Digital compensation with built-in self-testing nonlinear variable differential displacement sensor

Abstract

This invention relates to a digitally compensated, self-testing, nonlinear variable differential displacement sensing device, belonging to the fields of precision measuring instruments and avionics. The device includes a nonlinear variable differential transformer, a high signal-to-noise ratio analog front-end conditioning module, a direct digital frequency synthesis excitation module, a digital processing compensation module, and a built-in self-testing module. The core technology employs feedforward synchronous demodulation and laser interferometry calibration. It synchronously generates excitation and orthogonal reference signals through direct digital frequency synthesis, eliminating phase-locked loops. The signal processing link integrates a comprehensive algorithm engine encompassing analog front-end digital pre-distortion compensation, complex plane vector zero compensation, nonlinear inverse compensation, and adaptive temperature drift correction, achieving deterministic microsecond-level low-latency demodulation. The built-in self-testing module monitors the excitation source, sensor coil, signal link linearity, and compensation effectiveness in parallel. The device significantly improves measurement accuracy, dynamic response, and anti-interference capabilities.

Description

Technical Field

[0001] This application relates to the fields of precision measuring instruments and avionics technology, specifically to a nonlinear variable differential pressure displacement sensing device and its corresponding high-precision signal processing method, and particularly to a digitally compensated, self-testing nonlinear variable differential displacement sensing device and method. Background Technology

[0002] A linear variable displacement transformer (LVDT) is a classic electromechanical conversion element widely used in various precision displacement measurement scenarios due to its theoretically infinite resolution, lack of mechanical contact wear, high reliability, and long lifespan. Its basic working principle is as follows: when the primary coil is excited by a sinusoidal AC voltage of a certain frequency, the axial displacement of the core changes its magnetic coupling with the two secondary coils, generating a differential AC output voltage proportional to the displacement. However, in high-precision, high-dynamic applications, the differential AC output voltage proportional to the displacement is insufficient to adequately handle the nonlinearity of the entire signal chain. Nonlinear signal processing faces numerous challenges, including demodulation delay and phase uncertainty, insufficient compensation for non-ideal sensor characteristics, weak electromagnetic interference protection, and a lack of online self-monitoring and health monitoring.

[0003] As a precision displacement sensor, the nonlinear variable displacement transformer (NVDT) exhibits an inherent nonlinear relationship between its output voltage and core displacement, determined by the physical characteristics of the magnetic circuit. In high-precision measurement applications, this nonlinearity, along with distortion introduced by analog conditioning circuits and drift caused by environmental temperature variations, severely restricts the system's measurement accuracy and stability. Traditional signal processing solutions often employ phase-locked loop demodulation technology, which suffers from delay uncertainty and high phase tracking noise. Compensation for nonlinearity and temperature drift typically relies on simple piecewise linear or polynomial fitting, failing to provide unified modeling and compensation at the system level. Furthermore, related technologies generally lack comprehensive online monitoring capabilities for the health status of the sensor and signal link, making it difficult to meet the demands of high-reliability applications such as aerospace. Summary of the Invention

[0004] In view of the above problems, this application provides a digitally compensated, self-testing, nonlinear variable differential displacement sensing device and method to solve the technical problems of low measurement accuracy, poor stability, strong delay uncertainty, and large phase tracking noise.

[0005] According to the first aspect of this application, a digitally compensated, self-testing, nonlinear variable differential displacement sensing device is provided. The device includes: a nonlinear variable differential transformer, a high signal-to-noise ratio analog front-end conditioning module, a direct digital frequency synthesis excitation generation module, a digital processing compensation module, and a built-in self-testing module. The nonlinear variable differential transformer converts a linear displacement signal into a differential voltage signal with inherent nonlinear characteristics. The high signal-to-noise ratio analog front-end conditioning module amplifies, performs anti-aliasing filtering, and performs analog-to-digital conversion on the differential voltage signal to obtain a digital signal. The direct digital frequency synthesis excitation generation module generates a sinusoidal excitation signal to drive the primary coil of the nonlinear variable differential transformer and synchronously generates a sinusoidal excitation signal based on a synthesized clock source of the same frequency. The system includes in-phase and quadrature reference signals with the same frequency and strictly locked phase as the string excitation signal; a digital processing compensation module receives the digital signal and performs feedforward synchronous demodulation using the in-phase and quadrature reference signals. It employs a cascaded nonlinear inverse function based on laser interferometer calibration to perform analog front-end digital pre-distortion compensation, complex plane vector zero compensation, nonlinear inverse compensation, and adaptive temperature drift correction, outputting high-precision displacement values; a built-in self-test module monitors the excitation source, sensor coil, signal link linearity, and the effectiveness of digital compensation in parallel and in real time; the direct digital frequency synthesis excitation generation module and the digital processing compensation module together constitute a feedforward synchronous demodulation architecture, enabling phase tracking without a phase-locked loop during demodulation.

[0006] According to an embodiment of this application, the end-to-end cascaded nonlinear inverse function in the digital processing compensation module is obtained through laser interferometry calibration, including: using a laser interferometer to synchronously collect the true value of displacement and temperature, constructing a calibration dataset, and obtaining the laser interferometry calibration nonlinear inverse function through reverse identification; the laser interferometry calibration nonlinear inverse function is fitted with a piecewise polynomial and stored in the digital processing compensation module in the form of a lookup table.

[0007] According to an embodiment of this application, the analog front-end digital pre-distortion compensation in the digital processing compensation module includes: in the output path of the direct digital frequency synthesis excitation generation module, performing real-time pre-distortion processing on the original excitation digital waveform based on the inverse function of the calibrated analog front-end nonlinear transfer characteristics, so as to cancel the harmonic distortion and gain compression introduced by the analog conditioning stage at the source of the signal link, and the original excitation digital waveform corresponds to the sinusoidal excitation signal.

[0008] According to an embodiment of this application, the digital processing compensation module performs complex plane vector zero compensation, including: characterizing the zero-point residual error as a zero-point residual vector on the complex plane formed by feedforward synchronous demodulation; and subtracting the zero-point residual vector from the current measurement signal vector on the complex plane to achieve complex plane vector zero compensation.

[0009] According to an embodiment of this application, the digital processing compensation module performs nonlinear inverse compensation, including: based on a cascaded nonlinear inverse model from digital signal features to real displacement established through full-link system calibration, a piecewise polynomial fitting algorithm is used to perform real-time table lookup and interpolation calculation to achieve nonlinear inverse compensation.

[0010] According to an embodiment of this application, the digital processing compensation module performs adaptive temperature drift correction, including: deeply integrating temperature variables during the lookup table interpolation process of the cascaded nonlinear inverse model based on feedback data from the temperature sensor, and dynamically suppressing sensitivity and zero-point drift caused by changes in ambient temperature, so as to achieve adaptive temperature drift correction.

[0011] According to an embodiment of this application, the built-in self-test module includes: an excitation source monitoring unit, a sensor coil monitoring unit, a signal link linearity monitoring unit, and a compensation effectiveness monitoring unit. The excitation source monitoring unit is used to read back the sinusoidal excitation signal and determine whether the amplitude, frequency, and waveform distortion of the sinusoidal excitation signal exceed a preset threshold. The sensor coil monitoring unit is used to calculate the sum of the absolute values ​​of the output voltages of the two-stage coils of the nonlinear variable differential transformer and determine the coil fault based on whether the sum of the absolute values ​​deviates from a constant range. The signal link linearity monitoring unit analyzes the harmonic components in the signal under specific test excitation or normal operation to evaluate whether the linearity of the analog front-end amplifier and filter has degraded. The compensation effectiveness monitoring unit is used to monitor the statistical characteristics of the output residual after nonlinear inverse compensation and determine whether the compensation model of the digital processing compensation module is mismatched.

[0012] According to an embodiment of this application, the device employs an electromagnetic compatibility enhancement setup, which includes: a metal shielding shell for the sensor, a double-shielded connecting cable, an isolation power supply between analog and digital circuits, and a star-type single-point grounding system.

