A Smart IoT-based Transparent Inspection Management Method for Power Grid Materials

Abstract

This invention discloses a comprehensive management method for transparent inspection of power grid materials based on the Internet of Things (IoT), belonging to the field of power material management technology. The method includes the following steps: deploying a three-axis vector probe array at the material entry channel to collect multi-point spatiotemporal signals and calculate the field strength distribution, generating an electromagnetic disturbance spatiotemporal matrix, and outputting a radio frequency identification (RFID) tag resonance drift risk matrix. This invention models electromagnetic disturbance sources using a three-axis vector probe array, combines resonance registration fingerprinting and frequency band compensation to achieve dynamic correction of RFID frequency drift, and integrates phase-encoded backscattering and strain acoustic features to ensure identification consistency. By utilizing optical time reference synchronization and phase-programmable metasurfaces to reconstruct the near-field environment, an adaptively adjustable identification space is constructed, improving the identification stability and traceability reliability of power grid materials under high-disturbance environments, thus breaking through the bottlenecks of traditional identification technologies.

Description

Technical Field

[0001] This invention relates to the field of power material management technology, specifically to a method for transparent detection and full-process management of power grid materials based on the Internet of Things. Background Technology

[0002] "Transparent Inspection and Full-Process Management of Power Grid Materials Based on Smart IoT" refers to the real-time collection and uploading of multi-dimensional information such as material status, location, and environmental parameters through the integration of Internet of Things (IoT) sensing devices and smart terminals (such as RFID tags, QR codes, environmental sensors, and positioning chips) throughout the entire lifecycle of power grid materials, including procurement, warehousing, outbound, transportation, storage, use, and decommissioning. This is achieved through edge computing, cloud platforms, and big data analytics to construct a visualized, traceable, and intervention-enabled management system for the entire material process. "Transparent inspection" emphasizes automatically identifying the quantity, type, quality status, and transportation environment of materials through sensing data without requiring manual unsealing or inspection, ensuring intelligent, efficient, and real-time inspection. "Full-process management" refers to the digital recording and closed-loop tracking of all key nodes, from pre-receipt planning and allocation to post-use status assessment and decommissioning, thereby improving power grid operation and maintenance efficiency, ensuring material quality and safety, reducing the risk of human intervention, and achieving a true upgrade in intelligent material management.

[0003] The existing technology has the following shortcomings:

[0004] In existing technologies, RFID tags are commonly used as material identification carriers in the transparent inspection and management of power grid materials, in conjunction with wireless reading and writing devices to complete information collection and binding operations. However, in environments with high electromagnetic disturbances, such as substation construction areas, high-voltage transmission corridors, or scenarios involving the centralized delivery of large power equipment, there is significant interference from strong electric fields caused by arc discharge, frequent start-stop cycles, and cable leakage inductance. This interference can easily cause the RFID tag antenna structure to drift due to local electric field distortion. This drift will cause the tag's transmission frequency to deviate from the reader's preset receiving frequency, resulting in a mismatch between the system's reading frequency and the tag's actual transmission frequency. This can lead to abnormal states such as key materials not being identified, being incorrectly identified, or being repeatedly identified. Especially during the centralized warehousing phase of core equipment such as high-voltage circuit breakers and transformers, such identification anomalies will directly cause problems such as mismatch between accounts and goods, incorrect equipment identification binding, and chaotic entry and exit records. This can further lead to serious consequences such as a broken material traceability chain and failure of equipment lifecycle management. It not only undermines the integrity of the transparent inspection mechanism but also significantly increases safety hazards during power grid operation and maintenance.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a method for transparent detection and full-process management of power grid materials based on the Internet of Things, so as to solve the problems in the background art mentioned above.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for full-process management of transparent inspection of power grid materials based on the Internet of Things, comprising the following steps:

[0008] A three-axis vector probe array is deployed in the material entry channel to collect multi-point spatiotemporal signals and calculate the field strength distribution, generate an electromagnetic disturbance spatiotemporal matrix, and output the radio frequency identification tag resonance drift risk matrix.

[0009] Under the constraint of the resonance drift risk matrix of RFID tags, the resonance registration fingerprint of RFID tags and reading and writing devices is obtained. The complex reflection coefficient curve is established by frequency sweep excitation, the peak phase point is extracted and a set of resonance correction windows is formed.

[0010] Guided by the resonant correction window set, drift compensation matching parameters are obtained, and the variable capacitor matching plate is driven to adjust the load of the RFID tag antenna, outputting the frequency band pull amount and allowable error band.

[0011] Under the constraints of frequency band traction and allowable error band, phase-coded backscattering measurements are performed on the same material and the acoustic characteristics of the packaging surface strain are collected to construct a dual-verified identity criterion;

[0012] With the support of the identity criterion of dual verification, a time alignment baseline is obtained, and the optical time reference is called to realize the clock synchronization of edge nodes, and the identification stream record is mapped to a single time axis.

[0013] Guided by the time-aligned baseline, an adaptive time-space control solution is obtained. A phase-programmable metasurface array is deployed in the storage channel to reconstruct the near-field boundary in real time. Spectrum foldback traction is implemented and combined with field load inversion to form a closed-loop control mechanism that dynamically suppresses electromagnetic disturbance drift, thereby achieving stable identification of RFID tags.

[0014] Preferably, the steps for obtaining the spatiotemporal matrix of electromagnetic disturbance and outputting the resonant drift risk matrix of RFID tags are as follows:

[0015] Multiple triaxial vector probe groups are evenly deployed along the axial direction of the transportation path in the material receiving channel. Each probe group consists of three orthogonally placed electric field sensing units.

[0016] The voltage signals output by all probe sensing units are synchronously sampled using a unified clock trigger by the acquisition device to obtain an electric field intensity data volume with spatial positioning capability.

[0017] The triaxial electric field data is processed by the spatial vector synthesis method to form an electromagnetic disturbance spectrum sequence.

[0018] Based on the spectrum sequence, a perturbation spatiotemporal matrix is ​​constructed, and the resonance drift risk matrix is ​​calculated by combining the tag antenna electromagnetic compatibility mapping table, so as to realize the spatial positioning and risk prediction of tag frequency offset.

