil, a secondary coil, a core, a shell and a signal processing circuit. The primary coil and the secondary coil adopt a layered and segmented winding structure, an insulating layer is arranged between the layers, and a gap is arranged between the segments, so as to reduce the distributed capacitance and the eddy current loss; the core adopts a nanocrystalline alloy material and has high magnetic permeability and low loss; the signal processing circuit comprises a high-frequency excitation source of 10 kHz to 20 kHz, a lock-in amplifier and an LMS adaptive filtering module, and is used for suppressing noise. The measurement method comprises generating a high-frequency excitation, converting displacement into a differential signal, lock-in demodulation, adaptive filtering, temperature compensation and digital output. The dynamic response frequency of the present application is greater than or equal to 10 kHz, the response time is less than 1 ms, the precision is better than ±0.1% F.S., and the present application is suitable for high-frequency dynamic displacement measurement of aircraft control surfaces, engine vibration monitoring and the like.

Description

Technical Field

[0001] This invention belongs to the field of precision displacement measurement sensor technology, specifically relating to a high-frequency dynamic response LVDT (Linear Variable Differential Transformer) displacement sensor and its displacement measurement method, which is suitable for high-frequency dynamic displacement precision measurement in aerospace, high-end equipment, industrial automation and other fields. Background Technology

[0002] Displacement sensors are core detection components in modern industrial automation, intelligent manufacturing, and high-end equipment control, used to convert mechanical displacement into electrical signal output. LVDT displacement sensors, due to their advantages such as frictionless measurement, zero-position stability, high repeatability, and long lifespan, are widely used in aerospace vehicle control surface position feedback, engine vibration monitoring, hydraulic control systems, and industrial automation equipment.

[0003] However, with the rapid development of high-end equipment technology, traditional LVDT sensors are increasingly showing limitations in dynamic response performance. The excitation frequency of traditional LVDT sensors is usually 2kHz to 5kHz, and the dynamic response frequency (-3dB bandwidth) is generally no more than 2kHz to 3kHz, which is difficult to meet the needs of the following high-frequency dynamic measurement scenarios: (1) Modern aircraft adopt fly-by-wire or fly-by-light flight control technology, and the control surface frequency is greatly increased, requiring the position feedback sensor dynamic response frequency to be ≥5kHz, and some high-performance flight control systems require ≥10kHz; (2) Aircraft engine health monitoring requires high-frequency dynamic measurement of blade vibration and rotor displacement, and the frequency range usually covers 1kHz to 10kHz; (3) Closed-loop control of high-speed motion mechanisms (such as high-speed stamping equipment and precision CNC machine tool feed systems) requires displacement feedback with sub-millisecond response time.

[0004] The main reasons for the limited dynamic response of traditional LVDT sensors include: (1) Coil distributed capacitance and eddy current loss: Traditional LVDTs adopt a single-layer dense winding or simple layered winding structure, with a large distributed capacitance between windings. At high frequencies, the capacitive current increases, leading to energy loss. At the same time, the high-frequency alternating magnetic field induces eddy currents in conductive materials, generating additional losses and phase delays. (2) Insufficient high-frequency characteristics of magnetic core materials: Traditional LVDTs mostly use silicon steel sheets or permalloy as magnetic core materials. At high frequencies (>5kHz), the permeability decreases and the magnetic core loss increases sharply, resulting in a decrease in magnetic field transmission efficiency. (3) Low excitation frequency: Traditional designs are limited by the above factors, and the excitation frequency cannot be significantly increased, which restricts the dynamic response bandwidth. (4) Insufficient bandwidth of signal processing circuits: The bandwidth design of traditional phase-sensitive detector circuits and filter circuits is designed for low-frequency signals. The signal-to-noise ratio deteriorates when processing high-frequency signals.

[0005] To address these issues, researchers both domestically and internationally have conducted extensive research. For example, Honeywell in the United States developed a series of high-response LVDT products, improving the dynamic response to around 5kHz by modifying the coil structure and using special magnetic core materials; RDP Electronics in the United Kingdom used digital demodulation technology to replace traditional analog phase-sensitive detection, which improved high-frequency performance to some extent. However, these solutions still have shortcomings in areas such as further increasing the excitation frequency (>10kHz), optimizing magnetic core materials, and suppressing high-frequency noise.

[0006] Therefore, there is an urgent need to develop a high-frequency dynamic response LVDT displacement sensor with independent intellectual property rights, breaking through the high-frequency performance bottleneck of traditional LVDTs and achieving a dynamic response frequency ≥10kHz. Therefore, it is necessary to propose improvements. Summary of the Invention

[0007] The technical problem solved by this invention: The purpose of this invention is to provide a high-frequency dynamic response LVDT displacement sensor and its displacement measurement method. By systematically optimizing the coil structure, magnetic circuit design, excitation frequency and signal processing algorithm, the dynamic response frequency is ≥10kHz (-3dB bandwidth), the response time is <1ms, and the measurement accuracy is better than ±0.1% FS, thus meeting the stringent requirements of high-frequency dynamic displacement measurement.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: A high-frequency dynamic response LVDT displacement sensor includes a coil frame, a primary coil, a secondary coil, an iron core, a housing, and a signal processing circuit. The primary and secondary coils are wound on the coil frame. The iron core is movably disposed within a central through-hole of the coil frame. The coil frame is located inside the housing. The signal processing circuit is electrically connected to the primary and secondary coils. The primary coil and the secondary coil are respectively wound on the coil frame using a layered and segmented winding structure, with an insulating isolation layer between each layer of winding and an insulating gap between each segment; The core is made of a nanocrystalline alloy material with high magnetic permeability and low loss. The signal processing circuit includes a high-frequency excitation source, a lock-in amplifier, and an adaptive filtering module. The excitation frequency range of the high-frequency excitation source is 10kHz to 20kHz.