[0013] The second aspect of this application provides a digitally compensated, self-checking, nonlinear variable differential displacement sensing method using the aforementioned apparatus. The method includes: synchronously generating a sinusoidal excitation signal driving a nonlinear variable differential transformer, along with a synchronously generated in-phase reference signal and a quadrature reference signal with the same frequency and strictly locked phase as the sinusoidal excitation signal, using direct digital frequency synthesis technology; driving the nonlinear variable differential transformer with the sinusoidal excitation signal to acquire differential voltage signals in real time; conditioning the differential voltage signals and converting them into digital signals; performing feedforward synchronous demodulation on the digital signals using the in-phase and quadrature reference signals to obtain baseband in-phase and quadrature components; sequentially performing complex plane vector zero compensation, nonlinear inverse compensation, and adaptive temperature drift correction on the demodulated complex plane signal based on the baseband in-phase and quadrature components to obtain high-precision displacement values; and simultaneously performing real-time health monitoring and fault diagnosis of the excitation source, sensor coil, signal link linearity, and digital compensation effectiveness. Attached Figure Description

[0014] The above-mentioned contents, other objects, features and advantages of this application will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0015] Figure 1 This schematic diagram illustrates a module of a digitally compensated, self-testing, nonlinear variable differential displacement sensing device according to an embodiment of this application.

[0016] Figure 2 This illustration schematically shows a hardware architecture diagram of a digitally compensated, self-testing, nonlinear variable differential displacement sensing device according to an embodiment of this application.

[0017] Figure 3 This illustration schematically shows an algorithm diagram of a digitally compensated, self-testing, nonlinear variable differential displacement sensing device according to an embodiment of this application.

[0018] Figure 4 This schematically illustrates a parallel monitoring logic block diagram of the built-in self-test module of a digitally compensated built-in self-testing nonlinear variable differential displacement sensor according to an embodiment of this application.

[0019] Figure 5 The diagram schematically illustrates the digital filter response and predistortion compensation spectrum of the digitally compensated built-in self-testing nonlinear variable differential displacement sensing device according to an embodiment of this application.

[0020] Figure 6 This illustration schematically shows a static accuracy test error curve of a digitally compensated, self-testing, nonlinear variable differential displacement sensing device according to an embodiment of this application.

[0021] Figure 7This illustration schematically shows a temperature drift suppression and stable output curve of a digitally compensated, self-testing, nonlinear variable differential displacement sensor according to an embodiment of this application.

[0022] Figure 8 A flowchart illustrating a digitally compensated, self-testing, nonlinear variable differential displacement sensing method according to an embodiment of this application is shown. Detailed Implementation

[0023] The embodiments of this application will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of this application. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of this application for ease of explanation. However, it will be apparent that one or more embodiments may be implemented without these specific details. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concepts of this application.

[0024] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the scope of this application. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0025] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0026] When using expressions such as "at least one of A, B and C", they should generally be interpreted in accordance with the meaning that is commonly understood by those skilled in the art (e.g., "a system having at least one of A, B and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B and C, etc.).

[0027] Currently, common LVDT / NVDT signal processing solutions on the market mainly face the following technical bottlenecks. First, there is an inherent contradiction in the core dynamic performance of demodulation technology. Solutions based on dedicated analog phase-sensitive detector chips rely heavily on the accuracy and stability of analog components, making them susceptible to environmental temperature changes and component aging, leading to long-term drift, and making it difficult to achieve complex nonlinear compensation at the hardware level. While demodulation methods using digital phase-locked loops introduce the flexibility of digital processing, the inherent frequency acquisition and phase tracking process of the phase-locked loop inevitably introduces convergence time, resulting in uncertain demodulation delays and phase noise. Such uncertain delays and phase errors are unacceptable for modern high-dynamic, high-bandwidth closed-loop control systems (such as fly-by-wire flight control systems and distributed control surface actuation systems), directly deteriorating the system's phase margin, affecting stability, and potentially inducing oscillations. For optical precision mechanics and high-precision industrial automation equipment, dedicated analog phase-sensitive detectors and digital phase-locked loop demodulation also face nonlinearity, temperature drift, delay, and the potential performance and reliability problems arising from these issues.

[0028] Secondly, the inherent non-ideal characteristics of the sensor constitute the main source of error, and existing compensation methods are insufficiently systematic and comprehensive. These non-ideal characteristics mainly include: zero-point residual voltage, because the electrical and geometric parameters of the two secondary coils cannot be perfectly symmetrical, even when the core is at mechanical zero, the differential output is not zero, which usually manifests as a small signal with the same frequency as the excitation but a small amplitude and arbitrary phase; nonlinearity, due to factors such as magnetic circuit edge effects and saturation, the sensor's sensitivity is not constant throughout the entire measurement range, resulting in an S-shaped curve in the input (displacement)-output (voltage) relationship; and temperature drift, as changes in ambient temperature cause changes in parameters such as coil resistance and core permeability, leading to drift in sensitivity and zero point. Traditional methods for compensating for these errors are often isolated and coarse. For example, only a simple DC bias subtraction is performed on the zero-point residual voltage, ignoring its phase information, resulting in a nonlinear dead zone near the zero point; and a fixed temperature coefficient is used to correct the temperature drift, which cannot adapt to the nonlinear drift characteristics over a wide temperature range.

[0029] Furthermore, in practical engineering applications, especially in aerospace, precision optical instruments, and industrial automation, sensor systems are inevitably exposed to complex electromagnetic interference. This interference may originate from high-power motors, switching power supplies, radio frequency equipment, and the environment, and can couple into the sensor's signal and power lines through radiation or conduction. This can range from reducing the signal-to-noise ratio of the measurement signal to causing signal distortion, demodulation errors, or even damaging sensitive circuits or devices. Existing solutions often lack sufficient consideration in electromagnetic compatibility design, lacking systematic protection across the entire chain from the sensor probe to the signal processing unit.

[0030] Finally, existing solutions generally lack effective online health status monitoring and fault diagnosis capabilities. This means that the system cannot detect in real time critical states such as whether the sensor coil is open-circuited or short-circuited, whether the excitation source is working properly, or whether the front-end conditioning circuit has failed. Once a fault occurs, the system may not be able to provide timely warnings, making it difficult to provide a basis for decision-making for the upper-level control system to take fault-tolerant measures such as redundancy switching or safe shutdown, and thus failing to meet the stringent requirements for high safety integrity levels in fields such as aviation, aerospace, science, and industry.

[0031] To address the aforementioned technical problems, embodiments of this application provide a digitally compensated, self-testing, nonlinear variable differential displacement sensing device. The following will combine... Figures 1-7 The device is described in detail.

[0032] Figure 1 The schematic diagram illustrates a module of a digitally compensated, self-testing, nonlinear variable differential displacement sensing device according to an embodiment of this application.

[0033] exist Figure 1 In the illustrated embodiment, the digitally compensated, self-testing, nonlinear variable differential displacement sensing device 100 includes: a nonlinear variable differential transformer 110, a high signal-to-noise ratio analog front-end conditioning module 120, a direct digital frequency synthesis excitation generation module 130, a digital processing compensation module 140, and a built-in self-testing module 150. The nonlinear variable differential transformer 110 converts a linear displacement signal into a differential voltage signal with inherent nonlinear characteristics. The high signal-to-noise ratio analog front-end conditioning module 120 amplifies, performs anti-aliasing filtering, and performs analog-to-digital conversion on the differential voltage signal to obtain a digital signal. The direct digital frequency synthesis excitation generation module 130 generates a sinusoidal excitation signal to drive the primary coil of the nonlinear variable differential transformer and synchronously generates a signal based on a synthesized clock source of the same frequency. The system includes in-phase and quadrature reference signals with the same frequency and strictly locked phase as the sinusoidal excitation signal. The digital processing compensation module 140 receives the digital signal and performs feedforward synchronous demodulation using the in-phase and quadrature reference signals. It employs a cascaded nonlinear inverse function based on laser interferometer calibration to perform analog front-end digital pre-distortion compensation, complex plane vector zero compensation, nonlinear inverse compensation, and adaptive temperature drift correction, outputting a high-precision displacement value. The built-in self-test module 150 monitors the excitation source, sensor coil, signal link linearity, and the effectiveness of digital compensation in parallel and in real time. The direct digital frequency synthesis excitation generation module 130 and the digital processing compensation module 140 together constitute a feedforward synchronous demodulation architecture, eliminating the need for phase-locked loops for phase tracking during demodulation.

