New energy equipment digital twin construction method and system based on equipment code

Abstract

This invention discloses a method and system for constructing digital twins of new energy equipment based on device coding, belonging to the field of intelligent management and control of new energy equipment. The method includes: Step 1, constructing a resonant cavity, which comprises a static reference unit and a dynamic modulation unit. The static reference unit generates a reference peak, and the dynamic modulation unit couples dynamic physical quantities to a variable capacitor or inductor array to generate an offset peak; Step 2, exciting the resonant cavity with a frequency-modulated continuous wave, receiving the echo and performing a frequency domain transformation to obtain a composite spectrum containing the reference peak and the offset peak; Step 3, using the reference peak in the composite spectrum as the origin, and using the frequency difference between the offset peak and the reference peak as a radial vector. This invention can solve the problem that existing static device coding cannot carry real-time operational data, enabling one code to associate data throughout the entire lifecycle of the device, avoiding repeated database queries during digital twin model updates, improving the real-time performance of model updates, and reducing data binding errors.

Description

Technical Field

[0001] This invention relates to the field of intelligent management and control of new energy equipment, and more specifically, to a method and system for constructing digital twins of new energy equipment based on equipment coding. Background Technology

[0002] As the new energy industry rapidly develops towards intelligence and large-scale operation, digital twin technology has become a core support for improving the operation and maintenance efficiency of new energy equipment and ensuring its safe and stable operation. It is widely used in new energy scenarios such as wind farms and photovoltaic power stations to achieve precise control and optimized scheduling of equipment throughout its entire life cycle. The accurate construction and real-time updating of digital twin models cannot be separated from the collaborative drive of equipment coding and real-time dynamic data. The effective association between the two is a key prerequisite for achieving accurate mapping between virtual models and physical equipment.

[0003] Existing equipment codes are mostly static attribute tags, which can only record fixed basic information such as equipment model, specifications, and manufacturing date. They cannot carry data that changes dynamically over time, such as equipment operating status, health, and performance parameters. This static coding design results in the coding system and the dynamic data system being independent of each other in underlying physical logic. As a discrete and immutable symbol, the code lacks an effective correlation mechanism with the continuous and real-time changing sensor data stream, forming a loosely coupled state.

[0004] Existing codes can only serve as static identification tags for devices and cannot become dynamic carriers of real-time information. They cannot meet the requirement of linking data throughout the entire lifecycle with a single code. This problem directly leads to the need to repeatedly query external databases to obtain the mapping relationship between codes and dynamic data and rebind data streams when updating digital twin models. This not only significantly reduces the real-time update performance of the twin model but also easily causes problems such as data packet misalignment and data binding errors, affecting the model's accurate mapping of physical devices.

[0005] In new energy scenarios with extremely high real-time requirements, such as large-scale wind farms, which typically contain hundreds of wind turbines, their digital twin models need to be updated synchronously at millisecond speeds to quickly respond to changes in blade loads caused by sudden wind speed changes and ensure the safe and stable operation of the turbines. However, the pursuit of uniqueness in existing static coding systems exacerbates the above problems. Each time the twin model is refreshed, the coding-data mapping table of each wind turbine needs to be queried one by one and the sensor stream needs to be rebound, which will generate a cumulative delay of hundreds of milliseconds. Moreover, the model state is prone to errors due to misaligned data packets, which seriously affects the application effect of digital twin technology in the management and control of new energy equipment and restricts the digital and intelligent upgrading process of the new energy industry. Summary of the Invention

[0006] To address the problems existing in the prior art, the present invention aims to provide a method and system for constructing digital twins of new energy equipment based on equipment coding. This method can solve the problem that existing static equipment coding cannot carry real-time operating data, realize the association of one code with the entire life cycle data of the equipment, avoid repeated database queries when updating the digital twin model, improve the real-time performance of model updates, and reduce data binding errors.

[0007] To solve the above problems, the present invention adopts the following technical solution: The first aspect is a method for constructing digital twins of new energy equipment based on equipment coding, including: Step 1: Construct a resonant cavity, which includes a static reference unit and a dynamic modulation unit. The static reference unit generates a reference peak, and the dynamic modulation unit couples dynamic physical quantities to a variable capacitor or inductor array to generate an offset peak. Step 2: Excite the resonant cavity with a frequency-modulated continuous wave, receive the echo and perform frequency domain transformation to obtain a composite spectrum containing the reference peak and the offset peak; Step 3: Using the reference peak in the composite spectrum as the origin, the frequency difference between the offset peak and the reference peak is taken as the radial vector, and an ideal radial vector is preset. Step 4: Based on the deviation between the radial vector and the preset ideal radial vector, generate an anti-resonant electromagnetic pulse. The frequency of this anti-resonant electromagnetic pulse is aligned with the response bandwidth of the variable capacitor or inductor array. Step 5: Use anti-resonant electromagnetic pulses to deform the variable capacitor or inductor array to change the capacitance and shift the position of the subsequent shift peak. Step 6: Repeat steps 2 to 5 periodically to obtain an updated composite spectrum diagram with each frequency sweep, and generate the next anti-resonant electromagnetic pulse based on the radial vector deviation.

[0008] Further, step 1 includes: Construct a set of reflection structures on the resonant cavity substrate to generate discrete reference frequency peaks; On the same substrate and at a location that does not overlap with the space of the reflective structure, a resonant branch consisting of a variable capacitor or an inductor array is constructed, which generates a movable offset peak.

[0009] Furthermore, step 1 also includes: A nonlinear pre-distortion coupling path is established between dynamic physical quantities and the variable capacitor gap. This path makes the input displacement and the change in capacitor gap present a predetermined nonlinear functional relationship. The reflection structure and the resonant branch are encapsulated in the same cavity, and an electromagnetic absorption layer is set in the cavity to cover the transition frequency band between the reference peak and the offset peak.

[0010] Further, step 2 includes: Generate a continuous wave excitation signal whose instantaneous frequency changes logarithmically with time; Perform a fractional Fourier transform on the echo, with the transform order being the reciprocal of the base of the logarithmic function, to obtain a composite spectrum containing the reference peak and the offset peak.

[0011] Further, step 3 includes: In the composite spectrum diagram, the reference peak is selected by Lorentz line matching, and the weighted center frequency of the reference peak is used as the identity origin. The difference between the offset peak frequency and the identity origin is mapped to the physical quantity dimension according to the neighborhood of the reference peak to which the offset peak belongs, and after normalization, it forms a multidimensional radial vector. The ideal radial vector is preset based on the median of the radial vector obtained by multiple frequency sweeps during the initial commissioning of the equipment, and is fixed in the immutable storage area of ​​the resonant cavity.

[0012] Further, step 4 includes: The component deviation between the radial vector and the ideal radial vector is decomposed into amplitude deviation, sign deviation, and rate of change deviation relative to the previous frequency sweep, and encoded into amplitude code, phase code, and pulse width modulation code. Based on the amplitude code and pulse width modulation code, a baseband pulse train with an exponentially decaying envelope and a controlled pulse width is generated through nonlinear resistive-capacitive relaxation oscillation. The baseband pulse train is mixed with a carrier signal whose frequency is aligned with the response bandwidth of a variable capacitor or inductor array, and the phase of the mixing output is controlled according to the sign deviation.

