Scooter strength testing system and folding scooter based on mechanical sensor

Abstract

The present invention discloses a scooter strength testing system and a folding scooter based on a mechanical sensor, relating to the technical field of scooter strength testing. The system records a reference signal through a vibration loading module; a frequency marking module gradually adjusts the loading frequency and marks the resonance interval, and simultaneously collects aligned loading signals and response signals; a phase difference analysis module calculates phase difference fluctuations using fast Fourier transform; a nonlinear harmonic component analysis module further decomposes the resonance signal and quantifies the nonlinear proportion of the fundamental frequency and high-order harmonics; a comprehensive analysis module predicts test accuracy through a comprehensive evaluation of phase difference and nonlinear anomalies; a measurement result classification module performs hierarchical processing on the signal, and optimizes the loading conditions in combination with a prediction and adjustment module to ensure the reliability and consistency of the test signal, thereby significantly improving the accuracy and stability of the scooter structural mechanical test and providing a scientific basis for optimizing scooter design and manufacturing.

Description

Technical Field

[0001] The present invention relates to the technical field of scooter strength testing, and in particular to a scooter strength testing system based on a mechanical sensor and a folding scooter. Background Art

[0002] Scooter strength testing based on mechanical sensors utilizes mechanical sensor technology to assess the load-bearing capacity and durability of scooter structures. Existing mechanical sensors (such as pressure sensors, strain gauges, or accelerometers) can monitor key parameters of a scooter under load, such as stress, strain, and vibration frequency, in real time. This data, collected by the sensors, can be uploaded to a computing device for analysis to assess whether the scooter's strength meets design standards and safety regulations. When scooter manufacturers conduct quality control testing on new products, for example, they apply various simulated loads to the scooter's load-bearing components in a laboratory setting. Sensors are used to measure changes in key stress-bearing areas to ensure that the scooter avoids structural damage under high loads. Alternatively, sensors can be installed to monitor the load conditions of scooters in real-world usage scenarios, providing rental companies with maintenance recommendations to improve the safety and durability of the equipment.

[0003] The existing technology has the following shortcomings:

[0004] During the simulated load application process, if the vibration frequency of the test equipment coincides with the natural frequency of the scooter structure or the natural frequency of the sensor components, coupled resonance may be induced. This resonance not only amplifies the vibration amplitude, causing abnormal stress conditions on the scooter structure, but also can severely distort sensor signals or prevent the acquisition of critical data. Furthermore, false data generated under resonance conditions can mislead test results, preventing manufacturers from identifying the scooter's true strength or weaknesses. Summary of the Invention

[0005] The purpose of the present invention is to provide a scooter strength testing system based on a mechanical sensor and a folding scooter to solve the shortcomings of the background technology.

[0006] In order to achieve the above-mentioned object, the present invention provides the following technical solution: a scooter strength testing system based on a mechanical sensor, comprising a vibration loading module, a frequency marking module, a phase difference analysis module, a nonlinear harmonic component analysis module, a comprehensive analysis module, a measurement result classification module, and a prediction and adjustment module;

[0007] Vibration loading module: Multiple sensors are installed at different locations on the scooter, covering the loading point and monitoring areas away from the loading point. Vibration loading equipment is set up in each monitoring area, and the loading signals are recorded as reference signals.

[0008] Frequency Marking Module: This module applies a vibration load to the scooter, gradually adjusts the loading frequency, marks the intervals that exceed the structural resonance frequency, and synchronously collects the loading signal and sensor response signal within the marked interval to ensure time series alignment.

[0009] Phase difference analysis module: uses fast Fourier transform to extract the phase information of the loading signal and sensor signal on the main frequency component, analyzes the phase difference fluctuation between the loading signal and the sensor signal, and evaluates the consistency of the sensor signal with the reference signal;

[0010] Nonlinear harmonic component analysis module: If the sensor signal is inconsistent with the reference signal, the sensor output signal within the resonant frequency and its surrounding frequency range is decomposed into fundamental frequency components and higher-order harmonic components. The abnormal proportion of nonlinear components in the fundamental frequency components is analyzed to evaluate the authenticity of the mechanical response.

[0011] Comprehensive Analysis Module: This module evaluates the accuracy of the sensor in measuring the scooter's structural mechanical response under resonance conditions, based on the fluctuations in the phase difference between the loading signal and the sensor signal at different loading frequencies, as well as the abnormal proportion of nonlinear components in the fundamental frequency components.

[0012] Measurement result classification module: Based on the accuracy of the sensor in measuring the mechanical response of the scooter structure under resonance, the sensor measurement results are divided into accurate measurement results, incompletely accurate measurement results, and inaccurate measurement results, and corresponding processing is performed;

[0013] Prediction and Adjustment Module: For incomplete accuracy measurement results, a fixed measurement time period is used to analyze the sensor signal change trend under the resonant state and predict the accuracy attenuation of long-term testing. If the prediction results show that the signal distortion is uncontrollable, the loading frequency and sensor position are adjusted. If the signal distortion is controllable, the filtering and data compensation algorithms for subsequent tests are optimized.

[0014] Preferably, in the phase difference analysis module, the phase difference fluctuation between the loading signal and the sensor signal is analyzed to generate a phase difference fluctuation index, and the method for obtaining the phase difference fluctuation index is:

[0015] Obtain the loading signal x(t), which represents the vibration signal of the loading device, and the sensor signal y(t), which represents the vibration signal of the scooter structure response. Set the sampling frequency fs, set the sliding window function w(t), the window length Lw, and define the number of sampling points Sw for each sliding window movement;

[0016] Calculate the short-time Fourier transform of the loading signal, the expression is: ; Where X(t,f) is the spectrum of the loaded signal at time t and frequency f, is the frequency component in Fourier transform, is the original signal at time τ, It is a time-localized window function used to intercept a section of the signal for Fourier transform;

[0017] Calculate the short-time Fourier transform of the sensor signal, which is expressed as: ;in, is the spectrum of the sensor signal at time t and frequency f, in each time window , select the main frequency , the expression is: Where, is the amplitude of the loading signal at frequency f, which represents the signal energy distribution; is the amplitude of the sensor signal at frequency f, which represents the signal energy distribution;

[0018] At the main frequency Calculate the phase difference between the load signal and the sensor signal , the expression is: ; is the phase angle of the sensor signal at the main frequency, indicating the initial phase of the vibration; is the phase angle of the loading signal at the main frequency;

[0019] The phase difference sequence formed in each time window , k=1,2,…,M, where M is the total number of sliding windows; calculate the standard deviation and mean of the phase difference sequence; calculate the phase difference fluctuation index, the expression is: ; Where HMB is the phase difference fluctuation index, is the mean of the phase difference sequence.