[0034] As an example, the nonlinear variable differential transformer 110, the high signal-to-noise ratio analog front-end conditioning module 120, the direct digital frequency synthesis excitation generation module 130, the digital processing compensation module 140, and the built-in self-test module 150 are functional modules with corresponding functions, and each functional module is implemented based on a hardware architecture.

[0035] In this embodiment, the hardware system of the device adopts a highly integrated modular architecture design, mainly composed of three closely coordinated parts: a sensor probe module, an analog signal conditioning and acquisition board, and a digital processing and communication board. The nonlinear variable differential transformer 110 is integrated into the sensor probe module as part of it. The direct digital frequency synthesis excitation generation module 130 is integrated into the analog signal conditioning and acquisition board. The high signal-to-noise ratio analog front-end conditioning module 120, the digital processing compensation module 140, and the built-in self-test module 150 are integrated into the digital processing and communication board. Based on FPGA core, DDS excitation generation technology, and feedforward synchronous demodulation architecture, these modules achieve corresponding functions. For example, the entire cascaded nonlinear inverse function is pre-calibrated, identified, and stored via laser interferometry. The laser interferometry-calibrated nonlinear inverse function, combined with the current temperature and voltage amplitude, obtains a high-precision displacement estimate through bilinear interpolation.

[0036] Figure 2 The illustration shows a schematic diagram of the hardware architecture of a digitally compensated, self-testing, nonlinear variable differential displacement sensing device according to an embodiment of this application.

[0037] like Figure 2 As shown, in a more specific example, in the design and physical model construction of the sensor probe module, a high-precision five-wire nonlinear variable differential transformer with a range of ±900 mm can be selected. This high-precision five-wire nonlinear variable differential transformer has a range of 125 mm and uses a five-wire interface. The probe of this high-precision five-wire nonlinear variable differential transformer consists of a primary coil wound on a hollow frame and two secondary coils connected in reverse series. The iron core moves axially with the measured object within the frame. Considering the non-ideal characteristics of the actual physical environment, the influence of ambient temperature on the copper resistance of the primary and secondary coils, as well as the residual zero-point voltage caused by electrical asymmetry when the iron core is at mechanical zero position, can be clearly covered. To cope with the complex electromagnetic environment, the sensor is externally encapsulated with a permalloy shielding shell for permalloy shielding and connected to the back-end circuit through a double-shielded cable to suppress external interference to the greatest extent. Thus, the nonlinear variable differential transformer can convert linear displacement signals into differential voltage signals with inherent nonlinear characteristics with high precision.

[0038] This analog signal conditioning and acquisition board serves as a bridge connecting the physical world and the digital core, integrating a high-fidelity excitation drive link and a high signal-to-noise ratio (SNR) signal acquisition link. The excitation drive link uses a DAC + reconstruction filter + power operational amplifier approach to generate the sinusoidal excitation signal. In the excitation path, a field-programmable gate array (FPGA) directly synthesizes a sinusoidal sequence, which is then converted into an analog signal by a 16-bit high-precision digital-to-analog converter (DAC). This signal is smoothed by a second-order Butterworth reconstruction filter with a cutoff frequency of 10kHz, and then amplified by a high-power operational amplifier to drive the primary coil of the NVDT. In the signal acquisition link, the secondary differential signal output from the NVDT first passes through a common-mode choke to suppress common-mode noise, and then is converted into a single-ended signal by an instrumentation amplifier with a common-mode rejection ratio (CMRR) greater than 120dB. This single-ended signal undergoes anti-aliasing filtering, where a fourth-order Butterworth low-pass anti-aliasing filter removes high-frequency interference. Finally, it is synchronously sampled and digitized by a dual-channel 16-bit high-speed analog-to-digital converter (ADC) to obtain a digital signal. In addition, the analog signal conditioning and acquisition board integrates a high-precision digital temperature sensor and an isolated power supply module / isolated power supply. With the star-top single-point grounding design, it ensures the purity of the signal and the thermal stability of the system, thereby improving the stability of the digital signal obtained by the high signal-to-noise ratio analog front-end conditioning module in amplifying, anti-aliasing filtering and analog-to-digital conversion of differential voltage signals.

[0039] The digital processing and communication board is the core control hub of the entire sensing device, implemented based on the high-performance MicrosemiPolarFire series field-programmable gate array. Unlike traditional analog circuits or general-purpose processor solutions, this board constructs a fully digital feedforward synchronous architecture within the FPGA using logic gates. The integrated direct digital frequency synthesis excitation generation module uses the same clock source to drive the DDS module for excitation generation, producing a sinusoidal excitation signal, such as a 3kHz sinusoidal excitation signal, to drive the primary coil of the nonlinear variable differential transformer. Simultaneously, based on the same frequency synthesis clock source, it synchronously generates in-phase and quadrature reference signals with the same frequency and strictly locked phase as the sinusoidal excitation signal. This internally generates strictly phase-locked in-phase and quadrature reference sequences, thereby completing mixing and feedforward synchronous demodulation in the digital domain. The direct digital frequency synthesis excitation generation module and the digital processing compensation module together constitute a feedforward synchronous demodulation architecture, eliminating the need for phase-locked loops (PLLs) for phase tracking during demodulation. This architecture eliminates dependence on traditional PLLs and solves the problem of phase delay uncertainty. Furthermore, based on a full-link nonlinear compensation algorithm, it employs a cascaded nonlinear inverse function obtained from laser interferometer calibration to perform analog front-end digital pre-distortion compensation, complex plane vector zero compensation, nonlinear inverse compensation, and adaptive temperature drift correction, outputting high-precision displacement values. The final displacement data, after compensation by the FPGA's internal algorithm, is then output as high-precision displacement data via a high-speed SPI interface, RS-422, or ARINC 429 avionics bus. This modular design allows for independent selection and replacement of components according to different requirements, improving system flexibility and maintainability. Additionally, the built-in self-test module integrated into the intelligent built-in self-test (BIT) system provides parallel real-time monitoring of the excitation source, sensor coils, signal link linearity, and the effectiveness of digital compensation.

[0040] Based on this, the existing LVDT / NVDT displacement sensing solutions suffer from four major technical defects: 1) The demodulation technology relies on phase-locked loops or dedicated analog phase-sensitive detector chips, resulting in uncertain demodulation delays and high phase tracking noise, which cannot meet the requirements of high-dynamic closed-loop control systems; 2) The compensation methods for inherent nonlinearity of sensors, analog circuit distortion, and temperature drift are isolated and crude, without unified modeling at the system level, resulting in low measurement accuracy; 3) The lack of full-link electromagnetic compatibility design makes them susceptible to interference in complex electromagnetic environments such as aerospace and industrial automation, resulting in low signal-to-noise ratios or even distortion; 4) The lack of effective online health monitoring and fault diagnosis capabilities makes it impossible to detect sensor and signal link faults in real time, making it difficult to meet high safety integrity requirements.

[0041] The following beneficial effects are achieved: 1) The constructed feedforward synchronous demodulation architecture completely eliminates the delay uncertainty and phase noise caused by the phase-locked loop, realizing deterministic microsecond-level low-latency demodulation, solving the dynamic performance defects of traditional demodulation schemes from the source, and matching the phase margin and stability requirements of high dynamic control systems; 2) The modular hardware architecture and functional module division enable the coordinated work of signal acquisition, excitation generation, digital processing, and self-test monitoring. Each module can be independently selected, replaced, and upgraded, improving the engineering scalability and maintainability of the device; 3) The integration of the full-link nonlinear compensation algorithm and built-in self-test function integrates displacement measurement and health monitoring, significantly improving the measurement accuracy, environmental adaptability, and operational reliability of the device.