[0013] Furthermore, step 4 also includes: The RF pulse train output from the mixer is amplified by nonlinear saturation to flatten the pulse peak to a predetermined limiting level while retaining the exponential decay tail. The saturated amplified radio frequency pulse train is passed through a time delay line array to assign different arrival times to each physical quantity dimension, forming a time-division multiplexed anti-resonant electromagnetic pulse group.

[0014] Further, step 5 includes: Each pulse in the anti-resonant electromagnetic pulse group is applied sequentially to the corresponding variable capacitor or inductor array. The exponential decay tail of the pulse is used to excite forced vibration, generating transient displacement that is proportional to the pulse amplitude and whose phase determines the vibration direction. Transient displacement is converted into unidirectional stepping displacement through a mechanical ratchet, causing the equilibrium position to undergo an irreversible stepping change that is proportional to the pulse amplitude, and only in-phase pulses produce the stepping. The equilibrium position after the step is locked by electrostatic latching and temporarily unlocked when the next in-phase pulse arrives.

[0015] Further, step 6 includes: Based on the magnitude difference between the current radial vector and the previous radial vector, the waiting time interval for the next frequency sweep is dynamically adjusted. The magnitude and the interval are inversely correlated. When the magnitude is less than the dead zone threshold, the time interval is extended to the upper limit. At the end of the waiting interval, based on the comparison between the residual deviation magnitude of the current radial vector and the ideal radial vector and the compensation threshold, it is determined whether to skip steps 4 and 5, update the comparison benchmark to the current radial vector, and halve the compensation threshold when the number of consecutive skips reaches a predetermined upper limit.

[0016] Secondly, the present invention also provides a digital twin construction system for new energy equipment based on device coding, comprising: The resonant configuration module constructs a resonant cavity, which includes a static reference unit and a dynamic modulation unit. The static reference unit generates a reference peak, and the dynamic modulation unit couples dynamic physical quantities to a variable capacitor or inductor array to generate an offset peak. The frequency modulation conversion module excites the resonant cavity with a frequency-modulated continuous wave, receives the echo and performs frequency domain transformation to obtain a composite spectrum containing the reference peak and the offset peak. The vector calibration module uses the reference peak in the composite spectrum as the origin, takes the frequency difference between the offset peak and the reference peak as the radial vector, and presets an ideal radial vector. The anti-pulse generation module generates an anti-resonant electromagnetic pulse based on the deviation between the radial vector and the preset ideal radial vector. The frequency of the anti-resonant electromagnetic pulse is aligned with the response bandwidth of the variable capacitor or inductor array. The deformation peak shifting module uses anti-resonant electromagnetic pulses to deform a variable capacitor or inductor array to change the capacitance and shift the position of the subsequent shift peak. The periodic loop module periodically calls the frequency modulation conversion module, vector calibration module, anti-pulse generation module, and deformation peak shifting module to obtain an updated composite spectrum diagram for each frequency sweep, and generates the next anti-resonant electromagnetic pulse based on the radial vector deviation.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) This scheme constructs a resonant cavity containing a static reference unit and a dynamic modulation unit, and couples the static identity identifier with the dynamic physical quantity data. This solves the problem that the existing static device coding cannot carry real-time running data, realizes the association of one code with the entire life cycle data of the device, avoids repeated database queries when updating the digital twin model, improves the real-time performance of the model update, and reduces data binding errors.

[0018] (2) This scheme adopts logarithmic frequency modulation continuous wave excitation and resonant peak Lorentz line matching screening method, combined with fractional Fourier transform to process the echo signal, effectively eliminating clutter interference, ensuring clear separation of the reference peak and the offset peak, providing accurate spectral data for radial vector calculation, and ensuring accurate mapping between the digital twin model and the physical device.

[0019] (3) This scheme drives the deformation of the variable capacitor or inductor array through anti-resonant electromagnetic pulse, and achieves unidirectional stepping and stable locking of displacement with mechanical ratchet and electrostatic latch, so as to realize the precise adjustment of dynamic physical quantities, so that the offset peak is oriented closer to the ideal position, continuously corrects the deviation of the digital twin model, and improves the stability of model operation.

[0020] (4) This solution reduces invalid frequency sweeps and adjustment operations while ensuring adjustment accuracy by dynamically adjusting the frequency sweep waiting interval, adaptively optimizing the compensation threshold, and selectively skipping invalid adjustment steps. This reduces system energy consumption and equipment loss, adapts to the high real-time and low loss operation requirements of new energy equipment, and promotes the digital upgrade of the industry. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0022] Figure 1 This is a flowchart of the method for constructing a digital twin of new energy equipment based on equipment coding according to the present invention; Figure 2 This is a data flow diagram between various modules in the digital twin construction system for new energy equipment based on device coding, as described in this invention. Detailed Implementation

[0023] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0024] Example 1: Please see Figure 1 A method for constructing digital twins of new energy equipment based on equipment coding, the method includes: Step 1: Construct a resonant cavity, which includes a static reference unit and a dynamic modulation unit. The static reference unit generates a reference peak, and the dynamic modulation unit couples dynamic physical quantities to a variable capacitor or inductor array to generate an offset peak. The specific operations are as follows: First, a resonant cavity capable of simultaneously providing fixed identification and dynamic data transmission is constructed. This cavity adopts an integrated structural design, internally integrating two core functional units: a static reference unit and a dynamic modulation unit. These two units work collaboratively without interfering with each other. The core function of the static reference unit is to generate a stable and unchanging reference peak. This reference peak serves as the inherent identifier of the device, with fixed and unique frequency characteristics, unaffected by changes in various dynamic physical quantities during the operation of the new energy equipment. This provides a stable reference for subsequent offset peak positioning and deviation calculation. The dynamic modulation unit is responsible for realizing the coupling conversion between dynamic physical quantities and electrical parameters. Its core component is a variable capacitor or inductor array. When dynamic physical quantities such as temperature, vibration, and load change during the operation of new energy equipment, these dynamic physical quantities are coupled to the variable capacitor or inductor array, causing the capacitance or inductance of the array to change accordingly. This, in turn, causes the dynamic modulation unit to generate a frequency-shiftable offset peak. The reference peak and the offset peak together constitute the core spectral characteristics of the resonant cavity. The reference peak is used to identify the equipment, while the offset peak is used to carry the dynamic operating data of the equipment. The synergistic effect of the two realizes the transformation of equipment coding from static identification to dynamic information carrier, providing a data association basis for the real-time update of the subsequent digital twin model.

[0025] Step 1 also includes the following steps: Step 11: Construct a set of reflection structures that generate discrete reference frequency peaks on the resonant cavity substrate. The specific operations are as follows: The resonant cavity substrate is made of a material with stable electromagnetic and mechanical properties to ensure the long-term stability of the frequency characteristics of the reflection structure, unaffected by environmental factors. The reflection structure adopts a specific geometric layout and size design. By etching or depositing conductive patterns of a specific shape on the substrate surface, a set of independent reflection units are formed, each corresponding to a fixed resonant frequency. These reflection units are distributed in space at a preset interval, and the resonant frequencies of each unit do not overlap. Thus, after the incident frequency-modulated continuous wave is reflected by each reflection unit, a set of discrete reference frequency peaks with fixed frequencies are formed. Each reference frequency peak corresponds to a specific reflection unit. All reference frequency peaks together constitute the inherent identification code of the device. Its frequency parameters are calibrated and fixed at the factory and will not change during subsequent operation.

