Intelligent substation equipment tag automatic identification system based on RFID

Abstract

The invention discloses an intelligent substation equipment label automatic identification system based on RFID, and relates to the technical field of substation equipment label identification, comprising the following steps: monitoring the electromagnetic noise level in real time, and determining to enter interference compensation according to the noise intensity; dynamically adjusting an antenna beam direction and a working protocol according to noise and label distribution; clustering label signal features with similar space; source separation and label analysis of overlapped signals are completed through an edge neural network; counting the decoding success rate and the response time delay of each cluster, and calculating a quality score according to a preset rule; for clusters with high noise or decoding failure, compensation parameters are dynamically adjusted; and writing all the successfully read labels and the position information thereof into a local database. Through electromagnetic environment sensing dynamic filtering, antenna array self-adaptive beam forming, multi-label concurrent decoding and source separation and HF / UHF frequency band protocol automatic switching, the label reading reliability, the coverage range and the identification efficiency in the transformer substation environment are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of substation equipment tag identification, and particularly relates to an intelligent substation equipment tag automatic identification system based on RFID. BACKGROUND

[0002] With the continuous advancement of smart grid construction, the asset management and state monitoring of substation equipment put forward higher requirements for automation and precision. The traditional equipment tag identification mainly relies on manual point-by-point scanning or fixed camera identification mode, which has the problems of low work efficiency, easy to miss scanning, difficult to adapt to complex environment, etc., and is difficult to meet the real-time and reliability requirements of modern substations.

[0003] In recent years, radio frequency identification (RFID) technology has gradually been applied to substation equipment tag automatic identification field due to its advantages of non-contact, high speed and batch reading. However, the existing RFID-based identification system still has some deficiencies in practical application: first, the substation environment is full of high-voltage live equipment and power frequency and pulse noise generated by switch action, and the existing system usually uses fixed threshold or simple filtering, which is difficult to dynamically compensate for environmental interference, resulting in high tag reading failure rate; second, the traditional reader relies on single antenna or simple antenna array, lacks dynamic beamforming mechanism based on noise and tag distribution, has many coverage dead angles and large blind area, and cannot guarantee seamless coverage of all station equipment; third, when tags are densely distributed, the existing method can only identify serially or rely on simple time division protocol, and it is difficult to realize concurrent decoding in high-density scenarios, and the source separation ability of overlapping signals is insufficient, which is easy to produce high decoding error rate; fourth, there are many types of field tags (such as coexistence of HF and UHF tags), and most systems can only use a single frequency band protocol fixedly, and cannot dynamically switch protocols according to environmental noise ratio and tag distribution, resulting in a significant decline in reading performance in a specific frequency band. Therefore, an intelligent identification system capable of real-time sensing of electromagnetic environment, dynamically adjusting antenna beam and communication protocol, and having efficient multi-tag clustering and source separation decoding capability is needed to improve the reliability and practicality of substation equipment tag automatic identification. SUMMARY

[0004] In view of the problems that the existing substation equipment tag identification system is unstable in reading in strong electromagnetic interference environment, has many antenna coverage dead angles and cannot adaptively beamform, has low concurrent decoding efficiency in tag dense scenarios and insufficient anti-collision and source separation capability, and lacks dynamic switching and optimization support of multi-frequency band protocol, resulting in that the reliability and real-time of the overall identification are difficult to meet the needs of intelligent substations, the present application is proposed.

[0005] Therefore, the problem to be solved by the present application is how to provide an intelligent substation equipment tag automatic identification system based on RFID.

