Bicycle brake performance testing method based on multivariate data

Abstract

The invention relates to the technical field of brake performance testing, in particular to a bicycle brake performance testing method based on multivariate data, which comprises the following steps of: performing brake testing on a sampled bicycle and determining initial test data; the sampled bicycles are divided into a plurality of test groups for a periodic cycle test, and a brake test is carried out after the test is finished; determining the test characterization trend of the test group according to the brake test result of each sampling bicycle in the single test group, and determining the brake state of the test group in response to the test characterization trend; and determining whether to determine the first maintenance time of the batch of assembled bicycles in combination with the test period according to the brake state of the test group. The problems that when a bicycle brake system is newly assembled, all parts are usually tightened according to standard torque, but in the long-term riding process, due to vibration, uneven stress and abrasion running-in among the parts, part of the parts are likely to loosen, and therefore the brake performance and riding safety are affected are solved.

Description

Technical Field

[0001] This invention relates to the field of brake performance testing technology, and in particular to a method for testing bicycle brake performance based on multivariate data. Background Technology

[0002] Traditional brake testing relies heavily on visual inspection and simple mechanical measurements. Inspectors use their experience to observe brake components for wear and looseness, check the tightness of screws with wrenches and other tools, and judge performance by feeling the force and travel of the brake lever. This method is highly subjective and has limited accuracy. With technological advancements, various advanced testing technologies are emerging. 3D machine vision technology can accurately capture micron- or even sub-millimeter-level dimensional differences in brake components, making it extremely effective for detecting brake pad thickness, shape, surface roughness, and material defects.

[0003] Application No. 202211655785.7 discloses a bicycle brake component performance testing device, belonging to the field of brake performance testing technology. It includes a support frame and a top plate. A brake simulation component is installed on the upper end of the top plate to simulate the effect of hand-operated braking and record the number of braking cycles. The movable end of the brake simulation component is made of a soft material. A translation component is installed at the inner bottom of the support frame to synchronously move a static detection component and a dynamic detection component to sequentially perform static and dynamic testing on the brake component. The translation component is equipped with a static detection component for static testing and a dynamic detection component for dynamic testing of the brake component. Through this method, the present invention can achieve continuous dynamic and static testing within a limited space. Simultaneously, during testing, the use of a soft material for the movable end of the brake simulation component facilitates the simulation of hand-operated braking, reducing the gap with the actual hand-operated braking effect and thus reducing testing errors.

[0004] Therefore, it is evident that the existing technology has the following problems:

[0005] When a bicycle braking system is newly assembled, the components are usually tightened to the standard torque. However, during long-term riding, due to vibration, uneven force, and wear and tear between parts, some parts may become loose, which can affect braking performance and riding safety. Summary of the Invention

[0006] To address this issue, the present invention provides a bicycle braking performance testing method based on multivariate data, which overcomes the problem in the prior art where bicycle braking systems are typically tightened to standard torque during new assembly, but loosening may occur in some parts during long-term riding due to vibration, uneven force, and wear and break-in between parts, thus affecting braking performance and riding safety.

[0007] To achieve the above objectives, the present invention provides a method for testing bicycle braking performance based on multivariate data, comprising:

[0008] Step S1: Sample the assembled bicycles from the same batch and perform brake tests on all sampled bicycles to determine initial test data based on the brake test results. The initial test data includes initial braking time, initial audio characteristics, and initial vibration curve.

[0009] Step S2: Divide the sampled bicycles into several test groups, conduct periodic cyclic tests on each test group, and conduct brake tests after the test. The number of test groups is at least 3.

[0010] Step S3, determining the test characterization trend of the test group based on the brake test results of each sampled bicycle in a single test group, and determining the braking state of the test group in response to the test characterization trend, including:

[0011] Determine the test data for the corresponding test group and compare it with the initial test data, so as to determine the braking state of the corresponding test group based on the comparison results;

[0012] Alternatively, determine that the braking status of this test group is abnormal;

[0013] The test characterization trend includes a consistent characterization trend and a inconsistent characterization trend; the test data includes test braking time, test audio characteristics, and test vibration curve; and the braking state includes a normal state and an abnormal state.