[0026] Step 12: On the same substrate and at a location that does not overlap with the space of the reflective structure, construct a resonant branch consisting of a variable capacitor or inductor array. This resonant branch generates a movable offset peak. The specific operation is as follows: This step is used to construct the core component of the dynamic modulation unit, namely, a resonant branch composed of a variable capacitor or inductor array, located on the same resonant cavity substrate and in a position that does not spatially overlap with the reflective structure. The spatially non-overlapping layout design avoids electromagnetic interference between the reflective structure and the resonant branch, ensuring that their spectral characteristics can be clearly distinguished and do not affect each other. The resonant branch is composed of multiple variable capacitors or inductors connected in a preset array, with each element corresponding to a specific dynamic physical quantity monitoring dimension. The overall resonant frequency of the array is determined by the electrical parameters of each element. When various dynamic physical quantities change during the operation of the new energy equipment, these dynamic physical quantities are transmitted to the variable capacitor or inductor array through a preset coupling mechanism, causing changes in the capacitance or inductance of the corresponding elements in the array, thereby shifting the overall resonant frequency of the resonant branch and forming a movable shift peak. The frequency position of the shift peak has a clear correspondence with the amount of change in the dynamic physical quantity. The greater the change in the dynamic physical quantity, the greater the frequency shift of the shift peak relative to the reference peak, thus realizing the transformation of dynamic physical quantities into spectral characteristics, providing a carrier for subsequent dynamic data extraction and twin model updates.

[0027] Step 13: Establish a nonlinear pre-distortion coupling path from the dynamic physical quantity to the variable capacitor gap. This path ensures a predetermined nonlinear functional relationship between the input displacement and the change in the capacitor gap. The specific operation is as follows: Because the dynamic physical quantities change over a wide range during the operation of new energy equipment, and there is an inherent nonlinear relationship between the gap change and capacitance change of the variable capacitor, direct coupling will lead to distortion in the mapping relationship between the dynamic physical quantities and the offset peak frequency, affecting the accuracy of subsequent deviation calculations. Therefore, a nonlinear pre-distortion coupling path is designed to pre-distort the input of the dynamic physical quantities, so that the input displacement and the capacitor gap change present a predetermined nonlinear functional relationship. This nonlinear functional relationship can be determined through experimental calibration based on the actual range of dynamic physical quantity changes and the characteristic parameters of the variable capacitor. Its function is to compensate for the nonlinear characteristics of the variable capacitor itself, so that a linearized equivalent mapping relationship is formed between the dynamic physical quantities and the capacitor gap change, thereby ensuring that the frequency offset of the offset peak can accurately reflect the actual changes of the dynamic physical quantities. Assuming the input displacement is x and the capacitor gap change is y, the predetermined nonlinear functional relationship can be expressed as y = a*x. 2The formula +b*x+c is a quadratic function obtained by fitting the gap-displacement response characteristics of a variable capacitor through multiple experimental measurements. Here, a, b, and c are preset calibration coefficients determined by fitting experimental data. c represents the initial capacitor gap, typically 0.2 to 0.5 mm, corresponding to the initial spacing when the variable capacitor has zero displacement. b ranges from 0.7 to 1.0, used for linear compensation of the main displacement segment. a takes a negative value to offset the inherent nonlinearity of the capacitor gap-capacitance ratio, usually between -0.15 and -0.8. x represents the input displacement corresponding to the dynamic physical quantity generated during the operation of the new energy equipment, and y represents the change in the variable capacitor gap. This function enables precise coupling between the input displacement and the change in the capacitor gap, compensating for nonlinear distortion.

[0028] Step 14: Encapsulate the reflection structure and the resonant branch in the same cavity, and set an electromagnetic absorption layer inside the cavity that covers the transition frequency band between the reference peak and the offset peak. The specific operation is as follows: Encapsulating the reflective structure and resonant branch within the same cavity effectively isolates them from external environmental factors such as dust, moisture, and electromagnetic interference. This protects the stability and electromagnetic characteristics of the internal structure, preventing frequency drift of the reference peak and offset peak caused by external interference, and ensuring the uniqueness of the encoded identifier and the accuracy of dynamic data transmission. Simultaneously, an electromagnetic absorption layer is placed within the cavity. This absorption layer's absorption band is precisely designed to cover the transition band between the reference peak and the offset peak. Since the transition band between the reference peak and the offset peak generates clutter signals, these clutter signals can interfere with the identification accuracy of the reference peak and offset peak, making it impossible to clearly distinguish the two peaks after subsequent frequency domain transformation, thus affecting the accuracy of dynamic data extraction. The electromagnetic absorption layer absorbs all clutter signals within the transition band, preventing clutter signals from interfering with the spectral characteristics of the reference peak and the offset peak, ensuring clear separation between the reference peak generated by the reflective structure and the offset peak generated by the resonant branch.

[0029] In a preferred embodiment of the present invention, step 2 is further included: exciting the resonant cavity with a frequency-modulated continuous wave, receiving the echo and performing a frequency domain transformation to obtain a composite spectrum containing a reference peak and an offset peak. The specific operation is as follows: The resonant cavity has been constructed in step 1. Its internal static reference unit and dynamic modulation unit have their own resonant characteristics. They need to be excited by an external excitation signal to generate reflected echoes in order to obtain the corresponding spectral information. Frequency-modulated continuous wave is chosen as the excitation signal because it has the characteristics of continuously adjustable frequency, high signal stability, and strong anti-interference ability. It can accurately cover the frequency range of the reference peak generated by the static reference unit and the offset peak generated by the dynamic modulation unit, ensuring that both units can be effectively excited and generate clear reflected signals. After the excitation signal is applied to the resonant cavity, the static reference unit will reflect a signal corresponding to its fixed resonant frequency, and the dynamic modulation unit will reflect a signal corresponding to its current resonant frequency. The two reflected signals are superimposed to form an echo signal, which is captured by the receiving device. Since the echo signal is a continuous signal in the time domain, it is not possible to directly distinguish the frequency characteristics of the reference peak and the offset peak. Therefore, it is necessary to perform a frequency domain transformation on the received echo signal to convert the time domain signal into a frequency domain signal, and finally obtain a composite spectrum containing the reference peak and the offset peak. This spectrum clearly shows the frequency positions of the two peaks, providing an intuitive signal basis for subsequent calculation of the offset peak deviation with the reference peak as the origin.

[0030] Step 2 also includes the following steps: Step 21: Generate a continuous wave excitation signal whose instantaneous frequency varies logarithmically with time. The specific operation is as follows: Traditional linear frequency modulated continuous waves have a constant frequency change rate. When the frequency interval between the reference peak and the offset peak of the resonant cavity is large and the frequency offset range of the dynamic modulation unit is wide, problems such as incomplete coverage of the excitation signal or insufficient excitation of some frequency bands can easily occur, resulting in blurred peaks in the echo signal and affecting the accuracy of subsequent spectrum analysis. Therefore, a continuous wave with an instantaneous frequency that changes logarithmically with time is used as the excitation signal. The frequency change characteristics of the logarithmic function can make the frequency change rate of the excitation signal in the low-frequency and high-frequency bands match the distribution characteristics of the reference peak and the offset peak, ensuring that the excitation signal can fully cover the frequency range of the two peaks, while making the intensity of the reflected signal corresponding to the two peaks uniform, which is convenient for clear identification after subsequent frequency domain transformation.