[0009] Furthermore, the layered and segmented winding structure is as follows: the primary coil and the secondary coil are wound in N layers, N≥3, each layer contains M segments, M≥2; the winding directions between adjacent layers are opposite, and an insulation gap of 0.5mm to 1.5mm is provided between adjacent segments in the same layer; the insulation isolation layer is made of polyimide film or polytetrafluoroethylene film with a thickness of 0.05mm to 0.15mm.

[0010] Furthermore, the nanocrystalline alloy material is an Fe-based nanocrystalline alloy, with the following composition by mass percentage: Fe 70%–85%, Si 8%–15%, B 5%–10%, Cu 0.5%–1.5%, and Nb 2%–5%; the initial permeability μi of the iron core is ≥80000, the saturation magnetic induction Bs is ≥1.2T, the coercivity Hc is ≤1.0A / m, and the core loss in the frequency range of 10kHz–20kHz is less than 50mW / cm. 3 The iron core surface is coated with an insulating layer with a thickness of 2μm to 10μm and an interlayer resistance ≥100MΩ.

[0011] Furthermore, the high-frequency excitation source adopts direct digital frequency synthesis (DDS) technology, with an adjustable excitation frequency range of 10kHz to 20kHz, a frequency resolution of 0.01Hz, an excitation waveform of sine wave, a total harmonic distortion (THD) of ≤0.1%, an adjustable excitation amplitude range of 1Vpp to 10Vpp, and an amplitude stability of ≤0.05%.

[0012] Furthermore, the lock-in amplifier includes a quadrature demodulation circuit, a low-pass filter, and an automatic gain control (AGC) circuit; the quadrature demodulation circuit uses two reference signals with a 90° phase difference for synchronous detection; the cutoff frequency of the low-pass filter is 1 / 10 to 1 / 5 of the excitation frequency, used to extract the effective signal components; the automatic gain control circuit is used to compensate for signal amplitude changes caused by displacement changes.

[0013] Furthermore, the adaptive filtering module is used to suppress environmental electromagnetic interference and measurement noise in real time. It adopts the LMS (Least Mean Square) adaptive algorithm, and the reference input signal is taken from the output of the high-frequency excitation source as the phase reference of the desired signal. The filter order is adjustable from 16 to 128, the step size factor μ is adaptively adjusted from 0.001 to 0.1, the convergence time is ≤10ms, and the steady-state mean square error is ≤0.1%.

[0014] Furthermore, the secondary coil includes a primary coil and a secondary coil, which are connected in reverse series to form a differential voltage output; the signal processing circuit also includes an instrumentation amplifier, which amplifies the differential voltage output before outputting it, with a common-mode rejection ratio (CMRR) ≥ 80dB; the output signal is converted into a digital signal by an A / D converter, with a resolution of 16-bit or 24-bit and a sampling rate ≥ 100kS / s.

[0015] Furthermore, it also includes a temperature compensation module, which includes a built-in temperature sensor (PT100 or NTC thermistor) and a temperature compensation algorithm unit. The temperature sensor is located inside the coil frame and is used to monitor the temperature of the coil and the magnetic core in real time. The temperature compensation algorithm unit adopts a polynomial fitting or neural network method, and the compensation temperature range is -55℃ to +150℃, with a temperature drift of ≤±0.02% FS / ℃ across the entire temperature range.

[0016] Furthermore, the outer shell is made of stainless steel or titanium alloy, with a protection level of IP67 or above; the coil frame and the outer shell are sealed using a vacuum potting process, with epoxy resin or silicone rubber as the potting material; the leads are led out through glass sintered sealed terminals to ensure airtightness and electromagnetic shielding performance.

[0017] A displacement measurement method based on the aforementioned high-frequency dynamic response LVDT displacement sensor includes the following steps: S1: A sinusoidal excitation signal of 10kHz to 20kHz is generated by the high-frequency excitation source and applied to the primary coil; S2: Using the primary coil and the secondary coil wound in layers and segments, in conjunction with the iron core, the displacement of the iron core is converted into a differential voltage signal; S3: The lock-in amplifier is used to synchronously detect and quadrature demodulate the output signal of the secondary coil to extract the effective displacement signal; S4: The adaptive filtering module is used to suppress high-frequency noise and electromagnetic interference in real time; S5: Real-time compensation for temperature drift is achieved through a temperature compensation module; S6: Converts the signal to a digital signal via an A / D converter and outputs the displacement measurement result through a digital interface (RS-422 / 485, CAN, or SPI); The method achieves a dynamic response frequency ≥10kHz (-3dB bandwidth), measurement accuracy better than ±0.1% FS, and response time <1ms.