[0014] Step S4: Determine whether to combine the test cycle to determine the first maintenance time for this batch of assembled bicycles based on the braking status of the test group.

[0015] As a preferred technical solution for a bicycle braking performance testing method based on multivariate data, the process of determining initial test data based on the braking test results in step S1 includes:

[0016] Step S11: Determine the initial braking time based on the average braking time of each assembled bicycle;

[0017] Step S12: Draw a spectrum diagram based on the brake audio information of a single assembled bicycle to determine the energy distribution and frequency peak of the spectrum diagram of the assembled bicycle. Determine the initial audio characteristics based on the average value of the energy distribution and frequency peak of the spectrum diagrams of each assembled bicycle. The initial audio characteristics include the initial energy distribution and the initial frequency peak.

[0018] Step S13: Determine the initial vibration curve based on the same spectral characteristics of the brake vibration curves of each assembled bicycle.

[0019] As a preferred technical solution for a bicycle braking performance testing method based on multivariate data, in step S1, the braking test process includes:

[0020] Step A1: Install sensor groups on each of the sampled bicycles;

[0021] Step A2: Fix the sampling bicycles to the roller test stand, drive each sampling bicycle to accelerate to the standard speed and maintain the standard time.

[0022] Step A3: Control the brake lever to the preset lever travel and obtain braking duration, brake audio information and brake vibration curve.

[0023] As a preferred technical solution for a bicycle braking performance testing method based on multivariate data, in step S2, the cycle period of each test group is different.

[0024] As a preferred technical solution for a bicycle braking performance testing method based on multivariate data, step S3, the process of determining the test characterization trend of a single test group based on the braking test results of each sampled bicycle in the test group, includes:

[0025] Step S31: Determine the duration fluctuation parameter based on the ratio of the average deviation to the average value of the test braking time of each sampled bicycle.

[0026] Step S32: Based on the calculation results of the duration fluctuation parameter, determine whether to determine the test characterization trend of the test group according to the test audio characteristics and test vibration curve, wherein:

[0027] In response to the duration fluctuation parameter being greater than the preset fluctuation parameter, the test characterization trend of this experimental group is determined to be a characterization inconsistency trend;

[0028] Step S33: In response to the duration fluctuation parameter being less than or equal to the preset fluctuation parameter, the feature similarity and curve similarity are determined by the machine learning model based on the test audio features and test vibration curves of each sampled bicycle.

[0029] Step S34: Determine the test characterization trend of the experimental group based on the feature similarity and curve similarity, wherein:

[0030] When both feature similarity and curve similarity are greater than or equal to the corresponding preset thresholds, the test characterization trend of this experimental group is determined to be a characterization consistency trend.

[0031] Conversely, the test characterization trend of this experimental group is determined to be a characterization inconsistency trend.

[0032] As a preferred technical solution for a bicycle braking performance testing method based on multivariate data, in step S3, the process of determining the braking state of the test group in response to the test characterization trend includes:

[0033] In response to the determination result representing a consistent trend, the test data of the corresponding test group are determined and compared with the initial test data, so as to determine the braking state of the corresponding test group based on the comparison result;

[0034] Alternatively, in response to the judgment results that characterize different trends, the braking state of the test group is determined to be an abnormal state.

[0035] As a preferred technical solution for a bicycle braking performance testing method based on multivariate data, in step S3, the braking state of the corresponding test group is determined according to the comparison result between the experimental test data and the initial test data, including:

[0036] In response to the fact that the test data and the initial test data meet similarity conditions, the braking state of the corresponding test group is determined to be normal.

[0037] In response to the fact that the test data and the initial test data do not meet the similarity condition, the braking state of the corresponding test group is determined to be an abnormal state.