[0031] The instantaneous frequency changes with time according to a logarithmic function, and its relationship is determined by three preset parameters: the frequency scaling factor aw, the base of the logarithmic function bw, and the initial frequency cw. This relationship is a logarithmic function obtained through experimental fitting based on the frequency distribution range of the resonant cavity's reference peak and offset peak, combined with the excitation efficiency requirements of the excitation signal. The instantaneous frequency is represented by f(t), where t is the time variable of the excitation signal, and the specific value of f(t) is obtained by combining time t with the three parameters. The frequency scaling factor aw is used to adjust the frequency coverage range of the excitation signal to accurately match the resonant cavity's spectral range. The base of the logarithmic function, bw, is used to adjust the rate of frequency change over time, ensuring that the excitation signal can uniformly cover the frequency range of the reference peak and the offset peak. The initial frequency, cw, is used to determine the starting frequency of the excitation signal, ensuring that the starting frequency of the excitation signal matches the lowest resonant frequency of the resonant cavity. The three parameters, aw, bw, and cw, can be determined through experimental calibration. Specifically, the frequency range of the reference peak and the offset peak of the resonant cavity is measured, and different parameter values ​​are substituted for excitation tests. The parameter combination that makes the two peaks in the echo signal clearest is selected and fixed, thereby ensuring that the excitation signal can stably and efficiently excite the resonant cavity to generate a clear echo signal.

[0032] Step 22: Perform a fractional Fourier transform on the echo. The transform order is the reciprocal of the base of the logarithmic function, resulting in a composite spectrum containing the reference peak and the offset peak. The specific operation is as follows: The specific process of frequency domain transformation of the received echo signal involves converting the time-domain echo signal into a composite frequency-domain spectrum using a fractional Fourier transform. The transform order is the reciprocal of the base of the logarithmic function in step 21 to achieve optimal focusing and peak separation of the echo signal. The received echo signal is a superposition of the reflected signals from the static reference unit and the dynamic modulation unit, and is a non-stationary signal. Traditional Fourier transform is poor at processing non-stationary signals, easily leading to spectral spread and peak overlap, making it difficult to clearly distinguish between the reference peak and the offset peak, thus affecting the normal execution of subsequent steps. The fractional Fourier transform, as a generalized form of the traditional Fourier transform, can adjust the transform order according to the characteristics of the signal, achieving accurate frequency domain analysis of non-stationary signals, and is particularly suitable for processing echo signals corresponding to logarithmic frequency modulation excitation signals.

[0033] Since the instantaneous frequency of the excitation signal in step 21 changes logarithmically with time, its signal characteristics have a matching relationship with a fractional Fourier transform of a specific order. Setting the transform order to be the reciprocal of the base of the logarithmic function enables the echo signal to achieve optimal focusing in the frequency domain, effectively suppressing spectral spread, making the outlines of the reference peak and the offset peak clearer, and the frequency position more accurate. If the base of the logarithmic function in step 21 is b, then the order of the fractional Fourier transform is set to 1 / b. This order selection was determined through multiple transform experiments. When the order is the reciprocal of the base of the logarithm, the separation between the reference peak and the offset peak in the echo signal is the highest, and the spectral distortion is the smallest. After processing by this fractional Fourier transform, the echo signal in the time domain is converted into a composite spectrum in the frequency domain. The graph clearly shows two independent peaks, corresponding to the reference peak generated by the static reference unit and the offset peak generated by the dynamic modulation unit, respectively.

[0034] In a preferred embodiment of the present invention, step 3 is further included: taking the reference peak in the composite spectrum as the origin, using the frequency difference between the offset peak and the reference peak as the radial vector, and pre-setting an ideal radial vector. The specific operation is as follows: Step 2 has yielded a composite spectrum diagram containing a reference peak and an offset peak through frequency domain transformation. The reference peak serves as the inherent identifier of the device, with fixed frequency characteristics, making it suitable as the origin for positioning and calibration. The offset peak carries the device's dynamic operating data, and its frequency position shifts with changes in dynamic physical quantities. Using the reference peak in the composite spectrum diagram as the origin establishes a stable reference coordinate system, ensuring that all subsequent dynamic data extraction and calculation are based on a unified benchmark, avoiding data deviations caused by reference benchmark fluctuations. The frequency difference between the offset peak and the reference peak is used as a radial vector because the frequency difference directly corresponds to the change in dynamic physical quantities, and the vector form can simultaneously reflect the magnitude and direction of the change, adapting to the monitoring needs of multi-dimensional dynamic physical quantities. A preset ideal radial vector provides a clear comparison benchmark for subsequent judgments on whether the dynamic physical quantity corresponding to the offset peak is within the normal range and whether adjustment is needed. This ideal vector corresponds to the dynamic data under normal operating conditions of the device. By comparing it with the actual radial vector, device operating deviations can be quickly identified, providing a basis for accurate updates of the digital twin model and device operation and maintenance.

[0035] Step 3 also includes the following steps: Step 31: In the composite spectrum diagram, a reference peak is selected by Lorentz line matching, and the weighted center frequency of the reference peak is used as the origin of identity. The specific operation is as follows: In addition to the reference peak and the offset peak, the composite spectrum may also contain a small amount of clutter signals. These clutter signals can interfere with the identification of the reference peak. If the signal with the highest peak value is directly selected as the reference peak, misjudgment may occur, affecting the accuracy of subsequent data calculations. Since the reference peak generated by the static reference unit is a resonant signal, its spectrum curve conforms to the Lorentz line shape. This characteristic is the key basis for distinguishing the reference peak from clutter and offset peaks. Therefore, the Lorentz line shape matching method is used to screen the reference peak. The specific process is to fit and compare the signal curve in the composite spectrum with the preset Lorentz line shape, remove signals with a fitting degree lower than the preset threshold, and retain the signal with the highest fitting degree as the reference peak. This method can effectively eliminate clutter interference and ensure that the screened reference peak is accurate. The frequency of the reference peak is not a single fixed value. Its spectrum curve has a certain width. If only the frequency corresponding to the peak value is selected as the identity origin, the reference deviation will be caused by the spectrum width. Therefore, the weighted center frequency of the reference peak is used as the identity origin.

[0036] The calculation of the weighted center frequency needs to consider the signal strength at each frequency point of the reference peak spectrum curve. The greater the signal strength, the higher the weight of the corresponding frequency. The specific calculation process needs to be carried out in two steps: The first step is to select the starting frequency f1 and the ending frequency f2 of the reference peak spectrum, multiply each frequency point f in the interval with its corresponding signal intensity I(f), and then integrate all the product results to obtain the sum of the integrals as the numerator of the calculation. The second step is to directly integrate the signal strength I(f) corresponding to all frequency points within the spectrum interval, and use the sum of the integrals as the denominator for calculation. Finally, divide the numerator integral result by the denominator integral result, and the resulting value is the weighted center frequency f0 of the reference peak, which is the frequency corresponding to the identity origin.

