Anti-theft Locking Control Method for Electric Bicycles Based on Worm Gear Drive

The anti-theft locking control method for electric bicycles driven by worm gear transmission utilizes spectrum analysis and risk scoring models, combined with impact displacement weights, to distinguish between environmental disturbances and prying behavior. This solves the misjudgment problem in the anti-theft control of electric bicycles and achieves higher anti-theft monitoring accuracy.

CN123104823APending Publication Date: 2026-11-17ZHONGYIN INTELLIGENT TECHNOLOGY (WUXI) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611322429.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-28
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing anti-theft locking control methods for electric bicycles are prone to misinterpreting environmental disturbances as lock-picking behavior in complex environments, resulting in low accuracy of anti-theft monitoring and control.

Method used

An anti-theft locking control method for electric bicycles using worm gear transmission is proposed. By synchronously acquiring triaxial acceleration signals and environmental characteristic signals, setting a time window, performing spectrum analysis and risk scoring models, and combining impact displacement weights, the method distinguishes between environmental disturbances and prying behavior, thereby improving identification accuracy.

Benefits of technology

It effectively reduced the false alarm rate, improved the accuracy of identifying real lock-picking behavior, and enhanced the anti-theft control precision of electric bicycles.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123104823A_ABST
    Figure CN123104823A_ABST
Patent Text Reader

Abstract

This application relates to the field of electric bicycle anti-theft control technology, specifically to an electric bicycle anti-theft locking control method based on worm gear transmission. The method includes: setting a time window; obtaining the environmental noise intensity based on the discreteness and number of abrupt changes of the triaxial acceleration signal within the initial time window; determining the risk window; dividing the frequency band; performing spectral analysis on the triaxial acceleration signal of the initial time window to obtain the environmental spectral vector; for the risk window, obtaining a spectral deviation risk index based on the difference between the real-time spectral vector and the environmental spectral vector; calculating the impact displacement signal of the triaxial acceleration signal; calculating the impact displacement weight based on the correlation between the impact displacement signal and various environmental characteristic signals, combined with the spectral deviation risk index; and introducing the impact displacement weight into the anti-theft risk scoring to achieve hierarchical control. This application can improve the accuracy of anti-theft risk scoring, thereby improving the precision of anti-theft control for electric bicycles.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of electric bicycle anti-theft control technology, specifically to an electric bicycle anti-theft locking control method based on worm gear transmission. Background Technology

[0002] Existing anti-theft methods mainly include wheel hub locks, expansion brake locks, disc brake locks, or traditional mechanical locks. Among them, traditional mechanical locks generally have problems such as exposed lock body, easy to cut and pry, high rate of mutual opening, and lack of alarms; wheel hub locks, expansion brake locks, disc brake locks, and other solutions that directly lock rotating parts may cause the wheels to suddenly lock up if the locking action is performed while the vehicle is in motion due to accidental activation, remote control interference, controller malfunction, or software abnormality, posing a serious safety risk.

[0003] To address the aforementioned issues, worm gear transmission devices are commonly used to lock electric bicycles. Specifically, a motor drives the worm gear mechanism to push the locking tongue using forward and reverse rotation. Reliable locking and post-locking anti-theft monitoring are achieved through closed-loop judgment of the vehicle's status. This method offers advantages such as a large reduction ratio, high output torque, compact structure, and strong locking force. However, the anti-theft detection of this type of solution still relies on simple threshold judgment followed by multi-source weighted scoring. When the index exceeds the threshold, the system triggers an anti-theft alarm. However, in the actual parking environment of electric bicycles, there is significant environmental disturbance. Low-frequency ground vibrations caused by large vehicles passing by, collisions with neighboring vehicles, and vibrations from wind and rain can easily be mistaken for lock-picking behavior, resulting in low accuracy of anti-theft monitoring and control. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides an anti-theft locking control method for electric bicycles based on worm gear transmission, thereby resolving the existing issues.