[0020] Preferably, in the nonlinear harmonic component analysis module, after analyzing the abnormality of the nonlinear component ratio in the fundamental frequency component, a nonlinear component ratio abnormality index is generated. The method for obtaining the nonlinear component ratio abnormality index is:

[0021] The nonlinear component ratio sequence obtained from the nonlinear harmonic component analysis module is recorded as , the frequency range is [fstart, fend]; the proportion sequence Divide the sliding window into multiple subsequences with a fixed length of m: ; The window sliding step is set to s;

[0022] In each subsequence, define the embedding dimension m and similarity tolerance r: ; Count other subsequences and the current The degree of similarity, the number of templates that meet the conditions is recorded as A(m,r): The maximum distance between the set templates is , the expression is: ; Increase the embedding dimension to m+1, count the number of matching templates A(m+1,r), and calculate the abnormal index of nonlinear component proportion. The expression is: ; Where GQS is the abnormal index of nonlinear component proportion.

[0023] Preferably, in the comprehensive analysis module, the phase difference fluctuation index and the nonlinear component ratio abnormality index are converted into a comprehensive feature vector, and the comprehensive feature vector is used as the input of the machine learning model. The machine learning model uses each set of comprehensive feature vectors to predict the accuracy value label of the sensor measuring the mechanical response of the scooter structure under resonance as the prediction target, and takes minimizing the sum of the prediction errors of the accuracy value labels of the sensor measuring the mechanical response of the scooter structure under all resonance conditions as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence, and the model training is stopped. The accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance is determined according to the model output result, wherein the machine learning model is a polynomial regression model.

[0024] Preferably, in the measurement result classification module, the sensor measurement results are divided into accuracy measurement results, incomplete accuracy measurement results and inaccuracy measurement results according to the accuracy of the sensor measuring the mechanical response of the scooter structure under resonance conditions, specifically:

[0025] Comparing the obtained accuracy value of the sensor measuring the mechanical response of the scooter structure under the resonance condition with a gradient standard threshold, where the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and comparing the accuracy value of the sensor measuring the mechanical response of the scooter structure under the resonance condition with the first standard threshold and the second standard threshold respectively;

[0026] If the accuracy value of the sensor measuring the mechanical response of the scooter structure under the resonance condition is greater than the second standard threshold, it indicates that the accuracy of the sensor measuring the mechanical response of the scooter structure under the resonance condition is high, and a high accuracy signal is generated, and the sensor measurement result is classified as an accuracy measurement result;

[0027] If the accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the accuracy of the sensor measuring the mechanical response of the scooter structure under resonance is moderate, and a moderate accuracy signal is generated, and the sensor measurement result is classified as an incomplete accuracy measurement result;

[0028] If the accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance is less than the first standard threshold, it means that the accuracy of the sensor measuring the mechanical response of the scooter structure under resonance is low. At this time, a low accuracy signal is generated, and the sensor measurement result is classified as an inaccurate measurement result.

[0029] Preferably, in the prediction and adjustment module, for the incomplete accuracy measurement results, that is, the accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance generated within a fixed measurement time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the accuracy values ​​of the sensor measuring the mechanical response of the scooter structure under resonance that are greater than or equal to the first standard threshold and less than or equal to the second standard threshold in subsequent measurement time periods are collected, and a data set is constructed. The trend of sensor signal changes in the resonance state is analyzed according to the mean and standard deviation of the data set, and the accuracy attenuation of long-term testing is predicted.

[0030] Preferably, if the mean accuracy value in the data set is greater than or equal to the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is less than the reference threshold of the standard deviation of the accuracy value, the signal change trend is stable, the accuracy value is high and the fluctuation is small, it means that the signal distortion is controllable, the existing test conditions are maintained, and the filtering and data compensation algorithms are optimized for subsequent tests to further improve the data quality;

[0031] If the mean accuracy value is greater than or equal to the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is greater than or equal to the reference threshold of the standard deviation of the accuracy value, the accuracy value is high, but the signal fluctuation is large, and there are unstable factors, indicating that the signal distortion is uncontrollable. Optimize the layout of the sensor and filter the signal in the high fluctuation area to reduce the error;

[0032] If the mean accuracy value is less than the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is greater than or equal to the reference threshold of the standard deviation of the accuracy value, the accuracy value is low and fluctuates greatly, and the signal distortion is serious, indicating that the signal distortion is uncontrollable and the loading frequency range needs to be adjusted to avoid the resonant frequency;

[0033] If the mean accuracy value is less than the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is less than the reference threshold of the standard deviation of the accuracy value, the accuracy value is low, but the signal fluctuation is small and the distortion is stable, indicating that the signal distortion is controllable and the loading conditions are optimized.

[0034] A folding scooter comprises a bracket fixedly connected to a scooter chassis, a lower rotating rod and an upper rotating rod for controlling the steering of the scooter's front wheel, wherein the lower rotating rod is rotatably mounted inside the bracket, and the upper end of the lower rotating rod is rotatably connected to the lower end of the upper rotating rod via an adapter to achieve folding of the scooter;

[0035] A wheel frame is provided at the lower end of the lower rotating rod, and at least two groups of fixing screws distributed along the axial direction are provided between the wheel frame and the lower rotating rod.

[0036] Preferably, a shaft rod is provided in the adapter, a connecting piece is rotatably mounted on the surface of the shaft rod, and a second fixing screw is installed at the connection between the connecting piece and the lower end of the upper rotating rod.

[0037] Preferably, a connecting rod is rotatably mounted on the surface of the shaft, and a clamping block is provided on the upper surface of the connecting rod;

[0038] A clamping hole is provided on the surface of the upper rotating rod, and when the connecting rod is rotated upward, the clamping block can be matched with the clamping hole.