[0037] This calculation method is derived from the distribution characteristics of the reference peak spectrum, which can reflect the overall frequency characteristics of the reference peak and avoid the deviation caused by a single peak frequency. Here, f0 is the weighted center frequency of the reference peak, that is, the frequency corresponding to the identity origin; f is any frequency point within the spectrum range of the reference peak; I(f) is the signal strength corresponding to frequency f; f1 and f2 are the start frequency and end frequency of the reference peak spectrum, respectively, which can be determined by Lorentz line fitting; the identity origin calculated by this method can accurately reflect the inherent frequency characteristics of the reference peak.

[0038] Step 32: Map the difference between the offset peak frequency and the origin of identity to the physical quantity dimension according to the neighborhood of the reference peak to which the offset peak belongs. After normalization, a multidimensional radial vector is formed. The specific operation is as follows: Multiple offset peaks may exist in a composite spectrum, each corresponding to a dynamic physical quantity of a specific dimension. Each offset peak has a spatial and frequency neighborhood association with a specific reference peak; this neighborhood association is the key basis for determining the dimension of the physical quantity corresponding to the offset peak. First, the difference between the frequency of each offset peak and the identity origin (i.e., the weighted center frequency of the reference peak) is calculated. This difference directly reflects the degree of change in the corresponding dynamic physical quantity; the larger the difference, the more significant the deviation of the dynamic physical quantity from its normal state. Then, based on the neighborhood of the reference peak to which the offset peak belongs, the frequency difference is mapped to the corresponding physical quantity. The dimension of a quantity refers to the conversion of frequency differences into specific physical quantity changes through a pre-defined mapping relationship. For example, the frequency difference in one neighborhood is mapped to the vibration displacement of a wind turbine blade, and the frequency difference in another neighborhood is mapped to the operating temperature of the equipment, ensuring that each frequency difference accurately corresponds to a physical quantity dimension. Since different physical quantity dimensions have different ranges and dimensions of change, directly combining the frequency differences of each dimension into a vector would result in excessively large differences in the magnitude of the values, affecting the accuracy of subsequent deviation calculations. Therefore, it is necessary to normalize the mapping results of each dimension. The normalization process is carried out in two steps: The first step is to calculate the frequency difference between the offset peak corresponding to the i-th physical quantity dimension and the identity origin. fi, minus the minimum frequency difference corresponding to the dimension of that physical quantity. The first step is to calculate the minimum frequency difference between the two values, which is the minimum frequency difference when the equipment is operating normally. The second step is to calculate the maximum frequency difference corresponding to this physical quantity dimension. fimax and minimum value The difference of fimin is the maximum frequency difference within the allowable operating range of the device. Finally, the difference obtained in the first step is divided by the difference obtained in the second step, and the resulting value is the normalized vector component vi of the i-th physical quantity dimension. This processing method is derived to eliminate the influence of the dimensions of different physical quantities and to map the frequency differences of each dimension to the same order of magnitude. Here, vi is the normalized vector component of the i-th physical quantity dimension; and is the frequency difference between the offset peak and the origin corresponding to the i-th physical quantity dimension. fimin is the minimum frequency difference corresponding to the dimension of this physical quantity; fimax is the maximum frequency difference corresponding to the dimension of this physical quantity; by combining the vector components after normalizing all dimensions of physical quantities, a multidimensional radial vector can be formed, which can comprehensively and accurately reflect the operating status of dynamic physical quantities in various dimensions of the equipment.

[0039] Step 33: Based on the median of the radial vector obtained from multiple frequency sweeps during the initial commissioning of the equipment, a pre-set ideal radial vector is established and stored in the immutable memory region of the resonant cavity. The specific operation is as follows: The ideal radial vector corresponds to the dynamic physical quantity data of various dimensions under normal operating conditions of the equipment. Its accuracy directly affects the subsequent deviation calculation and dynamic adjustment effect. Therefore, it needs to be determined based on the initial state data when the equipment is first put into operation. When the equipment is first put into operation, it is in a normal state with minimal external interference and no long-term operating wear. At this time, multiple frequency sweeps can obtain multiple sets of radial vector data. Multiple frequency sweeps can effectively avoid random errors that may occur during a single frequency sweep, ensuring the representativeness and accuracy of the data. The number of frequency sweeps must meet the preset quantity requirements to ensure that the obtained radial vector data can cover the normal fluctuation range when the equipment is first put into operation, avoiding deviation of the ideal radial vector due to insufficient data. The median value is used to determine the ideal radial vector from the radial vector data obtained from multiple frequency sweeps. The radial vector is chosen because the median effectively suppresses the influence of outliers and is more stable than the average. It accurately reflects the radial vector characteristics under normal operating conditions, avoiding deviations from the ideal reference caused by single abnormal frequency sweeps. After determining the ideal radial vector, it is fixed in the immutable storage area of ​​the resonant cavity. The design of the immutable storage area prevents the ideal radial vector from being accidentally tampered with or modified during equipment operation, ensuring that all subsequent deviation calculations are based on a fixed ideal reference, thus guaranteeing the accuracy and stability of digital twin model updates. The fixing process adopts a hardware-level locking mechanism, making the ideal radial vector an inherent parameter of the equipment. It can only be recalibrated using dedicated equipment during factory calibration or major overhauls and cannot be changed during daily operation, ensuring its long-term stability.

[0040] In a preferred embodiment of the present invention, step 4 is further included: generating an anti-resonant electromagnetic pulse based on the deviation between the radial vector and a preset ideal radial vector. The frequency of the anti-resonant electromagnetic pulse is aligned with the response bandwidth of the variable capacitor or inductor array. The specific operation is as follows: By comparing the radial vector corresponding to the actual operating state with the ideal radial vector corresponding to the ideal state of the equipment, the operating deviation between the two is obtained. Based on this deviation, a dedicated anti-resonant electromagnetic pulse is generated. The center frequency of the pulse is matched with the electromagnetic response bandwidth of the variable capacitor or inductor array to ensure that the pulse can effectively act on the adjustable elements of the resonant branch. The deviation between the radial vector and the ideal radial vector directly reflects the degree and direction of the deviation of the current operating physical quantity of the new energy equipment from the standard state. As an adjustment execution signal, the anti-resonant electromagnetic pulse can change the electrical parameters of the variable capacitor or inductor array through electromagnetic action, thereby correcting the frequency position of the offset peak, so that the spectral characteristics of the resonant cavity output gradually approach the ideal state, realizing the self-calibration of the equipment's dynamic coding and operating data, and providing accurate and synchronized mapping data for the digital twin model.

[0041] Step 4 also includes the following steps: Step 41: Decompose the component deviation between the radial vector and the ideal radial vector into amplitude deviation, sign deviation, and rate of change deviation relative to the previous frequency sweep, and encode them into amplitude code, phase code, and pulse width modulation code. The specific operation is as follows: The deviations between the radial vector and the ideal radial vector in each physical quantity dimension are decomposed into multi-dimensional values ​​and converted into corresponding digital codes. First, the difference between the actual radial vector component and the ideal radial vector component is calculated dimension by dimension. The absolute value of this difference is extracted as the amplitude deviation, used to reflect the magnitude of the deviation of the current physical quantity from the ideal state. The positive or negative attribute of this difference is defined as the sign deviation, used to distinguish the adjustment direction of the physical quantity being too high or too low. Next, the rate of change between the current component deviation and the component deviation obtained from the previous frequency sweep is calculated to obtain the rate of change deviation, used to reflect the dynamic trend of the deviation. After completing the decomposition of the three types of deviations, the amplitude deviation is converted into an amplitude code, the value of which corresponds to the magnitude of the amplitude deviation. The sign deviation is converted into a phase code, used to identify the phase attribute of subsequent electromagnetic pulses. The rate of change deviation is converted into a pulse width modulation (PWM) code, used to constrain the duration of subsequent pulses. The three types of codes independently carry different characteristics of the deviation and do not interfere with each other.

