Battery identification and battery replacement method based on RFID

Abstract

The present application relates to battery swap cabinet electromagnetic compatibility and RFID identification technical field, especially in kind based on the battery identification battery swap method of RFID, this method includes: according to the rising amplitude, duration and waveform steepness of signal strength sequence determines the interference of battery swap cabinet capture pulse train, determines the interference intensity;When the interference intensity is strong pulse, according to the voltage sequence and signal strength subsequence determines the cross-correlation peak;When the pulse train interference is successfully coupled to the sampling circuit, determine the positive feedback level;Determine the anti-interference level to execute the anti-interference strategy.The present application is through the three parameter sliding window grading of pulse train interference, the cross-correlation coupling verification of received signal strength indication sequence and battery management system voltage and the collaborative analysis of common mode current positive feedback index, effectively solve the technical problem that the anti-interference response is blind and the battery swap process is easily interrupted by harmless interference due to the fact that only the existence of pulse group is detected without verifying whether the interference is truly coupled to the key sampling circuit in the prior art.

Description

Technical Field

[0001] This invention relates to the fields of electromagnetic compatibility and RFID identification technology for battery swapping cabinets, and particularly to an RFID-based battery identification and swapping method. Background Technology

[0002] With the rapid expansion of battery swapping networks for two-wheeled electric vehicles, the deployment scenarios for battery swapping cabinets have expanded from traditional storefronts to densely populated areas such as transportation hubs, underground parking lots, and centralized charging stations. The commercial deployment of high-power wireless charging piles (3kW-10kW), the switching operations of numerous variable frequency chargers in centralized charging stations, the start-up and shutdown of welding equipment in industrial areas, and substation switching operations all generate electromagnetic pulse trains with microsecond-level rise times, high amplitude, and wide frequency spectrum characteristics. The peak field strength of these pulse trains can reach tens of volts per meter, covering the commonly used operating frequency bands for RFID. These pulses couple into the RFID readers and antennas, battery metal casings, and the battery management system sampling circuits within the battery swapping cabinets through both conduction and radiation. This causes momentary suppression of the RFID reader's received signal strength indication, false transient drops in the battery management system voltage sampling, and positive feedback interference from the common-mode current induced in the battery casing, thus reducing the success rate of battery swapping and operational safety. Therefore, how to achieve reliable battery identification and anti-interference control in battery swapping cabinets under strong electromagnetic pulse interference environments is a pressing technical problem that needs to be solved in the current battery swapping infrastructure field.

[0003] Chinese Patent Application Publication No. CN119864929A discloses a detection and control system for electrical fast transient bursts (EFTs) in a device. The system includes a motherboard, a main control unit, and onboard relays. The onboard relays are located at pins on the motherboard, which connects to the device's components via these pins. The onboard relays provide electrical isolation between the motherboard's control circuit and external circuits; under normal conditions, the onboard relays are in a closed state. The main control unit is used for detecting and responding to EFT interference in the operating device and for providing crash protection control for the host device after a shutdown.

[0004] Therefore, the existing technology has the following problems: it relies on pulse group presence detection but does not extract the rise amplitude, duration, and waveform steepness characteristic parameters of the pulse train, which can easily lead to the anti-interference strategy being unable to distinguish the interference intensity, resulting in over-response to weak interference or under-response to strong interference; it relies on triggering isolation upon detecting interference but does not verify whether the pulse train is successfully coupled to the sampling circuit of the battery management system, which can easily cause unnecessary isolation actions, affecting the continuity and efficiency of the battery swapping process; it relies on only detecting the presence of the pulse group but does not monitor the common-mode current and positive feedback effect between the battery casing and the grounding terminal of the battery swapping cabinet, which can easily lead to the inability to quantify the positive feedback deterioration cycle of interference and common-mode current, making it difficult for the isolation response to block the positive feedback, and the actual anti-interference effect is limited. Summary of the Invention

[0005] To address this, the present invention provides an RFID-based battery identification and swapping method. This method overcomes the technical problems in the prior art, which suffer from blind anti-interference response and easy interruption of the swapping process due to the lack of verification of whether the interference is actually coupled to the key sampling circuit, by using a cross-correlation coupling verification mechanism between pulse train interference and the voltage sequence of the battery management system.

[0006] To achieve the above objectives, the present invention provides an RFID-based battery identification and swapping method, comprising: Based on the rise amplitude, duration, and waveform steepness of the signal strength sequence within the current sliding window, the determination result of the battery swapping cabinet capturing pulse train interference is determined, and the interference intensity of the pulse train interference is determined. Given that the interference intensity is a strong pulse, the cross-correlation peak value is determined based on the voltage sequence and signal strength subsequence of the battery management system in the battery swapping cabinet, wherein the signal strength subsequence is determined based on the signal strength sequence; Based on the determination result of the successful coupling of the burst interference to the sampling circuit of the battery management system, the positive feedback level of the common-mode current is determined, wherein the successful coupling of the burst interference is determined based on the cross-correlation peak value; Based on the battery swapping success rate within a preset analysis period, the frequency distribution of each interference intensity, and the frequency distribution and direction of change of risk level within a preset observation period, the anti-interference level is determined to execute the corresponding battery swapping cabinet anti-interference strategy. The risk level is determined based on the interference intensity and the positive feedback level.

[0007] Furthermore, the process of determining whether the battery swapping cabinet captures pulse train interference includes: The system determines whether the rise rate of the signal strength sequence within the current sliding window is greater than a preset amplitude threshold, whether the duration is greater than a preset duration threshold, and whether the waveform steepness is greater than a preset steepness threshold. When all three conditions are met, the system determines that the system is capturing pulse train interference. The signal strength sequence is determined based on the received signal strength indication of the RFID reader in the battery swapping cabinet.

[0008] Furthermore, the process of determining the interference intensity of a burst of interference includes: The system determines whether the rise amplitude of the pulse train interference is greater than a preset strong amplitude threshold, whether the duration is greater than a preset strong duration threshold, and whether the waveform steepness is greater than a preset strong steepness threshold. Only when all three are greater than their respective thresholds is the interference intensity determined to be a strong pulse; otherwise, it is determined to be a weak pulse.

[0009] Furthermore, the process of determining the cross-correlation peak includes: A time-domain waveform similarity analysis is performed on the signal strength subsequence and the voltage sequence. By comparing the matching degree of the two sequences at different time offsets, the value corresponding to the maximum matching degree is taken as the cross-correlation peak value.

[0010] Furthermore, the process of determining whether the burst interference has been successfully coupled to the sampling circuit of the battery management system includes: When the cross-correlation peak value is greater than the preset coupling threshold, it is determined that the pulse train interference has been successfully coupled to the sampling circuit of the battery management system; otherwise, it is determined that the coupling has not been successful.

[0011] Furthermore, the process of determining the positive feedback level of the common-mode current includes: The positive feedback index is determined based on the changing trends of the cross-correlation peak value between adjacent sliding windows and the changing trends of the effective value of the common-mode current. The positive feedback level of the common-mode current is determined based on the comparison results between the positive feedback index and the preset first positive feedback threshold and the preset second positive feedback threshold. The preset first positive feedback threshold is less than the preset second positive feedback threshold.

[0012] Furthermore, the process of determining the positive feedback index includes: The relative change in interference coupling strength is determined based on the multiple relationship between the peak values ​​of the cross-correlation between the current sliding window and the previous sliding window. The relative change in common-mode current is determined based on the multiple relationship between the current sliding window and the effective value of the common-mode current of the previous sliding window. The positive feedback index is determined based on the relative change in interference coupling strength and the relative change in common-mode current.

[0013] Furthermore, the process of determining the risk level based on the interference intensity and the positive feedback level includes: Map the interference intensity to a first threat level; Map the positive feedback level to the second threat level; The risk level is determined by the higher of the first threat level and the second threat level.

[0014] Furthermore, the process of determining the interference immunity level includes: Based on the battery swapping success rate and the frequency distribution of various interference intensities within the preset analysis period, the long-term correction factor is determined. Based on the frequency distribution and direction of change of the risk levels within the preset observation period, short-term risk factors are determined; The level of resistance to interference is determined based on the long-term correction factor and the short-term risk factor.