[0006] To solve the above technical problems, the present application provides the following technical solutions:

[0007] In a first aspect, the embodiments of the present application provide an RFID-based intelligent substation equipment tag automatic identification method, comprising: monitoring the electromagnetic noise level in the substation in real time, and determining whether to enter the interference compensation mode according to the noise intensity; dynamically adjusting the antenna array beam direction and the protocol type according to the noise condition and the tag distribution condition, and adjusting the directivity pattern and the protocol selection to realize the optimal tag coverage rate; clustering the tags close in space by using the tag signal characteristics to ensure that the number of tags in each cluster is within the decoding capability range; transmitting the reading command to each tag cluster in turn or in parallel, performing source separation and tag decoding on the overlapping signals by using the edge neural network, and calculating the decoding error rate of each cluster to determine whether it is a “difficult cluster”; calculating the quality score by combining the preset scoring rules with the identification success rate and the response time delay of all clusters; dynamically adjusting the compensation parameters such as phase, frequency offset and gain according to the monitored frequency offset and noise characteristics, and re-reading after completing signal compensation; and writing the tag information and position information of all successfully read tags into the database, and uploading the difficult cluster data to the cloud for subsequent optimization.

[0008] As a preferred scheme of the RFID-based intelligent substation equipment tag automatic identification method, the sensor performs integral operation on the noise power spectral density in the working bandwidth to obtain the total noise power, and compares the total power with the preset noise threshold; the integral operation accumulates and sums the noise intensity of each frequency point, and the threshold judgment compares the accumulation result of the noise intensity with the environmental tolerance limit to determine whether to enter the interference compensation mode.

[0009] As a preferred scheme of the RFID-based intelligent substation equipment tag automatic identification method, the optimal beamforming weight is constructed according to the array element response and the environmental noise covariance; the optimal beamforming weight construction combines the array response vector and the noise covariance matrix by using the maximum signal-to-noise ratio criterion to derive the weight vector, which is used to enhance the target direction signal and suppress the noise.

[0010] As a preferred scheme of the RFID-based intelligent substation equipment tag automatic identification method, the spatial coordinates and signal intensity characteristics of the tags are subjected to distance measurement and density clustering; the distance measurement takes the Euclidean distance between the tags as the clustering basis, and dynamically divides the clusters according to the point density and the minimum cluster size parameter to ensure that the number of tags in each cluster is within the decoding capability range.

[0011] As a preferred scheme of the RFID-based intelligent substation equipment tag automatic identification method, the method comprises the following steps: performing source separation on the overlapping signals by a lightweight neural network in the edge computing unit and counting the decoding result; the decoding error rate calculation is considered as a ratio operation of the number of error analysis tags and the total number of requested tags, which is used to measure the performance of the current network model on the cluster, and whether to mark as a difficult cluster is determined according to the performance.

[0012] As a preferred scheme of the RFID-based intelligent substation equipment tag automatic identification method, the method comprises the following steps: the decoding success rate and the average time delay of each cluster are weighted and summarized according to a preset weight to obtain a comprehensive quality score; the quality score is constructed by regarding the success rate as a positive indicator and the time delay as a negative indicator, and the two indicators are added after being multiplied by corresponding weights to form a total evaluation value, which is compared with a score threshold to determine whether to execute a rollback strategy.

[0013] As a preferred scheme of the RFID-based intelligent substation equipment tag automatic identification method, the method comprises the following steps: frequency offset correction is performed based on the deviation of the reference tag carrier and the design frequency, and the latest noise power is used to dynamically adjust the receiving chain gain and the band pass filtering parameters; the correction and gain adjustment take the frequency offset error as the phase correction amount and take the noise power as the gain control basis, which are used to maintain the expected signal-to-noise ratio.

[0014] In a second aspect, to further solve the problems existing in the substation equipment tag identification, the embodiment provides an RFID-based intelligent substation equipment tag automatic identification system, which comprises: a noise monitoring module, which is used to collect electromagnetic noise signals in real time at the substation site and judge the environmental noise level; a beam control module, which is used to dynamically adjust the beam direction and communication protocol of the multi-antenna array based on the noise level and tag distribution information; a clustering processing module, which is used to cluster the tags adjacent in space according to the collected tag signal characteristics; a decoding module, which is used to transmit a reading command to each tag cluster and separate the overlapping signal sources and analyze the tag ID by using the neural network in the edge computing unit; a quality evaluation module, which is used to calculate the comprehensive quality score according to the preset weight based on the decoding success rate and the response time delay of each cluster; a compensation adjustment module, which is used to dynamically adjust the receiving chain parameters such as phase, frequency offset and gain for the high-noise or decoding failure cluster; and a data management module, which is used to write the successfully identified tag information and spatial coordinates into a local database and upload the difficult cluster data to the cloud.