[0042] In this embodiment, the end-to-end cascaded nonlinear inverse function in the digital processing compensation module is obtained through laser interferometry calibration, which includes: using a laser interferometer to synchronously collect the true value of displacement and temperature, constructing a calibration dataset, and obtaining the laser interferometry calibration nonlinear inverse function through reverse identification; the laser interferometry calibration nonlinear inverse function is fitted with a piecewise polynomial and stored in the digital processing compensation module in the form of a lookup table.

[0043] In a specific example, consider an NVDT model that incorporates non-ideal factors; the secondary differential output voltage can be expressed as:

[0044] ;

[0045] in, For the displacement of the iron core, For displacement and temperature The relevant sensitivity function, and The amplitude and frequency of the excitation signal, This is due to the inherent phase shift. and These represent the amplitude and phase of the zero-point residual voltage, respectively. For noise. The goal of demodulation is to remove... Estimating with medium to high accuracy .

[0046] The core of the theory is to abstract the entire displacement measurement chain into a single process starting from mechanical displacement. To digital output code Deterministic cascaded nonlinear systems:

[0047] ;

[0048] in Ambient temperature. The inverse function of the system was identified through high-precision system-level calibration.

[0049] ;

[0050] This inverse transformation is performed in real time in the digital domain, thereby enabling active correction of nonlinearity across the entire link.

[0051] A deterministic cascaded nonlinear system is calibrated at the system-wide level using a laser interferometer, and its inverse function is identified in reverse. The specific calibration process is as follows:

[0052] First, a calibration system was built, using... A laser interferometer, serving as the true displacement reference, is synchronously mounted on a precision displacement stage along with an NVDT sensor and placed within a temperature-controlled chamber (-40℃ to +125℃). Data is then synchronously acquired, and the displacement measured by the laser interferometer is analyzed at multiple temperature points. Using this as a reference, the sensor output voltage is recorded synchronously. With temperature Construct the calibration dataset:

[0053] ;

[0054] Next, inverse function identification is performed. Based on the calibration data, a piecewise polynomial fitting method is used to establish the inverse mapping function:

[0055] ;

[0056] Where the coefficient The value is obtained through least squares fitting and stored as a two-dimensional lookup table. Based on this, an inverse transformation is performed in real time in the FPGA, and a high-precision displacement estimate can be obtained by combining bilinear interpolation with the current temperature and voltage amplitude.

[0057] Figure 3 The illustration shows a schematic diagram of the algorithm of the digitally compensated built-in self-testing nonlinear variable differential displacement sensing device according to an embodiment of this application.

[0058] exist Figure 3 In the example shown on the right, the acquisition and implementation of the end-to-end cascaded nonlinear inverse function in the digital processing compensation module is based on the laser interferometer calibration system, that is, it is completed through the end-to-end system-level calibration of the laser interferometer. The calibration process and function storage / call are shown in Figure 3. The specific steps are as follows:

[0059] Setting up a calibration system: A Renishaw XL-80 laser interferometer was used as the true displacement reference. It was installed synchronously with the NVDT sensor of this invention on a precision displacement stage and placed in a temperature-controlled chamber at -40℃ to +125℃. A high-precision digital temperature sensor was connected to collect the ambient temperature.

[0060] Calibration data acquisition: Multiple typical temperature points (covering the entire operating temperature range of the device) are set in the temperature control chamber. The NVDT core is driven by a precision displacement stage to complete the full range (±900mm) displacement movement. The displacement value measured by the laser interferometer is used as the true displacement value. The output voltage signal of the NVDT sensor and the ambient temperature data are recorded simultaneously to construct a three-dimensional calibration dataset containing "displacement-voltage-temperature". ,in For the true value of displacement, The sensor output voltage, For the corresponding temperature;

[0061] Inverse function identification / inverse model construction: Based on a calibration dataset, a piecewise polynomial fitting method is used, and the fitting coefficients are solved by the least squares method to inversely identify the full-link cascaded nonlinear inverse function from "voltage-temperature" to "displacement". This function covers the inherent nonlinearity of NVDT sensors, the distortion characteristics of the analog front-end conditioning module, and the impact of temperature on the sensing link;

[0062] Function storage and retrieval: The identified inverse function is stored in the non-volatile memory of the FPGA in the digital processing compensation module in the form of a two-dimensional lookup table (LUT). The index of the lookup table is the voltage amplitude and the temperature value. During the real-time measurement process of the device, the FPGA takes the currently acquired voltage amplitude and the temperature value fed back by the temperature sensor as input, and quickly obtains the high-precision displacement estimate from the lookup table through bilinear interpolation.

[0063] Based on this, the technical problem that related technologies only use simple piecewise linear or single polynomial fitting for nonlinear compensation of the sensing link, without incorporating the inherent nonlinearity of the sensor, analog circuit distortion, and temperature drift into a unified system-level model, and without a high-precision displacement truth benchmark for calibration, is that the compensation model has a low degree of matching with the actual sensing link, poor measurement accuracy across the entire temperature range and the entire measurement range, and cannot meet the technical requirements of high-precision displacement measurement. Using a laser interferometer as the true displacement reference, a full-link system-level calibration is achieved. The calibration results cover the nonlinear characteristics of all links from the NVDT sensor to the digital output, and the accuracy of the compensation model is far higher than that of traditional local calibration methods. Piecewise polynomial fitting is used to construct the inverse function, which can more accurately fit the S-shaped nonlinear input-output characteristics of the sensing link compared with single polynomial fitting, thus improving the accuracy of nonlinear compensation. The inverse function is stored in the form of a two-dimensional lookup table and combined with bilinear interpolation to achieve real-time retrieval. It can be completed in the FPGA through a fixed-point computation pipeline, with low computational latency and fast operation speed, matching the microsecond-level demodulation requirements of the device. The inverse function incorporates temperature variables, providing a model basis for subsequent adaptive temperature drift correction, realizing the synergy of nonlinear compensation and temperature drift compensation, and ensuring measurement stability across the entire temperature range.

[0064] In this embodiment, the analog front-end digital pre-distortion compensation in the digital processing compensation module includes: in the output path of the direct digital frequency synthesis excitation generation module, according to the inverse function of the calibrated analog front-end nonlinear transfer characteristics, performing real-time pre-distortion processing on the original excitation digital waveform to cancel the harmonic distortion and gain compression introduced by the analog conditioning stage at the source of the signal link, so that the original excitation digital waveform corresponds to the sinusoidal excitation signal.

[0065] As an example, the specific implementation of analog front-end digital predistortion compensation in the digital processing compensation module is integrated into the excitation signal output path of the direct digital frequency synthesis excitation generation module, as follows: Figure 3 As shown on the left, digital predistortion processing is performed after DDS excitation generation. Specifically, in terms of analog front-end digital predistortion collaborative compensation, to optimize end-to-end performance and suppress distortion at its source, this embodiment introduces digital predistortion technology into the DDS excitation generation path. The core digital processing involves real-time predistortion processing of the original ideal sine wave sequence generated by the DDS based on the inverse characteristics of the nonlinear transfer characteristics of the calibrated analog front-end, generating the excitation waveform. After this predistorted waveform passes through a non-ideal analog front-end, the harmonic distortion and gain compression of its output signal are preemptively canceled. This strategy of reducing the burden of front-end predistortion and ensuring accuracy through back-end inverse transformation achieves hierarchical and collaborative management of end-to-end nonlinearity.

[0066] Thus, the inherent nonlinearity of the analog front-end conditioning circuits (DAC, amplifier, filter, etc.) in existing sensing solutions introduces harmonic distortion and gain compression, leading to distortion of the excitation signal driving the sensor and thus affecting the accuracy of the sensor's output voltage. Furthermore, traditional solutions do not compensate for the distortion at the source of the analog front-end, but only perform simple filtering in the back-end signal processing, which cannot completely eliminate the impact of distortion on measurement accuracy. Subsequently, pre-distortion compensation is performed at the source of the excitation signal generation, fundamentally offsetting the nonlinear distortion and gain compression of the analog front end. Compared with the passive suppression method of back-end filtering, the distortion elimination effect is more thorough, significantly improving the purity of the excitation signal. Moreover, the pre-distortion processing is fully digitally implemented and integrated into the DDS excitation generation stage of the FPGA, without the need for additional analog devices. The hardware cost is low and easy to implement. Furthermore, the pre-distortion parameters can be adjusted by modifying the digital algorithm to adapt to analog front ends with different characteristics. This effectively suppresses the harmonic distortion of the analog front end, reduces the interference of the distorted signal on subsequent displacement demodulation and compensation, provides a pure excitation signal foundation for high-precision displacement measurement, reduces the filtering and nonlinear compensation burden of the back-end digital processing compensation module, and allows the back-end algorithm to focus more on the compensation of the sensor's inherent nonlinearity and temperature drift, thus improving the overall compensation efficiency.