[0015] Furthermore, the process of executing the corresponding anti-interference strategy for the battery swapping cabinet according to the anti-interference level includes: According to the anti-interference level, a strategy action of corresponding strength is selected from the preset strategy action set and executed. Among them, the low anti-interference level corresponds to standby, the medium anti-interference level corresponds to reducing the charging power, switching the RFID working channel or increasing the RFID transmission power, and the high anti-interference level corresponds to restarting the RFID reader, suspending battery swapping or reporting an alarm.

[0016] Compared with existing technologies, the advantages of this invention lie in its ability to accurately distinguish between strong pulses and weak interference through a sliding window classification of three parameters: rise amplitude, duration, and waveform steepness, thus avoiding ineffective responses to harmless interference. It determines whether interference has successfully coupled to the battery management system sampling circuit based on cross-correlation peak values, overcoming the limitation of only detecting the existence of pulse groups and solving the problems of blind anti-interference response and easy interruption of battery swapping. The introduction of common-mode current positive feedback levels quantifies the mutual excitation degree between interference and common-mode current, providing a basis for risk classification. Furthermore, by fusing long-term correction factors based on battery swapping success rate and strong pulse frequency statistics with short-term risk factors based on the direction of risk level changes and high-risk frequencies to generate anti-interference levels, the strategy execution adapts to different electromagnetic environments, achieving a balance between long-term statistical stability and short-term response sensitivity. This significantly improves the identification reliability and battery swapping success rate of battery swapping cabinets under strong electromagnetic pulse train environments, effectively solving the technical problems of blind anti-interference response and easy interruption of the battery swapping process due to only detecting the existence of pulse groups without verifying whether the interference is truly coupled to the key sampling circuit in existing technologies. Attached Figure Description

[0017] Figure 1 This is a flowchart of the RFID-based battery identification and swapping method in this embodiment; Figure 2 A flowchart for determining the positive feedback index in this embodiment; Figure 3 A flowchart for determining the risk level in this embodiment; Figure 4 A flowchart for determining the anti-interference level in this embodiment. Detailed Implementation

[0018] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0019] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0020] In battery swapping cabinet applications, RFID readers are used to read battery tags for authentication, and their received signal strength indicators reflect the quality of the communication link in real time. When electromagnetic pulse (EMB) interference couples to the reader's receiving channel, the received signal strength indicator exhibits an abnormal waveform with a sharp increase in amplitude, a steep leading edge, and a duration of a certain period. To continuously and comprehensively capture the temporal characteristics of transient pulse trains, this embodiment employs a sliding window technique to traverse the time series of the received signal strength indicator. The sliding window has a preset window length and a preset sliding step size, sequentially extracting local subsequences along the time axis for analysis, and ensuring that adjacent windows overlap to guarantee that any pulse train occurring at any given time is covered by at least one complete window, avoiding missed detections due to window boundary alignment. Considering the typical duration of EFBs in the battery swapping cabinet environment (tens to hundreds of microseconds) and the real-time processing capability, this application sets the sampling rate to 1 MHz, the window length to 200 microseconds, the sliding step size to 50 microseconds, and the overlap rate of adjacent windows to 75%. This setting can fully cover the pulse train waveform, meet the calculation requirements of rise amplitude, duration and waveform steepness, and control the positioning error of the pulse interference occurrence time within ±100 microseconds.

[0021] Please see Figure 1 The diagram shown is a flowchart of the RFID-based battery identification and swapping method of this embodiment. This embodiment provides an RFID-based battery identification and swapping method, including: Step S1: Based on the rise amplitude, duration and waveform steepness of the signal strength sequence within the current sliding window, determine the judgment result of the battery swapping cabinet capturing pulse train interference, determine the interference intensity of the pulse train interference, so as to distinguish between the strong pulses of real threats and the harmless weak interference, thereby avoiding ineffective responses to harmless interference.

[0022] In this embodiment, the signal strength sequence is determined by performing high-speed analog-to-digital conversion and continuous sampling on the I / Q signals after demodulation by the RFID reader, and calculating the instantaneous linear power value at each sampling point, so as to characterize the time-domain waveform of the received signal strength changing with time.

[0023] In this embodiment, the rise amplitude is determined based on the difference between the maximum and minimum values ​​of the received signal strength indication within the sliding window, to characterize the rise amplitude of the burst interference.

[0024] In this embodiment, the duration is determined based on the time taken for the received signal strength indication within the sliding window to first exceed a preset reference value and then fall back below that reference value, in order to characterize the duration of the burst interference.

[0025] In this embodiment, a preset reference value is used to determine the start and end times of pulse train interference, to characterize whether the received signal strength indication shows a transient rise caused by pulse interference. Based on the background noise statistical characteristics of the RFID reader in a noise-free environment, the preset reference value is typically set between μ+2σ and μ+4σ. If the preset reference value is less than μ+2σ, frequent false triggering of pulses will occur due to random noise fluctuations; if the preset reference value is greater than μ+4σ, weaker real pulse trains will be missed. Statistical analysis shows that μ+3σ corresponds to a false alarm probability of less than 0.3% and can effectively capture real pulse trains, simultaneously meeting the requirements of anti-false triggering and detection sensitivity. Therefore, in this embodiment, the preset reference value is set to μ+3σ.

[0026] In this embodiment, the waveform steepness is determined based on the maximum rate of change of the rising edge of the received signal strength indicator within the sliding window, in order to characterize the waveform steepness of the burst interference.

[0027] Step S2: Based on the case where the interference intensity is a strong pulse, determine the cross-correlation peak value according to the voltage sequence and signal intensity subsequence of the battery management system in the battery swapping cabinet, so as to characterize the waveform matching degree of the two sequences in the time domain.

[0028] In this embodiment, the signal strength subsequence is determined by extracting from the signal strength sequence based on the occurrence time of the captured burst interference and the position of the sliding window, so as to characterize the local received signal strength indication waveform segment corresponding to the burst interference.

[0029] In this embodiment, the voltage sequence is determined based on the voltage data synchronously sampled by the battery management system within the same time period corresponding to the same sliding window, so as to characterize the waveform of the change in battery terminal voltage within that time period.

[0030] Step S3: Based on the determination result that the pulse interference is successfully coupled to the sampling circuit of the battery management system, determine the positive feedback level of the common-mode current to characterize the severity of the mutual excitation enhancement between the pulse interference and the common-mode current.

[0031] Step S4: Based on the battery swapping success rate within the preset analysis period, the frequency distribution of each interference intensity, and the frequency distribution and change direction of the risk level within the preset observation period, determine the anti-interference level and execute the corresponding battery swapping cabinet anti-interference strategy.

[0032] In this embodiment, the preset analysis period is determined based on the dual constraints of statistical reliability and response timeliness. If the period is shorter than 12 hours, the sample size of the battery swapping success rate and the frequency of interference intensity is insufficient, and the standard error of the statistical mean exceeds 0.05, resulting in frequent fluctuations in the long-term correction factor. If the period is longer than 48 hours, the response delay to gradual changes in the electromagnetic environment or equipment aging exceeds two calendar days, making it impossible to adjust the anti-interference level in a timely manner. Therefore, the preset analysis period is usually set between 12 hours and 48 hours. Based on the analysis of the distribution of the average daily number of battery swapping requests and the rate of change of the electromagnetic interference environment, 24 hours is the minimum period required for the cumulative statistics to be greater than stable. At this time, the standard error is controlled within 0.05 and the delay does not exceed one calendar day. Therefore, in this embodiment, the preset analysis period is set to 24 hours.

[0033] In this embodiment, the battery swapping success rate is the ratio of the number of successful battery swaps to the total number of battery swapping requests within a preset analysis period. It is obtained by the battery swapping cabinet controller based on the battery swapping logs to characterize the impact of long-term pulse interference on the battery swapping service.

[0034] In this embodiment, the preset observation duration is determined based on the dual constraints of trend capture sensitivity and anti-jitter performance. If the observation duration is less than 10 seconds, the sample size for risk levels is insufficient, and the linear fitting slope is easily affected by random fluctuations, leading to frequent misjudgments of the direction of change. If the observation duration is longer than 60 seconds, the response delay to sudden deterioration of interference exceeds two pulse train fluctuation cycles, making it impossible to adjust short-term risk factors in a timely manner. Therefore, the preset observation duration is usually set between 10 and 60 seconds. Based on the measured statistics of electromagnetic pulse train interference fluctuation cycles in the battery swapping cabinet environment and the analysis of sliding window overlap rate, 30 seconds can contain a sufficient number of sliding windows, approximately 600, to ensure a stable fitting slope while keeping the response delay within an acceptable range for interference trend changes. Therefore, this embodiment sets the preset observation duration to 30 seconds.