[0038] The similarity conditions include the ratio of the test braking time to the initial braking time being within a preset range, the test audio characteristics being the same as the initial audio characteristics, and the test vibration curve being similar to the initial vibration curve.

[0039] As a preferred technical solution for a bicycle braking performance testing method based on multivariate data, in step S3, the vibration range is determined according to the initial vibration curve, and the test vibration curve is judged to be similar to the initial vibration curve based on the result that the test vibration curve is within the vibration range.

[0040] As a preferred technical solution for a bicycle braking performance testing method based on multivariate data, in step S4, the first maintenance time of the assembled bicycles in this batch is determined as the standard time based on the judgment that the braking status of each test group is in a normal state.

[0041] As a preferred technical solution for a bicycle braking performance testing method based on multivariate data, in step S4, based on the determination results of abnormal braking states in each test group, it is determined that the first maintenance time of the assembled bicycles in this batch is less than the standard time, combined with the test cycle.

[0042] Compared with existing technologies, the beneficial effects of this invention are as follows: The bicycle braking performance testing method based on multivariate data provided by this invention improves the scientific rigor and reliability of the test through innovative methods such as multivariate data acquisition, grouped cyclic testing, and trend analysis. In the initial stage of testing, multivariate data such as initial braking duration, audio characteristics, and vibration curves are collected to provide a precise benchmark for subsequent comparisons. In the grouped cyclic testing, different cycles simulate actual usage scenarios, dynamically capturing the trend of braking performance changes over time. The braking state is judged based on the test characteristics, avoiding the limitations of single data and quickly identifying batch product quality differences to avoid misjudgment. Simultaneously, by optimizing the model through multi-batch data, the braking system state can be accurately predicted. Finally, combined with the test cycle, the first maintenance time is scientifically determined, effectively reducing the risk of brake failure, ensuring riding safety, and providing data support for production and maintenance, thereby improving the overall quality and management efficiency of bicycle products.

[0043] In particular, this invention establishes a precise performance benchmark at the time of initial assembly by collecting multi-dimensional data such as initial braking duration, audio characteristics, and vibration curves, providing a reference for subsequent monitoring of performance changes caused by component loosening. Multiple groups undergo cyclical tests of different periods to simulate complex working conditions such as vibration and stress during long-term riding, dynamically capturing the performance evolution of the braking system during use and promptly identifying problems such as loose screws caused by component wear and break-in, and uneven stress. Based on the test characteristics, the brake status is judged, which can identify performance consistency changes caused by component loosening and also distinguish anomalies caused by batch product quality differences, avoiding missed or incorrect judgments. The initial maintenance time is determined by combining the test cycle, allowing for early intervention maintenance before component loosening causes serious safety problems, reducing the risk of brake failure caused by loose brake components, effectively ensuring riding safety, and optimizing maintenance strategies to improve the reliability of the bicycle braking system throughout its entire life cycle.

[0044] In particular, through multi-dimensional data collection and scientific analysis, a precise benchmark is established for monitoring bicycle brake performance: In determining braking time, the initial data is determined by the average value, which can effectively avoid individual errors, and based on the correlation between brake lever travel and braking time, the potential risk of brake loosening is accurately captured; in terms of audio feature analysis, audio is collected by a microphone sensor and converted into a spectrum diagram, and the faulty component is accurately located based on the unique frequency and energy distribution characteristics caused by the loosening of different brake components; when acquiring vibration curves, the triaxial acceleration data of the handle is integrated by an accelerometer to construct an initial curve that comprehensively reflects the vibration state of the handle, which can keenly detect abnormal vibrations caused by component loosening; the three work together to construct a complete performance profile of the brake system at the time of assembly from three key perspectives: braking efficiency, sound signal, and vibration state, providing a reliable reference for subsequent long-term monitoring of component loosening and prediction of changes in brake performance, greatly improving the accuracy and timeliness of brake system fault warning;