[0015] In a third aspect, the embodiment of the present application provides a computer device comprising a memory and a processor, and the memory stores a computer program, wherein when the computer program is executed by the processor, any step of the RFID-based intelligent substation equipment tag automatic identification method according to the first aspect of the present application is implemented.

[0016] In a fourth aspect, an embodiment of the present application provides a computer readable storage medium having stored thereon a computer program, wherein the computer program, when executed by a processor, implements any step of the RFID-based intelligent substation equipment tag automatic identification method according to the first aspect of the present application.

[0017] The present application has the following beneficial effects:

[0018] 1. The present application can enter the "interference compensation mode" in time through continuous monitoring of electromagnetic noise in the substation and comparison of the total noise power with the preset threshold, and through fine adjustment of compensation parameters such as phase correction and gain adjustment, the reading failure rate in a high-noise environment is reduced, thereby improving the stability and reliability of the system under complex power frequency and switching flashover noise conditions;

[0019] 2. The system can dynamically adjust the beam direction and opening width of the antenna array based on the beamforming weight vector and array factor calculated according to the maximum signal-to-noise ratio criterion, thereby effectively eliminating the blind area and achieving high coverage rate reading under different tag distribution conditions, and ensuring seamless identification of all substation equipment tags;

[0020] 3. The system simultaneously detects HF and UHF tags, and automatically switches between the two frequency bands according to the environmental noise ratio and the protocol switching failure rate, so that the optimal communication protocol can be selected under different frequency band interference conditions, thereby improving the overall success rate of tag reading and the anti-interference ability;

[0021] 4. The present application clusters spatially close tags based on the density clustering of RSSI and azimuth angle, and uses a lightweight convolutional network to perform source separation and parallel decoding on overlapping signals, thereby achieving concurrent reading of multiple tags, improving decoding throughput, and ensuring high-precision identification results, effectively reducing the time delay and error rate of traditional single serial decoding;

[0022] 5. The system calculates a comprehensive quality score based on the reading success rate and average time delay of each cluster, and when the score is lower than the preset threshold, it intelligently rolls back to the beamforming or cluster division stage for retry, and at the same time uploads all "difficult clusters" and their compensation history to the cloud platform for offline digital twin arrangement optimization and maintenance decision support, thereby realizing adaptive optimization and continuous iteration of system performance. BRIEF DESCRIPTION OF DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor. Among them:

[0024] Figure 1 The flow chart for implementing the present application in Example 1.

[0025] Figure 2 The system module chart for the present application in Example 2. DETAILED DESCRIPTION

[0026] In order to make the above objectives, features and advantages of the present application more apparent, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0027] In the following description, a lot of specific details are set forth in order to provide a thorough understanding of the present application, however, the present application can be practiced in other different manners than those described herein, and those skilled in the art can make similar generalization without departing from the spirit and scope of the present application, therefore, the present application is not limited to the specific embodiments disclosed below.

[0028] Secondly, the "one embodiment" or "embodiment" referred to herein means that the specific features, structures or characteristics can be included in at least one implementation of the present application. The "in one embodiment" appearing in different places in the specification does not mean the same embodiment, nor is the embodiment alone or selectively exclusive of other embodiments.

[0029] Example 1

[0030] Reference Figure 1 For the first embodiment of the present application, the embodiment provides an RFID-based intelligent substation equipment tag automatic identification method, including the following steps:

[0031] Step S1: Real-time monitoring of electromagnetic noise level in the substation, and determining whether to enter the interference compensation mode according to the noise intensity, which includes the following sub-steps:

[0032] S1-1: Reading the electromagnetic noise sensor data deployed around the reader in the bandwidth (such as 26MHZ) per second, and obtaining the power spectral density PSD(f) in the current frequency band.