[0019] Preferably, the steps for obtaining the resonant registration fingerprints of the RFID tag and the reader / writer and forming a set of resonant correction windows are as follows:

[0020] High-risk areas are identified in the radio frequency identification tag resonance drift risk matrix, and a tag-read / write environment simulation scenario is constructed.

[0021] The tag is excited by a frequency sweep signal and the reflected signal is collected to obtain the tag's reflection response data under electromagnetic disturbance environment;

[0022] The reflection response data is analyzed to extract the main resonant frequency, phase change frequency, and half-power bandwidth, forming a resonant registration fingerprint.

[0023] Based on the resonance registration fingerprint, a frequency offset tolerance range is set and a resonance correction window set is constructed to provide a basis for frequency adjustment.

[0024] Preferably, the steps for obtaining drift compensation matching parameters and adjusting the RFID tag antenna load to output band pull and allowable error band are as follows:

[0025] Extract frequency response characteristics based on the resonant correction window set and output a set of frequency band traction parameters;

[0026] The capacitance of the variable capacitor matching board is adjusted step by step according to the frequency band traction parameter set, and a reactance load balance combination is constructed by matching it with a standard inductor.

[0027] Repeated identification tests and frequency drift data were collected in an electromagnetic disturbance simulation environment, and the final output frequency band pull and allowable error band were determined.

[0028] Preferably, the steps for performing phase-encoded backscattering measurements and collecting the acoustic characteristics of the packaging surface strain on the same material to construct a dual-verification identity criterion are as follows:

[0029] The excitation frequency is set based on the frequency band traction amount, and phase-encoded backscattering measurement is performed to obtain the tag phase fingerprint;

[0030] Collect strain acoustic characteristics of the material packaging surface to form a set of structural response characteristic parameters;

[0031] A dual identity criterion is constructed by comparing the phase fingerprint with the acoustic features;

[0032] The verification results are written to the tag user data area as encrypted status markers for subsequent identification process verification.

[0033] Preferably, the steps for obtaining the time alignment baseline, using the optical time reference to synchronize the edge node clocks, and mapping the identification stream records to a single time axis are as follows:

[0034] The identification node is synchronized locally by injecting an optical time reference signal into the identification node through a fiber optic distributed time synchronization device.

[0035] The time correction factor of each identification node is invoked to perform a unified standard time mapping on the local timestamps generated by each identification record;

[0036] Sort all identification records by the corrected time value and construct a standard timeline identification stream; finally, output a structured record table containing standard timestamps and identification status for full-process identification traceability and strategy scheduling.

[0037] Preferably, the following steps are taken to obtain the adaptive time-space air conditioning control solution, deploy a phase-programmable metasurface array in the inlet channel, reconstruct the near-field boundary in real time, implement spectral foldback traction and combine it with field load inversion to form a closed-loop control mechanism that dynamically suppresses electromagnetic disturbance drift:

[0038] Extract the time periods and spatial locations of failed identifications to construct the target control area;

[0039] A phase-programmable metasurface array is deployed within the target control region, and a reflected phase control signal is applied to reconstruct the perturbation boundary.

[0040] Collect spectral offset and calculate spectral traction parameters to perform spectral return traction;

[0041] The system sets up an electric field detector to collect feedback data and performs field load inversion to dynamically correct array phase parameters. Finally, a closed-loop control cycle is constructed to achieve stable identification of RFID tags.

[0042] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0043] This invention achieves multi-point spatiotemporal modeling of disturbance sources by introducing a three-axis vector probe array. Combined with resonant registration fingerprinting and frequency band drift compensation, it dynamically corrects the frequency drift problem in tag identification. Furthermore, it ensures consistency between physical objects and tag identification results through joint modeling of phase-encoded backscattering and strain acoustic features. Further, it achieves real-time intervention and control of the disturbance field through optical time-reference-driven global temporal alignment and a near-field electromagnetic environment reconstruction mechanism constructed by a phase-programmable metasurface array. This forms a sustainable and adaptive identification environment, comprehensively ensuring the identifiable, verifiable, and traceable management capabilities of core power grid materials during the warehousing stage. This effectively reduces the risk of identification failure, avoids mismatch between records and goods and identity anomalies, and improves transparent detection efficiency and material flow security. Technically, this method does not rely on traditional static shielding or single-point redundant identification methods. Instead, it overcomes the technical bottleneck of easy failure of RFID systems in high-disturbance scenarios through multi-dimensional information fusion and spatiotemporal feedback coordination. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0045] Figure 1 This is a flowchart of the whole-process management method for transparent inspection of power grid materials based on smart IoT of the present invention. Detailed Implementation

[0046] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, they are provided so that the description of this disclosure will be more complete and fully convey the concept of the exemplary embodiments to those skilled in the art.

[0047] This invention provides, for example Figure 1 The illustrated method for transparent inspection and management of power grid materials based on the Internet of Things includes the following steps:

[0048] A three-axis vector probe array is deployed in the material entry channel to collect multi-point spatiotemporal signals and calculate the field strength distribution, generate an electromagnetic disturbance spatiotemporal matrix, and output the radio frequency identification tag resonance drift risk matrix.

[0049] To address the issue of identification failure caused by the resonant frequency drift of RFID tags in environments with high electromagnetic disturbance, it is first necessary to accurately model the electromagnetic environment of the inbound channel and output disturbance risk prediction results. The specific steps are as follows:

[0050] Multiple triaxial vector probe groups are evenly deployed along the axial direction of the transport path within the material receiving channel. Each probe group consists of three orthogonally placed electric field sensing units, used to collect electric field change signals in the horizontal (X-axis), vertical (Y-axis), and transport direction (Z-axis) of the channel, respectively. Each electric field sensing unit adopts an electrostatic induction structure, with an embedded high-impedance electrode plate and low-drift operational amplifier circuit, capable of detecting electric field intensity changes in the range of 10mV / m to 20V / m. To ensure sampling accuracy, each probe group is mounted on a high-dielectric-constant insulating support, with a graphite-coated shielding plate at the bottom to prevent stray charge coupling between the probe and the ground. The spacing between probe groups is set to 0.5 meters, covering the entire length of the channel, with the density increased to 0.25 meters at channel intersections, shelf dead corners, and near metal walls. All probe outputs are connected to a 16-channel high-synchronization acquisition device via a voltage isolation converter, with the sampling rate per channel set to 5MS / s, and synchronous sampling triggered by a unified clock signal. Throughout the probe deployment process, the goal is to maintain an equal distribution of sampling in the three-dimensional electric field space, ensuring that all spatial points have complete vector signal coverage and forming a continuous and dense sampling network. Compared to the traditional method of using only fixed-point unidirectional probes, this deployment method exhibits higher sensitivity and accuracy when dealing with abrupt changes in electric field vector direction or drift of strong interference sources.