[0018] Advantages of this invention compared to existing technologies: 1. The dynamic response performance of this solution is greatly improved: Through the synergistic optimization of the layered and segmented winding coil structure and the nanocrystalline alloy magnetic core material, combined with the 10kHz~20kHz high-frequency excitation, the dynamic response frequency is ≥10kHz (-3dB bandwidth) and the response time is <1ms, which is 3 to 5 times better than the traditional LVDT (bandwidth 2kHz~3kHz), meeting the demanding application requirements of aircraft control surface rapid control, engine high-frequency vibration monitoring, etc. 2. This solution significantly reduces high-frequency losses: the layered and segmented winding structure reduces distributed capacitance to less than 30% of that of traditional structures, and the nanocrystalline alloy core exhibits losses of less than 50mW / cm in the 10kHz–20kHz range. 3 (Traditional permalloy approximately 200 mW / cm) 3 This effectively solves the problems of energy loss and heat generation under high-frequency excitation, ensuring the long-term stable operation of the sensor; 3. Improved signal-to-noise ratio and measurement accuracy: The signal processing scheme combining lock-in amplification and LMS adaptive filtering effectively suppresses high-frequency electromagnetic interference and measurement noise, improving the signal-to-noise ratio by more than 15dB. Combined with 16-bit / 24-bit high-resolution A / D conversion, it achieves a measurement accuracy better than ±0.1% FS. 4. This solution has strong environmental adaptability: the temperature compensation module achieves a temperature drift of ≤±0.02% FS / ℃ across the entire temperature range of -55℃ to +150℃; the hermetic encapsulation structure (IP67 protection rating) meets the requirements for use in harsh environments such as high vibration, high shock, humid heat, and salt spray, making it particularly suitable for extreme environment applications such as aerospace. 5. This solution has independent intellectual property rights: the core technologies such as the layered and segmented winding process, the nanocrystalline alloy magnetic core formula, and the high-frequency signal processing algorithm all have independent intellectual property rights, which eliminates the dependence on imported high-end LVDT products and is of great significance to ensuring the security of the equipment supply chain in key areas. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the overall structure of the high-frequency dynamic response LVDT displacement sensor of the present invention; Figure 2 This is a cross-sectional schematic diagram of the layered and segmented winding coil structure of the present invention; Figure 3 This is a diagram showing the structure and magnetic performance curves of the nanocrystalline alloy magnetic core of this invention; Figure 4 This is a schematic block diagram of the signal processing circuit of the present invention; Figure 5 This is a flowchart of the lock-in amplification and adaptive filtering signal processing of the present invention; Figure 6 This is a schematic diagram illustrating the principle of the temperature compensation algorithm of the present invention; Figure 7 This is the frequency response characteristic curve of the sensor of the present invention; Figure 8 The above are the dynamic response test curves of the sensor in a typical application scenario in this embodiment of the invention. Detailed Implementation

[0020] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Please see Figures 1 to 8 The embodiments of the present invention are described in detail below.

[0022] Example 1: Structural Design of a High-Frequency Dynamic Response LVDT Displacement Sensor like Figure 1 As shown, the high-frequency dynamic response LVDT displacement sensor in this embodiment mainly includes: coil frame 1, primary coil 2, secondary coil 3 (first primary coil and second secondary coil), iron core 4, housing 5, signal processing circuit board 6, lead cable 7 and connector 8.

[0023] The coil bobbin 1 is made of high-strength engineering plastic (such as PEEK or PPS), which has high mechanical strength and temperature resistance (long-term operating temperature ≥180℃). The coil bobbin 1 has a central through hole for the reciprocating motion of the iron core 4. The outer surface of the coil bobbin 1 is machined with precision winding grooves, with a groove spacing accuracy of ±0.02mm, ensuring accurate winding positioning.

[0024] like Figure 2 As shown, the primary coil 2 and secondary coil 3 adopt a layered and segmented winding structure. Taking the primary coil 2 as an example, it is wound in 4 layers (N=4), and each layer contains 3 segments (M=3). The winding parameters are as follows: the enameled wire is φ0.05mm high-strength enameled copper wire, and the insulation varnish layer is heat resistant to 180℃ (H class); each layer has approximately 500 turns; the interlayer insulation uses polyimide film (thickness 0.08mm) and adopts a semi-overlapping wrapping method; the inter-segment insulation gap is 1.0mm, which is filled with epoxy resin for fixation; the winding direction is clockwise for the first layer → counterclockwise for the second layer → clockwise for the third layer → counterclockwise for the fourth layer, arranged alternately.

[0025] During layered and segmented winding, a four-level precision control system—comprising a self-positioning skeleton structure (positioning boss), CNC winding fixtures, closed-loop tension control, and visual inspection and calibration—ensures the axial position accuracy of each coil segment is controlled within ±0.05mm, guaranteeing symmetrical magnetic field distribution and balanced differential output. Specifically: The coil bobbin is integrally injection molded from high-temperature resistant polyimide material. N-1 annular positioning bosses are precisely machined along the axial direction on the outer wall of the bobbin, forming N equidistant winding segments. During winding, the enameled wire fits tightly against the sidewalls of the positioning bosses, achieving automatic axial positioning. The positioning bosses also serve as the supporting frame for the conductive shielding layer, with the shielding copper foil embedded in the grooves between adjacent bosses.

[0026] The winding process uses a precision CNC automatic winding machine, equipped with a special positioning fixture. The fixture includes: a front center and a rear center, an axial positioning reference block, a pneumatic clamping mechanism, and a CNC wire laying system.

[0027] Before winding, input the winding parameters for each coil segment into the CNC system: starting point axial coordinate, ending point axial coordinate, number of turns, tension value, and wire pitch. The winding machine automatically executes the winding of each segment in sequence. During the transition between segments, the guide pin quickly moves to the starting point of the next segment at a speed of 100mm / s to 200mm / s to avoid dragging the wire on the bobbin surface and causing positional deviation.

[0028] Winding tension control: A closed-loop tension control system is adopted, with a tension range of 50g to 150g (adjusted according to wire diameter) and tension fluctuation ≤ ±5%. Appropriate tension ensures that the enameled wire is tightly attached to the surface of the bobbin and the side wall of the positioning boss, preventing axial movement caused by winding slack.

[0029] After winding is completed, a high-resolution industrial camera is used to perform axial visual inspection of the coil frame. The inspection includes: the actual axial position of each coil segment, the spacing between segments, and the uniformity of the coil outer diameter. The vision system and the CNC winding machine form a closed-loop feedback: if the position deviation of a certain coil segment exceeds ±0.1mm, the system automatically marks the coil segment as unqualified and prompts for adjustment of tooling parameters or replacement of the frame.

[0030] Through the above three-level precision control, the axial position accuracy of each coil segment can be stably controlled within ±0.05mm, meeting the strict requirements of high-frequency dynamic response LVDT for magnetic field distribution symmetry.

[0031] After layered and segmented winding, each segment of the primary coil and each segment of the secondary coil are connected in series; the two sets of secondary coils are connected in series differentially with opposite polarity. The fully series connection ensures uniform current, superimposed voltage, and symmetrical magnetic field, which significantly improves the linearity and zero-point stability of the LVDT compared to the parallel connection.