[0033] S1-2: Integrating the power spectral density PSD(f) in the entire working bandwidth to obtain the total noise power P noise ;

[0034] Specifically, the total noise power P noise can be embodied by the following formula:

[0035]

[0036] In the formula, f1, f2 respectively represent the frequency band, for example, f1=902MHz, f2=928MHz.

[0037] S1-3: compare the total noise power P noise with the preset noise threshold P th , and make a decision.

[0038] Specifically, set the threshold P th = -60 dBm, if the total noise power P noise > the noise threshold P th , jump to step S6 to perform "interference compensation parameter fine-tuning"; otherwise, proceed to step S2.

[0039] It should be noted that the above steps continuously monitor the field power frequency interference and switch flashover noise intensity through electromagnetic noise sensors distributed in the substation equipment area, calculate the noise power and compare it with the preset noise threshold. If the noise is too high, it directly enters the subsequent compensation module, otherwise it continues to initialize the antenna beam and protocol configuration to ensure the feasibility of subsequent environmental reading;

[0040] For example, the measured PSD is approximately constant at 5 × 10 -9 W / Hz throughout the bandwidth, then P noise = 5 × 10 -9 W / Hz × (928-902) × 10 6 Hz = 5 × 10 -9 × 26 × 10 6 = 1.3 × 10 -1 W = 0.13 W, then 0.13 > 1 × 10 -6 (-60 dBm), it is determined as "high noise" and needs to enter the compensation process.

[0041] Step S2: dynamically adjust the antenna beam direction and communication protocol according to the environmental noise and tag distribution to maximize the tag coverage, which includes the following sub-steps:

[0042] S2-1: read the antenna array parameters (element number M, spacing d) and noise covariance matrix R nn , and calculate the beamforming weight vector w according to the maximum signal-to-noise ratio criterion

[0043] Specifically, the above beamforming weight vector w can be embodied by the following formula:

[0044]

[0045] In the formula, a(θ) is the array response vector, which is composed of a vector of each element phase / amplitude response; α is the normalization coefficient, which is taken to make ‖w‖ = 1.

[0046] S2-2: Calculate the array factor AF(θ) using the weighted value w after the shaping, and estimate the proportion CR of the number of tags in the beam coverage area;

[0047] Specifically, the above parameters can be embodied by the following formula:

[0048]

[0049] In the formula, d=c / 2f represents the distance between adjacent antenna units, c represents the speed of light, and f represents the working frequency (such as 915 MHz); λ=c / f represents the wavelength; θ represents the antenna pointing angle; represents the phase difference between the mth unit and the first unit.

[0050] For example, take M=8 element array, θ=30°, λ=c / f≈0.328m, d=c / 2f≈0.164m, then the phase difference between the first unit and the first unit is (radian), respectively, the phase difference when m=1,...,8 is calculated, and AF(30°)=1-1+1-1+1-1+1-1=0 is obtained, which indicates that the signal is completely cancelled in this direction, and a beam null direction appears, that is, no signal radiation or receiving ability; if other angles are used, such as θ=0°, then represents that the main beam direction is at 0°, which meets our expectation of strengthening the beam in the "forward direction";

[0051]

[0052] In the formula, CR represents the coverage rate; N cov represents the number of tags that can be detected in the beam coverage area, which is counted by pre-scanning or historical heat map statistics; N tot represents the total number of tags in the target area, which is obtained by system configuration registration table or full inventory before inspection.

[0053] S2-3: Simultaneously detect the on-site coaxial coil tag HF (13.56 MHz) and UHF tag (915 MHz), and automatically switch the optimal working protocol according to the proportion of environmental noise and the preset proportion;

[0054] For example, if P HF >0.3P noise , select the tag HF; otherwise, select the UHF tag.

[0055] S2-4: Set the coverage rate threshold CR th (such as 0.9), compare the coverage rate with the coverage rate threshold, and make a decision;

[0056] Specifically, if the coverage rate CR thIf the protocol switching failure rate is over the limit (e.g., > 5%), roll back to step S1.