[0035] In this embodiment, the risk level is determined based on the interference intensity and the positive feedback level to characterize the degree of safety threat posed by pulse train interference to the battery swapping process.

[0036] Strong pulses are identified by jointly considering three parameters: rise amplitude, duration, and waveform steepness. Strong interference must simultaneously possess high energy, long duration, and a steep leading edge; a single parameter anomaly is insufficient to pose a threat. Cross-correlation analysis is performed between the received signal strength indication corresponding to the strong pulse and the battery management system voltage sequence. The coupling state is determined using the normalized cross-correlation peak value. The same interference source generates synchronous waveforms in both channels, and irrelevant noise cannot form this synchronization. After confirming coupling, a common-mode current is introduced. The positive feedback exponent is obtained by multiplying its effective value change rate by the cross-correlation peak change rate, which is used to quantify the mutual excitation cycle between interference and common-mode current. Based on the principle of safety priority, the higher of the interference intensity and the positive feedback level is selected to determine the risk level. A long-term correction factor and a short-term risk factor are integrated to generate an anti-interference level, enabling the strategy to adapt to both gradual environmental changes and sudden deterioration. This solves the problems of blind anti-interference and easy interruption during battery swapping caused by existing technologies that only detect the existence of pulse groups without verifying coupling.

[0037] Specifically, the process of determining whether the battery swapping cabinet has captured pulse train interference includes: The system determines whether the rise rate of the signal strength sequence within the current sliding window is greater than a preset amplitude threshold, whether the duration is greater than a preset duration threshold, and whether the waveform steepness is greater than a preset steepness threshold. When all three conditions are met, the system determines that the pulse train interference has been captured.

[0038] In this embodiment, the preset amplitude threshold is determined based on the dual constraints of anti-false triggering and detection sensitivity. If the threshold is less than 3dB, background noise fluctuations are easily misjudged as pulse train interference, leading to frequent false triggers; if the threshold is greater than 7dB, weaker real pulse trains may be missed, affecting the integrity of interference capture. Therefore, the preset amplitude threshold is usually set between 3dB and 7dB. Based on the actual measurement statistics of the background noise fluctuation range and the instantaneous rise amplitude of electromagnetic pulse trains when the power swapping cabinet is working normally, 5dB is the midpoint between the upper limit of noise fluctuation and the lower limit of pulse train rise. At this point, the false alarm probability is less than 0.5% and the weak pulse capture rate is greater than 95%. Therefore, in this embodiment, the preset amplitude threshold is set to 5dB.

[0039] In this embodiment, the preset duration threshold is determined based on the dual constraints of filtering out isolated noise and retaining true pulse trains. If the threshold is less than 5 microseconds, ordinary spike noise is easily misjudged as pulse train interference, leading to frequent false triggers; if the threshold is greater than 20 microseconds, short-duration true pulse trains may be missed, affecting the integrity of interference detection. Therefore, the preset duration threshold is usually set between 5 and 20 microseconds. Based on the measured and statistical analysis of the typical duration of electrical fast transient bursts and the duration of ordinary spike noise, 10 microseconds is the balance point between the upper limit of noise duration and the lower limit of pulse train duration. This effectively eliminates atypical pulse train interference such as isolated narrow spikes while ensuring that the duration of true pulse trains is fully captured. Therefore, in this embodiment, the preset duration threshold is set to 10 microseconds.

[0040] In this embodiment, the preset steepness threshold is determined based on the dual constraint of distinguishing between steep pulses and gentle disturbances. If the threshold is less than 1.0 dB / µs, gentle rising edges such as power fluctuations or device switching are easily misjudged as electromagnetic pulse train interference, leading to frequent false triggers. If the threshold is greater than 3.0 dB / µs, real pulse trains with steep leading edges but slightly low rates of change may be missed, affecting the capture rate of threatening interference. Therefore, the preset steepness threshold is usually set between 1.0 dB / µs and 3.0 dB / µs. Based on the measured statistics of the rising edge characteristics of electrical fast transient pulse groups and the rising edges of common interference sources, under the conditions of a sampling rate of 1 MHz and a differential interval of 1 µs, 1.5 dB / µs is the balance point between the lower limit of steep leading edges and the upper limit of gentle disturbances. It can effectively distinguish between fast-rising-edge electromagnetic pulse trains and slowly changing environmental disturbances, while also taking into account the detection sensitivity of broadband pulse trains. Therefore, in this embodiment, the preset steepness threshold is set to 1.5 dB / µs.

[0041] The rise amplitude, duration, and waveform steepness correspond to the rise amplitude, duration, and rate of change of the leading edge of the interference, respectively. Only when all three exceed their respective preset thresholds can the pulse train interference be identified as being captured. A real pulse train must simultaneously possess the physical characteristics of high energy, long duration, and steep leading edge in the time domain. Exceeding the limit of a single parameter may originate from non-threatening events such as electrostatic discharge or power fluctuations. The combined determination of the three parameters can reduce false alarms caused by background noise and occasional spikes.

[0042] Specifically, the process of determining the intensity of burst interference includes: The system determines whether the rise amplitude of the pulse train interference is greater than a preset strong amplitude threshold, whether the duration is greater than a preset strong duration threshold, and whether the waveform steepness is greater than a preset strong steepness threshold. Only when all three are greater than their respective thresholds is the interference intensity determined to be a strong pulse; otherwise, it is determined to be a weak pulse.

[0043] In this embodiment, the preset strong amplitude threshold is determined based on the dual constraints of anti-false triggering and detection sensitivity. If the threshold is less than 3dB, background noise fluctuations are easily misjudged as strong interference, leading to frequent false triggers; if the threshold is greater than 15dB, strong pulses with higher energy may be missed, affecting the accuracy of intensity grading. Therefore, the preset strong amplitude threshold is usually set between 3dB and 15dB. Based on the measured statistics of the background fluctuation range of the received signal strength indication during normal operation of the battery swapping cabinet and the instantaneous rise amplitude caused by electrical fast transient pulse groups, 10dB is the balance point between the upper limit of background fluctuation and the lower limit of strong pulse rise. This avoids misjudging background noise as strong interference while ensuring that pulse trains with energy reaching the intervention intensity are correctly identified. Therefore, this embodiment sets the preset strong amplitude threshold to 10dB. An amplitude rise of less than 10dB is classified as weak interference, and an amplitude of 10dB or more is classified as strong interference.

[0044] In this embodiment, the preset strong duration threshold is determined based on the dual constraints of filtering out short-term noise and retaining valid pulse trains. If the threshold is less than 20 microseconds, ordinary spike noise is easily misjudged as strong interference, leading to frequent false triggering of intensity classification; if the threshold is greater than 100 microseconds, short-duration real strong pulses may be missed, affecting the integrity of the classification. Therefore, the preset strong duration threshold is usually set between 20 and 100 microseconds. Based on the measured and statistical analysis of the typical duration of electrical fast transient pulse trains and the duration of ordinary spike noise, 50 microseconds is the balance point between the upper limit of noise duration and the lower limit of pulse train duration. This effectively excludes non-threatening events such as isolated narrow spikes while ensuring that pulse trains with sufficient duration are correctly classified as strong interference. Therefore, this embodiment sets the preset strong duration threshold to 50 microseconds. Durations less than 50 microseconds are classified as weak interference, and durations of 50 microseconds or more are classified as strong interference.