[0032] The leads of the primary and secondary coils are safely and reliably led out in the multi-layer structure through the axial routing grooves and radial lead-out holes prefabricated in the skeleton, combined with layered isolation and insulating sleeve protection, while meeting the requirements of high frequency and low inductance.

[0033] The first and second secondary coils are symmetrically arranged on both sides of the primary coil 2, and their winding parameters are similar to those of the primary coil 2, but the number of turns is adjusted according to the output sensitivity requirements. The two secondary coils 3a and 3b are connected in series in opposite directions to form a differential output.

[0034] Tests have verified that the distributed capacitance of this layered and segmented winding structure is 25% of that of the traditional single-layer dense winding structure. At an excitation frequency of 15kHz, the coil quality factor Q is increased from 35 to 82 in the traditional structure, and the high-frequency loss is reduced by 62%.

[0035] Example 2: Preparation and Performance Testing of Nanocrystalline Alloy Cores The iron core 4 is made of Fe-based nanocrystalline alloy material, with the following specific composition ratio (mass percentage): Fe 78%, Si 10%, B 7%, Cu 1%, Nb 4%. The manufacturing process is as follows: (1) Amorphous alloy strips were prepared by single-roll rapid quenching method. The strips were about 25 μm thick and 10 mm wide. (2) The amorphous thin strip is wound into a ring-shaped magnetic core (outer diameter φ8mm, inner diameter φ4mm, height 5mm). The design basis for the matching relationship between the magnetic core height and the coil length, and the design basis for the magnetic core stroke and the sensor range are as follows: Matching relationship between magnetic core height and coil length 1) Basic Design Formula The total length L_coil of the LVDT sensor coil frame is determined by the following formula: L_coil = 2×S_range + h_core + 2×δ In the formula: S_range is the sensor range (±10mm, i.e., 10mm on one side); h_core is the magnetic core height (5mm); δ is the margin at both ends (to prevent the magnetic core from coming out of the effective area of ​​the coil, usually taken as 2mm to 5mm).

[0036] Substituting the parameters into this embodiment: L_coil = 2×10 + 5 + 2×2.5 = 30mm. That is, the total length of the coil frame is designed to be 30mm, and the magnetic core reciprocates within a 30mm long coil area for 20mm, with a 5mm margin at each end to ensure that the magnetic core is within the effective magnetic field range of the coil at any position.

[0037] 2) The ratio of core height to coil length In LVDT sensor design, the ratio of core height to total coil length (h_core / L_coil) is a critical design parameter, directly affecting sensitivity, linearity, and measurement range. Engineering practice shows that: When h_core / L_coil = 1 / 3 to 1 / 5: linearity is optimal (≤0.5%FS), but sensitivity is moderate; When h_core / L_coil = 1 / 5 to 1 / 8: the sensitivity is high and the linearity is good (≤1%FS), which is the commonly used range in engineering. When h_core / L_coil < 1 / 8: the sensitivity is very high, but the linear range is narrowed, and it is only suitable for precision measurements with small ranges.

[0038] In this embodiment, h_core / L_coil = 5 / 30≈1 / 6, which is within the optimal engineering range (1 / 5 to 1 / 8), balancing high sensitivity and good linearity.

[0039] Design basis for magnetic core stroke and sensor range 1) LVDT Measurement Principle and Core Stroke The measurement principle of LVDT sensors is based on electromagnetic coupling: the primary coil is excited by AC to generate an alternating magnetic field; the magnetic core, as a high-permeability magnetic circuit, couples the primary magnetic field to the secondary coils on both sides. When the magnetic core is displaced axially, the coupled magnetic flux of the secondary coils on both sides changes, generating a differential voltage output.

[0040] The key is that the magnetic core does not need to cover the entire secondary coil; differential output can be generated simply by simultaneously coupling the two secondary coils during the movement. Even if the magnetic core height (5mm) is much smaller than the total stroke (20mm), continuous linear measurement can be achieved as long as the magnetic core can form magnetic coupling with the secondary coils on both sides at any position.

[0041] 2) Physical constraints on core stroke and range The physical limit of the sensor's measurement range S_range is determined by the following factors: Core travel limit: The core center must not move beyond the effective area of ​​the coil, i.e., |S_range|≤(L_coil - h_core) / 2 -δ; Linearity constraint: Within the measurement range, the coupling area between the magnetic core and the two secondary coils should maintain an approximately linear relationship, requiring the magnetic core height to match the length of the secondary coil segment; Magnetic field uniformity: The magnetic field generated by the primary coil is not completely uniform in the axial direction. The range should be limited to the region where the magnetic field strength changes by less than 10%.

[0042] This embodiment verifies that with L_coil = 30mm, h_core = 5mm, δ = 2.5mm, the theoretical maximum range S_max = (30 - 5) / 2 - 2.5 = 10mm. The actual range is designed to be ±10mm, which exactly reaches the theoretical limit, resulting in a compact design.

[0043] 3) Mechanism of the influence of core height on linearity The smaller the core height (relative to the range), the smaller the coupling area between the core and the secondary coil (S2) at its extreme position (e.g., +10mm), and the more pronounced the nonlinearity of the output voltage. However, this invention compensates for this by the following measures: Layered and segmented winding: The secondary coil is divided into multiple segments (M=4), each segment is relatively short (about 5mm to 6mm), and the core height (5mm) is matched with the length of a single secondary coil segment to ensure that the core always completely covers at least one secondary coil segment within a range of ±10mm, maintaining effective coupling; Differential connection: Group S1 and Group S2 are connected in series with opposite polarities, which partially cancels out nonlinear errors and improves the equivalent linearity by more than 30%. Digital nonlinearity correction: The FPGA adaptive algorithm performs piecewise linearization compensation on the output, with residual nonlinearity <0.3%FS.

[0044] Therefore, although the core height of 5mm is less than the measurement range (10mm), a linear measurement range of ±10mm can be fully achieved through the comprehensive optimization of segmented winding structure, differential connection and digital compensation.