[0005] The anti-theft locking control method for electric bicycles based on worm gear transmission in this application adopts the following technical solution:

[0006] One embodiment of this application provides an anti-theft locking control method for electric bicycles based on worm gear transmission, the method comprising the following steps:

[0007] S1, synchronously acquire the triaxial acceleration signal of the electric bicycle and the environmental characteristic signal; set a time window, and obtain the environmental noise intensity based on the discreteness and number of abrupt changes of the triaxial acceleration signal within the initial time window;

[0008] S2, perform threshold analysis on the environmental noise intensity to determine the risk window; divide the frequency band and perform spectrum analysis on the acceleration triaxial signal of the initial time window to obtain the environmental spectrum vector; for the risk window, obtain the spectrum deviation risk index based on the gain of the real-time spectrum vector relative to the environmental spectrum vector in each frequency band over a continuous time period.

[0009] S3, for the risk window, calculate the continuous displacement distance of the triaxial acceleration signal to obtain the impact displacement signal; based on the correlation between the impact displacement signal and various environmental characteristic signals, obtain the risk correlation index; combine the risk correlation index and the spectrum deviation risk index to calculate the impact displacement weight;

[0010] S4 uses the impact displacement weight as the weight of the impact displacement signal, and combines it with the change in environmental characteristic signals to construct an anti-theft risk scoring model. Based on the anti-theft risk score, electric bicycles are subject to graded control.

[0011] Furthermore, the environmental noise intensity is obtained based on the discreteness and number of abrupt changes of the triaxial acceleration signal within the initial time window, including:

[0012] Calculate the combined standard deviation of the triaxial acceleration signals within the initial time window;

[0013] Three-dimensional clustering was performed on the triaxial acceleration signals to count the number of outliers; the ratio of the number of outliers to the length of the time window was calculated as the environmental impact frequency.

[0014] The normalized result of the product of the environmental impact frequency and the comprehensive standard deviation is taken as the environmental noise intensity.

[0015] Furthermore, the process for determining the risk window includes: if the environmental noise intensity of the current time window is greater than a preset impact threshold, then the current time window is determined to be a risk window.

[0016] Furthermore, the process of obtaining the spectrum deviation risk index includes:

[0017] The vibration frequency is divided into three continuous frequency bands. The energy of the triaxial acceleration signal in each frequency band is obtained by using the fast Fourier transform algorithm. The vector composed of the energy values ​​of each frequency band is used as the environmental spectrum vector.

[0018] Within the risk window of real-time monitoring, the vector composed of the energy values ​​of each frequency band is denoted as the real-time spectrum vector; the gain of the energy of each frequency band relative to the environmental spectrum vector in the real-time spectrum vector is calculated as the energy increment of each frequency band.

[0019] Calculate the sum of energy increments for each frequency band within m consecutive time windows, including the current risk window, and use this sum as the cumulative energy increment value for each frequency band.

[0020] By assigning a preset weight to the cumulative energy increment value of each frequency band, the weighted average of the cumulative energy increment values ​​of all frequency bands is calculated to obtain the spectrum deviation risk index.

[0021] Furthermore, the formula for calculating the energy increment of each frequency band is as follows:

[0022]

[0023] in, For the i-th frequency band The frequency band energy increment; The energy of the real-time spectrum vector in the i-th frequency band; Let i be the energy of the environmental spectrum vector in the i-th frequency band; To prevent small constants from being divided by zero.

[0024] Furthermore, the calculation process of the impact displacement signal includes:

[0025] The spatial Euclidean distance between adjacent moments is calculated from the triaxial acceleration signal to obtain a continuous acceleration-displacement signal; the acceleration-displacement signal is then normalized to obtain the impact displacement signal.

[0026] Furthermore, the process of obtaining the risk correlation index includes: calculating the correlation coefficient between the impact displacement signal and each environmental characteristic signal, and taking the average of all the correlation coefficients as the risk correlation index.

[0027] Furthermore, the method for calculating the impact displacement weight is as follows: calculate the product of the risk correlation index and the spectrum deviation risk index, and use the normalized value of the product as the impact displacement weight.

[0028] Furthermore, the environmental characteristic signals include: latch position signal, motor terminal voltage signal, and wheel speed signal.

[0029] Furthermore, the method of using the impact displacement weight as the weight of the impact displacement signal, combined with the change in environmental characteristic signals, constructs an anti-theft risk scoring model, and performs graded control of electric bicycles based on the anti-theft risk score, including:

[0030] For any environmental feature signal, obtain the absolute value sequence of the difference between the environmental feature signal in the current risk window and the environmental feature signal in the initial time window; use the average value of the absolute value sequence of the difference as the fluctuation value of the environmental feature signal;

[0031] The formula for calculating the theft risk score is:

[0032]

[0033] in, Assess the risk of theft. Impact displacement weight; The number of types of environmental characteristic signals; This represents the average value of the impact displacement signal; Let be the fluctuation value of the k-th environmental characteristic signal;

[0034] Threshold analysis is performed on the anti-theft risk score to determine the current risk level, and alarms and mechanical controls are activated on the electric bicycles based on the risk level.