[0039] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0040] 1. Through a multi-module collaborative design, this invention overcomes the existing problems of signal distortion and test result deviation caused by resonance, significantly improving the accuracy and reliability of scooter structural strength testing. The combination of the vibration loading module and the frequency marking module enables accurate collection of loading signals from key parts of the scooter and marks the resonance interval to ensure the integrity of the test data. The phase difference analysis module and the nonlinear harmonic component analysis module provide in-depth analysis of signal consistency and nonlinear characteristics, effectively evaluating the authenticity of the mechanical response. The comprehensive analysis module predicts signal accuracy through machine learning, providing a scientific basis for data classification and optimization processing, further ensuring the accuracy of the test.

[0041] 2. In the prediction and adjustment module, the present invention analyzes and processes trends in incompletely accurate data, proposing targeted optimization measures such as adjusting loading frequency, optimizing sensor position, and enhancing data compensation algorithms to ensure that signal distortion is within a controllable range or completely eliminated. Furthermore, the classification module manages test results in a hierarchical manner, adopting corresponding processing strategies for data of different quality levels, thereby improving data utilization efficiency and test robustness. The present invention's overall systematic and intelligent testing process not only provides high-quality mechanical performance data support for scooter design and manufacturing, but also significantly reduces production costs and design risks caused by testing errors.

[0042] 3. The present invention provides multiple sets of axially distributed fixing screws 1 to fix the lower turning rod and the wheel frame, thereby improving the strength and stability of the connection between the lower turning rod and the wheel frame. The design of the fixing screws 2 can improve the stability of the connection between the upper turning rod and the lower turning rod, thereby improving the stability of the steering control of the front wheel of the folding scooter. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments described in the present invention. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0044] Figure 1 It is a system module diagram of the present invention;

[0045] Figure 2 The foldable scooter of the present invention has a three-dimensional structure. Figure 1 ;

[0046] Figure 3 The foldable scooter of the present invention is a three-dimensional structure Figure 2 ;

[0047] Figure 4 For the present invention Figure 3 Schematic diagram of the enlarged structure at A in the middle;

[0048] Figure 5 This is a schematic diagram of the front structure of the folding scooter of the present invention;

[0049] Figure 6 For the present invention Figure 5 Schematic diagram of the enlarged structure at point B in the middle.

[0050] In the figure: 1, bracket; 2, lower rotating rod; 201, wheel frame; 3, upper rotating rod; 301, clamping hole; 4, adapter; 401, connecting piece; 5, fixing screw 1; 6, fixing screw 2; 7, connecting rod; 701, clamping block. DETAILED DESCRIPTION

[0051] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0052] For examples, see Figure 1 As shown, the scooter strength testing system based on mechanical sensors described in this embodiment includes a vibration loading module, a frequency marking module, a phase difference analysis module, a nonlinear harmonic component analysis module, a comprehensive analysis module, a measurement result classification module, and a prediction and adjustment module;

[0053] Vibration loading module: Multiple sensors are installed at different locations on the scooter, covering the loading point and monitoring areas away from the loading point. Vibration loading equipment is set up in each monitoring area, and the loading signals are recorded as reference signals.

[0054] Frequency Marking Module: This module applies a vibration load to the scooter, gradually adjusts the loading frequency, marks the intervals that exceed the structural resonance frequency, and synchronously collects the loading signal and sensor response signal within the marked interval to ensure time series alignment.

[0055] Phase difference analysis module: uses fast Fourier transform to extract the phase information of the loading signal and sensor signal on the main frequency component, analyzes the phase difference fluctuation between the loading signal and the sensor signal, and evaluates the consistency of the sensor signal with the reference signal;

[0056] Nonlinear harmonic component analysis module: If the sensor signal is inconsistent with the reference signal, the sensor output signal within the resonant frequency and its surrounding frequency range is decomposed into fundamental frequency components and higher-order harmonic components. The abnormal proportion of nonlinear components in the fundamental frequency components is analyzed to evaluate the authenticity of the mechanical response.

[0057] Comprehensive Analysis Module: This module evaluates the accuracy of the sensor in measuring the scooter's structural mechanical response under resonance conditions, based on the fluctuations in the phase difference between the loading signal and the sensor signal at different loading frequencies, as well as the abnormal proportion of nonlinear components in the fundamental frequency components.

[0058] Measurement result classification module: Based on the accuracy of the sensor in measuring the mechanical response of the scooter structure under resonance, the sensor measurement results are divided into accurate measurement results, incompletely accurate measurement results, and inaccurate measurement results, and corresponding processing is performed;

[0059] Prediction and Adjustment Module: For incomplete accuracy measurement results, a fixed measurement time period is used to analyze the sensor signal change trend under the resonant state and predict the accuracy attenuation of long-term testing. If the prediction results show that the signal distortion is uncontrollable, the loading frequency and sensor position are adjusted. If the signal distortion is controllable, the filtering and data compensation algorithms for subsequent tests are optimized.

[0060] In the vibration loading module, determine the structural parts of the scooter that need to be monitored: loading points: such as the connection points between the handlebars and the frame, the wheel bearings, etc. These locations are the main stress-bearing areas.

[0061] Areas away from the loading point: such as the middle of the pedal, support rod, etc. These areas can reflect the force transmission and attenuation characteristics.

[0062] Based on the structural complexity of the scooter, the entire scooter is divided into several monitoring areas (such as the front, middle, and rear) to ensure comprehensive coverage of the mechanical response. Sensors and loading devices are rationally distributed within each monitoring area to ensure representative monitoring data.

[0063] Select the appropriate sensor type based on the parameters to be measured: Strain gauge sensors: Measure local stress or strain changes, used to identify the stress state of the structure. Accelerometers: Measure vibration response, particularly suitable for dynamic load testing. Displacement sensors: Measure the deformation or displacement of the scooter's local structure.

[0064] Load point area: Secure the sensor directly to the load-bearing area (such as a connecting shaft or weld point) to ensure full contact with the surface. Use high-performance adhesives or welding to prevent it from falling out due to vibration. Away from the load point area: Install it in the critical force transmission path of the structure (such as the mid-section of the frame or at the ends of the pedals) to monitor force fluctuations or attenuation. Use a fixing clamp or protective case to prevent damage from impact and vibration.

[0065] Use different types of vibration loading equipment: Electromagnetic vibration tables: Precisely control the loading frequency and amplitude, suitable for dynamic testing over a wide frequency range. Pneumatic or hydraulic loading equipment: Suitable for simulating actual cyclic or impact loads. Customized vibration exciters: Adaptable to the scooter's geometry for localized loading of specific areas.