[0045] In this embodiment, the preset steepness threshold is determined based on the dual constraints of identifying steep pulses and filtering out gentle disturbances. If the threshold is less than 1.5 dB / µs, gentle rising edges such as power fluctuations or device switching are easily misjudged as strong interference, leading to frequent false triggering of intensity grading. If the threshold is greater than 5 dB / µs, real strong pulses with steep leading edges but slightly low rates of change may be missed, affecting the accuracy of grading. Therefore, the preset steepness threshold is usually set between 1.5 dB / µs and 5 dB / µs. Based on the measured statistics of the rising edge characteristics of electrical fast transient pulse groups and the rising edges of common interference sources, under the conditions of a sampling rate of 1 MHz and a differential interval of 1 µs, 3 dB / µs is the balance point between the lower limit of steep leading edges and the upper limit of gentle disturbances. It can effectively identify threatening pulses with steep leading edges and avoid misjudging gently changing environmental disturbances as strong interference. Therefore, in this embodiment, the preset steepness threshold is set to 3 dB / µs. Waveform steepness less than 3dB / microsecond is classified as weak interference, while steepness of 3dB / microsecond or more is classified as strong interference.

[0046] The response is triggered only when the rise amplitude, duration, and waveform steepness all exceed the strong level. This approach leverages the inherent physical properties of a real strong pulse, which must possess high energy, long duration, and a steep leading edge. Exceeding a single parameter may originate from non-threatening events such as electrostatic discharge or power fluctuations. The combined determination of these three factors can suppress false triggering caused by background noise and occasional spikes without sacrificing sensitivity to real strong pulses. This reduces unnecessary power swapping interruptions and improves operational continuity and safety redundancy.

[0047] Specifically, the process of determining the peak cross-correlation value includes: A time-domain waveform similarity analysis is performed on the signal strength subsequence and the voltage sequence. By comparing the matching degree of the two sequences at different time offsets, the value corresponding to the maximum matching degree is taken as the cross-correlation peak value.

[0048] In this embodiment, considering the unknown propagation delay of the electromagnetic pulse train coupling from the RFID antenna to the battery management system sampling circuit, and the potential fixed deviation in the sampling times of the two signal channels, the received signal strength indication subsequence and the voltage sequence may not be strictly aligned on the time axis. Therefore, a time-domain matching method is employed. The voltage sequence is successively shifted relative to the signal strength subsequence by different sampling offsets, and the matching degree of the two sequence waveforms is calculated at each offset. When the offset precisely compensates for the propagation delay and sampling deviation, the two sequence waveforms achieve optimal alignment, and the matching degree is taken as the maximum value. This maximum value is the cross-correlation peak value, used to characterize the waveform similarity of the two sequences in the optimal alignment state. Comparing the cross-correlation peak value with a preset coupling threshold determines whether the pulse train interference has successfully coupled to the battery management system sampling circuit. This process, by automatically searching for the optimal offset, eliminates the influence of unknown delays and ensures the accuracy of the coupling determination.

[0049] It is understood that this embodiment assumes that the signal strength subsequence and the voltage sequence have the same window length and that the sampling points correspond one-to-one. In actual engineering, if the sampling rates of the two sequences are different, they can be unified by interpolation resampling or downsampling; if there is a time offset, time alignment can be achieved by clock synchronization of the battery swapping cabinet or fixed delay compensation. The above processing is a conventional technical means in this field and will not be elaborated here.

[0050] By calculating the cross-correlation peak values ​​of the received signal strength indication and the battery management system voltage sequences, and utilizing the physical characteristic that the same interference source will inevitably produce synchronous waveforms in the RF channel and sampling circuit, coupled interference exhibits a high cross-correlation peak value, while uncoupled interference or independent noise shows no significant correlation. This peak value is sensitive to waveform shape but not amplitude; occasional spikes will not produce high values ​​due to low overall similarity. This accurately distinguishes whether interference has actually entered the battery sampling circuit, avoiding blindly triggering anti-interference actions simply by detecting the presence of interference, and providing a reliable coupling criterion for subsequent positive feedback quantization.

[0051] The same interference source will inevitably generate a synchronization waveform between the RF channel and the battery sampling circuit. When the burst interference successfully couples to the sampling circuit of the battery management system, the cross-correlation peak of the received signal strength indication sequence and the voltage sequence increases significantly; when there is no coupling, the two sequences remain low due to the lack of noise correlation.

[0052] Specifically, the process of determining whether the burst interference has been successfully coupled to the sampling circuit of the battery management system includes: When the cross-correlation peak value is greater than the preset coupling threshold, it is determined that the pulse train interference has been successfully coupled to the sampling circuit of the battery management system; otherwise, it is determined that the coupling has not been successful.

[0053] In this embodiment, the preset coupling threshold is determined based on the dual constraints of avoiding false positives and preventing false negatives. If the preset coupling threshold is less than 0.6, noise fluctuations in the uncoupled state are easily misjudged as successful coupling, leading to frequent triggering of unnecessary anti-interference actions; if the preset coupling threshold is greater than 0.8, true coupling in the strongly coupled state may be missed, affecting the timeliness of the interference response. Therefore, the preset coupling threshold is usually set between 0.6 and 0.8. Based on the measured statistical analysis of the cross-correlation peak values ​​of the power swapping cabinet in the uncoupled and strongly coupled states, the average peak value in the uncoupled state is approximately 0.25, and the average peak value in the strongly coupled state is approximately 0.72. 0.65 is the separation point between the two distribution ranges, which can effectively suppress noise false positives while ensuring reliable identification of true coupling. Therefore, in this embodiment, the preset coupling threshold is set to 0.65.

[0054] By comparing the cross-correlation peak value with a preset coupling threshold, when the cross-correlation peak value reaches or exceeds the preset coupling threshold, it is determined that the interference has invaded the sampling circuit; otherwise, it is determined that the interference only exists in the receiving channel. This avoids performing isolation or interruption operations on harmless interference that only exists in the receiving channel, and makes the anti-interference action accurately focus on the coupling state that truly threatens battery sampling, thereby improving the pertinence of the anti-interference strategy and the operational reliability of the battery swapping cabinet.

[0055] When a burst of interference successfully couples to the battery management system's sampling circuit, it induces a common-mode current on the battery casing. This common-mode current further deteriorates the RFID reader's receiving environment, leading to an increase in subsequent cross-correlation peaks. The increased cross-correlation peaks, in turn, indicate stronger interference coupling, thus forming a bidirectionally reinforced closed loop. This embodiment uses positive feedback levels to quantify this self-deteriorating dynamic process into discrete threat levels, providing a basis for subsequent risk level determination and differentiated anti-interference strategies.

[0056] Specifically, the process of determining the positive feedback level of the common-mode current includes: The positive feedback index is determined based on the changing trends of the cross-correlation peak value between adjacent sliding windows and the changing trends of the effective value of the common-mode current. The positive feedback level of the common-mode current is determined based on the comparison results between the positive feedback index and the preset first positive feedback threshold and the preset second positive feedback threshold.

[0057] In this embodiment, the preset first positive feedback threshold is determined based on the dual constraints of suppressing false noise and retaining the true positive feedback attenuation. If the threshold is less than 0.8, the exponential downward deviation caused by measurement noise is easily misjudged as positive feedback attenuation, leading to frequent decreases in the positive feedback level; if the threshold is greater than 0.95, the exponential decrease during the true attenuation of positive feedback may be missed, affecting the timeliness of the level response. Therefore, the preset first positive feedback threshold is usually set between 0.8 and 0.95. Statistical analysis of the normal fluctuation range of the positive feedback index in the field measurement of the battery swapping cabinet shows that the downward fluctuation of the index near the benchmark value of 1 is usually within -10%. Taking 0.9 as the lower boundary can effectively filter out false attenuation caused by noise and reliably capture the true weakening trend of positive feedback. Therefore, in this embodiment, the preset first positive feedback threshold is set to 0.9.

[0058] In this embodiment, the preset second positive feedback threshold is determined based on the dual constraints of suppressing false noise and capturing true positive feedback enhancement. If the threshold is less than 1.05, the upper deviation caused by measurement noise is easily misjudged as positive feedback enhancement, leading to frequent increases in the positive feedback level; if the threshold is greater than 1.2, the exponential increase when the positive feedback is truly enhanced may be missed, affecting the timely response to serious threats. Therefore, the preset second positive feedback threshold is usually set between 1.05 and 1.2. Statistical analysis of the normal fluctuation range of the positive feedback index in the field measurement of the battery swapping cabinet shows that the upper fluctuation range of the index near the benchmark value of 1 is usually within +10%. Taking 1.1 as the upper boundary can effectively filter out false increases caused by noise and accurately identify the true amplification of positive feedback. Therefore, in this embodiment, the preset second positive feedback threshold is set to 1.1.