[0057] It should be noted that the above protocol switching failure rate is obtained as follows: in the running process, the reader records the total number of switching and the number of failures at each protocol switching, and performs real-time statistics on the two data in a sliding window to obtain the current switching failure rate.

[0058] S2-5: When the coverage and switching meet the requirements, go to step S3; otherwise, roll back or retry as needed.

[0059] It should be noted that the above step is based on the noise and label distribution information obtained by preprocessing, and the controller dynamically adjusts the beam direction and width of the multi-antenna array, and automatically switches between UHF and HF protocols according to the on-site label type and noise proportion, to achieve maximum coverage and optimal reading success rate. The entire process is internally provided with a fast feedback loop, which can immediately fine-tune or roll back when the coverage or protocol switching fails.

[0060] Step S3: Cluster the tags with similar spatial features using tag signal characteristics to ensure that the number of tags in each cluster is within the decoding capability range, including the following sub-steps:

[0061] S3-1: Collect the RSSI value r of each tag i , azimuth angle θ1...θ n and other signal characteristics, map to spatial coordinates, and establish a feature vector.

[0062] S3-2: Calculate the Euclidean distance d ij between the tags using the distance threshold ε (e.g., 0.8 m) and the minimum number of points in the cluster minPts (e.g., 3) as parameters;

[0063] Specifically, the above Euclidean distance d ij can be represented by the following formula:

[0064]

[0065] In the formula, x i , y i represent the planar coordinates of tag i; x j , y j represent the planar coordinates of tag j; and the above coordinates are obtained by the RSSI+TDOA / angle estimation algorithm after beamforming, or are recorded by calibration during installation;

[0066] For example, if the coordinates of tag i are (2.0, 3.0) m and the coordinates of tag j are (2.6, 3.4) m, then i, j belong to the same cluster core neighborhood, and the number of tags is recorded.

[0067] S3-3: Record the number of tags N c in each cluster, and compare it with the upper limit of the number of tags N max , and make a decision.

[0068] Specifically, if the number of tags N c is greater than the upper limit of the number of tags N max , re-perform clustering and subdivision within the cluster until all clusters meet N c ≤ 20.

[0069] S3-4: If the total number of clusters exceeds the system concurrent decoding capability, return to step S2 to reduce the coverage area of each cluster; otherwise, go to step S4.

[0070] It should be noted that in the above steps, after the beam configuration is completed, the system automatically clusters and clusters the tags that are spatially adjacent and similar in signal characteristics according to the strength and angle of arrival of the real-time received signal. If the number of tags in a cluster is too large or the number of clusters exceeds the concurrent processing capability, the controller will immediately redivide the clusters or adjust the coverage area to ensure that each cluster can be efficiently and reliably decoded in parallel in the subsequent steps.

[0071] Step S4: Transmit read commands to each tag cluster in sequence or in parallel, and perform source separation and decoding on the overlapping signals through the edge neural network, including the following sub-steps:

[0072] S4-1: According to the cluster division, transmit read commands to each tag in the cluster, and collect baseband signal samples;

[0073] S4-2: Run a lightweight convolutional network on the edge computing unit to perform source separation on the multi-tag overlapping signal and parse out the tag ID.

[0074] S4-3: Count the number of decoding errors and the total number of reads for each cluster, and calculate the error rate ER i .

[0075] Specifically, the error rate ER i can be represented by the following formula:

[0076]

[0077] In the formula, E i represents the total number of tags initiated by cluster i for reading, which is recorded when reading is scheduled; N i represents the number of tags in cluster i that fail to decode, which is compared with the registration table after decoding to count the number of mismatches or CRC failures.

[0078] S4-4: Set the error rate threshold ER n(such as 0.1), the error rate ER i and the error rate threshold ER n Compare and make decisions;

[0079] Specifically, if the error rate ER i >Error rate threshold ER n , then mark the cluster as a “difficult cluster” and jump to step S6; otherwise, cache the result and continue to process the next cluster.