[0042] Step 42: Based on the amplitude code and pulse width modulation code, a baseband pulse train with an exponentially decaying envelope and controlled pulse width is generated through nonlinear resistor-capacitor relaxation oscillation. The specific operation is as follows: Using amplitude code and pulse width modulation code as control parameters, a nonlinear resistive-capacitive relaxation oscillation circuit generates a baseband pulse train that meets the requirements. The operating parameters of the nonlinear resistive-capacitive relaxation oscillation circuit are jointly set by amplitude code and pulse width modulation code. Amplitude code determines the initial excitation intensity of the circuit oscillation, thereby controlling the amplitude reference of the output pulse. Pulse width modulation code regulates the charging and discharging cycle of the circuit, thereby limiting the duration and interval of the pulse. During operation, the circuit naturally forms an oscillating waveform whose amplitude decays exponentially with time. After shaping, this waveform forms a baseband pulse train. Each single pulse in the pulse train has an exponentially decaying envelope characteristic, and the pulse width strictly follows the setting of the pulse width modulation code, so that the time domain characteristics of the baseband pulse train are adapted to the amplitude and trend of the equipment's operating deviation.

[0043] Step 43: Mix the baseband pulse train with a carrier signal whose frequency is aligned with the response bandwidth of the variable capacitor or inductor array, and control the phase of the mixing output according to the sign deviation. The specific operation is as follows: The baseband pulse train is mixed with the target carrier signal to form an RF pulse, and the phase of the mixed output signal is controlled by the sign deviation. First, a carrier signal with a center frequency that coincides with the center of the response bandwidth of the variable capacitor or inductor array is selected. This frequency setting ensures that the mixed RF signal can be effectively responded to by the adjustable element. The baseband pulse train and the carrier signal are simultaneously input into the mixing unit. The characteristics of the baseband pulse are loaded onto the carrier through frequency shifting processing to form an RF pulse train. During the mixing process, the phase of the mixed output signal is adjusted according to the sign deviation obtained in step 41. When the sign deviation is positive, a positive phase is output, and when the sign deviation is negative, a negative phase is output. The phase difference corresponds to the adjustment direction of the subsequent variable capacitor or inductor array, so that the RF pulse train carries complete information of both the deviation amplitude and the adjustment direction.

[0044] Step 44: The RF pulse train output from the mixer is amplified by nonlinear saturation to flatten the pulse peak to a predetermined limiting level while retaining the exponentially decaying tail. The specific operation is as follows: The RF pulse train output from the mixer undergoes nonlinear saturation amplification, limiting the pulse peak value while preserving the exponentially decaying tail. The RF pulse train is then input into the nonlinear saturation amplification circuit, which clips the pulse peak value according to a preset amplitude limit, preventing overdrive or structural damage to the variable capacitor or inductor array due to excessive pulse amplitude. During peak clipping, the circuit fully preserves the original exponentially decaying tail of the pulse. This attenuation structure provides continuous and stable excitation energy for the forced vibration of subsequent array elements, preventing a decrease in adjustment accuracy caused by sudden strong impacts. The amplitude of the RF pulse after saturation amplification is within a safe and controllable range, while retaining the time-domain waveform characteristics required for precise adjustment.

[0045] Step 45: The saturated amplified radio frequency pulse train is passed through a time delay line array to assign different arrival times to each physical quantity dimension, forming a time-division multiplexed anti-resonant electromagnetic pulse group. The specific operation is as follows: The time-delayed linear array is used to time-distribute the saturated amplified radio frequency pulses, forming a time-division multiplexed anti-resonant electromagnetic pulse group. The time-delayed linear array contains multiple independent delay channels, each corresponding to a physical quantity dimension, and each channel is set with a different signal transmission delay. The radio frequency pulses corresponding to each dimension are sent to the matched delay channels respectively. The time delay difference between different channels will cause the pulses of each dimension to be output sequentially at different times, avoiding multiple pulses acting on the array at the same time, which would cause electromagnetic interference and adjustment disorder. The radio frequency pulses output sequentially according to the time sequence together form the time-division multiplexed anti-resonant electromagnetic pulse group. This pulse group can excite the corresponding units of the variable capacitor or inductor array one by one according to the preset time sequence, realize the independent and orderly adjustment of multiple physical quantity dimensions, and ensure the accuracy of resonant cavity parameter correction.

[0046] In a preferred embodiment of the present invention, step 5 is further included, in which the variable capacitor or inductor array is deformed by an anti-resonant electromagnetic pulse to change the capacitance and shift the position of the subsequent shift peak. The specific operation is as follows: Based on the anti-resonance electromagnetic pulse generated in the previous steps, closed-loop adjustment of the electrical parameters of the resonant branch is achieved. The electromagnetic pulse drives the variable capacitor or inductor array to generate controllable structural deformation, directly changing the equivalent capacitance or inductance of the array, thereby changing the overall resonant characteristics of the resonant branch. This causes the position of the offset peak generated under the next frequency-modulated continuous wave excitation to shift directionally. This adjustment process is based on the radial vector deviation. The spectral position of the offset peak is corrected by adjusting the electrical parameters at the physical level, so that the dynamic data carried by the equipment encoding gradually conforms to the ideal operating state. This continuously ensures the accurate mapping relationship between the digital twin model and the physical new energy equipment, and avoids model mapping failure due to data deviation.

[0047] Step 5 also includes the following steps: Step 51: Apply each pulse in the anti-resonant electromagnetic pulse group to the corresponding variable capacitor or inductor array in chronological order. Utilize the exponentially decaying tail of the pulses to excite forced vibration, generating a transient displacement that is proportional to the pulse amplitude and whose phase determines the vibration direction. The specific operation is as follows: Following the time-division multiplexing sequence of the anti-resonant electromagnetic pulse group, each pulse is sequentially applied to the variable capacitor or inductor array unit corresponding to different physical quantity dimensions, ensuring that each pulse precisely acts on the preset adjustment object. After the anti-resonant electromagnetic pulse is applied, the continuous electromagnetic energy carried by its exponentially decaying tail interacts with the conductive structure of the array, exciting the elastic support component of the array to generate forced vibration. The transient displacement amplitude generated during the vibration is proportional to the pulse amplitude; the larger the pulse amplitude, the greater the corresponding transient displacement. At the same time, the phase characteristics of the pulse directly determine the direction of motion of the forced vibration. Different phases correspond to opposite displacement directions. Through the dual controllable output of displacement magnitude and direction, a precise driving basis is provided for the displacement conversion of the subsequent mechanical structure.