[0051] After the probe array is deployed and connected to the acquisition device, data acquisition initialization is performed. The specific process includes: configuring the sampling window width to 5000 sampling points per frame and the time resolution to 200 nanoseconds; setting the sampling period to continuous acquisition for 60 seconds, covering the entire process of normal material warehousing. During acquisition, the voltage signal output by each probe sensing unit is pre-filtered with a cutoff frequency set between 1MHz and 2GHz to remove baseline drift caused by low-frequency mechanical vibration and high-frequency environmental noise interference. The filtered signal is then converted into a digital electric field strength value through high-precision analog-to-digital conversion and enters the data processing stage. For the spatial location of different probes, the system assigns absolute spatial coordinates to each sampling point and records the timestamp and spatial location binding relationship in each frame of data, thus forming an electric field strength data volume with spatial positioning capabilities. Based on this data volume, a spatial vector synthesis method is used to process each triaxial data point, calculating the synthesized electric field strength and principal direction angle at each point to obtain the electric field spatial distribution vector map of a certain moment within the channel. Each image can be considered a snapshot of the electromagnetic disturbance of the channel at that instant, and multiple consecutive images constitute a sequence of electric field evolution trajectories. Unlike existing methods, this invention does not perform time window averaging during data processing to avoid diluting the disturbance peaks. At the same time, an inter-axis interference compensation term is introduced when solving the main electric field direction to correct the error caused by the probe offset angle, thereby obtaining electric field characteristic data that is closer to the real scene.

[0052] Based on the obtained electric field vector map sequence, an electromagnetic disturbance spatiotemporal matrix was constructed in temporal and spatial order. This matrix uses the sampling time frame as the column index and the probe location number as the row index, with each matrix element representing the composite electric field intensity value collected by that probe in the corresponding time frame. To enhance spatial resolution, all matrix elements are accompanied by the electric field direction angle vector of their corresponding location. By performing time series analysis on the data in the matrix, abnormal patterns of drastic fluctuations in electric field intensity within a short period can be identified in certain regions. Furthermore, correlation analysis is performed between the time series corresponding to each probe and its spatial neighboring probes to extract the disturbance propagation direction, velocity, and diffusion range, constructing a disturbance propagation path model. Based on this, a disturbance intensity differential statistical method is introduced, defining signal sequences with a change amplitude exceeding 3V / m and lasting longer than 10ms as high-disturbance events. The occurrence time, location, and duration are recorded, and the distribution density of these events in the channel is statistically analyzed to form a disturbance density map. This map, combined with the main electric field direction vector field, generates a complete electromagnetic disturbance flow trajectory map, providing a basis for identifying potentially high-risk areas. Unlike traditional methods that only focus on the instantaneous peak electric field intensity, this matrix introduces parameters for disturbance persistence and propagation, thereby improving the comprehensive identification capability of resonant drift induction factors.

[0053] Based on the spatiotemporal matrix and perturbation density map of electromagnetic disturbances, this study further analyzes their impact on the resonant performance of RFID tag antennas. First, resonant frequency shift data for each type of tag antenna at various electric field strength levels are acquired. Through prior electromagnetic compatibility testing in the laboratory, a mapping table between electric field strength and tag resonant frequency is established. Then, each electric field strength value from the perturbation matrix is ​​substituted into the mapping table to predict the possible resonant frequency shift value of the corresponding tag. The prediction results are summarized according to spatial location and time order to form an RFID tag resonant drift risk matrix. Each element in the matrix represents the possible resonant frequency shift of the tag at a certain time and location, in MHz, along with the shift direction and maximum estimated error range. Based on practical engineering experience, a tolerance range for the identification receiving frequency band is set; any predicted shift exceeding this range is marked as a high-risk point. Finally, this risk matrix can serve as input for subsequent identification frequency band adjustment, compensation device activation, and environmental reconstruction operations, enabling accurate positioning and visual early warning of electromagnetic interference risks before materials are stored, significantly improving identification accuracy and the stability of the transparent detection system.

[0054] Under the constraint of the resonance drift risk matrix of RFID tags, the resonance registration fingerprint of RFID tags and reading and writing devices is obtained. The complex reflection coefficient curve is established by frequency sweep excitation, the peak phase point is extracted and a set of resonance correction windows is formed.

[0055] To ensure that RFID tags can still be accurately identified by reading and writing devices under disturbed environments, it is necessary to establish their resonant frequency registration characteristics under the current electromagnetic conditions, and based on this, construct a set of correction windows that can be used for subsequent frequency compensation. The specific steps are as follows:

[0056] In the generated RFID tag resonance drift risk matrix, spatial locations and corresponding time periods belonging to high drift levels were identified. Using these high-risk identification areas as the target, a tag-reading / writing environment simulation experimental scenario was constructed at the corresponding locations. This scenario involved placing a set of RFID tag samples and using an electromagnetic background arrangement consistent with the warehousing channel, including similar metal shelves, insulation layers, cable routing structures, and typical packaging materials. Each tag was installed on the surface of the actual packaging, maintaining the same pasting method, antenna orientation, and position layout as the actual warehousing process, ensuring that the experimental data fully reflects the identification status in the actual disturbance environment. A set of sweep frequency signal transmitting antennas was fixedly installed in a straight line with the tag position and 20 cm from the front of the tag to excite the tag operation. This transmitting antenna was directly connected to the sweep frequency signal source, with the signal source output frequency range set to 860MHz to 960MHz, a step interval of 100kHz, and a constant output power of +5dBm to cover the commonly used operating frequency band of RFID tags. During tag excitation, no multi-metallic surface reflectors were allowed around the tag area to prevent additional signal echoes from interfering with the actual measurement results.