[0066] Loading equipment is installed within each monitoring area to ensure that force is effectively transmitted to the area monitored by the sensor. Loading point area: The vibration loading equipment directly contacts the scooter's main load-bearing components (such as the wheel axle and handlebar connection) to achieve precise force application. Areas away from the loading point: The loading equipment applies force indirectly through the frame or connectors to ensure that the overall force characteristics of the structure are reflected.

[0067] Use high-precision data acquisition equipment to record the input signal (reference signal) from the loading device. This includes: Loading frequency: Sweep gradually from low to high frequencies (e.g., 0.1Hz-100Hz). Loading amplitude: Control the amplitude according to design requirements to simulate different load intensities in actual use. Time series: Ensure that the loading signal and sensor response signal are collected synchronously.

[0068] Equipped with a multi-channel data acquisition system, simultaneously record sensor output signals from all monitoring areas of the scooter. Ensure a sufficiently high sampling rate (e.g., over 1000 samples per second) to capture details of the dynamic vibration response. Ensure time series alignment so that the loading signal and sensor response data accurately correspond.

[0069] Before formal testing, calibrate the sensor to ensure its response is proportional to the actual force. Apply a known load to the sensor and compare the output signal to verify its accuracy. Verify the linkage between the loading device and the sensor: Ensure that the vibration amplitude and frequency of the loading device can be accurately sensed by the sensor. Check the system for time lag or signal distortion, and perform debugging if necessary.

[0070] In the frequency marking module, apply a vibration load to the scooter, gradually adjusting the loading frequency to mark intervals exceeding the structural resonance frequency. Synchronously acquire the loading signal and sensor response signal within the marked interval to ensure time series alignment. Select a suitable vibration loading device (such as an electromagnetic shaker or hydraulic loader) and connect it to the scooter's primary load-bearing areas (such as the frame and pedal connections). Set the initial loading frequency (typically starting at a low frequency, such as 0.1 Hz) and select an appropriate loading amplitude range based on the scooter's designed load capacity.

[0071] During the loading process, gradually increase the loading frequency to cover the range of possible resonant frequencies of the structure: Sweep Mode: Linear increments (e.g., 0.1 Hz increments) or logarithmically increasing the frequency range (e.g., 0.1 Hz, 0.3 Hz, 1 Hz, etc.). Frequency Range: Depending on the characteristics of the scooter structure, the test range is typically 0.1 Hz to 100 Hz or higher. Step Time: Maintain a certain loading time (e.g., 10 seconds) at each frequency point to ensure the sensor records a stable response.

[0072] During the test, the scooter's sensor signals and the loading device's feedback were monitored to identify resonant frequencies. Resonance: At certain frequencies, the sensor output signal (such as vibration amplitude or strain response) will be significantly amplified, indicating that the scooter's structure has reached resonance. The frequency range that causes the amplified response during the frequency change is recorded as the resonant region.

[0073] Mark the frequency range outside the resonant frequency range for focused analysis in subsequent tests: Resonant range: Record the start frequency fstart and end frequency fend where resonance occurs. Frequencies outside the resonant range, below fstart and above fend, are marked as non-resonant regions. This is used to evaluate the linearity of the loading signal and sensor response.

[0074] During the loading process, the software monitors sensor signal amplification in real time and automatically marks frequencies outside the resonance range. After the test is complete, the frequencies outside the resonance range are marked by analyzing the spectral or time domain characteristics of the loading and response signals.

[0075] Use a unified clock source to synchronize the loading signal and sensor response signal to ensure consistent timestamps for all data. Data acquisition equipment should support multi-channel synchronous acquisition to avoid inconsistent time series due to sampling delays. Ensure that the sampling frequency is sufficiently high (typically 10 times or more of the loading frequency; for example, if the loading frequency is 100 Hz, the sampling frequency should be at least 1000 Hz). A sampling frequency that is too low can result in loss or distortion of critical signal information.

[0076] Recording loading device signals, including loading frequency, amplitude, and direction, serves as a reference signal for subsequent comparative analysis. Examples include the excitation signal voltage from a vibration table or the pressure changes from a hydraulic load. Sensor signal types include: Strain signals: Reflect the local force and deformation characteristics of the scooter's structure. Vibration signals: Record the amplitude and frequency characteristics of the overall or local vibration response of the scooter. Displacement signals: Monitor the movement and displacement of the scooter's local structure.

[0077] Phase difference analysis module: uses fast Fourier transform to extract the phase information of the loading signal and sensor signal on the main frequency component, analyzes the phase difference fluctuation between the loading signal and the sensor signal, and evaluates the consistency of the sensor signal with the reference signal.

[0078] The loading signal (reference signal) is provided by the loading device (e.g., the input signal from a vibration table). Sensor signals (e.g., strain or acceleration signals) are collected by sensors installed on various structural parts of the scooter. Low-pass or band-pass filtering is used to remove high-frequency noise and DC components, enhancing signal quality. Time series synchronization between the loading signal and the sensor signal is ensured.

[0079] Perform an FFT on the loading signal x(t) and the sensor signal y(t) to obtain a frequency domain representation. Extract the amplitude and phase information of the dominant frequency: the dominant frequency is the frequency component with the highest amplitude in the signal spectrum, and the phase angle at that dominant frequency. Calculate the phase difference between the loading signal and the sensor signal. If the phase difference is greater than 0, the sensor signal lags behind the loading signal; if the phase difference is less than 0, the sensor signal leads the loading signal. Gradually adjust the loading frequency and record the phase difference at each frequency. Plot a curve of the loading frequency versus phase difference to analyze how the phase difference fluctuates with frequency.

[0080] After analyzing the phase difference fluctuation between the loading signal and the sensor signal, a phase difference fluctuation index is generated to evaluate the consistency between the sensor signal and the reference signal. The phase difference fluctuation index is obtained as follows:

[0081] Obtain the loading signal x(t), representing the vibration signal of the loading device, and the sensor signal y(t), representing the vibration signal of the scooter's structural response. Set the sampling frequency fs: Ensure the sampling frequency satisfies the Nyquist sampling theorem and is typically selected to be at least 10 times the main frequency of the loading signal. Set the sliding window function w(t): A Hanning, Hamming, or Gaussian window is commonly used, with a window length of Lw. Define the number of sampling points Sw per sliding window movement.