[0048] Step 52: The transient displacement is converted into a unidirectional stepping displacement through a mechanical ratchet, causing an irreversible stepping change in the equilibrium position that is proportional to the pulse amplitude. Only pulses in the same phase generate the stepping. The specific operation is as follows: The transient displacement generated by forced vibration is transmitted to a mechanical ratchet mechanism integrated into the array structure. Utilizing the unidirectional transmission characteristic of the mechanical ratchet, the reciprocating transient displacement is converted into a stepping displacement along a fixed direction. The change in equilibrium position caused by the stepping displacement is proportional to the amplitude of the transient displacement; the larger the pulse amplitude, the greater the shift in equilibrium position caused by a single step. The mechanical ratchet only effectively transmits transient displacements caused by in-phase pulses, forming an irreversible stepping change. Transient displacements caused by out-of-phase pulses are canceled out by the ratchet's engagement structure, preventing effective displacement transmission. This avoids adjustment backsliding caused by reverse displacement and ensures that the position change of the array structure always advances along the correction direction.

[0049] Step 53: Lock the equilibrium position after stepping using electrostatic latch until the next in-phase pulse arrives, then temporarily unlock it. The specific operation is as follows: After the mechanical ratchet completes a unidirectional stepping displacement, the electrostatic latch structure fixes the array's equilibrium position after the stepping. The electrostatic latch relies on the adsorption force generated by the electrostatic field to constrain the array's plates or inductive support structure, keeping the current position stable and preventing displacement due to equipment vibration or internal residual stress. This locked state is maintained until the next in-phase anti-resonant electromagnetic pulse arrives at the resonant cavity. At this time, the electrostatic latch receives a pulse trigger signal, temporarily releasing the constraint and allowing the array to perform another stepping displacement. After the current pulse-driven stepping action is completed, the electrostatic latch resumes its working state, continuously locking the new equilibrium position, achieving stable maintenance of the array's electrical parameters after adjustment, and providing a fixed physical reference for the next spectrum detection and deviation correction.

[0050] In a preferred embodiment of the present invention, step 6 is further included: periodically repeating steps 2 to 5, so that each frequency sweep obtains an updated composite spectrum, and generating the next anti-resonant electromagnetic pulse based on the radial vector deviation, wherein step 6 further includes: Step 61: Based on the magnitude difference between the current radial vector and the previous radial vector, dynamically adjust the waiting time interval for the next frequency sweep. The magnitude and the interval are inversely correlated. When the magnitude is less than the dead zone threshold, the interval is extended to the upper limit. The specific operation is as follows: The current radial vector reflects the actual operating state of the dynamic physical quantities of the device in various dimensions, while the previous radial vector reflects the operating state of the previous cycle. The magnitude of the difference between the two is used to quantify the change in the device's operating state between two frequency sweeps. The larger the magnitude of the difference, the more drastic the change in the device's dynamic physical quantities and the faster the deviation from the stable operating state. In order to capture such rapid changes in a timely manner and avoid the deviation from continuing to expand and affecting the accurate mapping of the digital twin model, it is necessary to shorten the waiting time interval for the next frequency sweep. By conducting more frequent frequency sweeps, we can quickly obtain information on state changes and make timely adjustments.

[0051] Conversely, when the difference magnitude is less than the preset dead zone threshold, it indicates that the equipment's operating state is stabilizing and the state change between two frequency sweeps is minimal. In this case, extending the waiting time interval to the preset upper limit reduces unnecessary frequency sweeps, lowers system energy consumption and equipment mechanical and electromagnetic losses, and ensures that critical state changes are not missed due to excessively long intervals. The dead zone threshold, the upper limit of the waiting time interval, and the lower limit must all be determined through experimental calibration based on the type of new energy equipment, operating conditions, and the allowable fluctuation range of dynamic physical quantities. This ensures that the interval adjustment logic is compatible with the actual operating requirements of the equipment and can be directly calibrated and reused according to specific equipment parameters. The calculation of the difference magnitude follows the conventional calculation logic of multidimensional vector magnitudes. It is obtained by taking the square root of the sum of the squares of the differences between the corresponding components of the two radial vectors. No additional complex calculation models are needed; accurate quantification can be achieved through conventional mathematical operations alone. Step 62: At the end of the waiting interval, based on the comparison between the residual deviation magnitude of the current radial vector and the ideal radial vector and the compensation threshold, decide whether to skip steps 4 and 5, update the comparison benchmark to the current radial vector, and halve the compensation threshold when the number of consecutive skips reaches a predetermined upper limit. The specific operation is as follows: After the waiting time interval set in step 61 ends, the residual deviation magnitude between the current radial vector and the ideal radial vector is first calculated. This residual deviation magnitude is the quantified value of the deviation between the equipment's operating state and the ideal state after adjustments in steps 4 and 5, directly reflecting the effect of the previous adjustments. The residual deviation magnitude is compared with a preset compensation threshold. If the residual deviation magnitude is less than or equal to the compensation threshold, it indicates that the current equipment operating state is close to the ideal state, and the deviation is within an acceptable range. Therefore, steps 4 and 5 (anti-resonant electromagnetic pulse generation and array adjustment) do not need to be performed; these two steps can be skipped to reduce energy consumption and equipment damage caused by ineffective adjustments. Simultaneously, the current radial vector is updated as a new comparison reference. The deviation calculation after the next frequency sweep will use this updated reference as a reference, avoiding over-adjustment caused by small deviations between the ideal radial vector and the actual stable state. If the residual deviation magnitude is greater than the compensation threshold, steps 4 and 5 are executed normally to continue adjusting the variable capacitor or inductor array until the deviation drops to an acceptable range.

[0052] In addition, the number of times steps 4 and 5 are skipped is counted. When the number of consecutive skips reaches the preset upper limit, it indicates that the current compensation threshold may be too high, causing some minor deviations to not be adjusted in time. At this time, the compensation threshold is halved to improve the adjustment sensitivity and ensure that even if there are small residual deviations, they can be detected in time and adjusted in a targeted manner to avoid the accumulation of deviations affecting the mapping accuracy of the digital twin model. The initial value of the compensation threshold and the upper limit of the number of consecutive skips are determined through experimental calibration and can be adjusted according to the actual accuracy requirements of the equipment operation.

[0053] Example 2: Please see Figure 2 Based on Example 1, this example provides a digital twin construction system for new energy equipment based on device coding, including: The resonant configuration module constructs a resonant cavity, which includes a static reference unit and a dynamic modulation unit. The static reference unit generates a reference peak, and the dynamic modulation unit couples dynamic physical quantities to a variable capacitor or inductor array to generate an offset peak. The frequency modulation conversion module excites the resonant cavity with a frequency-modulated continuous wave, receives the echo and performs frequency domain transformation to obtain a composite spectrum containing the reference peak and the offset peak. The vector calibration module uses the reference peak in the composite spectrum as the origin, takes the frequency difference between the offset peak and the reference peak as the radial vector, and presets an ideal radial vector. The anti-pulse generation module generates an anti-resonant electromagnetic pulse based on the deviation between the radial vector and the preset ideal radial vector. The frequency of the anti-resonant electromagnetic pulse is aligned with the response bandwidth of the variable capacitor or inductor array. The deformation peak shifting module uses anti-resonant electromagnetic pulses to deform a variable capacitor or inductor array to change the capacitance and shift the position of the subsequent shift peak. The periodic loop module periodically calls the frequency modulation conversion module, vector calibration module, anti-pulse generation module, and deformation peak shifting module to obtain an updated composite spectrum diagram for each frequency sweep, and generates the next anti-resonant electromagnetic pulse based on the radial vector deviation.