[0045] In particular, through a standardized and systematic brake testing process, accurate and comprehensive collection of brake performance data was achieved; a sensor group including a timer, microphone sensor, and accelerometer was installed to cover the key parameters of brake performance collection from multiple dimensions; the sampled bicycles were uniformly accelerated to a standard speed and maintained using a roller test bench to simulate real riding conditions and ensure the validity and comparability of the test data; the brake lever was controlled to a preset stroke to accurately trigger the maximum braking force, and braking duration, brake audio information, and brake vibration curve were acquired simultaneously. From the three levels of braking efficiency, sound characteristics, and vibration state, a complete performance baseline for the brake system at the time of assembly was established, providing a scientific and reliable reference for subsequent monitoring of component loosening and prediction of performance degradation, effectively improving the accuracy and timeliness of brake system fault warning;

[0046] In particular, step S3 achieves efficient and accurate assessment of the bicycle braking system status through rigorous data processing and logical judgment. In determining the trend of test characteristics, braking time is the primary criterion. By calculating the duration fluctuation parameter, differences in braking performance within groups are quickly identified. When the fluctuation exceeds a preset value, it is directly judged as a trend of inconsistent characteristics, reducing subsequent complex analysis and improving efficiency. When the fluctuation is within a reasonable range, a machine learning model is introduced to perform similarity analysis on audio features and vibration curves. This multi-dimensional data is used to comprehensively consider the braking system status, avoiding misjudgment based on a single indicator. In the braking status determination stage, the test data is compared with the initial data using multiple parameters. Strict similarity conditions are set from three dimensions: braking time, audio features, and vibration curves, accurately distinguishing between normal and abnormal states. This ensures the accuracy of the judgment and allows for the location of specific loose parts based on audio features in abnormal states, providing precise guidance for repair. The entire process is interconnected, balancing efficiency and accuracy, effectively improving the reliability and specificity of braking system fault diagnosis, and providing strong support for ensuring riding safety.

[0047] In particular, step S4, based on scientific test result analysis, achieves precise and dynamic management of bicycle brake system maintenance strategies: when the brakes of each test group are in normal condition, a standardized maintenance time is uniformly determined, providing a standardized reference for brake system maintenance under normal usage scenarios, avoiding over-maintenance or under-maintenance, and balancing maintenance costs and safety; in the case of abnormal conditions, the initial maintenance time is flexibly shortened according to the abnormal conditions of different test groups and the test cycle. For example, based on the cycle differences of abnormal test groups, the maintenance time is precisely set to 2 to 3 months, which can intervene in maintenance in the early stage of brake system problems, effectively reducing the safety risks caused by the decline in brake performance due to loose parts; at the same time, the assembly process is optimized in reverse for the abnormal conditions of the first test group. By manually tightening and adjusting the robot assembly parameters, the potential for loosening caused by subsequent assembly problems is reduced from the source, forming a closed-loop management model of "test-feedback-optimization", which significantly improves the reliability and safety of the bicycle brake system throughout its entire life cycle and also provides a basis for continuous improvement in the manufacturing process. Attached Figure Description

[0048] Figure 1 This is a flowchart illustrating the steps of a bicycle braking performance testing method based on multivariate data, as described in an embodiment of the present invention.

[0049] Figure 2 This is a flowchart for determining the braking state of a single test group in an embodiment of the present invention. Detailed Implementation

[0050] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0051] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0052] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0053] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0054] Please see Figure 1 The diagram illustrates the steps of a bicycle braking performance testing method based on multivariate data, as described in an embodiment of the present invention. This embodiment provides a bicycle braking performance testing method based on multivariate data, comprising:

[0055] Step S1: Sample the assembled bicycles from the same batch and perform brake tests on all sampled bicycles to determine initial test data based on the brake test results. The initial test data includes initial braking time, initial audio characteristics, and initial vibration curve. It can be understood that the initial test data is the normal state data measured when the bicycles are just assembled and all fixing screws related to the brake system are tightened. This data can be used as a reference for subsequent tests to detect any abnormalities in the subsequent data.