[0080] It's important to note that in the above steps, for each tag cluster, the reader / writer issues read commands sequentially or in parallel. The collected overlapping signals are then source-separated and decoded by a lightweight neural network within the edge computing unit, accurately identifying the IDs of all tags. If the decoding error rate for a cluster exceeds the system's tolerance threshold, it is marked as a "problematic cluster" and transferred to the subsequent compensation process.

[0081] Step S5: Calculate the quality score based on the read success rate and latency of each cluster. If the score does not meet the standard, roll back the previous optimization steps. This step includes the following sub-steps:

[0082] S5-1: Summarize the read success rate SR of all successful clusters i =1-ER i and the average delay T i , define the quality score Q;

[0083] Specifically, the quality score Q can be expressed by the following formula:

[0084]

[0085] Where, It is expressed as the average success rate of all clusters; It is expressed as the average delay of all clusters; T max Indicates the maximum delay allowed by the system, which is set according to the system performance requirements; w1 and w2 are weight coefficients, which are collaboratively configured based on the importance of the service side (success rate vs. delay) (for example, w1 = 0.7, w2 = 0.3);

[0086] S5-2: Set the quality score threshold Q th (such as 0.8), the quality score Q and the quality score threshold Q th Compare and make decisions;

[0087] Specifically, if the quality score Q ≥ the quality score threshold Q th , proceed to step S7;

[0088] Otherwise, categorize by main cause (low success rate or excessive latency):

[0089] Rollback strategy one: if the main factor is cluster decoding failure, rollback to step S3 to re-cluster;

[0090] Rollback strategy two: if it is coverage or beam reason, rollback to step S2 to re-shaping.

[0091] It should be noted that in the above steps, the system aggregates the decoding success rate, average read latency and other indicators of each cluster, calculates the comprehensive quality score, and judges whether the overall recognition meets the standard based on this. If the score is low, it will intelligently fall back to the grouping or beam configuration step for retry until the recognition quality meets the expected standard.

[0092] Step S6: dynamically adjust the compensation parameters such as phase, frequency offset and gain according to the monitored frequency offset and noise characteristics, and retry reading, including the following sub-steps:

[0093] S6-1: read the actual carrier f of the reference tag meas , calculate the frequency offset Δf = f meas -f0, and apply phase correction y[n] to the received signal x[n];

[0094] Specifically, the above phase correction y[n] can be embodied by the following formula:

[0095]

[0096] In the formula, f0 represents the design frequency (such as 915MHz); T s =1 / B represents the sampling interval.

[0097] S6-2: adjust the receive chain gain G and bandpass filter parameters according to the latest noise power P noise , to maintain the desired signal-to-noise ratio.

[0098] S6-3: after compensation, return to step S4 to retry reading for "difficult clusters";

[0099] Specifically, if it still fails to pass the bit error rate threshold for N times (for example, 3 times), it will roll back to step S1 to wait for environmental changes or switch antenna modes; otherwise, it will return to step S4 normally.

[0100] It should be noted that the above steps are aimed at the high noise condition or "difficult cluster" problem identified in steps S1 and S4. The system performs frequency offset correction, phase compensation and gain control and other digital signal processing on the received signal to suppress environmental interference in real time. After compensation is completed, the decoding of the cluster is immediately retried. If the quality requirement is still not met, the system will fall back to the environment detection or beam shaping step.

[0101] Step S7: write all successfully read tag information and position information into the database, and upload the difficult cluster data to the cloud for subsequent optimization, including the following sub-steps:

[0102] S7-1: organize all successfully read tag IDs, cluster center coordinates (x, y, z), and success rate SR i and average latency T i .

[0103] S7-2: write the summary data into the central database, update the digital twin model and the visualization interface, and generate the identification report for this time.

[0104] S7-3: upload all "difficult clusters" and their compensation history to the cloud for offline arrangement optimization and maintenance decision support.

[0105] It should be noted that after all clusters pass the decoding quality evaluation, the system writes the unique identification of each tag, the spatial position cluster center, the decoding quality, and the latency information into the central database, and displays it in real time on the visualization interface. At the same time, for the "difficult cluster" data that cannot be read with high quality, it is simultaneously uploaded to the cloud platform for subsequent digital twin analysis and tag arrangement optimization.