[0059] When the positive feedback index is less than or equal to the preset first positive feedback threshold, it is determined to be positive feedback level I, indicating that no significant positive feedback has been formed between the interference and the common-mode current or the positive feedback is decaying; when the positive feedback index is greater than the preset first positive feedback threshold and less than the preset second positive feedback threshold, it is determined to be positive feedback level II, indicating that there is a certain degree of positive feedback but it has not yet reached a dangerous level; when the positive feedback index is greater than or equal to the preset second positive feedback threshold, it is determined to be positive feedback level III, indicating that the positive feedback has been significantly enhanced and poses a serious threat.

[0060] By setting a first positive feedback threshold and a second positive feedback threshold to form a symmetrical hysteresis comparison interval centered on 1, small fluctuations in the index around 1 will not cause frequent level jumps. This is because measurement noise and environmental disturbances only cause local random walks. Only when the index continuously rises above the second positive feedback threshold or continuously falls below the preset first positive feedback threshold will a level change be triggered. This filters out spurious changes while preserving the true positive feedback trend, improving the stability of level determination and preventing the anti-interference strategy from over-responding due to meaningless jumps.

[0061] The positive feedback index combines the changing trends of the cross-correlation peak between adjacent windows and the effective value of the common-mode current into a single value, so as to quantitatively characterize the strength of the positive feedback loop between the interference and the common-mode current.

[0062] Please see Figure 2 As shown, this is a flowchart for determining the positive feedback index in this embodiment.

[0063] Specifically, the process of determining the positive feedback exponent to characterize the degree of synchronous enhancement of interference and common-mode current includes: Based on the multiple relationship between the cross-correlation peak values ​​of the current sliding window and the previous sliding window, the relative change in interference coupling strength is determined to characterize the increasing or decreasing trend of interference coupling strength between adjacent windows. Based on the multiple relationship between the effective value of the common-mode current of the current sliding window and the previous sliding window, the relative change of the common-mode current is determined to characterize the enhancement or decay trend of the common-mode current between adjacent windows. Based on the relative change in interference coupling strength and the relative change in common-mode current, a positive feedback index is determined to characterize whether a synchronously enhanced positive feedback loop is formed between the interference and the common-mode current and its severity.

[0064] Since positive feedback exhibits a progressively increasing synergistic enhancement over time, it is necessary to compare data from adjacent windows to capture dynamic trends. Therefore, it is required to obtain the RMS value of the common-mode current and the peak value of the cross-correlation between the current window and the previous window. In this embodiment, the RMS value of the common-mode current is obtained by calculating the root mean square value of the acquired common-mode current sequence, and the sampling rate is consistent with the received signal strength indication; the peak value of the cross-correlation is directly obtained from the previous steps.

[0065] To eliminate the interference of absolute amplitude differences between different windows on trend judgment and to make the changes comparable in a normalized manner, this embodiment calculates the ratio of the current cross-correlation peak value to the previous cross-correlation peak value as the relative change in interference coupling strength; and calculates the ratio of the current effective value of common-mode current to the effective value of common-mode current in the previous window as the relative change in common-mode current.

[0066] The physical nature of positive feedback requires that the interference coupling strength and common-mode current must increase synchronously. This necessitates combining the trends of these two dimensions into a continuous index using a logical AND operation. In this embodiment, the positive feedback index is obtained by multiplying the relative change in interference coupling strength and the relative change in common-mode current.

[0067] By multiplying the relative change in interference coupling strength with the relative change in common-mode current, and leveraging the sensitivity of multiplication to synchronous changes in two dimensions, the positive feedback exponent only increases significantly when both relative changes are greater than 1. This accurately distinguishes between genuine positive feedback loops and single-dimensional random disturbances. This method provides a reliable numerical basis for classifying positive feedback levels, avoids misjudgments caused by random fluctuations in one dimension, reduces unnecessary anti-interference actions, and improves the system's response accuracy and the continuity of the battery swapping process.

[0068] To quantify the overall risk level of current pulse train interference to battery swapping safety, it is necessary to integrate the two independent threats of interference intensity and positive feedback level into a comprehensive index, providing a unified risk basis for subsequent anti-interference level decisions. Please refer to [link / reference]. Figure 3 As shown, it is a flowchart for determining the risk level in this embodiment.

[0069] Specifically, the process of determining the risk level based on the interference intensity and the positive feedback level includes: The interference intensity is mapped to a first threat level to characterize the direct threat that the energy, duration, and steepness of the pulse train interference itself pose to battery swapping safety. The positive feedback level is mapped to a second threat level to characterize the severity of positive feedback degradation caused by the mutual excitation of interference and common-mode current. The risk level is determined by the higher of the first threat level and the second threat level to characterize the overall security risk after merging the two independent threats.

[0070] Since interference intensity directly reflects the energy, duration, and steepness of the pulse train itself, it is the most direct source of threat and therefore needs to be quantified into a threat level first. In this embodiment, interference intensity is mapped to a first threat level. Specifically, strong pulses correspond to a high level, medium pulses to a medium level, and weak pulses to a low level. A strong pulse indicates that the interference has reached a strong level in at least one dimension of energy, duration, or steepness, posing a direct threat to the battery swapping cabinet; a medium pulse indicates that the interference has a certain intensity but has not yet reached a dangerous level; and a weak pulse indicates that the interference is weak and has basically no impact on the normal operation of the battery swapping cabinet.

[0071] The positive feedback level characterizes the degree of deterioration caused by the mutual excitation of interference and common-mode current. It is independent of the interference intensity and therefore needs to be quantified separately as a separate threat level. In this embodiment, the positive feedback level is mapped to a second threat level. Specifically, positive feedback level III corresponds to a high level, positive feedback level II corresponds to a medium level, and positive feedback level I corresponds to a low level. Positive feedback level III indicates that interference and common-mode current have formed a significant bidirectional reinforcing cycle, and the threat continues to escalate; positive feedback level II indicates the existence of some positive feedback but it has not yet deteriorated; positive feedback level I indicates that positive feedback has not formed or is attenuating.

[0072] Since the two dimensions are independent and either dimension reaching a high level is sufficient to trigger an accident, adopting the higher-order approach prioritizes safety in reflecting the overall risk. In this embodiment, the first threat level and the second threat level are compared, and the higher of the two is taken as the final risk level. This comparison method only needs to determine the level of discrete threats, without complex calculations, ensuring the real-time response capability of the battery swapping cabinet's embedded system. Reaching a high level in either dimension triggers a high risk, ensuring that strong anti-interference strategies are executed promptly when serious threats occur, thereby enhancing the safety protection capability of the battery swapping cabinet in strong electromagnetic pulse train environments.

[0073] By taking the higher of the two threat levels, interference intensity and positive feedback level, as the risk level, a high-risk safety judgment logic can be triggered by any independent hazard factor. This ensures that both the direct impact of a strong pulse and the continuous deterioration of positive feedback can promptly trigger a high-risk response, thereby preventing serious threats from being missed.

[0074] Neither long-term statistics nor short-term dynamic quantities can fully characterize the changing patterns of the electromagnetic environment. Long-term statistics can smooth out random fluctuations and reflect the environmental baseline, but their response is lagging; short-term dynamic quantities can sensitively capture sudden deterioration, but are easily affected by random disturbances. Therefore, it is necessary to integrate the two into a comprehensive anti-interference level to balance stability and sensitivity, thereby guiding the battery swapping cabinet to implement a hierarchical adaptive anti-interference strategy.

[0075] Please see Figure 4 As shown, it is a flowchart for determining the anti-interference level in this embodiment.

[0076] Specifically, the process of determining the anti-interference level includes: Based on the battery swapping success rate and frequency distribution of each interference intensity within the preset analysis period, a long-term correction factor is determined to characterize the average severity of the long-term electromagnetic environment and the deviation of detection sensitivity. Based on the frequency distribution and direction of change of the risk level within the preset observation period, short-term risk factors are determined to characterize the dynamic changes and intensity of short-term risks. Based on the long-term correction factor and the short-term risk factor, the anti-interference level is determined to characterize the comprehensive risk level after integrating long-term statistics and short-term dynamics, and is used for the implementation of graded anti-interference strategies.