[0045] Design Parameter Comparison Table

[0046] Annealing was performed under a nitrogen protective atmosphere: heating rate 5℃ / min, annealing temperature 560℃, holding time 1 hour, and furnace cooling.

[0047] In this embodiment, the annealing treatment of the nanocrystalline alloy magnetic core is performed using a horizontal vacuum tube annealing furnace or a vertical box-type protective atmosphere annealing furnace. Both furnace types can meet the crystallization annealing requirements of nanocrystalline alloys. The specific choice depends on the production batch and the size of the magnetic core. In this embodiment, the length-to-diameter ratio of the magnetic core is relatively small, so a horizontal vacuum tube annealing furnace is recommended for batch processing.

[0048] Nitrogen protective atmosphere parameters Nitrogen purity requirements: High-purity nitrogen (N2) must be used, with a purity ≥99.99% (4N grade), oxygen content ≤10ppm, and water content (dew point) ≤-40℃. Trace amounts of oxygen and water vapor in nitrogen will react with the Fe element in the nanocrystalline alloy at high temperatures, forming Fe2O3 or Fe3O4 oxide layers. This leads to damage to the insulating layer on the magnetic core surface and deterioration of magnetic properties (a decrease in permeability of 10%–20%). Therefore, nitrogen purity is a critical control parameter in the annealing process.

[0049] Nitrogen flow control: The nitrogen flow rate is determined based on the furnace volume and the number of air exchanges. Furnace volume V (for a tube furnace Φ80mm×1000mm, V≈5L): Recommended nitrogen flow rate 20~30 L / min; Furnace volume V (box furnace 400×400×400mm, V≈64L): Recommended nitrogen flow rate 40~60 L / min; Air exchange rate: ≥5 times / hour during the heating stage and ≥3 times / hour during the heat preservation stage, ensuring that the oxygen content in the furnace is always <50ppm.

[0050] This embodiment uses a horizontal tubular annealing furnace (V≈5L) with a nitrogen flow rate of 25 L / min, corresponding to approximately 300 air changes per hour, which is far higher than the minimum requirement, ensuring a pure atmosphere inside the furnace.

[0051] Furnace pressure control: The furnace is maintained under a slight positive pressure, ranging from 0.02 MPa to 0.05 MPa (gauge pressure). The purpose of this slight positive pressure is to prevent outside air (including oxygen and water vapor) from seeping into the furnace through any gaps in the furnace seal; to maintain the nitrogen flow direction from the inside out, forming a stable gas curtain for protection; and to promptly remove trace amounts of volatile organic compounds (such as adsorbed gases released during residual stress relaxation) released by the magnetic core during annealing.

[0052] Pressure control method: A fine-tuning valve and a U-tube manometer (or digital pressure sensor) are installed at the gas outlet of the furnace tube. The valve opening is adjusted manually or automatically to stabilize the furnace pressure at 0.03 MPa ± 0.01 MPa. If the pressure is lower than 0.01 MPa, the nitrogen flow rate is immediately increased or the gas outlet valve is closed; if the pressure is higher than 0.06 MPa, the gas outlet valve is opened appropriately to release pressure.

[0053] During the annealing process, the following atmosphere parameters should be monitored and recorded in real time: Oxygen content: A zirconia oxygen analyzer or an electrochemical oxygen sensor was used, with a measurement range of 0 to 1000 ppm and an accuracy of ±1%FS. The data was recorded every 10 minutes. Dew point temperature: A cold mirror dew point meter was used, with a measurement range of -60℃ to +20℃ and an accuracy of ±0.2℃. The temperature was recorded every 10 minutes. Furnace pressure: A digital pressure transmitter (0-0.1 MPa, accuracy 0.1%FS) is used for continuous recording.

[0054] Annealing process record sheets should be kept as batch files for magnetic core production for a period of ≥10 years to facilitate quality traceability.

[0055] (4) After annealing, a MgO-Al2O3 composite insulating coating (5μm thick) is coated on the surface of the magnetic core and cured at 300℃ for 1 hour; The key performance indicators of the nanocrystalline alloy magnetic core were tested and are as follows: initial permeability μi is 105000 (tested at 1kHz); saturation magnetic induction Bs is 1.25T; coercivity Hc is 0.8A / m; and core loss at 10kHz is 42mW / cm. 3 The core loss at 15kHz is 48mW / cm. 3 The core loss at 20kHz is 55mW / cm. 3 The Curie temperature Tc is 520℃.

[0056] like Figure 3 As shown, the left image displays the Fe-based nanocrystalline alloy toroidal magnetic core structure (outer diameter φ8mm, inner diameter φ4mm, height 5mm), coated with an MgO-Al2O3 insulating layer. The right image shows a comparison of the BH magnetization curves. The nanocrystalline alloy exhibits higher initial permeability (μi≥80000), higher saturation magnetic induction (Bs=1.25T), and lower coercivity (Hc≤1.0A / m) compared to permalloy 1J85, and its loss is significantly lower than that of conventional materials in the 10kHz-20kHz frequency band. This nanocrystalline alloy magnetic core maintains high permeability (μ>60000) and low loss in the 10kHz~20kHz range, significantly outperforming the comparative permalloy magnetic core (1J85, with a loss of approximately 210mW / cm at 15kHz). 3 (approximately 25,000 μ).

[0057] After applying the nanocrystalline alloy magnetic core to the LVDT sensor, the sensor sensitivity was improved by about 35% and the phase delay was reduced by about 40% at an excitation frequency of 15kHz, laying the magnetic circuit foundation for high-frequency dynamic response.

[0058] Example 3: Design and Implementation of Signal Processing Circuit like Figure 4 As shown, the signal processing circuit 6 mainly includes the following modules: (1) DDS high-frequency excitation source: The ADI AD9834 DDS chip, together with a low-pass filter and a power amplifier, is used to generate a stable high-frequency sinusoidal excitation signal. Main parameters: frequency 15kHz (programmable adjustment), amplitude 5Vpp, THD=0.08%, amplitude stability 0.03%.