[0035] This application has at least the following beneficial effects:

[0036] This application collects data on environmental interference factors after the locking command is executed. By analyzing the environmental vibration background after the vehicle is locked, vibration characteristics are extracted. Spectral analysis of the triaxial acceleration signal is performed to extract environmental characteristics. By comparing the difference between the real-time spectral vector and the environmental spectral vector, a spectral deviation risk index is obtained, preventing environmental vibration from being misjudged as lock picking and causing frequent false alarms. Further analysis of the high correlation between the impact displacement signal caused by continuous prying and environmental characteristic signals such as the micro-displacement of the lock tongue is conducted. Combined with the spectral deviation risk index, the impact displacement weight is calculated, improving the accuracy of continuous prying identification. Using the impact displacement weight as the weight of the impact displacement signal in the anti-theft risk score, while adjusting the weight of environmental characteristic signals, normal disturbances can be filtered out, accurately identifying theft behavior and improving the precision of anti-theft control for electric bicycles. Attached Figure Description

[0037] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 A flowchart illustrating the steps of the anti-theft locking control method for electric bicycles based on worm gear transmission provided in this application;

[0039] Figure 2 A flowchart illustrating the acquisition of impact displacement weights is provided for one embodiment of this application. Detailed Implementation

[0040] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. In the absence of conflict, the embodiments of this application and the technical features in the embodiments can be combined with each other. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0041] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0042] The following description, in conjunction with the accompanying drawings, details the specific scheme of the anti-theft locking control method for electric bicycles based on worm gear transmission provided in this application.

[0043] This application provides an embodiment of an anti-theft locking control method for electric bicycles based on worm gear transmission. Specifically, the method is described below. Please refer to [link to relevant documentation]. Figure 1 The method includes the following steps:

[0044] Step S1: Synchronously acquire the triaxial acceleration signal of the electric bicycle and the environmental feature signal; set a time window, and obtain the environmental noise intensity based on the discreteness and number of abrupt changes of the triaxial acceleration signal within the initial time window.

[0045] One specific implementation scenario of this application is an anti-theft control scenario for an electric bicycle after locking, based on a worm gear drive. The locking command execution process is as follows: when the locking motor rotates forward, it drives the worm wheel through the worm gear. The worm wheel pushes the locking tongue out and inserts it into the locking hole via a shift fork, rack, cam, or connecting rod. After locking is completed, the worm gear mechanism maintains the locking tongue position by relying on its self-locking characteristic.

[0046] To further eliminate the interference of environmental disturbances on anti-theft detection, such as Figure 2 As shown, this embodiment first collects data on interference factors in the environment. By analyzing the environmental vibration background after the vehicle is locked, vibration characteristics are extracted and then correlated and fused with the micro-displacement of the lock tongue. In the multi-source weighted scoring, a dynamic weight of impact displacement is set to distinguish between environmental disturbances and continuous prying behavior, thereby reducing the false alarm rate and improving the accuracy of identifying real lock-picking behavior.

[0047] In this embodiment, a 2-minute delay sampling interval is set after the electric bicycle is successfully locked to avoid data contamination caused by the user moving or adjusting the bicycle immediately after locking it. Acceleration triaxial signals and environmental characteristic signals are collected simultaneously. The collected environmental characteristic signals include: latch position signal, motor terminal voltage signal, and wheel speed signal. The latch position signal is used to determine whether environmental vibrations will cause the latch to vibrate naturally; the motor terminal signal is used to determine whether there are minor back-dragging disturbances or power attack behaviors in the worm gear transmission chain under normal conditions; the wheel speed signal is used to exclude cases where the vehicle is pushed; and the acceleration signal is used to establish the environmental vibration background.

[0048] To ensure data quality and the accuracy of subsequent calculations, the data collected in the above steps needs to be preprocessed accordingly. Specifically: missing values ​​are filled in using linear interpolation to ensure data integrity; to eliminate random noise interference from the sensor itself, the collected data is smoothed using a moving average filter with a window size of 10; and to eliminate the influence of different value ranges, the data of different dimensions is normalized using the max-min normalization method. The linear interpolation, moving average filter, and max-min normalization method are well-known techniques, and their specific implementation processes will not be elaborated upon.