[0082] Calculate the short-time Fourier transform of the loading signal, the expression is: ; Where X(t,f) is the spectrum of the loaded signal at time t and frequency f, is the frequency component in Fourier transform, is the original signal at time τ, It is a time-localized window function used to intercept a section of the signal for Fourier transform;

[0083] Calculate the short-time Fourier transform of the sensor signal, which is expressed as: ;in, is the spectrum of the sensor signal at time t and frequency f, in each time window , select the main frequency , the expression is: Where, is the amplitude of the loaded signal at frequency f, which represents the signal energy distribution; is the amplitude of the sensor signal at frequency f, which represents the signal energy distribution;

[0084] At the main frequency Calculate the phase difference between the load signal and the sensor signal , the expression is: ; is the phase angle of the sensor signal at the main frequency, indicating the initial phase of the vibration; is the phase angle of the loaded signal at the main frequency; the phase difference sequence formed in each time window , k=1,2,…,M, where M is the total number of sliding windows; calculate the standard deviation and mean of the phase difference sequence; calculate the phase difference fluctuation index, the expression is: ; Where HMB is the phase difference fluctuation index, is the mean of the phase difference sequence.

[0085] The phase difference fluctuation index reflects the stability of the phase difference between the sensor signal and the reference signal. A larger phase difference fluctuation index indicates inconsistency between the sensor signal and the reference signal at different loading frequencies or time windows. This can manifest as significant phase difference fluctuations, potentially due to resonance effects, signal lag, or noise and distortion during sensor measurement. In such cases, the sensor signal may not accurately reflect the true mechanical response of the scooter structure, reducing the reliability of the data.

[0086] A smaller phase difference fluctuation index indicates better consistency between the sensor signal and the reference signal, and the phase difference between the signals remains stable during dynamic loading. This indicates that the sensor accurately captures the characteristics of the loading signal, with minimal signal distortion and high measurement reliability. This generally indicates good test conditions, accurately reflecting the response characteristics of the scooter structure, and providing high-quality data for subsequent mechanical analysis.

[0087] Nonlinear harmonic component analysis module: If the sensor signal is inconsistent with the reference signal, the sensor output signal within the resonant frequency and its surrounding frequency range is decomposed into fundamental frequency components and high-order harmonic components. The abnormal proportion of nonlinear components in the fundamental frequency components is analyzed to evaluate the authenticity of the mechanical response.

[0088] First, through a previous step (such as FFT or phase difference analysis), the scooter's resonant frequency range [fstart, fend] is identified. Signals within this range are susceptible to resonance and distortion. The sensor output signal's spectrum data at and around the resonant frequency is extracted as the target signal for analysis.

[0089] The sensor output signal y(t) is decomposed into its frequency domain representation Y(f) using Fourier transform: ; is the kernel function of the Fourier transform. The fundamental frequency component Y(f1) (the fundamental frequency, f1) is the primary energy concentration point of the signal and represents the scooter's linear response to the loading signal. The higher-order harmonic components Y(2f1), Y(3f1), ... are nonlinear responses to the loading signal and are typically caused by nonlinear structural behavior or signal distortion.

[0090] Reconstruct the fundamental frequency and higher-order harmonic signals back into the time domain: ;in, is the amplitude of each harmonic, is the phase, and N is the total number of phases.

[0091] Calculate the energy of the fundamental frequency component , the expression is: ; Calculate the total energy of higher-order harmonics, the expression is: ; Calculate the proportion of higher-order harmonics in the total signal energy, which is defined as the proportion of non-linear components , and the expression is: ; The higher the value, the more serious the non-linear distortion of the signal.

[0092] After analyzing the abnormal situation of the proportion of non-linear components in the fundamental frequency component, generate the abnormal index of the proportion of non-linear components to evaluate the authenticity of the mechanical response. The method for obtaining the abnormal index of the proportion of non-linear components is as follows:

[0093] The sequence of the proportion of non-linear components obtained from the non-linear harmonic component analysis module is denoted as , and the frequency range is [fstart, fend]; divide the proportion sequence into multiple subsequences by a sliding window with a fixed length of m: ; The window sliding step is set to s, usually s < m, to cover more data points.

[0094] In each subsequence, define the embedding dimension m and the similarity tolerance r: ; Count the similarity degree between other subsequences and the current , and the number of template satisfying the condition is denoted as A(m, r): Set the maximum distance of the quantitative template as , and the expression is: ; Increase the embedding dimension to m + 1, repeat the above steps, count the number of matching templates A(m + 1, r), and calculate the abnormal index of the proportion of non-linear components. The expression is: ; In the formula, GQS is the abnormal index of the proportion of non-linear components.

[0095] When the abnormal index of the proportion of non-linear components is larger, it indicates that the randomness and complexity of the proportion of higher-order harmonics in the sensor signal increase, and the non-linear behavior is significant. This usually indicates that the mechanical response of the scooter structure deviates from the ideal linear characteristics under resonance or non-linear distortion conditions. The authenticity of the mechanical response is relatively low in this case, and it may not accurately reflect the actual stress state of the scooter structure, suggesting potential problems in the test conditions or the structure itself that need to be further optimized or corrected.

[0096] When the abnormal index of the proportion of non-linear components is smaller, it indicates that the proportion of higher-order harmonics in the sensor signal is lower and the change rule is stable, and the mechanical response has strong linear characteristics, and the signal is stable and predictable. This means that the mechanical response of the scooter structure can more realistically reflect the actual stress state, and the credibility of the test results is relatively high, providing reliable data support for the mechanical performance evaluation and optimization design.

[0097] Comprehensive Analysis Module: Evaluates the accuracy of the sensor's measurement of the scooter's structural mechanical response under resonance conditions based on the fluctuations in the phase difference between the loading signal and the sensor signal at different loading frequencies, as well as the abnormal proportion of nonlinear components in the fundamental frequency components.

[0098] The phase difference fluctuation index and the nonlinear component proportion anomaly index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses the accuracy value label of the sensor measuring the mechanical response of the scooter structure under resonance as the prediction target for each set of comprehensive feature vectors, and takes minimizing the sum of the prediction errors of the accuracy value labels of the sensor measuring the mechanical response of the scooter structure under all resonance conditions as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence, and the model training is stopped. The accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance is determined according to the model output results, wherein the machine learning model is a polynomial regression model.

[0099] The accuracy of the sensor's measurement of the scooter's structural mechanical response under resonance is obtained by obtaining the corresponding function expression from the comprehensive feature vector training data of the trained machine learning model: Where, is the output function of the model, HMB is the phase difference fluctuation index, GQS is the nonlinear component ratio abnormality index, The accuracy value of the sensor in measuring the mechanical response of the scooter structure under resonance conditions.