[0056] Step S2: Divide the sampled bicycles into several test groups, conduct periodic cyclic tests on each test group, and conduct brake tests after the test. The number of test groups is at least 3. It can be understood that different cycle periods for each group can help understand the braking performance and the state of the fixing screws at different stages of the batch of bicycles. Furthermore, the data can be used to train a model of the loose screw state to predict the state of the batch of bicycles.

[0057] In practice, subsequent batches of bicycles can be supplemented with test data to optimize and update the model, making it more accurate in predicting the corresponding batches of bicycles.

[0058] Step S3, determining the test characterization trend of the test group based on the brake test results of each sampled bicycle in a single test group, and determining the braking state of the test group in response to the test characterization trend, including:

[0059] Determine the test data for the corresponding test group and compare it with the initial test data, so as to determine the braking state of the corresponding test group based on the comparison results;

[0060] Alternatively, determine that the braking status of this test group is abnormal;

[0061] The test characterization trend includes a consistent characterization trend and a inconsistent characterization trend; the test data includes test braking time, test audio characteristics, and test vibration curve; and the braking state includes a normal state and an abnormal state.

[0062] It is understandable that the test characterization trend indicates the consistency of the test results of each sampled bicycle in the test group. It is only necessary to determine the braking status when the test results of the sampled bicycles in the group are consistent. If the test results of the sampled bicycles in the group are not consistent, it means that the braking performance of the sampled bicycles is different (i.e. the quality of the batch of bicycles is inconsistent), and the corresponding braking status should be directly defined as an abnormal state.

[0063] Step S4: Determine whether to combine the test cycle to determine the first maintenance time for this batch of assembled bicycles based on the braking status of the test group.

[0064] Specifically, in step S1, the process of determining the initial test data based on the brake test results includes:

[0065] Step S11: Determine the initial braking time based on the average braking time of each assembled bicycle. It can be understood that when the brake lever travel is the same, the longer the braking time, the looser the braking device is (i.e., there are loose fixing screws in the braking system).

[0066] Step S12: Draw a spectrum diagram based on the brake audio information of a single assembled bicycle to determine the energy distribution and frequency peak of the spectrum diagram of the assembled bicycle. Determine the initial audio characteristics based on the average value of the energy distribution and frequency peak of the spectrum diagrams of each assembled bicycle. The initial audio characteristics include the initial energy distribution and the initial frequency peak.

[0067] In practice, the braking audio information during the braking test is acquired through a microphone sensor; then, the time-domain signal is converted into a frequency-domain signal using Fourier transform to obtain a spectrum; the spectrum is plotted with frequency on the horizontal axis and signal energy (amplitude) on the vertical axis.

[0068] In practice, the collision sound of the brake caliper shaking has energy mainly concentrated in the low frequency range (50Hz~500Hz); the "squeaking" sound of the brake cable rubbing has prominent high-frequency components, with energy concentrated in the 1000Hz~5000Hz range; the harsh sound of abnormal friction of the brake pads has a higher frequency and may have obvious peaks in the 3000Hz~10000Hz range; therefore, the problem of which brake component is determined by the audio characteristics.

[0069] Step S13: Determine the initial vibration curve based on the same spectral characteristics of the brake vibration curves of each assembled bicycle.

[0070] It is understandable that the brake vibration curve refers to the vibration frequency of the brake lever during the brake test. The vibration curve is plotted with time on the horizontal axis and vibration acceleration on the vertical axis. In practice, an acceleration sensor is installed on the sampled bicycle handlebar to measure the acceleration changes of the brake lever in the X, Y, and Z axes. The sensor then integrates the electrical signals from the three axes into a vibration acceleration signal before outputting it.

[0071] Specifically, in step S1, the brake test process includes:

[0072] Step A1: Install sensor groups on each of the sampling bicycles; it is understood that the sensor groups include timers, microphone sensors, and accelerometers; wherein, the microphone sensors and timers do not need to be installed on the sampling bicycles, but can be placed around the roller test bench to collect audio information and record time.