[0077] The long-term electromagnetic environment exhibits gradual variability; for example, seasonal changes, equipment aging, or the increase or decrease of surrounding interference sources can all alter the baseline level of interference. Single short-term detection cannot reflect this long-term trend, and the battery swapping success rate directly reflects the actual impact of interference on business operations. In this embodiment, a smoothing estimate is performed based on statistical data within a preset analysis period to obtain a factor capable of correcting the long-term baseline of the interference resistance level, namely, the long-term correction factor.

[0078] Specifically, since the battery swapping success rate directly reflects the actual impact of service interference, and the frequency of strong pulses characterizes the severity of long-term interference, relying solely on a single indicator cannot distinguish between environmental degradation and detection sensitivity deviation. Therefore, it is necessary to simultaneously statistically analyze both as inputs for calculating the long-term correction factor and set corresponding threshold ranges. In this embodiment, the preset analysis period is set to 24 hours, based on the average daily battery swapping requests, to ensure statistical significance. The battery swapping success rate is obtained as the ratio of successful battery swaps to the total number of battery swapping requests. The frequency of strong pulses is obtained by statistically analyzing the interference intensity judgment results of all sliding windows within the preset analysis period. Specifically, after completing the interference intensity judgment in each sliding window, if the judgment result is a strong pulse, the event is counted once. At the end of the preset analysis period, the total number of accumulated strong pulse events is taken as the strong pulse frequency within that period.

[0079] To prevent frequent fluctuations of the long-term correction factor near the boundary and to achieve step-by-step smooth adjustment, a baseline operating point needs to be set, and the correction factor is adjusted proportionally based on the deviation between the actual statistical value and the baseline. In this embodiment, the operating baseline is set when the battery swapping success rate is a preset first success rate threshold and the strong pulse frequency is a preset first frequency threshold, corresponding to the critical state requiring intervention. At this point, the long-term correction factor is set to 1. In this embodiment, historical operational data of multiple battery swapping cabinets are collected over multiple preset analysis periods. Three variables are recorded for each period: the deviation of the battery swapping success rate from the operating baseline, the deviation of the strong pulse frequency from the operating baseline, and the change in the battery swapping success rate in the next period. Using the battery swapping success rate deviation and the strong pulse frequency deviation as two independent variables, and the change in the battery swapping success rate in the next period as the dependent variable, a multiple linear regression is performed to fit the regression coefficients of the two independent variables. The regression coefficient of the battery swapping success rate deviation is 0.03, and the regression coefficient of the strong pulse frequency deviation is 0.02, and so on. To prevent excessive reduction of the correction factor in a clean environment, reduction coefficients of 0.02 and 0.05 are symmetrically set. These symmetry coefficients are obtained through empirical adjustment or another set of regression analyses to ensure that the anti-interference capability is not excessively reduced when the environment is favorable. In this embodiment, the initial value 1 is added to the above-mentioned increases and decreases, and the results are limited by upper and lower limits to obtain the long-term correction factor within the current preset analysis period.

[0080] Understandably, to prevent the long-term correction factor from deviating from a reasonable range and causing system malfunctions, an upper limit of 2.0 and a lower limit of 0.4 are set for the long-term correction factor. These upper and lower limits are calibrated through field tests: In extreme interference environments, such as when the battery swapping station is near a high-power wireless charging pile or inverter cluster, multiple sets of continuous, preset analysis period operational data are collected. When the correction factor exceeds 2.0, the anti-interference level remains consistently high, causing unnecessary reductions in charging power or channel switching, affecting battery swapping efficiency; when the correction factor is below 0.4, the anti-interference capability is insufficient, resulting in a significant increase in battery swapping failure rate. Multiple tests have verified that setting the upper and lower limits to 2.0 and 0.4 respectively can cover the adjustment needs under extreme environments while ensuring system stability.

[0081] In this embodiment, the preset first success rate threshold is a boundary value used to determine whether the battery swapping success rate is lower than the acceptable lower limit, thus characterizing whether the battery swapping process is significantly interfered with. In this embodiment, the preset first success rate threshold is determined based on the dual constraints of anti-false triggering and timely response. If the threshold is below 70%, occasional drops in battery swapping success rate will frequently trigger an increase in the long-term correction factor, leading to unnecessary anti-interference upgrades; if the threshold is above 85%, a drop in success rate caused by actual interference may be missed, delaying the anti-interference response. Therefore, the preset first success rate threshold is usually set between 70% and 85%. Statistical analysis of the acceptable lower limit of battery swapping success rate in the battery swapping cabinet industry's operational statistics shows that 80% is the critical point that can distinguish between occasional fluctuations and significant interference. A value below this indicates that the battery swapping process has been significantly interfered with, requiring an increase in the long-term correction factor. Therefore, in this embodiment, the preset first success rate threshold is set to 80%.

[0082] In this embodiment, the preset second success rate threshold is the boundary value for determining whether the battery swapping success rate has reached an excellent level, thus characterizing whether the battery swapping service has reached an excellent level and indirectly reflecting that the electromagnetic environment has no significant impact on the battery swapping success rate. In this embodiment, the preset second success rate threshold is determined based on the dual constraints of avoiding excessive intervention and maintaining sensitivity. If the threshold is below 90%, even slight fluctuations in the success rate will trigger a reduction in the long-term correction factor, which may lead to an excessive weakening of anti-interference capabilities; if the threshold is above 98%, the opportunity to reduce intervention in a clean environment will be missed, resulting in unnecessary strategy execution. Therefore, the preset second success rate threshold is usually set between 90% and 98%. Statistical analysis of the excellent level benchmark value in battery swapping cabinet operation shows that 95% is a reasonable boundary that can characterize a clean electromagnetic environment without excessive intervention. Values ​​higher than this, combined with low strong pulse frequencies, indicate a clean electromagnetic environment, and the long-term correction factor can be appropriately reduced. Therefore, in this embodiment, the preset second success rate threshold is set to 95%.

[0083] In this embodiment, the preset first frequency threshold is the boundary value used to determine whether the frequency of strong pulses within a preset analysis period exceeds the normal range, thus characterizing the density of long-term electromagnetic interference. The preset first frequency threshold is determined based on the dual constraints of anti-false triggering and timely response. If the threshold is lower than 5 times / day, occasional strong pulse events will frequently trigger an increase in the long-term correction factor, leading to unnecessary anti-interference upgrades; if the threshold is higher than 15 times / day, truly dense strong pulse interference may be missed, delaying the anti-interference response. Therefore, the preset first frequency threshold is usually set between 5 times / day and 15 times / day. Statistical analysis of the relationship between the frequency of strong pulses and the success rate of battery swapping in typical operational data of battery swapping cabinets shows that 10 times / day is the critical point that can distinguish between occasional interference and dense threats. When a strong pulse occurs on average every 2.4 hours, it is sufficient to have a significant impact on the success rate of battery swapping. Therefore, in this embodiment, the preset first frequency threshold is set to 10 times / day.

[0084] In this embodiment, the preset second frequency threshold is a boundary value used to determine whether the frequency of strong pulses within a preset analysis period is below a reasonable level, thus characterizing the sparseness of strong pulse interference and directly reflecting the cleanliness of the electromagnetic environment. This threshold is set based on the statistical characteristics of the frequency of strong pulses in a clean environment. If the threshold is too low, for example, below 2 times / day, insufficient sporadic strong pulse events will prevent the long-term correction factor from decreasing in a timely manner, missing the opportunity to reduce the intervention intensity. If the threshold is too high, for example, above 8 times / day, even a large number of strong pulse events will trigger the factor to decrease, potentially leading to excessive weakening of the anti-interference capability. Therefore, the preset second frequency threshold is typically set between 2 times / day and 5 times / day. Statistical analysis of the frequency of strong pulses in a clean environment within typical operating data of the battery swapping cabinet shows that when there are 2 or fewer strong pulse events per day, the electromagnetic environment has virtually no impact on the battery swapping success rate, and the intervention intensity can be appropriately reduced. In this embodiment, the preset second frequency threshold is set to 2 times / day. When the frequency of strong pulses is below this threshold and the battery swapping success rate is higher than the preset second success rate threshold, it indicates that the electromagnetic environment is clean, and the long-term correction factor should be reduced.

[0085] Short-term risk factors are used to capture the changing trends and intensity of risk levels within a short time window, enabling rapid response to sudden and deteriorating disruptions. Because short-term risk dynamics are characterized by suddenness and high intensity, it is necessary to examine both the direction of risk level changes and the frequency of high-risk occurrences.