[0049] In worm gear anti-theft lock scenarios, the locking mechanism has a high degree of mechanical self-locking capability. Actual prying behavior does not necessarily manifest as a strong impact, but may instead present as low-amplitude, continuous, repetitive, and directionally related mechanical disturbances. To prevent environmental vibrations from being misinterpreted as lock picking and causing frequent false alarms, it is necessary to monitor triaxial acceleration signals over a continuous time period. Environmental characteristics can be extracted through spectral analysis of these triaxial acceleration signals.

[0050] Set a time window of length T=10s. For the initial time window, calculate the composite standard deviation of the triaxial acceleration signals within the time window. The formula for calculating the composite standard deviation of three-dimensional data is a well-known technique, and the calculation process will not be elaborated here. The composite standard deviation represents the degree of instability of environmental vibration.

[0051] To extract the impact characteristics experienced by electric bicycles, three-dimensional clustering of the triaxial acceleration signals is performed. This embodiment employs the DBSCAN clustering algorithm to count the number of outliers. Outliers indicate abrupt changes in the data, representing occasional impacts in the environment. The ratio of the number of outliers to the length of the time window is calculated as the environmental impact frequency, representing the frequency of short-term impacts per unit time. The environmental noise intensity is calculated by combining the environmental impact frequency and the comprehensive standard deviation.

[0052] As an embodiment of this application, the formula for calculating environmental noise intensity is:

[0053]

[0054] Where S is the ambient noise intensity; The combined standard deviation of the three-axis acceleration signals; For environmental impact frequency; It is a linear normalization function.

[0055] The overall standard deviation indicates the degree of instability of environmental vibration. The more chaotic the environmental vibration, the higher the frequency of short-term impacts per unit time, indicating that the electric bicycle is in a complex environment and the environmental noise is greater.

[0056] Step S2: Perform threshold analysis on the environmental noise intensity to determine the risk window; divide the frequency bands and perform spectrum analysis on the acceleration triaxial signal of the initial time window to obtain the environmental spectrum vector; for the risk window, obtain the spectrum deviation risk index based on the gain of the real-time spectrum vector relative to the environmental spectrum vector in each frequency band within a continuous time period.

[0057] The intensity of ambient noise indicates the complexity of the environment in which an electric bicycle is located. If the ambient noise intensity is low, it means that the environment in which the electric bicycle is located is relatively quiet and the possibility of it being impacted is small. If the ambient noise intensity is high, it means that the electric bicycle has been impacted and there is a greater risk of it being stolen.

[0058] Therefore, this solution uses a preset impact threshold. Implementers can calculate and set appropriate thresholds based on multiple experiments in actual scenarios; if the environmental noise intensity of the current time window is greater than the preset impact threshold, the current time window is determined to be a risk window.

[0059] In the electric bicycle scenario, environmental disturbances and continuous prying may have similar amplitudes, but they differ in frequency distribution, energy persistence, main frequency stability, frequency band shift characteristics, and correlation with lock tongue displacement. Large vehicles passing by, ground resonance, and wind blowing across the vehicle body typically concentrate on low or low-to-mid frequencies; raindrops, light bumps from neighboring vehicles, and occasional impacts may generate short-duration high-frequency pulses; while prying the lock body with a tool may manifest as repetitive mid-frequency vibrations, high-frequency localized impacts, and non-random energy growth related to the micro-displacement of the lock tongue.

[0060] To distinguish vibration signals from different sources and separate short-duration high-frequency pulses in the environment to prevent interference with anti-theft monitoring, a spectral analysis of the triaxial acceleration signal within an initial time window is first performed. In this embodiment, the vibration frequency is divided into three continuous frequency bands: low frequency band: F1 = 0.5Hz to 5Hz; mid frequency band: F2 = 5Hz to 30Hz; and high frequency band: F3 = above 30Hz. The energy of the triaxial acceleration signal in each frequency band is obtained using a Fast Fourier Transform algorithm. Furthermore, the vector composed of the energy values ​​of each frequency band is used as the environmental spectrum vector.