[0100] Measurement result classification module: Based on the accuracy of the sensor in measuring the mechanical response of the scooter structure under resonance, the sensor measurement results are divided into accurate measurement results, incompletely accurate measurement results, and inaccurate measurement results, and corresponding processing is performed;

[0101] Comparing the obtained accuracy value of the sensor measuring the mechanical response of the scooter structure under the resonance condition with a gradient standard threshold, where the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and comparing the accuracy value of the sensor measuring the mechanical response of the scooter structure under the resonance condition with the first standard threshold and the second standard threshold respectively;

[0102] If the accuracy value of the sensor measuring the mechanical response of the scooter structure under the resonance condition is greater than the second standard threshold, it indicates that the accuracy of the sensor measuring the mechanical response of the scooter structure under the resonance condition is high, and a high accuracy signal is generated, and the sensor measurement result is classified as an accuracy measurement result;

[0103] If the accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the accuracy of the sensor measuring the mechanical response of the scooter structure under resonance is moderate, and a moderate accuracy signal is generated, and the sensor measurement result is classified as an incomplete accuracy measurement result;

[0104] If the accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance is less than the first standard threshold, it means that the accuracy of the sensor measuring the mechanical response of the scooter structure under resonance is low. At this time, a low accuracy signal is generated, and the sensor measurement result is classified as an inaccurate measurement result.

[0105] If a high-accuracy signal is generated, the data quality is reliable and can be used for further analysis or decision-making. If a medium-accuracy signal is generated, the data has some distortion and can be filtered or compensated as needed. If a low-accuracy signal is generated, the data quality is poor and requires troubleshooting or retesting.

[0106] Prediction and Adjustment Module: For incomplete accuracy measurement results, a fixed measurement time period is used to analyze the sensor signal change trend under the resonant state and predict the accuracy attenuation of long-term testing. If the prediction results show that the signal distortion is uncontrollable, the loading frequency and sensor position are adjusted. If the signal distortion is controllable, the filtering and data compensation algorithms for subsequent tests are optimized.

[0107] For the incomplete accuracy measurement results, that is, the accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance generated in a fixed measurement time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the accuracy values ​​of the sensor measuring the mechanical response of the scooter structure under resonance that are greater than or equal to the first standard threshold and less than or equal to the second standard threshold in subsequent measurement time periods are collected, and a data set is constructed. The change trend of the sensor signal under the resonance state is analyzed according to the mean and standard deviation of the data set, and the accuracy attenuation of long-term testing is predicted.

[0108] If the mean accuracy value within the data set is greater than or equal to the reference threshold for the mean accuracy value, and the standard deviation of the accuracy values ​​is less than the reference threshold for the standard deviation of the accuracy values, the signal trend is stable, the accuracy values ​​are high, and the fluctuations are small. This indicates that signal distortion is controllable and the existing test conditions can be maintained. For subsequent tests, optimize the filtering and data compensation algorithms to further improve data quality.

[0109] If the mean accuracy value is greater than or equal to the reference threshold for the mean accuracy value, and the standard deviation of the accuracy value is greater than or equal to the reference threshold for the standard deviation of the accuracy value, the accuracy value is high, but the signal fluctuation is large, indicating instability. This indicates that the signal distortion is uncontrollable. Optimize the sensor layout or anti-interference design. Filter the signal in the high-fluctuation area to reduce the error.

[0110] If the mean accuracy value is less than the reference threshold for the mean accuracy value, and the standard deviation of the accuracy value is greater than or equal to the reference threshold for the standard deviation of the accuracy value, the accuracy value is low and fluctuates widely, indicating severe signal distortion. This indicates that the signal distortion is uncontrollable and the loading frequency range needs to be adjusted to avoid resonant frequencies. Check the sensor status and replace or relocate the sensor if necessary.

[0111] If the mean accuracy value is less than the reference threshold for the mean accuracy value, and the standard deviation is less than the reference threshold for the standard deviation, the accuracy value is low, but the signal fluctuation is small and the distortion is stable. This indicates that the signal distortion is controllable and the loading conditions (such as the loading amplitude or load distribution) can be optimized. This can enhance the adaptability of subsequent data compensation algorithms and improve accuracy.

[0112] In subsequent tests, optimizing filtering algorithms can effectively reduce signal distortion caused by resonance or noise, improving data reliability and accuracy. For example, adaptive filters dynamically adjust filter parameters to precisely remove high-frequency noise or low-frequency interference based on the varying response frequencies of the scooter structure. Furthermore, bandpass filters can focus on critical frequency ranges outside the resonance range, avoiding attenuation of important components in the test signal and thus more accurately reflecting the structure's true mechanical response.

[0113] In terms of data compensation algorithms, compensation techniques based on machine learning or model prediction can be introduced. By learning from known distorted signals, they can correct for system errors caused by sensors or loading conditions. For example, nonlinear correction algorithms can dynamically compensate for nonlinear distortion based on the proportion of high-order harmonic components. Meanwhile, multi-sensor data fusion algorithms can eliminate deviations caused by single-point anomalies by integrating multi-point data verification. The optimized data processing method not only improves the robustness of the test but also provides higher-quality measurement data for scooter strength assessment.

[0114] This embodiment includes a vibration loading module, a frequency marking module, a phase difference analysis module, a nonlinear harmonic component analysis module, a comprehensive analysis module, a measurement result classification module, and a prediction and adjustment module, designed to comprehensively evaluate the mechanical response of the scooter structure under resonance and the sensor measurement accuracy. The vibration loading module places sensors at key locations on the scooter and applies loads, recording the loading signals as reference signals. The frequency marking module gradually adjusts the loading frequency, marks the resonance interval, and synchronously acquires signals. The phase difference analysis module uses fast Fourier transform to calculate the phase difference fluctuations between the loading signal and the sensor signal. The nonlinear harmonic component analysis module analyzes the abnormal ratio of the fundamental frequency and higher-order harmonics in the signal to assess the authenticity of the mechanical response. The comprehensive analysis module combines phase difference fluctuations and nonlinear abnormal ratios to predict measurement accuracy under resonance. The measurement result classification module classifies data into high, medium, and low accuracy based on accuracy values. The prediction and adjustment module analyzes the changing trends of incomplete accuracy results to predict signal attenuation and optimize loading conditions or data compensation to ensure test reliability and data quality.