[0059] (2) Signal conditioning circuit: including a preamplifier (AD8429, low noise, high CMRR) and a bandpass filter (center frequency 15kHz, bandwidth ±2kHz).

[0060] (3) Lock-in amplifier: An analog multiplier (AD835) is used to achieve quadrature demodulation. The two reference signals are directly provided by the DDS, with a precise phase difference of 90°. The low-pass filter is a 4th-order Butterworth active filter with a cutoff frequency of 1.5kHz.

[0061] The lock-in amplifier in this embodiment is implemented in a fully digital manner, and the signal sampling, quadrature demodulation, low-pass filtering and amplitude calculation are completed based on the FPGA (Field Programmable Gate Array) chip.

[0062] Advantages of digital quadrature demodulation: No temperature drift: Digital calculations are unaffected by temperature, and performance remains consistent across the entire temperature range of -55℃ to +150℃; High phase accuracy: The phase difference between the two digital reference signals is exactly 90°, with an error of <0.01°; Large dynamic range: Digital quantization bits ≥ 16 bits, dynamic range > 96dB, which is better than analog schemes (typically 60dB). Highly programmable: The filter parameters, integration time, and gain can be flexibly adjusted within the FPGA to adapt to different excitation frequencies and bandwidth requirements.

[0063] Generation of two 90° phase difference reference signals: 1) DDS (Direct Digital Synthesis) Principle The FPGA integrates a DDS core, which generates two orthogonal reference signals through a phase accumulator.

[0064] 2) Phase synchronization and frequency tracking The reference frequency f_ref of the DDS is strictly synchronized with the sensor excitation frequency, obtained by dividing the excitation source clock using the internal PLL (phase-locked loop) of the FPGA. When the excitation frequency deviates slightly (±1%) due to temperature or load changes, the FPGA's adaptive frequency tracking algorithm adjusts Δφ in real time to ensure that f_ref always equals the excitation frequency, guaranteeing the coherence of the phase-locked amplification. Frequency tracking accuracy: ±0.1Hz (relative accuracy ±0.001%).

[0065] 3) Reference signal parameters

[0066] The specific circuit topology of the low-pass filter: 1) Selection of digital low-pass filter type The low-pass filter in the lock-in amplifier is used to filter out the high-frequency components generated after mixing, retaining the DC (low-frequency) signal components. This invention uses an FIR (Finite Impulse Response) digital filter instead of an IIR (Infinite Impulse Response) filter for the following reasons: FIR filters have strictly linear phase characteristics and constant group delay, which will not cause signal phase distortion and ensure the accurate phase relationship between the I and Q signals; FIR filters are unconditionally stable, have no feedback loops, and quantization noise does not accumulate. FIR filters have symmetrical coefficients and can be efficiently implemented using FPGA multiplier resources and shift registers.

[0067] 2) FIR filter design parameters

[0068] 3) FPGA implementation structure of FIR filter The serial multiply-accumulate (MAC) structure is used, and it is implemented using the FPGA's built-in DSP48 slice: Shift register chain: stores the most recent 128 sample points x[n] to x[n-127], each sample point is 16 bits; Coefficient memory: ROM stores 128 filter coefficients h[0]~h

[127] , 16-bit fixed-point numbers; Multiply-accumulate: Performs one multiplication (16×16→32bit) and accumulation per clock cycle, completes one convolution operation every 128 clock cycles, and outputs one filtered result; Parallel processing: The I-path and Q-path each use a MAC unit for independent parallel computation, with an output rate of 15kS / s.

[0069] (4) LMS Adaptive Filter: Implemented using TI's TMS320F28379D digital signal processor. The filter order is 64, the initial step size is μ=0.05, and the normalized LMS (NLMS) algorithm is used to prevent divergence. The reference signal is taken from the DDS excitation source output, and correlation operation is performed with the sensor output signal to extract the effective displacement signal in real time and suppress broadband noise.

[0070] The phase synchronization solution in this embodiment is a digital phase synchronization based on quadrature demodulation and multiplexing. This invention has implemented digital quadrature demodulation (I / Q dual-channel demodulation) in the lock-in amplifier section. This module can be directly reused as the phase synchronization source of the LMS adaptive filter without the need for additional hardware circuitry.

[0071] (5) Temperature compensation module: such as Figure 6 As shown, a PT100 platinum resistance temperature sensor (placed inside the coil frame) is used. The sensor is embedded within the frame material. After being excited by a constant current source and amplified by an instrumentation amplifier, the temperature data is acquired by a DSP. The temperature compensation algorithm uses a 5th-order polynomial fitting: ΔVtemp = a0 + a1T + a2T2 + a3T 3 + a4T 4 + a5T 5 Where T is the measured temperature (°C), and the coefficients a0 to a5 are determined through a full-temperature-range calibration experiment. After compensation, the output drift across the entire temperature range (-55°C to +150°C) is ≤ ±0.015% FS / °C.

[0072] (6) A / D Conversion and Digital Output: A 24-bit Σ-Δ A / D converter (ADS1256) is used, with a sampling rate of 100kS / s and an effective resolution of 22 bits. The digital interface supports RS-485 (Modbus protocol), CAN (CANopen protocol) and SPI modes.

[0073] like Figure 5 As shown, the signal processing flow of lock-in amplification and adaptive filtering is as follows: The sensor output signal Vs first passes through preamplifier and bandpass filter, and after being amplified by preamplifier (AD8429), it enters the quadrature demodulation stage through bandpass filter (center frequency 15kHz); DDS provides two reference signals with a 90° phase difference, which are synchronously detected with the sensor signal to obtain the in-phase component I and the quadrature component Q; the I / Q components are low-pass filtered and then further suppressed by LMS adaptive filter to suppress residual noise; finally, after temperature compensation and linearization correction, a high-precision digital displacement is output.