[0067] In this embodiment, the digital processing compensation module performs complex plane vector zero compensation, including: characterizing the zero-point residual error as a zero-point residual vector on the complex plane formed by feedforward synchronous demodulation; and subtracting the zero-point residual vector from the current measurement signal vector on the complex plane to achieve complex plane vector zero compensation.

[0068] like Figure 3 As shown on the left, this embodiment abandons the traditional feedback phase-locked loop structure and adopts a feedforward synchronous demodulation architecture. A highly stable direct digital frequency synthesizer (DDS) module is integrated within the digital processing platform centered on a field-programmable gate array (FPGA). The unique feature of this DDS module is that it utilizes the same phase accumulator and clock source to synchronously generate a frequency of... =3 kHz excitation sequence And in-phase reference sequences and orthogonal reference sequences:

[0069] ;

[0070] Since the demodulation reference sequence is directly derived from the excitation source, its frequency and phase relationship are predetermined and strictly synchronized, thus eliminating the acquisition time, steady-state phase error, and delay uncertainty caused by the phase-locked loop.

[0071] To suppress noise To prevent sampling aliasing, this invention employs a combined analog and digital filtering strategy. The sensor signal obtained by the ADC sampling... Frequency mixing is achieved by performing digital multiplication operations with two orthogonal reference sequences respectively:

[0072] ;

[0073] After mixing, the signal undergoes downsampling and fine filtering through a cascaded digital filter chain consisting of multi-stage (e.g., 5-stage) cascaded integrator-comb (CIC) filters and compensated finite-length unit impulse response (FIR) filters. The first stage is a 5-stage cascaded integrator-comb (CIC) filter, which downsamples high-sampling-rate (e.g., 1MHz) data to a lower rate (e.g., 10kHz) to efficiently filter out harmonic components, thus providing initial anti-aliasing capability; its system function is denoted as... The second stage is a 15th-order linear-phase FIR filter, which compensates for passband attenuation caused by CIC, ensures flat gain within the passband, suppresses specific interference, and compensates for phase; its system function is extremely... Within the signal frequency band, the overall system response is:

[0074] ;

[0075] The signal is optimized to have a flat amplitude-frequency response and a linear phase-frequency response, thereby minimizing signal distortion and group delay variation. This collaborative filtering strategy minimizes signal distortion and provides high-quality baseband in-phase components for nonlinear inverse transform and temperature drift compensation. and orthogonal components .

[0076] against Traditional methods typically involve simple scalar subtraction after demodulating the amplitude. This embodiment proposes vector compensation on the complex plane (IQ plane) after digital quadrature demodulation. Specifically, on the complex plane formed by feedforward synchronous demodulation, the zero-point residual error is characterized as a zero-point residual vector. The current measurement signal vector on the complex plane is subtracted from this zero-point residual vector to achieve zero vector compensation on the complex plane. Specifically, as follows... Figure 3 As shown, after anti-aliasing filtering and analog-to-digital conversion sampling, the signal is represented in the discrete domain as follows: Compare them separately with and After multiplication and low-pass filtering, the in-phase component of the baseband is obtained. and orthogonal components They constitute a complex number. This refers to the current measurement signal vector on the complex plane. During system calibration, the steady-state output of the core at the NVDT mechanical zero point is accurately measured to obtain the zero-point residual vector. In real-time measurement, zero compensation of the complex plane vector can be expressed as:

[0077] ;

[0078] ;

[0079] This operation physically removes both the amplitude and phase effects of the residual voltage from the current measured signal vector, thus fundamentally solving the nonlinear distortion problem near the zero point. For the compensated vector... The modulus is extracted using the CORDIC (Coordinate Rotation Digital Computation) algorithm. and phase :

[0080] .

[0081] Based on this, the zero-point residual error is represented as a vector on the complex plane. By using vector subtraction, the amplitude and phase effects of the zero-point residual voltage are eliminated simultaneously, fundamentally solving the nonlinear distortion problem near the zero point caused by traditional scalar subtraction and significantly improving the measurement accuracy in the zero-point region. The compensation process is completed on the complex plane in the digital domain, implemented based on FPGA logic operations, which is fast and has low latency, matching the microsecond-level demodulation requirements of the device. The zero-point residual vector is fixed and stored after factory calibration, and real-time compensation only requires performing a simple vector subtraction, with low algorithm complexity and no need to occupy a large amount of FPGA computing resources. The effective signal after compensation has no zero-point residual error, providing a clean voltage amplitude basis for subsequent nonlinear inverse compensation, further improving the overall displacement measurement accuracy.

[0082] In this embodiment, the digital processing compensation module performs nonlinear inverse compensation, including: based on the cascaded nonlinear inverse model from digital signal features to real displacement established through full-link system calibration, a piecewise polynomial fitting algorithm is used to perform real-time table lookup and interpolation calculation to achieve nonlinear inverse compensation.

[0083] The digital processing compensation module performs adaptive temperature drift correction, including: deeply integrating temperature variables during the lookup table interpolation process of the cascaded nonlinear inverse model based on feedback data from the temperature sensor, dynamically suppressing sensitivity and zero-point drift caused by changes in ambient temperature, so as to achieve adaptive temperature drift correction.

[0084] exist Figure 3 In the example shown, for the sensitivity function To address the nonlinearity and temperature drift, this invention employs a cascaded nonlinear system inverse transformation and temperature drift co-compensation strategy. This transformation and compensation strategy is based on a multidimensional (displacement-temperature) lookup table (LUT) established through precise end-to-end system calibration. For nonlinear correction, the sensor is calibrated at a constant temperature to obtain accurate displacement-voltage data pairs across the entire measurement range. ,in The amplitude after demodulation and zero-point compensation.

[0085] Constructing the inverse model The model uses piecewise polynomials for high-precision fitting:

[0086] ;

[0087] coefficient Stored in a lookup table, enabling real-time and rapid correction.

[0088] Based on this, this paper addresses the technical problems in related technologies where nonlinear compensation of the sensing link only uses a global single polynomial or simple piecewise linear fitting, which cannot accurately match the S-shaped nonlinear input-output characteristics of NVDT caused by magnetic circuit edge effects and magnetic saturation. Furthermore, it fails to combine the nonlinear characteristics of the entire link for unified compensation, resulting in low nonlinear compensation accuracy and poor measurement linearity across the entire range. A nonlinear inverse model is constructed based on full-link system-level calibration, covering all nonlinear characteristics of the sensor and analog front end. Compared with traditional local nonlinear compensation, this significantly improves the comprehensiveness and accuracy of compensation. Piecewise polynomial fitting replaces global single polynomial fitting, which can more accurately fit the S-shaped nonlinear characteristics and effectively improve the nonlinear compensation accuracy across the entire range. The inverse model is stored in the form of a lookup table, and compensation is achieved by combining real-time lookup and interpolation calculations. This results in fast operation speed and low latency, and can be completed in an FPGA through a fixed-point computation pipeline, shortening the response time. The compensation process is fully digitally implemented, and the polynomial coefficients of the lookup table can be updated to adapt to NVDT sensors with different ranges and characteristics, enhancing adaptability.

[0089] Regarding temperature drift compensation, temperature The main effect is sensitivity drift. This is modeled into an inverse function, and is observed at multiple temperature points during high and low temperature chamber experiments. By subscribing, a series of inverse functions related to temperature are obtained. Finally, all parameters are stored in non-volatile memory indexed by temperature, forming a two-dimensional lookup table.