[0115] See also Figure 2-Figure 6 As shown, a foldable scooter comprises a bracket 1 fixedly connected to the scooter chassis, a lower rotating rod 2 and an upper rotating rod 3 for controlling the steering of the front wheel of the scooter, the lower rotating rod 2 is rotatably mounted inside the bracket 1, and the upper end of the lower rotating rod 2 is rotatably connected to the lower end of the upper rotating rod 3 through an adapter 4 to realize the folding of the scooter; a wheel frame 201 is provided at the lower end of the lower rotating rod 2, and at least two groups of axially distributed fixing screws 5 are provided between the wheel frame 201 and the lower rotating rod 2. By providing a plurality of axially distributed fixing screws 5, the connection strength between the lower rotating rod 2 and the wheel frame 201 is increased, and at the same time, the fixing screws 5 can be prevented from falling off, thereby improving the stability of the connection between the lower rotating rod 2 and the wheel frame 201.

[0116] On this basis, an axis rod is provided in the adapter 4, and the adapter 4 is fixedly installed on the upper end of the lower rotating rod 2. A connecting member 401 is rotatably installed on the surface of the axis rod, and a fixing screw 2 6 is installed at the connection between the connecting member 401 and the lower end of the upper rotating rod 3. By providing the fixing screw 2 6, the connection strength between the adapter 4 and the upper rotating rod 3 is improved, thereby improving the stability of the connection between the upper rotating rod 3 and the lower rotating rod 2.

[0117] On this basis, a connecting rod 7 is rotatably installed on the surface of the shaft rod, and a clamping block 701 is provided on the upper surface of the connecting rod 7; a clamping hole 301 is provided on the surface of the upper rotating rod 3. When the connecting rod 7 is rotated upward, the clamping block 701 can cooperate with the clamping hole 301. When the clamping block 701 cooperates with the clamping hole 301, the upper rotating rod 3 and the lower rotating rod 2 are coaxial and relatively fixed. When the clamping block 701 is separated from the clamping hole 301, the upper rotating rod 3 can be rotated downward relative to the lower rotating rod 2 to realize the folding action.

[0118] The above is only a specific implementation method of the present application, but the scope of protection of the present application is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in this application, which should be covered by the scope of protection of the present application.

Claims

1. The scooter strength testing system based on mechanical sensors is characterized by: It includes vibration loading module, frequency marking module, phase difference analysis module, nonlinear harmonic component analysis module, comprehensive analysis module, measurement result classification module and prediction adjustment module; Vibration loading module: Multiple sensors are installed at different locations on the scooter, covering the loading point and monitoring areas away from the loading point. Vibration loading equipment is set up in each monitoring area, and the loading signals are recorded as reference signals. Frequency Marking Module: This module applies a vibration load to the scooter, gradually adjusts the loading frequency, marks the intervals that exceed the structural resonance frequency, and synchronously collects the loading signal and sensor response signal within the marked interval to ensure time series alignment. Phase difference analysis module: uses fast Fourier transform to extract the phase information of the loading signal and sensor signal on the main frequency component, analyzes the phase difference fluctuation between the loading signal and the sensor signal, and evaluates the consistency of the sensor signal with the reference signal; Nonlinear harmonic component analysis module: If the sensor signal is inconsistent with the reference signal, the sensor output signal within the resonant frequency and its surrounding frequency range is decomposed into fundamental frequency components and higher-order harmonic components. The abnormal proportion of nonlinear components in the fundamental frequency components is analyzed to evaluate the authenticity of the mechanical response. Comprehensive Analysis Module: This module evaluates the accuracy of the sensor in measuring the scooter's structural mechanical response under resonance conditions, based on the fluctuations in the phase difference between the loading signal and the sensor signal at different loading frequencies, as well as the abnormal proportion of nonlinear components in the fundamental frequency components. Measurement result classification module: Based on the accuracy of the sensor in measuring the mechanical response of the scooter structure under resonance, the sensor measurement results are divided into accurate measurement results, incompletely accurate measurement results, and inaccurate measurement results, and corresponding processing is performed; Prediction and Adjustment Module: For incomplete accuracy measurement results, a fixed measurement time period is used to analyze the sensor signal change trend under the resonant state and predict the accuracy attenuation of long-term testing. If the prediction results show that the signal distortion is uncontrollable, the loading frequency and sensor position are adjusted. If the signal distortion is controllable, the filtering and data compensation algorithms for subsequent tests are optimized.

2. The scooter strength testing system based on a mechanical sensor according to claim 1, characterized in that: In the phase difference analysis module, the phase difference fluctuation between the loading signal and the sensor signal is analyzed to generate a phase difference fluctuation index. The method for obtaining the phase difference fluctuation index is as follows: Obtain the loading signal x(t), which represents the vibration signal of the loading device, and the sensor signal y(t), which represents the vibration signal of the scooter structure response. Set the sampling frequency fs, set the sliding window function w(t), the window length Lw, and define the number of sampling points Sw for each sliding window movement; Calculate the short-time Fourier transform of the loading signal, the expression is: ; Where X(t,f) is the spectrum of the loaded signal at time t and frequency f, is the frequency component in Fourier transform, is the original signal at time τ, It is a time-localized window function used to intercept a section of the signal for Fourier transform; Calculate the short-time Fourier transform of the sensor signal, which is expressed as: ;in, is the spectrum of the sensor signal at time t and frequency f, in each time window , select the main frequency , the expression is: Where, is the amplitude of the loading signal at frequency f, which represents the signal energy distribution; is the amplitude of the sensor signal at frequency f, which represents the signal energy distribution; At the main frequency Calculate the phase difference between the load signal and the sensor signal , the expression is: ; is the phase angle of the sensor signal at the main frequency, indicating the initial phase of the vibration; is the phase angle of the loading signal at the main frequency; The phase difference sequence formed in each time window , k=1,2,…,M, where M is the total number of sliding windows; calculate the standard deviation and mean of the phase difference sequence; calculate the phase difference fluctuation index, the expression is: ; Where HMB is the phase difference fluctuation index, is the mean of the phase difference sequence.