[0074] The performance of this signal processing circuit at a working frequency of 15kHz was tested and found to be as follows: signal-to-noise ratio (SNR) of 78dB (compared to approximately 62dB in conventional solutions); total harmonic distortion (THD) of -85dB; phase noise (1kHz offset) of -110dBc / Hz; response time (10%–90% step response) of 0.6ms; and dynamic response frequency (-3dB bandwidth) of 12.5kHz.

[0075] Example 4: Sensor Performance Testing and Verification A high-frequency dynamic response LVDT displacement sensor prototype (range ±10mm, outer diameter φ20mm, length 80mm) was manufactured according to the above embodiments one to three, and comprehensive performance testing was carried out.

[0076] (1) Static performance test: linearity is ±0.08% FS (within the range of ±10mm at full scale); sensitivity is about 200mV / mm / V (under 5V excitation); repeatability is ±0.03% FS; hysteresis is ±0.05% FS.

[0077] (2) Dynamic performance testing: A laser interferometer (Renishaw XL-80, resolution 0.1nm) was used as the displacement standard, and dynamic calibration was performed using a high-frequency vibration table (LDS V830, maximum frequency 20kHz). The test results are as follows: Figure 7 As shown in the graph: The amplitude-frequency response curve in the upper figure shows that the sensor of this scheme has a -3dB bandwidth of 12.8kHz and a -1dB bandwidth of 8.5kHz, which is significantly better than the imported Type A (6.5kHz) and the domestic Type B (2.8kHz). The phase-frequency response curve in the lower figure shows that the phase delay is <15° and the group delay variation is <5% within 10kHz; the rise time of the step response from 10% to 90% is 0.58ms and the overshoot is <2%; the frequency response flatness has a bandwidth of 9.2kHz within ±0.5dB. The test results show that the sensor's dynamic response frequency reaches and exceeds the design target of 10kHz, meeting the technical requirements for rapid control of aircraft control surfaces (response time <1ms).

[0078] (3) Environmental adaptability test: Temperature cycling -55℃↔+150℃, 1000 cycles, performance drift <±0.2%FS; High temperature storage +150℃ storage for 500 hours, performance recovery drift <±0.15%FS; Low temperature operation -55℃ operation for 4 hours, output sensitivity change <±2%; Vibration test random vibration 20g RMS (20Hz~2000Hz), sinusoidal vibration 30g (50Hz~500Hz), no structural damage, normal performance; Impact test 50g half sine wave, 11ms, 3 times on each of the three axes, no damage; EMC test conducted emission, radiation emission, conducted sensitivity, and radiation sensitivity tests according to the relevant GJB151B standards, all of which meet the requirements.

[0079] (4) Long-term reliability test: Continuous working life: 5000 hours of continuous operation at +85℃, performance drift < ±0.3% FS; Mechanical life: 10 million cycles of reciprocating motion of the iron core, wear amount < 1μm, and no significant deterioration of performance indicators.

[0080] Example 5: Verification of Typical Application Scenarios The sensor of this invention was applied to the control surface position feedback system of a certain type of aircraft, replacing the original traditional LVDT sensor, and flight test verification was carried out.

[0081] Application parameters: Control surface type is aileron; control surface deflection range is ±25°; maximum deflection rate is approximately 120° / s; control bandwidth requirement is ≥8Hz; position feedback accuracy requirement is ±0.1°.

[0082] Sensor installation method: The LVDT sensor body is fixed to the servo housing using a special clamp. The iron core is connected to the servo rocker arm via a connecting rod, converting the angular displacement of the servo surface into the linear displacement of the LVDT iron core (the transmission ratio is designed to be 1mm / °).

[0083] Flight test results: The sensor operated normally throughout the flight, with no signal loss or abnormal jumps; in terms of dynamic tracking performance, the sensor output delay was <0.7ms and the equivalent phase delay was <3° (@8Hz) at maximum deflection, meeting the stability requirements of the flight control system; in terms of accuracy verification and comparison with the control surface position calculated by the fiber optic gyroscope inside the flight control system, the sensor measurement error was <±0.08° (RMS), which is better than the ±0.1° requirement; in terms of environmental adaptability, the sensor output was stable under high temperature (+70℃) and high vibration (approximately 8g RMS measured in flight), with no drift exceeding the standard.

[0084] like Figure 8 As shown in the figure above, the step response characteristics are compared. The rise time (10%-90%) of this scheme is 0.58ms, which is significantly faster than the imported A-type (1.2ms) and the domestically produced B-type (3.5ms). The figure below shows the dynamic tracking curve of the aileron control surface during a flight test. The sensor outputs smooth tracking when the control surface changes direction rapidly (about 100° / s), without overshoot or oscillation. The measurement error is <±0.08° (RMS), and the dynamic tracking delay is <0.7ms, which fully verifies the high-frequency dynamic response performance and reliability.

[0085] Example 6: Comparison Test with Other Sensors The high-frequency dynamic response LVDT sensor of this invention (No.: this invention-Type A) was compared with three mainstream sensors on the market: Comparison sample 1 was a high-response LVDT (imported Type A) from a certain American company; Comparison sample 2 was a traditional LVDT (domestic Type B) from a certain domestic manufacturer; and Comparison sample 3 was an inductive encoder (imported Type C) from a certain German brand.

[0086] The results of the comparative tests are summarized in the table below:

[0087] Test conclusions: (1) The dynamic response frequency (12.8kHz) of the present invention-A type is significantly better than that of the imported A type (6.5kHz) and the domestic B type (2.8kHz), and slightly lower than that of the imported C type inductive encoder (15kHz), but the price is only 24% of that of the imported C type, which has a significant cost advantage; (2) The present invention-A type is significantly better than all the comparison samples in terms of temperature range (-55℃~+150℃) and vibration resistance (30g), and is particularly suitable for extreme environment applications such as aerospace; (3) Based on the comprehensive cost-effectiveness evaluation, the present invention-A type has the best cost-effectiveness in high temperature and high frequency application scenarios, and can completely replace the imported A type and the domestic B type, and can replace the imported C type in cost-sensitive scenarios.