[0090] Based on the aforementioned system calibration and inverse model, displacement is accurately estimated through real-time inverse transformation and bilinear interpolation. During real-time measurement, the system uses the current amplitude. and temperature readings Access the 2D LUT using the input coordinates. First, locate it on the temperature axis. The interval Then in and On the corresponding curve, according to Interpolation calculations were performed to obtain two displacement estimates. and Finally, linear interpolation is performed in the temperature dimension to obtain a high-precision displacement output that has undergone collaborative compensation for end-to-end nonlinearity and temperature drift.

[0091] ;

[0092] This process is implemented within the FPGA using a fixed-point arithmetic pipeline, resulting in a deterministic and extremely short computational delay. Actual measurements show that, under the given architecture and parameters, the fixed delay from signal acquisition to displacement output can be less than 50 microseconds.

[0093] Based on this, this paper addresses the technical problem that existing technologies only use a fixed temperature coefficient for simple temperature drift compensation, which cannot adapt to the nonlinear drift characteristics across a wide temperature range. Furthermore, the lack of coordinated temperature drift compensation and nonlinear compensation leads to significant drift in sensor sensitivity and zero point under high and low temperature environments, resulting in a substantial decrease in measurement accuracy and stability, failing to meet the requirements of wide temperature range applications such as aerospace. This paper addresses the issue by deeply integrating temperature variables into the interpolation process of the nonlinear inverse model, achieving coordinated nonlinear compensation and temperature drift correction rather than isolated temperature drift correction, thus significantly improving compensation accuracy across a wide temperature range. A two-dimensional lookup table is constructed based on full-temperature-range calibration, which can accurately match the nonlinear temperature drift characteristics of the sensing link. Compared to fixed temperature coefficient correction methods, the temperature drift suppression effect is more significant. Adaptive correction is completed in real-time in the FPGA through bilinear interpolation, resulting in low computational latency, fast response speed, and dynamic tracking of ambient temperature changes, avoiding the lag in temperature drift suppression.

[0094] In this embodiment, the built-in self-test module includes: an excitation source monitoring unit, a sensor coil monitoring unit, a signal link linearity monitoring unit, and a compensation effectiveness monitoring unit. The excitation source monitoring unit is used to read back the sinusoidal excitation signal and determine whether the amplitude, frequency, and waveform distortion of the sinusoidal excitation signal exceed a preset threshold. The sensor coil monitoring unit is used to calculate the sum of the absolute values ​​of the output voltages of the two-stage coils of the nonlinear variable differential transformer and determine the coil fault based on whether the sum of the absolute values ​​deviates from a constant range. The signal link linearity monitoring unit analyzes the harmonic components in the signal under specific test excitation or normal operation to evaluate whether the linearity of the analog front-end amplifier and filter has degraded. The compensation effectiveness monitoring unit is used to monitor the statistical characteristics of the output residual after nonlinear inverse compensation and determine whether the compensation model of the digital processing compensation module is mismatched.

[0095] Figure 4 The diagram schematically illustrates the parallel monitoring logic block diagram of the built-in self-test module of the digitally compensated built-in self-testing nonlinear variable differential displacement sensing device according to an embodiment of this application.

[0096] exist Figure 4In the example shown, the built-in self-test logic is implemented as an independent hardware process within the field-programmable gate array (FPGA). This demonstrates the logic and alarm output of four parallel channels in the built-in self-test module: excitation monitoring, coil monitoring, signal chain monitoring, and compensation effectiveness detection. The excitation source monitoring unit, sensor coil monitoring unit, signal chain linearity monitoring unit, and compensation effectiveness monitoring unit form a parallel monitoring architecture for real-time monitoring with microsecond-level response. The core logic of its parallel detection is implemented through an independent hardware process, employing N-to-M voting logic to avoid false alarms due to transient interference. Multi-channel data fusion is used for real-time health status assessment. Specifically, the excitation source monitoring unit monitors the excitation signal, with parameters including amplitude, frequency, and distortion. The monitoring condition involves comparison with a preset threshold. An alarm condition can be an over-limit exceeding the limit for five consecutive cycles. The excitation monitoring process of this excitation source monitoring unit analyzes the excitation readback signal and calculates the effective value. and base frequency If the amplitude exceeds the preset threshold for 5 consecutive cycles (amplitude exceeds [6.0V, 7.5V] or frequency deviation exceeds ±30 Hz), the excitation source is determined to be faulty.

[0097] The sensor coil monitoring unit monitors the output of the secondary coil. The monitoring principle is that the sum of the absolute values ​​remains essentially constant under normal conditions. An alarm condition can be a deviation from the window for 20 consecutive points. The sensor coil monitoring process in the sensor coil monitoring unit synchronously reads the ADC data from both secondary coils and calculates the sum of their absolute values.

[0098] ;

[0099] During a normal trip, The voltage should remain basically constant. If it deviates from the preset window for 20 consecutive points, for example [3.0V, 3.6V], it can be determined that the coil may have an open circuit or short circuit fault.

[0100] The signal link linearity monitoring unit monitors the ADC output data. The monitoring process involves analyzing harmonic components in signals under specific test excitation or during normal operation to assess whether the linearity of the analog front-end amplifier and filters has degraded. Simultaneously, it continuously monitors the main ADC's output data stream, using algorithms to detect abnormal states such as persistent saturation, fixed dead values, or a signal-to-noise ratio below 20 dB.

[0101] The compensation effectiveness monitoring unit monitors the output residual value after compensation, uses statistical characteristic analysis as the monitoring method, and uses the root mean square value of the residual as the judgment criterion. The alarm condition is model mismatch alarm. The compensation effectiveness monitoring process of this unit is used to monitor the statistical characteristics of the output residual after nonlinear inverse compensation. If the root mean square value of the residual and other statistical quantities continuously exceed the threshold, it is determined that the compensation model may be mismatched due to device aging or environmental changes, and an early warning is triggered. All monitoring channels adopt voting logic such as "N out of M" to avoid false alarms due to transient interference and generate a comprehensive equipment status word.

[0102] Correspondingly, comprehensive equipment status word output is implemented, which involves predictive maintenance based on the status summary of each monitoring channel, fault codes and level classifications, status indications (normal / early warning / fault), and alarm signal outputs. Based on this, a full-link, multi-dimensional parallel self-testing system is constructed, covering four core links: excitation source, sensor coil, signal link, and compensation model, realizing health status monitoring of the entire life cycle of the device. Each monitoring unit is integrated into the FPGA as an independent hardware process, running in parallel with the main signal path without interference or delay, achieving microsecond-level fault response and meeting the real-time requirements of highly dynamic systems. Voting logic and continuous sampling point / period judgment are adopted to effectively avoid false alarms caused by transient electromagnetic interference and improve the reliability of self-test results. The generation of comprehensive equipment status words and fault codes can accurately locate the fault location and type, providing a decision basis for redundancy switching and safe shutdown of the upper control system, while providing data support for predictive maintenance, significantly improving the operational reliability of the device and the safety of the target system. The compensation effectiveness monitoring unit can detect the mismatch of the compensation model in real time and prompt recalibration to ensure the measurement accuracy after long-term operation of the device, solving the problem of decreased compensation accuracy caused by device aging and environmental changes.

[0103] In this embodiment, the device employs an electromagnetic compatibility enhancement system, which includes: a metal shielding shell for the sensor, double-shielded connecting cables, an isolation power supply between analog and digital circuits, and a star-type single-point grounding system.