3. The scooter strength testing system based on a mechanical sensor according to claim 2, characterized in that: In the nonlinear harmonic component analysis module, the nonlinear component ratio abnormality in the fundamental frequency component is analyzed to generate a nonlinear component ratio abnormality index. The nonlinear component ratio abnormality index is obtained as follows: The nonlinear component ratio sequence obtained from the nonlinear harmonic component analysis module is recorded as , the frequency range is [fstart, fend]; the proportion sequence Divide the sliding window into multiple subsequences with a fixed length of m: ; The window sliding step is set to s; In each subsequence, define the embedding dimension m and similarity tolerance r: ; Count other subsequences and the current The degree of similarity, the number of templates that meet the conditions is recorded as A(m,r): The maximum distance between the set templates is , the expression is: ; Increase the embedding dimension to m+1, count the number of matching templates A(m+1,r), and calculate the abnormal index of nonlinear component proportion. The expression is: ; Where GQS is the abnormal index of nonlinear component proportion.

4. The scooter strength testing system based on a mechanical sensor according to claim 3, characterized in that: In the comprehensive analysis module, the phase difference fluctuation index and the nonlinear component proportion anomaly index are converted into comprehensive feature vectors, and the comprehensive feature vectors are used as the input of the machine learning model. The machine learning model uses each set of comprehensive feature vectors to predict the accuracy value label of the sensor measuring the mechanical response of the scooter structure under resonance as the prediction target, and takes minimizing the sum of the prediction errors of the accuracy value labels of the sensor measuring the mechanical response of the scooter structure under all resonance conditions as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence, and the model training is stopped. The accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance is determined according to the model output results. Among them, the machine learning model is a polynomial regression model.

5. The scooter strength testing system based on mechanical sensors according to claim 4, characterized in that: In the measurement result classification module, the sensor measurement results are divided into accurate measurement results, incompletely accurate measurement results, and inaccurate measurement results according to the accuracy of the sensor in measuring the mechanical response of the scooter structure under resonance conditions. Specifically: Comparing the obtained accuracy value of the sensor measuring the mechanical response of the scooter structure under the resonance condition with a gradient standard threshold, where the gradient standard threshold includes a first standard threshold and a second standard threshold, and the first standard threshold is less than the second standard threshold, and comparing the accuracy value of the sensor measuring the mechanical response of the scooter structure under the resonance condition with the first standard threshold and the second standard threshold respectively; If the accuracy value of the sensor measuring the mechanical response of the scooter structure under the resonance condition is greater than the second standard threshold, it indicates that the accuracy of the sensor measuring the mechanical response of the scooter structure under the resonance condition is high, and a high accuracy signal is generated, and the sensor measurement result is classified as an accuracy measurement result; If the accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, it means that the accuracy of the sensor measuring the mechanical response of the scooter structure under resonance is moderate, and a moderate accuracy signal is generated, and the sensor measurement result is classified as an incomplete accuracy measurement result; If the accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance is less than the first standard threshold, it means that the accuracy of the sensor measuring the mechanical response of the scooter structure under resonance is low. At this time, a low accuracy signal is generated, and the sensor measurement result is classified as an inaccurate measurement result.

6. The scooter strength testing system based on mechanical sensors according to claim 1, characterized in that: In the prediction and adjustment module, for the incomplete accuracy measurement results, that is, the accuracy value of the sensor measuring the mechanical response of the scooter structure under resonance generated within a fixed measurement time period is greater than or equal to the first standard threshold and less than or equal to the second standard threshold, the accuracy values ​​of the sensor measuring the mechanical response of the scooter structure under resonance that are greater than or equal to the first standard threshold and less than or equal to the second standard threshold in subsequent measurement time periods are collected, and a data set is constructed. The change trend of the sensor signal under the resonance state is analyzed based on the mean and standard deviation of the data set, and the accuracy attenuation of long-term testing is predicted.

7. The scooter strength testing system based on a mechanical sensor according to claim 6, characterized in that: If the mean accuracy value in the data set is greater than or equal to the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is less than the reference threshold of the standard deviation of the accuracy value, the signal change trend is stable, the accuracy value is high and the fluctuation is small, indicating that the signal distortion is controllable. Maintain the existing test conditions, optimize the filtering and data compensation algorithms for subsequent tests, and further improve the data quality; If the mean accuracy value is greater than or equal to the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is greater than or equal to the reference threshold of the standard deviation of the accuracy value, the accuracy value is high, but the signal fluctuation is large, and there are unstable factors, indicating that the signal distortion is uncontrollable. Optimize the layout of the sensor and filter the signal in the high fluctuation area to reduce the error; If the mean accuracy value is less than the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is greater than or equal to the reference threshold of the standard deviation of the accuracy value, the accuracy value is low and fluctuates greatly, and the signal distortion is serious, indicating that the signal distortion is uncontrollable and the loading frequency range needs to be adjusted to avoid the resonant frequency; If the mean accuracy value is less than the reference threshold of the mean accuracy value, and the standard deviation of the accuracy value is less than the reference threshold of the standard deviation of the accuracy value, the accuracy value is low, but the signal fluctuation is small and the distortion is stable, indicating that the signal distortion is controllable and the loading conditions are optimized.

8. A folding scooter, using the scooter strength testing system based on a mechanical sensor as claimed in any one of claims 1 to 7, comprising a bracket (1) fixedly connected to the scooter chassis, a lower turning rod (2) and an upper turning rod (3) for controlling the steering of the scooter front wheel, characterized in that: The lower rotating rod (2) is rotatably mounted inside the bracket (1), and the upper end of the lower rotating rod (2) is rotatably connected to the lower end of the upper rotating rod (3) via a connecting member (4) to achieve folding of the scooter; A wheel frame (201) is provided at the lower end of the lower rotating rod (2), and at least two groups of fixing screws (5) distributed along the axial direction are provided between the wheel frame (201) and the lower rotating rod (2).

9. The folding scooter according to claim 8, characterized in that: The adapter (4) is provided with a shaft, and a connecting piece (401) is rotatably mounted on the surface of the shaft. A second fixing screw (6) is mounted at the connection between the connecting piece (401) and the lower end of the upper rotating rod (3).

10. The folding scooter according to claim 9, characterized in that: A connecting rod (7) is rotatably mounted on the surface of the shaft, and a clamping block (701) is provided on the upper surface of the connecting rod (7); A clamping hole (301) is provided on the surface of the upper rotating rod (3), and when the connecting rod (7) is rotated upward, the clamping block (701) can cooperate with the clamping hole (301).

Images