[0061] The environmental spectrum vector clearly represents the energy distribution of various disturbance factors in the environment, providing a benchmark for real-time signal analysis during anti-theft monitoring. Based on this, within the risk window of real-time monitoring, the vector composed of the energy values ​​of each frequency band is denoted as the real-time spectrum vector; the gain of each frequency band's energy relative to the environmental spectrum vector is calculated and denoted as the energy increment of each frequency band. The specific calculation formula is as follows:

[0062]

[0063] in, For the i-th frequency band The frequency band energy increment; The energy of the real-time spectrum vector in the i-th frequency band; Let i be the energy of the environmental spectrum vector in the i-th frequency band; To prevent division by zero of small constants, this embodiment... .

[0064] When the energy increment of a certain frequency band is large, an alarm is not triggered directly. Instead, it is further determined whether the frequency band meets the characteristics of environmental disturbance or prying. If a high-frequency band suddenly rises, but the duration is very short and there is no bolt displacement or anti-drag signal, it can be determined as a short-term impact and will not affect the risk score. If the energy of a certain frequency band remains high within a continuous window and the cumulative value of the frequency band increment is large, the prying risk is high. Therefore, the sum of the energy increments of each frequency band within m consecutive time windows, including the current risk window, is calculated as the cumulative value of the energy increment of each frequency band. In this embodiment, m=5.

[0065] Furthermore, because thieves repeatedly pry open the lock using crowbars, screwdrivers, pliers, or metal tools, their actions often exhibit a repetitive rhythm, concentrating frequency band energy in the mid-frequency or mid-to-high frequency range. Therefore, the cumulative energy increments in the mid-frequency and high-frequency ranges can be given greater weight in calculating the spectrum deviation risk index. The formula for calculating the spectrum deviation risk index is:

[0066]

[0067] Where R is the spectrum deviation risk index; This represents the cumulative energy increment in the low-frequency band. The preset weight for the low-frequency band is set to 0.15 in this embodiment; This represents the cumulative energy increment in the mid-frequency band. The preset weight for the mid-frequency band is set to 0.5 in this embodiment; This represents the cumulative energy increment in the high-frequency band. The preset weight for the high-frequency band is set to 0.35 in this embodiment; it should be noted that the implementer must meet the following requirements when setting the weight. , , The sum of their values ​​equals 1.

[0068] Step S3: For the risk window, calculate the continuous displacement distance of the triaxial acceleration signal to obtain the impact displacement signal; based on the correlation between the impact displacement signal and various environmental characteristic signals, obtain the risk correlation index; and calculate the impact displacement weight by combining the risk correlation index and the spectrum deviation risk index.

[0069] Short-term impacts refer to abnormal vibrations that are short in duration, have a high peak energy, but lack continuity and correlation with the lock body. Examples include collisions with nearby vehicles, pedestrians bumping into handlebars, wind blowing over plastic objects, concentrated raindrop impacts, and instantaneous transmission of road vibrations. These vibrations typically have an unpredictable direction and are not stably correlated with the micro-displacement of the latch; the energy does not show a continuously increasing trend. Continuous prying, on the other hand, involves vibrations originating close to the lock body or latch, with vibration energy accumulating or repeating over a certain period. This is usually accompanied by micro-movements, rebounds, or trend-based shifts in the latch position, and by reverse dragging disturbances at the motor end.

[0070] Therefore, by calculating the spatial Euclidean distance between adjacent moments of the triaxial acceleration signal, a continuous acceleration-displacement signal is obtained; by normalizing the acceleration-displacement signal, an impact displacement signal is obtained.

[0071] The correlation between the impact displacement signal and various environmental characteristic signals within the current risk window is calculated, namely, the correlation with the latch position signal, motor terminal voltage signal, and wheel speed signal. In this embodiment, the correlation coefficient between the impact displacement signal and various environmental characteristic signals is calculated, and the average of all the correlation coefficients is used as the risk correlation index.

[0072] If the impact signal is highly correlated with the displacement of the latch, the reverse drag of the motor, and the change in wheel speed, it is a high-risk impact, and the risk of theft is greater; if it only manifests as overall vehicle body vibration without latch, reverse drag, or wheel speed change, it is determined to be an environmental impact.