[0088] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0089] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A high-frequency dynamic response LVDT displacement sensor, comprising a coil frame (1), a primary coil (2), a secondary coil (3), an iron core (4), a housing (5), and a signal processing circuit (6), wherein the primary coil (2) and the secondary coil (3) are wound on the coil frame (1), the iron core (4) is movably disposed in the central through hole of the coil frame (1), the coil frame (1) is disposed inside the housing (5), and the signal processing circuit (6) is electrically connected to the primary coil (2) and the secondary coil (3), characterized in that: The primary coil (2) and the secondary coil (3) are respectively wound on the coil skeleton (1) using a layered and segmented winding structure. An insulating isolation layer is provided between each layer of windings, and an insulating gap is provided between segments. The iron core (4) is made of nanocrystalline alloy material; The signal processing circuit (6) includes a high-frequency excitation source, a lock-in amplifier and an adaptive filtering module, wherein the excitation frequency range of the high-frequency excitation source is 10kHz to 20kHz.

2. The high-frequency dynamic response LVDT displacement sensor according to claim 1, characterized in that: The layered and segmented winding structure is as follows: the primary coil (2) and the secondary coil (3) are wound into N layers, N≥3, and each layer contains M segments, M≥2; the winding directions between adjacent layers are opposite, and an insulation gap of 0.5mm to 1.5mm is set between adjacent segments in the same layer; the insulation isolation layer is made of polyimide film or polytetrafluoroethylene film with a thickness of 0.05mm to 0.15mm.

3. The high-frequency dynamic response LVDT displacement sensor according to claim 1, characterized in that: The nanocrystalline alloy material is an Fe-based nanocrystalline alloy, and its composition by mass percentage is: Fe 70%~85%, Si 8%~15%, B 5%~10%, Cu 0.5%~1.5%, Nb 2%~5%; the initial permeability μi of the iron core (4) is ≥80000, the saturation magnetic induction intensity Bs≥1.2T, the coercivity Hc≤1.0A / m, and the core loss in the frequency range of 10kHz~20kHz is less than 50mW / cm. 3 The surface of the iron core (4) is coated with an insulating layer with a thickness of 2μm to 10μm.

4. The high-frequency dynamic response LVDT displacement sensor according to claim 1, characterized in that: The high-frequency excitation source adopts direct digital frequency synthesis technology, with an adjustable excitation frequency range of 10kHz to 20kHz, a frequency resolution of 0.01Hz, an excitation waveform of sine wave, a total harmonic distortion of ≤0.1%, an adjustable excitation amplitude range of 1Vpp to 10Vpp, and an amplitude stability of ≤0.05%.

5. The high-frequency dynamic response LVDT displacement sensor according to claim 1, characterized in that: The lock-in amplifier includes a quadrature demodulation circuit, a low-pass filter, and an automatic gain control circuit; the quadrature demodulation circuit uses two reference signals with a 90° phase difference for synchronous detection; the cutoff frequency of the low-pass filter is 1 / 10 to 1 / 5 of the excitation frequency; the automatic gain control circuit is used to compensate for signal amplitude changes caused by displacement changes.

6. The high-frequency dynamic response LVDT displacement sensor according to claim 1, characterized in that: The adaptive filtering module adopts the LMS adaptive algorithm. The reference input signal is taken from the output of the high-frequency excitation source. The filter order is adjustable from 16 to 128, the step size factor μ is adaptively adjusted from 0.001 to 0.1, the convergence time is ≤10ms, and the steady-state mean square error is ≤0.1%.

7. The high-frequency dynamic response LVDT displacement sensor according to claim 1, characterized in that: The secondary coil (3) includes a primary coil and a secondary coil, which are connected in series in reverse to form a differential voltage output; the signal processing circuit (6) also includes an instrumentation amplifier, which amplifies the differential voltage output and outputs it with a common-mode rejection ratio ≥80dB.

8. The high-frequency dynamic response LVDT displacement sensor according to claim 1, characterized in that: It also includes a temperature compensation module, which includes a built-in temperature sensor and a temperature compensation algorithm unit. The temperature sensor is located inside the coil frame (1) and is used to monitor the temperature of the coil and the magnetic core in real time. The temperature compensation algorithm unit adopts a polynomial fitting or neural network method, and the compensation temperature range is -55℃ to +150℃, with a temperature drift of ≤±0.02%FS / ℃ across the entire temperature range.

9. The high-frequency dynamic response LVDT displacement sensor according to claim 1, characterized in that: The outer shell (5) is made of stainless steel or titanium alloy and has a protection level of IP67 or above; the coil frame (1) and the outer shell (5) are vacuum potting process, and the potting material is epoxy resin or silicone rubber; the lead wire is led out through glass sintered sealed terminals.

10. A displacement measurement method based on the high-frequency dynamic response LVDT displacement sensor according to any one of claims 1 to 9, characterized in that, Includes the following steps: S1: A sinusoidal excitation signal of 10kHz to 20kHz is generated by the high-frequency excitation source and applied to the primary coil (2). S2: Using the primary coil (2) and the secondary coil (3) wound in layers and segments, in conjunction with the iron core (4), the iron core displacement is converted into a differential voltage signal; S3: The output signal of the secondary coil (3) is synchronously detected and quadrature demodulated using the lock-in amplifier to extract the effective displacement signal; S4: The adaptive filtering module is used to suppress high-frequency noise and electromagnetic interference in real time; S5: Real-time compensation for temperature drift is achieved through a temperature compensation module; S6: Converts the signal to a digital signal via an A / D converter and outputs the displacement measurement result through a digital interface; The method achieves a dynamic response frequency ≥10kHz, measurement accuracy better than ±0.1% FS, and response time <1ms.

Images