[0106] In summary, the present application collects the power frequency and pulse noise characteristics in the substation in real time through the electromagnetic environment perception module and dynamically adjusts the filtering parameters, effectively improving the reliability of tag reading under strong electromagnetic interference; using the antenna array beamforming algorithm, the transmission and reception beams are adaptively optimized according to the device distribution and environmental noise, achieving non-blind area coverage of all station devices; combined with multi-tag concurrent decoding and source separation technology, the decoding efficiency and anti-collision ability in tag dense scenarios are significantly improved; at the same time, the protocol switching module can dynamically switch between UHF and HF frequency bands according to the environmental noise distribution and tag type, ensuring the best reading performance of different frequency band tags, thereby greatly improving the stability, real-time performance and application range of the intelligent substation equipment tag automatic identification system as a whole.

[0107] Embodiment 2

[0108] Referring to Figure 2 , the second embodiment of the present application is different from the first embodiment in that it also provides an RFID-based intelligent substation equipment tag automatic identification system, comprising:

[0109] A noise monitoring module for real-time acquisition of electromagnetic noise signals and judgment of environmental noise level in the substation site;

[0110] A beam control module for dynamically adjusting the beam direction and communication protocol of the multi-antenna array based on the noise level and tag distribution information to maximize the tag coverage rate;

[0111] a clustering processing module configured to cluster tags that are spatially adjacent according to the collected tag signal features, so as to ensure that the number of tags in each cluster is within a system decodable range;

[0112] a decoding module configured to transmit a reading command to each tag cluster, and to separate overlapping signal sources and analyze tag IDs by using a neural network in the edge computing unit;

[0113] a quality evaluation module configured to calculate a comprehensive quality score according to a preset weight based on a decoding success rate and a response time delay of each cluster, so as to determine whether to rollback and optimize the previous steps;

[0114] a compensation adjustment module configured to dynamically adjust phase, frequency offset, gain and other receiving chain parameters for a high-noise or decoding-failed cluster and re-read;

[0115] a data management module configured to write successfully identified tag information and spatial coordinates into a local database, and upload difficult cluster data to the cloud for subsequent optimization and maintenance.

[0116] The modules work cooperatively through a controller to realize fully automatic and robust device tag identification and management.

[0117] The embodiment also provides a computer device suitable for the case of the RFID-based intelligent substation device tag automatic identification method, and the computer device includes a memory and a processor.

[0118] The computer device can be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for running the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is configured to perform wired or wireless communication with an external terminal. The wireless communication can be achieved through WIFI, a carrier network, NFC (near field communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen. The input device of the computer device can be a touch layer overlaid on the display screen, or a key, trackball or touchpad arranged on the shell of the computer device. In addition, the input device can be an external keyboard, touchpad or mouse, etc.

[0119] The embodiment also provides a storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method for automatically identifying an RFID-based intelligent substation equipment tag as proposed in the above embodiment. The storage medium can be implemented by any type of volatile or nonvolatile storage devices or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic storage, a flash memory, a magnetic disk, or an optical disk.

[0120] It should be noted that the above embodiment is only used to illustrate the technical solutions of the present application but not limit the present application. Although the present application is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application, and all of them should be covered in the scope of the claims of the present application.

Claims

1. A method for automatic identification of smart substation equipment tags based on RFID, characterized by: include: Step S1, real-time monitoring of the electromagnetic noise level in the substation, and judging whether to enter the interference compensation mode according to the noise intensity; Step S2: Dynamically adjust the antenna array beam direction and protocol type based on the noise situation and tag distribution, and achieve optimal tag coverage by adjusting the pattern and protocol selection; Step S3: Clustering spatially close tags using tag signal characteristics to ensure that the number of tags in each cluster is within the decoding capability. Step S4: Send read commands to each tag cluster sequentially or in parallel, perform source separation and tag decoding on the overlapping signals through the edge neural network, and calculate the decoding error rate of each cluster to determine whether it is a "difficult cluster"; Step S5: Count the recognition success rates and response delays of all clusters, and calculate the quality scores based on the preset scoring rules. Step S6: dynamically adjust compensation parameters such as phase, frequency offset, and gain according to the monitored frequency offset and noise characteristics, and re-read after completing signal compensation. In step S7, all successfully read tag information and location information are written into the database, and the difficult cluster data is uploaded to the cloud for subsequent optimization.