[0104] As an example, the sensor features a metal shielded housing: the NVDT sensor probe is externally encapsulated with a permalloy shield. Permalloy has high magnetic permeability, effectively shielding against external magnetic field interference while suppressing the sensor's own electromagnetic radiation, protecting the internal coil's magnetic circuit characteristics from external electromagnetic influences; double-shielded connection cable: a double-shielded cable connects the sensor probe to the analog signal conditioning and acquisition board. The inner shield suppresses differential-mode interference, and the outer shield suppresses common-mode interference. The cable's shield is reliably grounded, effectively preventing electromagnetic interference from coupling into the signal line via conduction; isolated power supply between analog and digital circuits: power is provided between the analog signal conditioning and acquisition board and the digital processing... In the communication board, isolated power modules are used to supply power to analog and digital circuits separately, achieving electrical isolation between the analog and digital power supplies. This prevents high-frequency switching noise from the digital circuits from coupling into the analog circuits through the power lines, ensuring the purity of analog signal acquisition and excitation generation. Star-shaped single-point grounding system: All hardware units of the device (sensors, analog boards, digital boards) adopt a star-shaped single-point grounding design, with all grounding terminals connected to the same common grounding point. This avoids ground loop interference caused by multiple grounding points, effectively suppresses electromagnetic interference conducted through the grounding lines, and ensures a consistent ground potential throughout the entire device.

[0105] The four electromagnetic compatibility enhancement features mentioned above work together to form a full-link electromagnetic protection system, enabling the device to maintain stable measurement performance even in complex electromagnetic environments such as high-power motors, switching power supplies, and radio frequency equipment.

[0106] Based on this, a comprehensive electromagnetic compatibility (EMC) protection system was constructed, achieving all-round suppression of EMC from four dimensions: sensors, transmission, power supply, and grounding. Compared with traditional partial shielding methods, this significantly improves anti-interference capabilities. The permalloy shielding shell and double-shielded cables effectively suppress radiated and conducted interference, the isolated power supply avoids crosstalk between digital circuit noise and analog circuits, and the star-type single-point grounding eliminates ground loop interference. These multiple measures work together to ensure signal purity. In environments with strong common-mode interference, the system improves anti-interference performance, enabling it to operate stably in complex electromagnetic environments such as aerospace and high-precision industrial automation. The EMC enhancement settings are passive protection at the hardware level, requiring no additional digital filtering algorithms, not occupying FPGA computing resources, and not increasing the device's processing latency. The EMC protection settings ensure the measurement stability and reliability of the device in complex environments, extend the lifespan of the components, and reduce the failure rate.

[0107] To verify the effectiveness of the present invention, a series of tests were conducted. Figure 5 The illustration schematically shows the digital filter response and predistortion compensation spectrum of the digitally compensated built-in self-testing nonlinear variable differential displacement sensing device according to an embodiment of this application.

[0108] like Figure 5 As shown, Figure 5 A shows that the amplitude gradually decreases with increasing frequency, exhibiting low-pass filtering characteristics, with significant attenuation at high frequencies. Analog filters can be used for anti-aliasing, signal smoothing, or bandwidth limiting, but their amplitude-frequency response is often not flat enough, especially exhibiting attenuation at the passband edges. Figure 5 b shows that the amplitude increases with frequency, exhibiting high-pass or gain-boosting characteristics, designed to compensate for the attenuation of analog filters in the high-frequency range. Digital filters, through predistortion or inverse filtering, flatten the non-flat response of analog filters, thereby achieving a flatter overall response in the system. Figure 5 c shows that the overall amplitude-frequency response of the cascaded analog filter and digital compensation filter remains flat across the entire frequency range, indicating that the digital compensation filter successfully cancels the amplitude-frequency distortion of the analog filter, giving the system an ideal flat amplitude-frequency characteristic. Figure 5 Figure d shows that after the excitation digital waveform with predistortion compensation passes through the DAC, filter, power operational amplifier, and other stages of the analog front-end conditioning module, the nonlinear distortion and gain compression of the analog front-end are canceled at the source, and the final output to the primary coil of the NVDT is an ideal sinusoidal analog excitation signal, effectively suppressing the second, third, and fourth harmonic distortions. The predistortion compensation spectrum comparison is as follows: Figure 5 As shown in d, the harmonic amplitude is significantly reduced after compensation.

[0109] Figure 6 The static accuracy test error curve of a digitally compensated, self-testing, nonlinear variable differential displacement sensor according to an embodiment of this application is illustrated. Figure 6 The static nonlinear error of the device in this embodiment and the traditional solution were compared with the curve graph over the entire range. The static accuracy test was based on a precision laser interferometer. The results show that the maximum nonlinear error of the device of the present invention is 0.038% FS, which is significantly better than the traditional method.

[0110] Figure 7 The schematic diagram illustrates the temperature drift suppression stable output curve of a digitally compensated, self-testing, nonlinear variable differential displacement sensor according to an embodiment of this application. Figure 7 The stability of the normalized output of the device of the present invention as a function of temperature during temperature cycling is illustrated by a graph. Temperature cycling tests show that, within the temperature range of -40℃ to +125℃, the maximum output drift of the device of the present invention, after compensation, is suppressed to within ±0.05% FS. Dynamic response tests show that the -3 dB bandwidth of the device of the present invention exceeds 150 Hz, and the group delay is constant. Under strong common-mode interference, the output fluctuation is less than 0.01% FS, demonstrating excellent anti-interference capability.

[0111] The above test results show that the digitally compensated built-in self-testing nonlinear variable differential displacement sensing device / device in the embodiments of the present invention is significantly superior to the traditional LVDT / NVDT signal conditioning scheme in terms of accuracy, temperature drift suppression, dynamic response, and anti-interference capability. It effectively solves the problems of magnetic saturation edge nonlinearity, system coupling nonlinearity, temperature drift, phase noise and delay uncertainty, and difficulty in real-time parallel built-in self-testing in LVDT / NVDT high-precision ranging technology and application.

[0112] Based on the aforementioned digitally compensated, self-testing, nonlinear variable differential displacement sensing device, embodiments of this application also provide a digitally compensated, self-testing, nonlinear variable differential displacement sensing method. The following will be combined with... Figure 8 The device is described in detail.

[0113] Figure 8 A flowchart illustrating a digitally compensated, self-testing, nonlinear variable differential displacement sensing method according to an embodiment of this application is shown.

[0114] like Figure 8 As shown, the digital compensation built-in self-testing nonlinear variable differential displacement sensing method in this embodiment of the application uses the sensing device described above. The method includes:

[0115] S810: Using direct digital frequency synthesis technology, a sinusoidal excitation signal driving the nonlinear variable differential transformer is synchronously generated, along with an in-phase reference signal and a quadrature reference signal that are at the same frequency and strictly locked in phase with the sinusoidal excitation signal; S820: The nonlinear variable differential transformer is driven by the sinusoidal excitation signal to acquire the differential voltage signal in real time; S830: The differential voltage signal is conditioned and converted into a digital signal; S840: The digital signal is synchronously demodulated using the in-phase reference signal and the quadrature reference signal to obtain the baseband in-phase component and the quadrature component; S850: Based on the baseband in-phase component and the quadrature component, the demodulated complex plane signal is sequentially subjected to complex plane vector zero compensation, nonlinear inverse compensation, and adaptive temperature drift correction to obtain a high-precision displacement value; and, in parallel, real-time health monitoring and fault diagnosis are performed on the linearity of the excitation source, sensor coil, signal link, and the effectiveness of digital compensation.

[0116] In this embodiment, the inverse nonlinear function of the entire cascade is obtained through laser interferometry calibration, which includes: using a laser interferometer to synchronously collect the true value of displacement and temperature, constructing a calibration dataset, and obtaining the laser interferometry calibration inverse nonlinear function through inverse identification; the laser interferometry calibration inverse nonlinear function is fitted with a piecewise polynomial and stored in the digital processing compensation module in the form of a lookup table.

[0117] In this embodiment, performing analog front-end digital predistortion compensation includes: in the output path of the direct digital frequency synthesis excitation generation module, performing real-time predistortion processing on the original excitation digital waveform according to the inverse function of the calibrated analog front-end nonlinear transfer characteristics, so as to cancel the harmonic distortion and gain compression introduced by the analog conditioning stage at the source of the signal link, wherein the original excitation digital waveform corresponds to the sinusoidal excitation signal.