[0057] During the excitation signal frequency sweep, reflected signals are synchronously acquired via a receiving antenna located near the front of the tag. This receiving antenna is connected to a vector network analysis device to acquire the complex reflection coefficient data of the tag across the entire frequency sweep range in real time. Each sampling point of the data contains two elements: the amplitude value of the reflected signal at the corresponding frequency and the phase angle change value at the corresponding frequency. This invention is particularly concerned with phase abrupt change points and rapid decrease points in the reflection coefficient, both of which often occur together when the tag antenna enters a resonant state. To improve data stability, each complete frequency sweep process is repeated three times, with each session lasting less than 5 seconds. The three sets of data are then averaged to remove accidental fluctuations caused by environmental interference. Simultaneously, a blank carrier sample (without the tag) is placed under the same environmental conditions as the tag and subjected to frequency sweep processing in the same frequency band to obtain reference reflection data. The tag-containing measurement data and the blank sample data are subtracted one-to-one to obtain the tag's unique reflection response data, effectively eliminating background clutter interference. This reflection response data contains the tag's actual resonant behavior in this environment and is the basic data for identifying the degree of tag frequency offset.

[0058] A thorough analysis of the acquired reflection response data is conducted to extract the frequency response characteristics of the tag under disturbance conditions, forming a resonance registration fingerprint. The specific analysis process includes the following three steps: First, extreme value identification is performed on the reflection amplitude curve to extract the frequency value of the minimum reflection point, which is taken as the main resonant frequency under the current environment. Second, differential analysis is performed on the phase angle change curve to find the center frequency of the steep phase change region. This frequency is cross-validated with the main resonant frequency; if the deviation is within 0.5MHz, the data is considered reliable. Third, the positions of the half-power points on both sides of the resonant frequency are calculated to obtain the bandwidth range. The phase change amplitude, reflection curve slope, and trough symmetry indices within this bandwidth are extracted. Finally, these are combined into a unique dataset, which is the resonance registration fingerprint of the tag under the disturbance conditions. The fingerprint data structure includes not only the absolute value of the main resonant frequency but also all frequency characteristic parameters within the resonant characteristic segment, ensuring its usability for subsequent multi-dimensional frequency band matching judgment.

[0059] Based on the tag's resonant registration fingerprint, a set of resonant correction windows for subsequent identification and adjustment is constructed. First, for the main resonant frequency in the registration fingerprint, a certain tolerable frequency offset range is extended to its left and right. This range is determined based on the actual receiving sensitivity of the reader / writer device; in this embodiment, it is set to ±3MHz. This frequency band is the first correction window where the tag can be stably identified under the current environment. Second, the registration fingerprints of the tag in different postures are constructed in parallel, forming multiple window sets corresponding to different directions. A union operation is performed on these sets to obtain all possible identification windows, forming a complete frequency response interval map. Then, an overlap analysis is performed between this response interval map and the receiving bandwidth of the reader / writer device to evaluate the degree of overlap between each window and the receiving bandwidth, and the minimum overlap frequency band, maximum overlap frequency band, and average overlap frequency band width are recorded. If the identification overlap ratio is less than 60%, it is determined that there is a significant risk of frequency mismatch in the current posture, requiring subsequent fine-tuning compensation. Finally, all the constructed correction window sets not only provide boundary constraints for reader frequency tuning but also provide a clear target frequency range for subsequently driving the adjustable matching board, possessing both engineering operability and real-time controllability.

[0060] Guided by the resonant correction window set, drift compensation matching parameters are obtained, and the variable capacitor matching plate is driven to adjust the load of the RFID tag antenna, outputting the frequency band pull amount and allowable error band.

[0061] To ensure that RFID tags always operate within the identifiable frequency band in environments with high electromagnetic disturbances, it is necessary to obtain frequency band pulling parameters that can be dynamically adjusted based on the previously established resonant correction window set, and to precisely adjust the tag antenna load to output the final frequency band pulling amount and allowable error band. The specific steps are as follows:

[0062] Based on the established resonance correction window set, the target identification frequency band range is selected, and the tag's current actual resonance frequency is compared. The current resonance frequency is obtained by using frequency sweep excitation and vector network analysis in the previous steps to clearly obtain the measured main resonance frequency of the tag under the current environment. This measured frequency value is compared with the center frequency of the target window. If there is a deviation of more than 1MHz between the two, it is determined to be a frequency band mismatch, and frequency band pulling adjustment is required. To determine the specific pulling parameters, the left and right bandwidth values, phase change rate, and asymmetry index of the main peak of the tag's current frequency response curve are further extracted from the resonance registration fingerprint. These parameters are used to determine the frequency response change trend of the tag under different capacitive loads, and based on this, the possible direction and magnitude of the resonant frequency shift when adjusting a specific capacitance level are predicted. For example, when the main resonance frequency is below the target frequency band, and the bandwidth on the right side of the curve is wider than that on the left side, and the phase angle changes more rapidly at the high frequency end, the adjustment direction is predicted to be pulling towards the high frequency, and the pulling magnitude is initially set to 110% of the measured deviation value to cope with the risk of re-drift caused by disturbances. The final output is a set of frequency band towing parameters for the current attitude of the tag, including direction (up or down), towing target value (in MHz), and maximum allowable error range (e.g., ±0.5MHz), which will serve as the input for subsequent adjustments.