[0073] In the electric bicycle scenario, the spectrum deviation risk index determines whether the current spectrum is similar to the environmental spectrum model. If the similarity is high and there are no lock-related features, it is judged as an environmental disturbance, and the weight of impact displacement needs to be reduced in the anti-theft risk score. If the current spectrum differs significantly from the environmental spectrum model, and the impact displacement signal is highly correlated with environmental feature signals such as lock tongue displacement, the anti-theft risk is high, and the weight of impact displacement needs to be increased in the anti-theft risk score.

[0074] Therefore, in this embodiment, the product of the risk correlation index and the spectrum deviation risk index is calculated, and the normalized value of the product is used as the impact displacement weight.

[0075] Step S4: The impact displacement weight is used as the weight of the impact displacement signal. Combined with the change in environmental characteristic signals, an anti-theft risk scoring model is constructed, and the electric bicycle is classified and controlled according to the anti-theft risk score.

[0076] The changes in environmental characteristic signals are incorporated into the anti-theft risk scoring model. For any environmental characteristic signal, the absolute value sequence of the difference between the environmental characteristic signal in the current risk window and the environmental characteristic signal in the initial time window is obtained; the average value of the absolute value sequence of the difference is taken as the fluctuation value of the environmental characteristic signal. Taking the latch position signal as an example, the a-th element in the absolute value sequence of the difference of the latch position signal... for: ;in, This is the a-th element of the latch position signal for the current risk window; This is the a-th element of the latch position signal during the initial time window; The absolute value sign is used. The fluctuation value of the latch position signal is the mean of all elements in the absolute value sequence of differences.

[0077] Furthermore, an anti-theft risk scoring model is constructed, threshold analysis is performed on the anti-theft risk score to determine the current risk level, and alarms and mechanical controls are activated on the electric bicycles based on the risk level.

[0078] The final formula for calculating the anti-theft risk score is:

[0079]

[0080] in, Assess the risk of theft. Impact displacement weight; The number of types of environmental characteristic signals; This represents the average value of the impact displacement signal; Let be the fluctuation value of the k-th environmental characteristic signal.

[0081] In this embodiment, if If it is, then it is judged to be of a low risk level; if If so, the risk level is determined to be medium; if If the risk level is high, it is determined to be a short-term environmental disturbance and the normal state is maintained; if the risk level is medium, it is recorded or briefly prompted; if the risk level is high, it is executed with a high-intensity buzzer, location upload, controller disabling, and abnormal log recording.

[0082] To address the issue of electric bicycle worm gear anti-theft locks easily misinterpreting environmental disturbances as continuous prying in complex parking environments, this paper proposes an anti-theft detection method that combines environmental background analysis, spectrum identification, risk correlation analysis, and dynamic weight fusion. This method uses environmental noise intensity to identify abnormal disturbances, spectrum and energy changes to identify abnormal mechanical actions, and micro-displacement of the lock tongue and reverse drag signals from the motor end to confirm the correlation between the lock-picking behavior and the actual prying action. Furthermore, it achieves a graded response through dynamic weighting and an alarm mechanism. Through this control method, this scheme can improve the detection accuracy of actual prying behavior while reducing false alarms, thereby enhancing the anti-theft control precision of electric bicycles.

[0083] Through the above description of the embodiments in conjunction with the accompanying drawings, those skilled in the art will understand that, for the sake of convenience and brevity, the above division of functional modules is only used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0084] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for anti-theft locking control of electric bicycles based on worm gear transmission, characterized in that, The method includes the following steps: S1, synchronously acquire the triaxial acceleration signal of the electric bicycle and the environmental characteristic signal; set a time window, and obtain the environmental noise intensity based on the discreteness and number of abrupt changes of the triaxial acceleration signal within the initial time window; S2, perform threshold analysis on the environmental noise intensity to determine the risk window; divide the frequency band and perform spectrum analysis on the acceleration triaxial signal of the initial time window to obtain the environmental spectrum vector; for the risk window, obtain the spectrum deviation risk index based on the gain of the real-time spectrum vector relative to the environmental spectrum vector in each frequency band over a continuous time period. S3, for the risk window, calculate the continuous displacement distance of the triaxial acceleration signal to obtain the impact displacement signal; based on the correlation between the impact displacement signal and various environmental characteristic signals, obtain the risk correlation index; combine the risk correlation index and the spectrum deviation risk index to calculate the impact displacement weight; S4 uses the impact displacement weight as the weight of the impact displacement signal, and combines it with the change in environmental characteristic signals to construct an anti-theft risk scoring model. Based on the anti-theft risk score, electric bicycles are subject to graded control.