[0118] In this embodiment, performing complex plane vector zero compensation includes: characterizing the zero-point residual error as a zero-point residual vector on the complex plane formed by feedforward synchronous demodulation; and subtracting the zero-point residual vector from the current measurement signal vector on the complex plane to achieve complex plane vector zero compensation. Performing nonlinear inverse compensation includes: based on a cascaded nonlinear inverse model established through end-to-end system calibration, from digital signal characteristics to actual displacement, using a piecewise polynomial fitting algorithm for real-time table lookup and interpolation calculations to achieve nonlinear inverse compensation. Performing adaptive temperature drift correction includes: based on feedback data from the temperature sensor, deeply integrating temperature variables during the lookup table interpolation process of the cascaded nonlinear inverse model to dynamically suppress sensitivity and zero-point drift caused by changes in ambient temperature, thereby achieving adaptive temperature drift correction.

[0119] In this embodiment, real-time health monitoring and fault diagnosis of the excitation source, sensor coil, signal link linearity, and digital compensation effectiveness are performed in parallel, including: reading back the sinusoidal excitation signal and determining whether the amplitude, frequency, and waveform distortion of the sinusoidal excitation signal exceed a preset threshold; calculating the sum of the absolute values ​​of the output voltages of the two-stage coils of the nonlinear variable differential transformer, and determining coil faults based on whether the sum of the absolute values ​​deviates from a constant range; evaluating whether the linearity of the analog front-end amplifier and filter degrades by analyzing the harmonic components in the signal under specific test excitation or normal operation; and monitoring the statistical characteristics of the output residual after nonlinear inverse compensation to determine whether the compensation model of the digital processing compensation module is mismatched.

[0120] In this embodiment, the digital compensation built-in self-testing nonlinear variable differential displacement sensing method is implemented based on the aforementioned digital compensation built-in self-testing nonlinear variable differential displacement sensing device. This device adopts electromagnetic compatibility enhancement settings, which include: a metal shielding shell for the sensor, a double-shielded connecting cable, an isolation power supply between analog and digital circuits, and a star-type single-point grounding system.

[0121] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems and methods according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0122] Those skilled in the art will understand that the features described in the various embodiments of this application can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, the features described in the various embodiments of this application can be combined and / or combined in various ways without departing from the spirit and teachings of this application. All such combinations and / or combinations fall within the scope of this application.

Claims

1. A digitally compensated, self-testing, nonlinear variable differential displacement sensing device, characterized in that, The device includes: a nonlinear variable differential transformer, a high signal-to-noise ratio analog front-end conditioning module, a direct digital frequency synthesis excitation generation module, a digital processing compensation module, and a built-in self-test module; The nonlinear variable differential transformer is used to convert linear displacement signals into differential voltage signals with inherent nonlinear characteristics; The high signal-to-noise ratio analog front-end conditioning module is used to amplify, anti-aliasing filter, and perform analog-to-digital conversion on the differential voltage signal to obtain a digital signal; The direct digital frequency synthesis excitation generation module is used to generate a sinusoidal excitation signal that drives the primary coil of the nonlinear variable differential transformer, and synchronously generates an in-phase reference signal and a quadrature reference signal with the same frequency and strictly locked phase as the sinusoidal excitation signal based on the same frequency synthesis clock source. The digital processing compensation module is used to receive the digital signal and perform feedforward synchronous demodulation using the in-phase reference signal and the quadrature reference signal. It uses the full-link cascaded nonlinear inverse function obtained based on laser interferometer calibration to perform analog front-end digital pre-distortion compensation, complex plane vector zero compensation, nonlinear inverse compensation and adaptive temperature drift correction, and outputs high-precision displacement value. The built-in self-testing module is used to monitor the excitation source, sensor coil, signal link linearity, and the effectiveness of digital compensation in parallel and in real time. The direct digital frequency synthesis excitation generation module and the digital processing compensation module together constitute a feedforward synchronous demodulation architecture, so that the demodulation process does not require phase tracking by a phase-locked loop.

2. The apparatus according to claim 1, characterized in that, The end-to-end cascaded nonlinear inverse function in the digital processing compensation module is obtained through laser interferometry calibration, including: The true values ​​of displacement and temperature are simultaneously acquired using a laser interferometer to construct a calibration dataset. The nonlinear inverse function of laser interferometry calibration is obtained through reverse identification. The laser interferometric calibration nonlinear inverse function is fitted using a piecewise polynomial and stored in the digital processing compensation module in the form of a lookup table.

3. The apparatus according to claim 1, characterized in that, The analog front-end digital pre-distortion compensation in the digital processing compensation module includes: In the output path of the direct digital frequency synthesis excitation generation module, the original excitation digital waveform is pre-distorted in real time according to the inverse function of the calibrated analog front-end nonlinear transfer characteristics, so as to cancel the harmonic distortion and gain compression introduced by the analog conditioning stage at the source of the signal link. The original excitation digital waveform corresponds to the sinusoidal excitation signal.

4. The apparatus according to claim 1, characterized in that, The digital processing compensation module performs complex plane vector zero compensation, including: On the complex plane formed by feedforward synchronous demodulation, the zero-point residual error is characterized as a zero-point residual vector; The zero-point residual vector is subtracted from the current measurement signal vector on the complex plane to achieve zero compensation of the complex plane vector.

5. The apparatus according to claim 1, characterized in that, The digital processing compensation module performs nonlinear inverse compensation, including: based on the cascaded nonlinear inverse model from digital signal features to real displacement established through full-link system calibration, a piecewise polynomial fitting algorithm is used to perform real-time table lookup and interpolation calculation to achieve nonlinear inverse compensation.

6. The apparatus according to claim 5, characterized in that, The digital processing compensation module performs adaptive temperature drift correction, including: deeply integrating temperature variables during the lookup table interpolation process of the cascaded nonlinear inverse model based on feedback data from the temperature sensor, dynamically suppressing sensitivity and zero-point drift caused by changes in ambient temperature, so as to achieve adaptive temperature drift correction.

7. The apparatus according to claim 1, characterized in that, The built-in self-testing module includes: an excitation source monitoring unit, a sensor coil monitoring unit, a signal link linearity monitoring unit, and a compensation effectiveness monitoring unit; The excitation source monitoring unit is used to read back the sinusoidal excitation signal and determine whether the amplitude, frequency and waveform distortion of the sinusoidal excitation signal exceed a preset threshold. The sensor coil monitoring unit is used to calculate the sum of the absolute values ​​of the output voltages of the two-stage coils of the nonlinear variable differential transformer, and to determine the coil fault based on whether the sum of the absolute values ​​deviates from a constant range. The signal link linearity monitoring unit analyzes the harmonic components in the signal under specific test excitation or normal operation to assess whether the linearity of the analog front-end amplifier and filter has degraded. The compensation effectiveness monitoring unit is used to monitor the statistical characteristics of the output residual after nonlinear inverse compensation and to determine whether the compensation model of the digital processing compensation module is mismatched.

8. The apparatus according to claim 1, characterized in that, The device employs electromagnetic compatibility enhancement features, which include: a metal shielded housing for the sensor, double-shielded connecting cables, an isolated power supply between analog and digital circuits, and a star-connected single-point grounding system.

9. A digitally compensated, self-checking, nonlinear variable differential displacement sensing method, characterized in that, The method employs the apparatus as described in any one of claims 1 to 8, characterized in that the method comprises: Using direct digital frequency synthesis technology, a sinusoidal excitation signal that drives a nonlinear variable differential transformer is synchronously generated, along with an in-phase reference signal and a quadrature reference signal that have the same frequency and are strictly locked in phase with the sinusoidal excitation signal. The nonlinear variable differential transformer is driven by the sinusoidal excitation signal to acquire the differential voltage signal in real time; The differential voltage signal is conditioned and converted into a digital signal; The digital signal is synchronously demodulated by feedforward using the in-phase reference signal and the quadrature reference signal to obtain the baseband in-phase component and the quadrature component; Based on the baseband in-phase and quadrature components, the demodulated complex plane signal is sequentially subjected to complex plane vector zero compensation, nonlinear inverse compensation, and adaptive temperature drift correction to obtain high-precision displacement values; and... It performs real-time health monitoring and fault diagnosis of the excitation source, sensor coil, signal link linearity, and digital compensation effectiveness in parallel.

Images