[0063] Based on the frequency band pulling parameter set, a variable capacitor matching plate connected to the tag antenna feed point is finely adjusted step by step. The matching plate consists of several switchable capacitor arrays, each with a capacitance value of 10 picofarads, 20 picofarads, 30 picofarads, 40 picofarads, 50 picofarads, and 60 picofarads. The capacitor arrays are switched sequentially via an electronic relay controller, making the total output capacitance of the matching plate discretely adjustable. In specific operation, the initial capacitance value is first set to 30 picofarads. The capacitor group is activated under power-on conditions while keeping the tag stationary. Subsequently, a frequency sweep excitation test is performed again, and the updated complex reflection coefficient curve is recorded to calculate the position of the tag's main resonant frequency after adjustment. If the adjusted frequency offset direction is correct, i.e., the frequency point tends towards the target identification frequency band, the capacitance is adjusted in step intervals, gradually changing towards the predicted capacitance direction. The above frequency test steps are repeated after each adjustment. When the capacitance is adjusted to a certain value, if the tag's resonant frequency falls within ±0.3MHz of the target identification window center and deviates by no more than ±0.1MHz in three consecutive tests, the adjustment is stopped and the matching board capacitance value is locked. To prevent the tag's resonant frequency from shifting again due to environmental disturbances, this implementation also includes an inductor compensation step. After the capacitance is adjusted, a set of standard inductors with capacitances ranging from 0.5 microhenries to 5 microhenries are connected, in 0.5 microhenry increments. A suitable inductor value is selected to match the current capacitance to construct a reactance-load balanced combination, making the overall antenna impedance characteristics of the tag more closely conform to the 50-ohm standard, thereby improving its resonant frequency stability.

[0064] After fine-tuning the antenna load, it is necessary to verify whether the matching adjustment results meet the requirements for long-term identification stability, and output the final frequency band pull and allowable error band. First, after the adjustment is completed, a complete sweep frequency reflection curve is re-acquired to confirm that the current tag resonant frequency is indeed located in the center of the target identification window, and the net change value from the original resonant frequency to the current frequency is calculated. This value is the frequency band pull. This pull is precisely labeled with the adjustment amplitude in MHz and is used as the initial setting basis for subsequent batch matching adjustments. Next, electromagnetic disturbance simulation conditions are set, and disturbance field sources of different intensities and directions are applied to the area around the tag. In specific implementation, a 5V / m electric field disturbance signal is generated by high-voltage cable pulse start-up, and a DC load cycle switching device is configured to operate within 50 cm of the tag to simulate the dynamic interference environment in the warehousing operation. The tag identification response is continuously acquired 100 times, and the identification success rate and the instantaneous drift range of the main resonant frequency are recorded. If the tag's main frequency remains within ±0.5MHz throughout the entire identification process, and the identification success rate exceeds 98%, then this round of adjustment is considered to have reached an acceptable error tolerance. The final output for the tag's allowable error band is ±0.5MHz, and the frequency band pull is the frequency difference before and after adjustment. This output serves as the frequency stability basis for the next stage of cross-modal identification and comparison, and can be used to support parameters for subsequent tag dynamic calibration, identification network allocation, and identification window synchronization strategies, greatly improving the robust identification capability of transparent detection under on-site interference conditions.

[0065] Under the constraints of frequency band traction and allowable error band, phase-coded backscattering measurements are performed on the same material and the acoustic characteristics of the packaging surface strain are collected to construct a dual-verified identity criterion;

[0066] After completing frequency band adjustment and obtaining a stable frequency identification benchmark, in order to ensure that the physical goods bound to the RFID tags have not been replaced, transferred, or damaged, it is necessary to perform cross-physical dimension consistency verification on the goods, which specifically includes the following steps:

[0067] Based on the obtained frequency band pull and allowable error band, a suitable frequency excitation source is selected, and phase-coded backscatter measurements are performed on the frequency-adjusted RFID tags. The excitation signal is output in the form of continuous wave pulses, with each pulse length set to 20 microseconds and an interval of 10 microseconds between adjacent pulses. 200 pulses are continuously output, and the excitation frequency is precisely set to the center value of the tag's resonant frequency after adjustment, with a deviation controlled within ±0.1MHz. The transmitting antenna is installed 30 cm directly in front of the tag, using a directional antenna to reduce environmental reflection interference. The tag receives the backscattered signal, which is received by the corresponding receiving antenna, and its phase response trajectory is obtained through frequency domain analysis. To ensure the repeatability and anti-interference of the measurement results, each measurement is repeated three times. An independent backscattered phase response curve is obtained in each round, and the results from the three rounds are averaged and fitted to extract the stable peak phase sequence, the frequency distribution of the main response point, and the phase fluctuation envelope. This sequence constitutes the unique phase backscatter fingerprint of the tag in its adjusted state, used for subsequent consistency comparison.

[0068] After label backscattering measurements are completed, non-destructive acoustic feature acquisition is performed on the outer packaging structure of the product to construct an independent identification mark at the material structure level. This step utilizes a piezoelectric ceramic sheet acoustic probe, measuring 10 mm × 10 mm with a frequency response range of 5 kHz to 25 kHz, fixedly adhered to the packaging surface near the label. The acoustic excitation source is a pulsed loudspeaker with a fixed emission frequency of 12 kHz and a single pulse duration of 50 milliseconds. Each measurement cycle includes 30 repeated transmissions, with each transmission spaced 300 milliseconds apart. The signal output from the acoustic probe is preamplified and then fed into a 24-bit precision analog-to-digital converter. The sampling frequency is set to 100 kHz to ensure clear capture of sound pressure changes caused by minute strains in the surface structure. The resulting raw acoustic waveform is digitally bandpass filtered, and the first harmonic peak, concentrated spectral energy density region, envelope attenuation trend, and time reflection delay structure are extracted to form a multi-dimensional acoustic feature parameter set. These characteristics are influenced by factors such as the thickness and density of the packaging material, the pressing state, and the adhesion of the printed layer. They are highly irreversible and unreproducible, and can be used to accurately characterize the structural response characteristics of the current material entity.

[0069] Cross-validation analysis was performed on the obtained RFID tag phase backscatter fingerprint and the acoustic strain characteristics of the packaging surface to construct a dual identity criterion. First, in the phase fingerprint dimension, the historical phase sequence of the tag in the previous identification record was extracted and overlapped with the current measurement value to determine if the main peak phase position remained within a deviation range of 1 degree. Simultaneously, the area ratio of the overlapping region between the phase envelopes was calculated; if the ratio exceeded 90%, the tag was considered not physically replaced. Second, in the acoustic feature dimension, the current main peak frequency of the acoustic spectrum was compared with the initial characteristics of the material upon entry into the registered database; an error not exceeding 200 Hz was considered consistent. Simultaneously, the shape overlap between the spectral energy density distribution patterns was compared, and morphological difference indicators were extracted using a standard contour fitting algorithm; a difference less than 10% was considered structurally consistent. Only when both dimensions of the criteria met the consistency requirements was a "completely consistent" verification result output; if only one dimension met the requirements, a "low-confidence consistency" flag was output; if neither dimension met the requirements, an "identification anomaly" status was output. These results were used for subsequent binding of confidence management, serving as a hard verification threshold before the identification flow entered the traceability system.