2. The RFID-based automatic identification method for smart substation equipment tags according to claim 1, characterized in that: In step S1, the sensor integrates the noise power spectral density within the working bandwidth to obtain the total noise power, and compares the total power with the preset noise threshold. The integration operation accumulates the noise intensity of each frequency point, and the threshold judgment compares the accumulated noise intensity result with the environmental tolerance limit to determine whether to enter the interference compensation mode.

3. The RFID-based automatic identification method for smart substation equipment tags according to claim 1, characterized in that: In step S2, an optimal shaping weight is constructed based on the array unit response and the environmental noise covariance; the optimal shaping weight is constructed by combining the array response vector and the noise covariance matrix through the maximum signal-to-noise ratio criterion to derive a weight vector for enhancing the target direction signal and suppressing the noise.

4. The RFID-based automatic identification method for smart substation equipment tags according to claim 1, characterized in that: In step S3, a distance metric is performed on the spatial coordinates and signal strength characteristics of the tags and density clustering is performed. The distance metric uses the Euclidean distance between tags as the clustering basis, and dynamically divides clusters according to the point density and the minimum cluster size parameter to ensure that the number of tags in each cluster is within the decoding capability.

5. The RFID-based automatic identification method for smart substation equipment tags according to claim 1, characterized in that: In step S4, the overlapping signals are source separated and the decoding results are counted through the lightweight neural network in the edge computing unit; the decoding error rate calculation is regarded as a ratio operation of the number of incorrectly parsed labels and the total number of requested labels, which is used to measure the performance of the current network model on the cluster and decide whether to mark it as a difficult cluster based on this.

6. The RFID-based automatic identification method for smart substation equipment tags according to claim 1, characterized in that: In step S5, the decoding success rate and average delay of each cluster are weighted and summarized according to preset weights to obtain a comprehensive quality score; the quality score is constructed by considering the success rate as a positive indicator and the delay as a negative indicator, multiplying them by the corresponding weights and adding them together to form an overall evaluation value, which is compared with the scoring threshold to determine whether to execute the rollback strategy.

7. The RFID-based automatic identification method for smart substation equipment tags according to claim 1, characterized in that: In step S6, frequency offset correction is performed based on the deviation between the reference tag carrier and the design frequency, and the receiving chain gain and bandpass filter parameters are dynamically adjusted in combination with the latest noise power. The correction and gain adjustment use the frequency offset error as the phase correction amount and the noise power as the basis for gain control to maintain the desired signal-to-noise ratio.

8. An RFID-based automatic identification system for smart substation equipment tags, based on the RFID-based automatic identification method for smart substation equipment tags according to any one of claims 1 to 7, characterized in that: include, Noise monitoring module, used to collect electromagnetic noise signals in real time at the substation site and determine the ambient noise level; A beam control module, which dynamically adjusts the beam direction and communication protocol of the multi-antenna array based on noise levels and tag distribution information; Clustering processing module, used to cluster spatially adjacent tags according to the collected tag signal characteristics; The decoding module is used to send read commands to each tag cluster and use the neural network in the edge computing unit to separate overlapping signal sources and resolve tag IDs; The quality assessment module is used to count the decoding success rate and response delay of each cluster and calculate the comprehensive quality score based on the preset weights; Compensation adjustment module, used to dynamically adjust receive chain parameters such as phase, frequency offset, and gain for clusters with high noise or decoding failures; The data management module is used to write the successfully identified tag information and spatial coordinates into the local database and upload the difficult cluster data to the cloud.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the RFID-based smart substation equipment tag automatic identification method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the RFID-based smart substation equipment tag automatic identification method according to any one of claims 1 to 7 are implemented.

Images