[0086] Since discrete risk levels cannot directly reflect the overall trend, the observation period needs to be divided into a preset number of sub-intervals. The average risk level within each sub-interval is calculated and made continuous, with low risk assigned a value of 1, medium risk assigned a value of 2, and high risk assigned a value of 3. This yields an average value sequence, which is then linearly fitted to obtain the slope. The slope is then further linearly fitted to the average value sequence to obtain the slope. The direction of change is determined based on the slope. Specifically, when the slope is greater than a preset slope threshold, it is determined to be an upward direction; when the slope is less than a negative preset slope threshold, it is determined to be a downward direction; otherwise, it is a stable direction.

[0087] In this embodiment, the preset number of sub-intervals refers to the number of consecutive equal-length sub-intervals into which the preset observation duration is divided. This is used to transform the discrete risk level sequence into a continuous average value sequence for linear fitting, thereby determining the direction of risk level change. The selection of this number is based on the dual constraints of fitting stability and response timeliness. If the number of sub-intervals is less than 5, there are insufficient data points for fitting, and the slope is easily affected by fluctuations in a single sub-interval, leading to misjudgments of the direction of change. If the number of sub-intervals is more than 10, the time of each sub-interval is too short, the average value fluctuation increases, and the computational load of fitting increases. Therefore, the preset number of sub-intervals is usually set between 5 and 10. Analysis of the number of sliding windows and fitting stability within a 30-second observation duration shows that 6 sub-intervals provide sufficient data points to ensure the statistical stability of the linear fitting, and each sub-interval contains a sufficient number of sliding windows to ensure the reliability of the average value. Therefore, this embodiment sets the preset number of sub-intervals to 6.

[0088] In this embodiment, the preset slope threshold is a boundary value that distinguishes between random fluctuations and continuous trends in the average risk level sequence, used to determine the direction of risk level change. This threshold is set based on the difference in the magnitude of changes in the average value sequence during normal fluctuations and continuous deterioration: under normal fluctuations, the random change in the average value of adjacent sub-intervals is usually no more than 0.1, while during continuous deterioration, the increment of the average value is generally more than 0.2. If the threshold is below 0.05, normal fluctuations are easily misjudged as continuous changes, leading to frequent jumps in the direction of change; if the threshold is above 0.2, the true deterioration trend may be missed, delaying the response. Therefore, the preset slope threshold is usually set between 0.05 and 0.2. In this embodiment, the preset slope threshold is set to 0.1, a value that matches the upper limit of normal fluctuations, effectively filtering out random fluctuations and truly capturing continuous upward or downward trends, thereby ensuring the reliability of the risk level change direction determination.

[0089] In this embodiment, the number of times a high-risk level occurs within a preset observation period is counted to characterize the density of short-term interference.

[0090] In this embodiment, the preset high-risk frequency threshold is a boundary value that distinguishes between occasional interference and dense threats, used to determine whether the high-risk level has reached a frequent occurrence state within a preset observation period. This threshold is set based on the total number of sliding windows within the preset observation period and the statistical characteristics of the critical point for dense interference occurrences. If the threshold is below 2 times, occasional high-risk events are easily misjudged as dense threats, leading to an excessive increase in short-term risk factors; if the threshold is above 5 times, genuine high-risk dense occurrences may be missed, delaying anti-interference response. Therefore, the preset high-risk frequency threshold is usually set between 2 and 5 times. Statistical analysis of approximately 600 sliding windows within a 30-second observation period shows that when high risk occurs 3 times, it occurs on average once every 10 seconds. Actual measurements indicate that the interference has entered a frequent occurrence state at this point, effectively distinguishing between occasional interference and dense threats. Therefore, this embodiment sets the preset high-risk frequency threshold to 3 times.

[0091] The short-term risk factor is used to capture short-term deterioration trends and concentrated risks. Its value is determined based on field test calibration results. For scenarios with an upward trend and the number of high-risk events reaching or exceeding a preset high-risk event threshold, candidate values ​​of 1.1, 1.2, and 1.3 are tested sequentially, with the improvement in battery swapping success rate and the number of strategy jitters used as evaluation indicators. Experimental results show that: a value of 1.1 indicates insufficient anti-interference strength and no significant improvement in battery swapping success rate; a value of 1.3 increases strategy jitter, and its product with the long-term correction factor easily exceeds the upper limit of the anti-interference level of 2.5, leading to over-intervention; a value of 1.2 achieves the best balance between improved battery swapping success rate and strategy stability. Therefore, the short-term risk factor for this scenario is set to 1.2. For scenarios with a downward trend and the number of high-risk events below a preset high-risk event threshold, candidate values ​​of 0.8, 0.9, and 1.0 are tested sequentially. Experimental results show that a factor of 0.9 effectively reduces unnecessary intervention without causing a slow response when the actual situation worsens due to an excessively low factor; a factor of 0.8 results in excessive response decay, while a factor of 1.0 indicates excessive intervention intensity. Therefore, the short-term risk factor for this scenario is set to 0.9. Values ​​for other combined scenarios—1.1 for rising and below the threshold, 1.0 for falling and reaching or exceeding the threshold, and 1.0 for stable conditions—are also based on experimental calibration. These values ​​are matched with the adjustment step size of the long-term correction factor, with a range limited to 0.8 to 1.3 to ensure that the anti-interference level changes smoothly after fusion and does not exceed the preset upper and lower limits.

[0092] Long-term correction factors and short-term risk factors characterize electromagnetic interference risk at different time scales. The long-term correction factor reflects the environmental baseline and detection sensitivity deviation, while the short-term risk factor captures sudden deterioration trends and intensity. They are independent and complementary; multiplicative fusion reflects a synergistic effect and avoids the smoothing effect of weighted averaging on extreme cases. Simultaneously, to prevent the fusion result from exceeding a reasonable range and causing system response malfunction, upper and lower limits need to be set for amplitude limitation. In this embodiment, the long-term correction factor and the short-term risk factor are multiplied to obtain the initial anti-interference level. The long-term correction factor provides the long-term baseline, while the short-term risk factor provides dynamic correction. To prevent numerical overflow, the upper limit of the anti-interference level is set to 2.5, and the lower limit to 0.5. These upper and lower limits are estimated based on the product of the upper and lower limits of the long-term correction factor and the short-term risk factor.

[0093] By using a long-term correction factor to statistically smooth the battery swapping success rate and the frequency of strong pulses within a preset analysis period, frequent fluctuations in the anti-interference level caused by a single battery swapping failure or a single strong pulse event are avoided. A short-term risk factor is used to rapidly respond to changes in the risk level and the frequency of high-risk occurrences within a preset observation period, ensuring timely upgrades to the anti-interference level when the risk level continues to rise and the number of high-risk occurrences reaches a preset threshold. By productizing the long-term correction factor and the short-term risk factor, a balance is achieved between long-term statistical stability and short-term dynamic sensitivity in the anti-interference level. This results in a reasonable trade-off between system stability and response sensitivity, significantly improving the battery swapping cabinet's adaptability and battery swapping success rate in strong electromagnetic pulse environments.

[0094] Because the anti-interference level integrates long-term environmental statistics and short-term dynamic risks, its numerical value directly represents the comprehensive degree of the current interference threat. The higher the threat, the greater the required anti-interference strength. The anti-interference level is divided into three levels—low, medium, and high—based on numerical ranges, and each level is mapped to a set of strategy actions with increasing intervention intensity to achieve a precise match between the degree of interference and the strength of the response.