[0070] The dual verification results are bound to the material's identification record as a trusted identity tag, ensuring that all identification actions during subsequent outbound, transportation, installation, and use are based on the verified identity. Specifically, three states—"Completely Consistent," "Low-Confidence Consistency," and "Identification Anomaly"—are appended to the user data area in the tag's memory using a status marker. This area is a 512-bit specific-length space, written using a special encryption format, and cannot be rewritten by ordinary read / write devices. During each identification, the read / write device first reads this status marker. Only if the identification state is "Completely Consistent" is the material allowed to proceed to the next step; if the state is "Low-Confidence Consistency," it enters the manual review stage; if it is "Identification Anomaly," the material is immediately isolated, and manual investigation is prompted. This mechanism effectively prevents identification information contamination caused by tag misidentification, impersonation, or replacement, ensuring that the identification identity remains consistent with the physical state throughout the entire material's lifecycle, constructing a robust and transparent detection mechanism based on dual physical characteristics.

[0071] With the support of the identity criterion of dual verification, a time alignment baseline is obtained, and the optical time reference is called to realize the clock synchronization of edge nodes, and the identification stream record is mapped to a single time axis.

[0072] To ensure consistent identification timing even when multiple identification nodes are operating simultaneously, an optical time reference is needed to synchronize the clocks of the edge identification nodes, and all identification records are projected onto a unified timeline. The specific implementation steps are as follows:

[0073] After completing the dual identity verification of the RFID tag and the goods, the identification timing information of the goods is extracted from each identification location as the basic data for unified timeline reconstruction. In each identification event, four types of time information are generated: the first is the trigger time when the excitation source emits a signal to the RFID tag, captured by the trigger circuit of the local read / write device; the second is the reception time when the backscattered signal returns from the RFID tag and is captured by the receiving antenna, determined by the identification threshold reached in the receiving circuit; the third is the acoustic excitation start time, i.e., the moment when the sound source first emits a sound wave to the packaging surface, initiated by a pulse signal sent by the sound source controller; and the fourth is the time when the peak of the strain acoustic characteristics appears, recorded by the piezoelectric ceramic receiver at the time point corresponding to the strongest strain signal. All four types of time are generated by the time reference provided by the local oscillation circuit within each identification node. Due to the different precision of the oscillation circuits between nodes, time errors caused by time drift or start-up delays exist, resulting in out-of-order, duplicate, and overwritten anomalies in cross-node identification records. To solve these problems, the local time of each identification node needs to be calibrated against the high-precision reference time as a prerequisite for establishing a unified timeline.

[0074] A high-precision optical time reference signal is continuously injected into each identification node through a fiber-optic distributed time synchronization device, achieving synchronous control of the local clock of the edge identification nodes. The optical time reference signal originates from a highly stable atomic clock configured in the central control device. The time stamp pulses emitted by this atomic clock are converted into optical pulse signals by a dual-channel laser and distributed to all identification nodes deployed in the identification scenario through single-mode fiber. A photoelectric conversion unit is deployed at the identification node, where a PIN photodiode converts the received optical pulses into high-level pulse electrical signals. This pulse serves as a reference timing input directly connected to the high-precision crystal oscillator control pin on the local identification control board. Upon pulse arrival, the identification node interrupts its local timing process, compares the current timing value with the reference pulse setting value, calculates the time offset, and stores this offset as a clock correction factor in the node control memory chip. Before each identification operation, the identification node calls the most recent time correction factor to correct all pending identification timestamps, ensuring millisecond-level time consistency accuracy even when edge nodes operate independently for hours or even days. To further improve stability, this invention designs a temperature compensation mechanism within the identification node. This mechanism measures the real-time temperature changes of the identification node's operating environment and uses a temperature drift error lookup table to dynamically adjust the local clock's operating speed. This ensures that the clock stability can be maintained for a short period even after the fiber optic connection is interrupted or the identification device is temporarily powered off and restarted.

[0075] After completing the local clock calibration of all edge identification nodes, all identification records from different times, devices, and locations are uniformly mapped to a standard timeline. First, a corresponding time correction factor is added to each identification record, and its local timestamp is weighted and corrected to a unified standard time value, with a correction accuracy controlled within ±0.5 milliseconds. Then, all identification records are sorted according to the corrected time values ​​to ensure that the chronological order matches the actual flow of materials. After sorting, the interval between consecutive identification events for each material is checked. If the time difference between two events is less than 500 milliseconds and the identification tag codes are identical, it is determined to be multiple repeated readings of the same material at the same stage. In this case, such identification records are merged into a single identification segment, retaining the earliest and latest time points as the start and end boundaries of this identification segment; other records are no longer stored separately. For events with an identification interval exceeding 500 milliseconds and different identification location numbers, it is determined to be cross-node identification and needs to be explicitly retained in the unified timeline. Finally, after the timeline of all identification events is constructed, a complete material identification flow record is formed. Each record includes fields such as: standard timestamp, material number, tag identification code, identification node number, phase backscatter status, acoustic consistency status, frequency band traction parameters, and signal strength level, and is saved as a standard format structure table for subsequent full-process traceability and dynamic identification strategy scheduling. This identification flow record format not only ensures that all identification behaviors generated by materials at any stage of the identification process can be clearly displayed on the timeline, but also provides a unique time series reference for dynamic identification accuracy and equipment identification responsibility traceability. In actual deployment, it effectively solves the problems of time misordering, identification coverage, and status conflicts caused by asynchronous equipment in traditional identification processes, significantly improving the overall accuracy and security of multi-point collaborative identification.