[0054] The above description is merely a preferred embodiment of the present invention; however, the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and its improved concepts, should be covered within the scope of protection of the present invention.

Claims

1. A method for constructing a digital twin of a new energy device based on device coding, characterized in that, include: Step 1: Construct a resonant cavity, which includes a static reference unit and a dynamic modulation unit. The static reference unit generates a reference peak, and the dynamic modulation unit couples dynamic physical quantities to a variable capacitor or inductor array to generate an offset peak. Step 2: Excite the resonant cavity with a frequency-modulated continuous wave, receive the echo and perform frequency domain transformation to obtain a composite spectrum containing the reference peak and the offset peak; Step 3: Using the reference peak in the composite spectrum as the origin, the frequency difference between the offset peak and the reference peak is taken as the radial vector, and an ideal radial vector is preset. Step 4: Based on the deviation between the radial vector and the preset ideal radial vector, generate an anti-resonant electromagnetic pulse. The frequency of this anti-resonant electromagnetic pulse is aligned with the response bandwidth of the variable capacitor or inductor array. Step 5: Use anti-resonant electromagnetic pulses to deform the variable capacitor or inductor array to change the capacitance and shift the position of the subsequent shift peak. Step 6: Repeat steps 2 to 5 periodically to obtain an updated composite spectrum diagram with each frequency sweep, and generate the next anti-resonant electromagnetic pulse based on the radial vector deviation.

2. The new energy equipment digital twin construction method based on device coding according to claim 1, characterized in that, Step 1 includes: Construct a set of reflection structures on the resonant cavity substrate to generate discrete reference frequency peaks; On the same substrate and at a location that does not overlap with the space of the reflective structure, a resonant branch consisting of a variable capacitor or an inductor array is constructed, which generates a movable offset peak.

3. The new energy equipment digital twin construction method based on equipment coding according to claim 2, characterized in that, Step 1 further includes: A nonlinear pre-distortion coupling path is established between dynamic physical quantities and the variable capacitor gap. This path makes the input displacement and the change in capacitor gap present a predetermined nonlinear functional relationship. The reflection structure and the resonant branch are encapsulated in the same cavity, and an electromagnetic absorption layer is set in the cavity to cover the transition frequency band between the reference peak and the offset peak.

4. The new energy equipment digital twin construction method based on equipment coding according to claim 3, characterized in that, Step 2 includes: Generate a continuous wave excitation signal whose instantaneous frequency changes logarithmically with time; Perform a fractional Fourier transform on the echo, with the transform order being the reciprocal of the base of the logarithmic function, to obtain a composite spectrum containing the reference peak and the offset peak.

5. The new energy equipment digital twin construction method based on equipment coding according to claim 4, characterized in that, Step 3 includes: In the composite spectrum diagram, the reference peak is selected by Lorentz line matching, and the weighted center frequency of the reference peak is used as the identity origin. The difference between the offset peak frequency and the identity origin is mapped to the physical quantity dimension according to the neighborhood of the reference peak to which the offset peak belongs, and after normalization, it forms a multidimensional radial vector. The ideal radial vector is preset based on the median of the radial vector obtained by multiple frequency sweeps during the initial commissioning of the equipment, and is fixed in the immutable storage area of ​​the resonant cavity.

6. The device code-based new energy device digital twin construction method according to claim 1, characterized in that, Step 4 includes: The component deviation between the radial vector and the ideal radial vector is decomposed into amplitude deviation, sign deviation, and rate of change deviation relative to the previous frequency sweep, and encoded into amplitude code, phase code, and pulse width modulation code. Based on the amplitude code and pulse width modulation code, a baseband pulse train with an exponentially decaying envelope and a controlled pulse width is generated through nonlinear resistive-capacitive relaxation oscillation. The baseband pulse train is mixed with a carrier signal whose frequency is aligned with the response bandwidth of a variable capacitor or inductor array, and the phase of the mixing output is controlled according to the sign deviation.

7. The method for constructing a digital twin of new energy equipment based on equipment coding according to claim 6, characterized in that, Step 4 also includes: The RF pulse train output from the mixer is amplified by nonlinear saturation to flatten the pulse peak to a predetermined limiting level while retaining the exponential decay tail. The saturated amplified radio frequency pulse train is passed through a time delay line array to assign different arrival times to each physical quantity dimension, forming a time-division multiplexed anti-resonant electromagnetic pulse group.

8. The method for constructing a digital twin of new energy equipment based on equipment coding according to claim 7, characterized in that, Step 5 includes: Each pulse in the anti-resonant electromagnetic pulse group is applied sequentially to the corresponding variable capacitor or inductor array. The exponential decay tail of the pulse is used to excite forced vibration, generating transient displacement that is proportional to the pulse amplitude and whose phase determines the vibration direction. Transient displacement is converted into unidirectional stepping displacement through a mechanical ratchet, causing the equilibrium position to undergo an irreversible stepping change that is proportional to the pulse amplitude, and only in-phase pulses produce the stepping. The equilibrium position after the step is locked by electrostatic latching and temporarily unlocked when the next in-phase pulse arrives.

9. The method for constructing a digital twin of new energy equipment based on equipment coding according to claim 8, characterized in that, Step 6 includes: Based on the magnitude difference between the current radial vector and the previous radial vector, the waiting time interval for the next frequency sweep is dynamically adjusted. The magnitude and the interval are inversely correlated. When the magnitude is less than the dead zone threshold, the interval is extended to the upper limit. At the end of the waiting interval, based on the comparison between the residual deviation magnitude of the current radial vector and the ideal radial vector and the compensation threshold, it is decided whether to skip steps 4 and 5, update the comparison benchmark to the current radial vector, and halve the compensation threshold when the number of consecutive skips reaches a predetermined upper limit.

10. A digital twin construction system for new energy equipment based on equipment coding, applied to the digital twin construction method for new energy equipment based on equipment coding as described in any one of claims 1-9, characterized in that, include: The resonant configuration module constructs a resonant cavity, which includes a static reference unit and a dynamic modulation unit. The static reference unit generates a reference peak, and the dynamic modulation unit couples dynamic physical quantities to a variable capacitor or inductor array to generate an offset peak. The frequency modulation conversion module excites the resonant cavity with a frequency-modulated continuous wave, receives the echo and performs frequency domain transformation to obtain a composite spectrum containing the reference peak and the offset peak. The vector calibration module uses the reference peak in the composite spectrum as the origin, takes the frequency difference between the offset peak and the reference peak as the radial vector, and presets an ideal radial vector. The anti-pulse generation module generates an anti-resonant electromagnetic pulse based on the deviation between the radial vector and the preset ideal radial vector. The frequency of the anti-resonant electromagnetic pulse is aligned with the response bandwidth of the variable capacitor or inductor array. The deformation peak shifting module uses anti-resonant electromagnetic pulses to deform a variable capacitor or inductor array to change the capacitance and shift the position of the subsequent shift peak. The periodic loop module periodically calls the frequency modulation conversion module, vector calibration module, anti-pulse generation module, and deformation peak shifting module to obtain an updated composite spectrum diagram for each frequency sweep, and generates the next anti-resonant electromagnetic pulse based on the radial vector deviation.

Images