[0095] In this embodiment, actual operating data of multiple battery swapping cabinets under typical electromagnetic environments were selected, and at least 50 sets of anti-interference level samples were collected for each of the clean environment, moderate interference environment, and severe interference environment. Specifically, a clean environment is defined as a battery swapping success rate higher than a preset second success rate threshold and a strong pulse frequency lower than a preset second frequency threshold; a moderate interference environment is defined as a battery swapping success rate between a preset first success rate threshold and a preset second success rate threshold, and a strong pulse frequency between a preset second frequency threshold and a preset first frequency threshold; a severe interference environment is defined as a battery swapping success rate lower than a preset first success rate threshold, a strong pulse frequency higher than a preset first frequency threshold, and a positive feedback level reaching a preset third positive feedback level. Statistical analysis shows that the mean value of the anti-interference level samples in the clean environment is lower than 1.0, and the maximum value does not exceed 1.0; the mean value of the anti-interference level samples in the moderate interference environment is between 1.0 and 1.8, with over 90% of the samples falling within the 1.0 to 1.8 range; and the mean value of the anti-interference level samples in the severe interference environment is higher than 1.8, and the minimum value is not lower than 1.8. Therefore, using 1.0 as the dividing threshold between low and medium levels, and 1.8 as the dividing threshold between medium and high levels, can ensure that the classification intervals correspond well with the risk levels under typical electromagnetic environments.

[0096] The anti-interference level is a comprehensive index obtained by combining the product of long-term correction factors and short-term risk factors. Its value directly represents the comprehensive degree of electromagnetic interference threatening battery swapping safety.

[0097] Specifically, the process of implementing the corresponding anti-interference strategy for the battery swapping cabinet according to the anti-interference level includes: According to the anti-interference level, a strategy action of corresponding strength is selected from the preset strategy action set and executed. Among them, the low anti-interference level corresponds to standby, the medium anti-interference level corresponds to reducing the charging power, switching the RFID working channel or increasing the RFID transmission power, and the high anti-interference level corresponds to restarting the RFID reader, suspending battery swapping or reporting an alarm.

[0098] Specifically, when the anti-interference level is in the low range, it indicates that the current electromagnetic environment is clean or the interference is weak, and it has no substantial impact on the battery swapping process. Therefore, only routine monitoring needs to be maintained, and no active intervention is required. In this embodiment, the low range corresponds to an anti-interference level of less than 1.0, and a standby maintenance strategy is implemented.

[0099] Specifically, when the anti-interference level is in the medium range, it indicates that there is interference of a certain intensity but it has not yet formed positive feedback deterioration. Mild anti-interference actions are needed to improve the identification environment. In this embodiment, the medium range corresponds to an anti-interference level greater than or equal to 1.0 and less than 1.8. At least one action is selected from a preset set of low-intensity strategy actions, including reducing charging power, switching the RFID working channel, and increasing RFID transmission power. The specific action can be flexibly determined according to the real-time status and the on-site electromagnetic environment. For example, switching the working channel can be tried first; if this is ineffective, then reducing the charging power or increasing the transmission power can be considered.

[0100] Specifically, when the anti-interference level is in the high range, it indicates severe interference or a positive feedback loop that has worsened the situation, posing an urgent threat to battery swapping safety and necessitating mandatory measures. In this embodiment, the high-level range corresponds to an anti-interference level greater than or equal to 1.8. At least one of the preset high-intensity strategy actions is selected for execution. This strategy action set includes restarting the RFID reader, suspending the current battery swapping process, and sending an alarm to the operation and maintenance platform. Typically, suspending battery swapping and reporting an alarm are executed simultaneously, and a restart operation is performed as needed.

[0101] By dividing the anti-interference level into three ranges—low, medium, and high—and mapping them to different intervention intensities, the intervention intensity monotonically increases with the threat level. This hierarchical mapping maintains monitoring only when the threat is minor to avoid over-response and reduced battery swapping efficiency, while enforcing mandatory measures promptly when the threat is severe to ensure battery swapping safety, thereby improving the execution efficiency of the anti-interference strategy and the success rate of battery swapping.

[0102] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A battery identification and swapping method based on RFID, characterized in that, include: Based on the rise amplitude, duration, and waveform steepness of the signal strength sequence within the current sliding window, the determination result of the battery swapping cabinet capturing pulse train interference is determined, and the interference intensity of the pulse train interference is determined. Given that the interference intensity is a strong pulse, the cross-correlation peak value is determined based on the voltage sequence and signal strength subsequence of the battery management system in the battery swapping cabinet, wherein the signal strength subsequence is determined based on the signal strength sequence; Based on the determination result of the successful coupling of the burst interference to the sampling circuit of the battery management system, the positive feedback level of the common-mode current is determined, wherein the successful coupling of the burst interference is determined based on the cross-correlation peak value; Based on the battery swapping success rate within a preset analysis period, the frequency distribution of each interference intensity, and the frequency distribution and direction of change of risk level within a preset observation period, the anti-interference level is determined to execute the corresponding battery swapping cabinet anti-interference strategy. The risk level is determined based on the interference intensity and the positive feedback level.

2. The RFID-based battery identification and swapping method according to claim 1, characterized in that, The process of determining whether the battery swapping cabinet can capture pulse train interference includes: The system determines whether the rise rate of the signal strength sequence within the current sliding window is greater than a preset amplitude threshold, whether the duration is greater than a preset duration threshold, and whether the waveform steepness is greater than a preset steepness threshold. When all three conditions are met, the system determines that the system is capturing pulse train interference. The signal strength sequence is determined based on the received signal strength indication of the RFID reader in the battery swapping cabinet.

3. The RFID-based battery identification and swapping method according to claim 2, characterized in that, The process of determining the intensity of burst interference includes: The system determines whether the rise amplitude of the pulse train interference is greater than a preset strong amplitude threshold, whether the duration is greater than a preset strong duration threshold, and whether the waveform steepness is greater than a preset strong steepness threshold. Only when all three are greater than their respective thresholds is the interference intensity determined to be a strong pulse; otherwise, it is determined to be a weak pulse.

4. The RFID-based battery identification and swapping method according to claim 1, characterized in that, The process of determining the peak cross-correlation value includes: A time-domain waveform similarity analysis is performed on the signal strength subsequence and the voltage sequence. By comparing the matching degree of the two sequences at different time offsets, the value corresponding to the maximum matching degree is taken as the cross-correlation peak value.

5. The RFID-based battery identification and swapping method according to claim 1, characterized in that, The process of determining whether the burst interference has been successfully coupled to the sampling circuit of the battery management system includes: When the cross-correlation peak value is greater than the preset coupling threshold, it is determined that the pulse train interference has been successfully coupled to the sampling circuit of the battery management system; otherwise, it is determined that the coupling has not been successful.

6. The RFID-based battery identification and swapping method according to claim 1, characterized in that, The process of determining the positive feedback level of the common-mode current includes: The positive feedback index is determined based on the changing trends of the cross-correlation peak value between adjacent sliding windows and the changing trends of the effective value of the common-mode current. The positive feedback level of the common-mode current is determined based on the comparison results between the positive feedback index and the preset first positive feedback threshold and the preset second positive feedback threshold. The preset first positive feedback threshold is less than the preset second positive feedback threshold.

7. The RFID-based battery identification and swapping method according to claim 6, characterized in that, The process of determining the positive feedback exponent includes: The relative change in interference coupling strength is determined based on the multiple relationship between the peak values ​​of the cross-correlation between the current sliding window and the previous sliding window. The relative change in common-mode current is determined based on the multiple relationship between the current sliding window and the effective value of the common-mode current of the previous sliding window. The positive feedback index is determined based on the relative change in interference coupling strength and the relative change in common-mode current.

8. The RFID-based battery identification and swapping method according to claim 1, characterized in that, The process of determining the risk level based on the interference intensity and the positive feedback level includes: Map the interference intensity to a first threat level; Map the positive feedback level to the second threat level; The risk level is determined by the higher of the first threat level and the second threat level.

9. The RFID-based battery identification and swapping method according to claim 1, characterized in that, The process of determining the interference immunity level includes: Based on the battery swapping success rate and the frequency distribution of various interference intensities within the preset analysis period, the long-term correction factor is determined. Based on the frequency distribution and direction of change of the risk levels within the preset observation period, short-term risk factors are determined; The level of resistance to interference is determined based on the long-term correction factor and the short-term risk factor.

10. The RFID-based battery identification and swapping method according to claim 1, characterized in that, The process of implementing the corresponding anti-interference strategy for the battery swapping cabinet according to the anti-interference level includes: According to the anti-interference level, a strategy action of corresponding strength is selected from the preset strategy action set and executed. Among them, the low anti-interference level corresponds to standby, the medium anti-interference level corresponds to reducing the charging power, switching the RFID working channel or increasing the RFID transmission power, and the high anti-interference level corresponds to restarting the RFID reader, suspending battery swapping or reporting an alarm.

Images