[0076] Guided by the time-aligned baseline, an adaptive time-space control solution is obtained. A phase-programmable metasurface array is deployed in the storage channel to reconstruct the near-field boundary in real time. Spectrum foldback traction is implemented and combined with field load inversion to form a closed-loop control mechanism that dynamically suppresses electromagnetic disturbance drift, thereby achieving stable identification of RFID tags.

[0077] To ensure stable identification performance of RFID tags in environments with high electromagnetic disturbances, a closed-loop control mechanism for dynamically suppressing disturbances needs to be constructed through active intervention and adaptive regulation of the electromagnetic field. The specific steps are as follows:

[0078] Based on the unified timeline information of the identification stream records, key time segments with significantly decreased identification success rates are extracted from the identification results. The identification device number, spatial location code, and tag category corresponding to these time segments are extracted as the initial basis for setting the target control area. Using time segments in the identification stream where the identification success rate is below a set threshold as input, a pre-constructed electromagnetic disturbance risk matrix is ​​superimposed. Through spatiotemporal matching, regions where identification nodes with disturbance intensity exceeding the median and whose disturbance frequency overlaps with the RFID tag frequency band are identified, delineating the spatial boundary where electromagnetic disturbance has a significant impact. Within this boundary, a spatial lattice network model is set at 10-centimeter intervals to simulate the propagation trend of disturbance energy within the boundary. The main direction and frequency distribution of the disturbance are reconstructed through vector superposition, providing a basis for subsequent control array layout.

[0079] Within the disturbance boundary space, a phase-programmable metasurface array is arranged according to the vector disturbance direction. This array consists of regularly arranged subwavelength-level reflective units, each comprising an electrically controlled capacitor layer, a conductive pattern layer, and a substrate insulating layer. Each unit measures 5 mm × 5 mm and is capable of responding to control signals at time intervals on the order of tens of nanoseconds. The array is installed in a two-dimensional grid on the top, side walls, and ground reflection areas of the inlet channel, ensuring complete coverage of the electromagnetic wave propagation path. Each unit is connected to the central controller via a low-latency signal channel. The controller calculates the required spatial phase distribution of the reflected wavefront based on the dominant frequency and disturbance direction generated by the aforementioned failed identification records. Upon receiving a voltage control signal, each unit adjusts its internal electrically controlled capacitor value to achieve real-time control of the incident electromagnetic wave reflection phase, thereby synthesizing the reflected wavefront in the desired direction and reconstructing the propagation boundary within the disturbance area.

[0080] A spectrum return-pull operation is performed on the offset information between the received identification frequency band and the preset resonant frequency band. The specific process is as follows: First, the spectrum scan data returned by the RF reader / writer when identification fails is collected, and the difference between the main peak frequency position in the tag's response spectrum and the standard resonant frequency is extracted. Second, a functional relationship model between the spectrum offset and the reflection phase is established based on this frequency difference, and the spectrum pull is converted into phase modulation parameters, which are then allocated to each unit of the metasurface array. When the array performs phase modulation, the energy wave that originally drifted to the non-identification frequency band will be reflected and returned to the set frequency band, causing the tag's response frequency to fall back into the reader / writer's receiving window, achieving "spectrum return." Unlike existing technologies that attempt to compensate for frequency band drift by increasing transmission power or adding identification redundancy, this method actively guides the interference wave by changing the electromagnetic propagation path, ensuring the tag is always in optimal resonant operating conditions.

[0081] Based on the field load inversion method, real-time closed-loop feedback adjustment is performed on the electromagnetic disturbance control effect. At least three electric field detectors are deployed within the channel, positioned at the center of the top reflection region, the channel centerline, and the lower boundary, respectively, to collect local electric field intensity changes and spectral characteristics. Each detector captures a spectral snapshot every second and compares it with the previous second's record. If the energy value of a certain frequency band rises above a set threshold, an inversion calculation is triggered. The inversion process constructs a three-dimensional disturbance energy distribution map based on multi-point electric field intensity values ​​and compares it with the previous field model to determine whether the current disturbance source has shifted or experienced energy enhancement. If significant drift or enhancement is observed, the control parameters of the metasurface array units are readjusted to maintain the minimum interference angle between the reflection direction and the disturbance direction, achieving dynamic adaptive compensation of the target area. This method is completely different from traditional static shielding or fixed interference frequency band filtering. It not only responds to frequency changes in electromagnetic disturbances in real time but also performs rapid spatial reconfiguration for spatial drift of the disturbance source, significantly improving the dynamic stability of the identification space.

[0082] The various control steps are integrated into a continuous, closed-loop control logic, constructing a dynamically adjustable identification space to ensure that RFID tags maintain normal identification functionality even in complex interference environments such as arc discharge, high-frequency pulses, and inductive coupling disturbances. This closed-loop control process performs a full-process readjustment every 1 second, comprising four stages: spectrum re-acquisition, phase offset update, electric field detection feedback, and phase control execution. The system automatically identifies areas with high failure event density and centralizes array control resources for frequency band reguidance, shortening the identification failure window time. Field verification showed that when the electric field disturbance amplitude in the storage channel exceeds 20 volts per meter, the conventional identification failure rate is as high as 18%. However, after deploying the closed-loop phase metasurface control scheme of this invention, the failure rate is reduced to less than 3%, and the number of repeated identification events is reduced by more than half, significantly improving the practical usability and security of transparent detection in complex environments.

[0083] This invention achieves multi-point spatiotemporal modeling of disturbance sources by introducing a three-axis vector probe array. Combined with resonant registration fingerprinting and frequency band drift compensation, it dynamically corrects the frequency drift problem in tag identification. Furthermore, it ensures consistency between physical objects and tag identification results through joint modeling of phase-encoded backscattering and strain acoustic features. Further, it achieves real-time intervention and control of the disturbance field through optical time-reference-driven global temporal alignment and a near-field electromagnetic environment reconstruction mechanism constructed by a phase-programmable metasurface array. This forms a sustainable and adaptive identification environment, comprehensively ensuring the identifiable, verifiable, and traceable management capabilities of core power grid materials during the warehousing stage. This effectively reduces the risk of identification failure, avoids mismatch between records and goods and identity anomalies, and improves transparent detection efficiency and material flow security. Technically, this method does not rely on traditional static shielding or single-point redundant identification methods. Instead, it overcomes the technical bottleneck of easy failure of RFID systems in high-disturbance scenarios through multi-dimensional information fusion and spatiotemporal feedback coordination.