2. The electric bicycle anti-theft locking control method based on worm gear transmission as described in claim 1, characterized in that, The environmental noise intensity is obtained based on the discreteness and number of abrupt changes of the triaxial acceleration signal within the initial time window, including: Calculate the combined standard deviation of the triaxial acceleration signals within the initial time window; Three-dimensional clustering was performed on the triaxial acceleration signals to count the number of outliers; the ratio of the number of outliers to the length of the time window was calculated as the environmental impact frequency. The normalized result of the product of the environmental impact frequency and the comprehensive standard deviation is taken as the environmental noise intensity.

3. The electric bicycle anti-theft locking control method based on worm gear transmission as described in claim 1, characterized in that, The process of determining the risk window includes: if the environmental noise intensity of the current time window is greater than a preset impact threshold, then the current time window is determined to be a risk window.

4. The electric bicycle anti-theft locking control method based on worm gear transmission as described in claim 1, characterized in that, The process of obtaining the spectrum deviation risk index includes: The vibration frequency is divided into three continuous frequency bands. The energy of the triaxial acceleration signal in each frequency band is obtained by using the fast Fourier transform algorithm. The vector composed of the energy values ​​of each frequency band is used as the environmental spectrum vector. Within the risk window of real-time monitoring, the vector composed of the energy values ​​of each frequency band is denoted as the real-time spectrum vector; the gain of the energy of each frequency band relative to the environmental spectrum vector in the real-time spectrum vector is calculated as the energy increment of each frequency band. Calculate the sum of energy increments for each frequency band within m consecutive time windows, including the current risk window, and use this sum as the cumulative energy increment value for each frequency band. By assigning a preset weight to the cumulative energy increment value of each frequency band, the weighted average of the cumulative energy increment values ​​of all frequency bands is calculated to obtain the spectrum deviation risk index.

5. The electric bicycle anti-theft locking control method based on worm gear transmission as described in claim 4, characterized in that, The formula for calculating the energy increment of each frequency band is as follows: in, For the i-th frequency band The frequency band energy increment; The energy of the real-time spectrum vector in the i-th frequency band; Let i be the energy of the environmental spectrum vector in the i-th frequency band; To prevent small constants from being divided by zero.

6. The anti-theft locking control method for electric bicycles based on worm gear transmission as described in claim 1, characterized in that, The calculation process of the impact displacement signal includes: The spatial Euclidean distance between adjacent moments is calculated from the triaxial acceleration signal to obtain a continuous acceleration-displacement signal; the acceleration-displacement signal is then normalized to obtain the impact displacement signal.

7. The electric bicycle anti-theft locking control method based on worm gear transmission as described in claim 1, characterized in that, The process of obtaining the risk correlation index includes: calculating the correlation coefficient between the impact displacement signal and various environmental characteristic signals, and taking the average of all the correlation coefficients as the risk correlation index.

8. The anti-theft locking control method for electric bicycles based on worm gear transmission as described in claim 1, characterized in that, The method for calculating the impact displacement weight is as follows: calculate the product of the risk correlation index and the spectrum deviation risk index, and use the normalized value of the product as the impact displacement weight.

9. The anti-theft locking control method for electric bicycles based on worm gear transmission as described in claim 1, characterized in that, The environmental characteristic signals include: latch position signal, motor terminal voltage signal, and wheel speed signal.

10. The anti-theft locking control method for electric bicycles based on worm gear transmission as described in claim 1, characterized in that, The method involves using the impact displacement weight as the weight of the impact displacement signal, combining it with changes in environmental characteristic signals, to construct an anti-theft risk scoring model. Based on this anti-theft risk score, the electric bicycle is subject to graded control, including: For any environmental feature signal, obtain the absolute value sequence of the difference between the environmental feature signal in the current risk window and the environmental feature signal in the initial time window; use the average value of the absolute value sequence of the difference as the fluctuation value of the environmental feature signal; The formula for calculating the theft risk score is: in, Assess the risk of theft. Impact displacement weight; The number of types of environmental characteristic signals; This represents the average value of the impact displacement signal; Let be the fluctuation value of the k-th environmental characteristic signal; Threshold analysis is performed on the anti-theft risk score to determine the current risk level, and alarms and mechanical controls are activated on the electric bicycles based on the risk level.