[0084] The foregoing has only described certain exemplary embodiments of the present invention by way of illustration. Undoubtedly, those skilled in the art can modify the described embodiments in various ways without departing from the spirit and scope of the present invention. Therefore, the foregoing drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

Claims

1. A method for transparent inspection and full-process management of power grid materials based on the Internet of Things, characterized in that, Includes the following steps: A three-axis vector probe array is deployed in the material entry channel to collect multi-point spatiotemporal signals and calculate the field strength distribution, generate an electromagnetic disturbance spatiotemporal matrix, and output the radio frequency identification tag resonance drift risk matrix. Under the constraint of the resonance drift risk matrix of RFID tags, the resonance registration fingerprint of RFID tags and reading and writing devices is obtained. The complex reflection coefficient curve is established by frequency sweep excitation, the peak phase point is extracted and a set of resonance correction windows is formed. Guided by the resonant correction window set, drift compensation matching parameters are obtained, and the variable capacitor matching plate is driven to adjust the load of the RFID tag antenna, outputting the frequency band pull amount and allowable error band. Under the constraints of frequency band traction and allowable error band, phase-coded backscattering measurements are performed on the same material and the acoustic characteristics of the packaging surface strain are collected to construct a dual-verified identity criterion; With the support of the identity criterion of dual verification, a time alignment baseline is obtained, and the optical time reference is called to realize the clock synchronization of edge nodes, and the identification stream record is mapped to a single time axis. Guided by the time-aligned baseline, an adaptive time-space control solution is obtained. A phase-programmable metasurface array is deployed in the storage channel to reconstruct the near-field boundary in real time. Spectrum foldback traction is implemented and combined with field load inversion to form a closed-loop control mechanism that dynamically suppresses electromagnetic disturbance drift. The steps are as follows: Extract the time periods and spatial locations of failed identifications to construct the target control area; A phase-programmable metasurface array is deployed within the target control region, and a reflected phase control signal is applied to reconstruct the perturbation boundary. Collect spectral offset and calculate spectral traction parameters to perform spectral return traction; The system sets up an electric field detector to collect feedback data and performs field load inversion to dynamically correct array phase parameters. Finally, a closed-loop control cycle is constructed to achieve stable identification of RFID tags.

2. The method for full-process management of transparent power grid material inspection based on smart IoT as described in claim 1, characterized in that, The steps to obtain the spatiotemporal matrix of electromagnetic disturbance and output the resonance drift risk matrix of RFID tags are as follows: Multiple triaxial vector probe groups are evenly deployed along the axial direction of the transportation path in the material receiving channel. Each probe group consists of three orthogonally placed electric field sensing units. The voltage signals output by all probe sensing units are synchronously sampled using a unified clock trigger by the acquisition device to obtain an electric field intensity data volume with spatial positioning capability. The triaxial electric field data is processed by the spatial vector synthesis method to form an electromagnetic disturbance spectrum sequence. Based on the spectrum sequence, a perturbation spatiotemporal matrix is ​​constructed, and the resonance drift risk matrix is ​​calculated by combining the tag antenna electromagnetic compatibility mapping table, so as to realize the spatial positioning and risk prediction of tag frequency offset.

3. The method for full-process management of transparent power grid material inspection based on smart IoT as described in claim 2, characterized in that, The steps for obtaining the resonance registration fingerprints of the RFID tag and the reader / writer device and forming a resonance correction window set are as follows: High-risk areas are identified in the radio frequency identification tag resonance drift risk matrix, and a tag-read / write environment simulation scenario is constructed. The tag is excited by a frequency sweep signal and the reflected signal is collected to obtain the tag's reflection response data under electromagnetic disturbance environment; The reflection response data is analyzed to extract the main resonant frequency, phase change frequency, and half-power bandwidth, forming a resonant registration fingerprint. Based on the resonance registration fingerprint, a frequency offset tolerance range is set and a resonance correction window set is constructed to provide a basis for frequency adjustment.

4. The method for full-process management of transparent power grid material inspection based on smart IoT as described in claim 3, characterized in that, The steps for obtaining drift compensation matching parameters and adjusting the RFID tag antenna load to determine the output band pull and allowable error band are as follows: Extract frequency response characteristics based on the resonant correction window set and output a set of frequency band traction parameters; The capacitance of the variable capacitor matching board is adjusted step by step according to the frequency band traction parameter set, and a reactance load balance combination is constructed by matching it with a standard inductor. Repeated identification tests and frequency drift data were collected in an electromagnetic disturbance simulation environment, and the final output frequency band pull and allowable error band were determined.

5. The method for full-process management of transparent power grid material inspection based on smart IoT as described in claim 4, characterized in that, The steps for constructing a dual-verification identity criterion by performing phase-encoded backscattering measurements and collecting the surface strain acoustic characteristics of the packaging for the same material are as follows: The excitation frequency is set based on the frequency band traction amount, and phase-encoded backscattering measurement is performed to obtain the tag phase fingerprint; Collect strain acoustic characteristics of the material packaging surface to form a set of structural response characteristic parameters; A dual identity criterion is constructed by comparing the phase fingerprint with the acoustic features; The verification results are written to the tag user data area as encrypted status markers for subsequent identification process verification.

6. The method for full-process management of transparent power grid material inspection based on smart IoT as described in claim 5, characterized in that, The steps to obtain the timing alignment baseline, use the optical time reference to synchronize the edge node clocks, and map the identification stream records to a single time axis are as follows: The identification node is synchronized locally by injecting an optical time reference signal into the identification node through a fiber optic distributed time synchronization device. The time correction factor of each identification node is invoked to perform a unified standard time mapping on the local timestamps generated by each identification record; Sort all identification records by the corrected time value and construct a standard timeline identification stream; finally, output a structured record table containing standard timestamps and identification status for full-process identification traceability and strategy scheduling.

Images