[0073] Step A2: Fix the sampling bicycles to the roller test stand, drive each sampling bicycle to accelerate to the standard speed and maintain the standard time; it should be understood that any existing technology can be used as the roller test stand.

[0074] Step A3: Control the brake lever to the preset lever travel distance, and obtain the braking duration, brake audio information, and brake vibration curve. It can be understood that the preset lever travel distance is typically the distance the lever is squeezed when the brake reaches maximum braking force.

[0075] Specifically, in step S2, the cycle period of each test group is different.

[0076] It is understandable that there should be at least 10 cycles between any two experimental groups;

[0077] In practice, a single cycle must pass through four road sections: normal road section, bumpy road section, water crossing section, and normal road section; each road section must include at least three speed ranges: low speed (below 10km / h), medium speed (10km / h to 25km / h), and high speed (above 25km / h); each speed range must be maintained for at least 5 minutes and at least 5 stops must be made;

[0078] In practice, it is preferable to set the cycle period of the first test group to 30, the cycle period of the second test group to 50, and the cycle period of the third test group to 100.

[0079] Understandably, setting up test groups and different cycle periods aims to systematically simulate the real-world usage scenarios of a bicycle braking system under complex operating conditions, accurately monitoring its performance changes and component status. Multiple test groups, through different cycle periods, can cover the braking system's condition under varying usage durations and wear levels, simulating the effects of vibration, uneven stress, and component wear and break-in during long-term riding. Each test group's cycle includes various road sections (normal, bumpy, water crossing) and speed ranges (low, medium, high speed), with each speed range maintained for a certain duration and multiple braking stops, simulating diverse riding environments and braking frequency, comprehensively examining the braking system's performance under different road conditions and braking intensities. The combination of these two approaches can capture the performance degradation trend of the braking system over time and also detect abnormal changes caused by component loosening, providing data support for predicting braking system failures and determining reasonable maintenance intervals, effectively improving the safety and reliability of the braking system.

[0080] Please see Figure 2 The diagram shows a flowchart illustrating the determination of the braking state of a single test group according to an embodiment of the present invention. Specifically, in step S3, the process of determining the test characteristic trend of the test group based on the braking test results of each sampled bicycle in the single test group includes:

[0081] Step S31: Determine the duration fluctuation parameter based on the ratio of the average deviation to the average value of the test braking time of each sampled bicycle; it is understood that braking time is the fastest and most intuitive data to reflect the performance of the bicycle braking system.

[0082] Step S32: Based on the calculation results of the duration fluctuation parameter, determine whether to determine the test characterization trend of the test group according to the test audio characteristics and test vibration curve, wherein:

[0083] In response to the duration fluctuation parameter being greater than the preset fluctuation parameter, the test characterization trend of this experimental group is determined to be a characterization inconsistency trend;

[0084] It is understandable that if the braking times of the sampled bicycles are not uniform, it means that the performance of the braking system of this group of sampled bicycles is very different. Therefore, it is not necessary to use other data to directly determine that it is a characteristic of the difference.

[0085] In implementation, the preset fluctuation parameter is set to [0.05, 0.1]. The smaller the preset fluctuation parameter, the greater the deviation between the sampled bicycles representing a consistent trend; preferably, it is set to 0.1.

[0086] Step S33: In response to the duration fluctuation parameter being less than or equal to the preset fluctuation parameter, the feature similarity and curve similarity are determined by the machine learning model based on the test audio features and test vibration curves of each sampled bicycle.

[0087] In practice, the machine learning model is trained on prior data and can be used directly.

[0088] Step S34: Determine the test characterization trend of the experimental group based on the feature similarity and curve similarity, wherein:

[0089] When both feature similarity and curve similarity are greater than or equal to the corresponding preset thresholds, the test characterization trend of this experimental group is determined to be a characterization consistency trend.

[0090] Conversely, the test characterization trend of this experimental group is determined to be a inconsistent characterization trend;

[0091] In practice, the preset thresholds for both similarity scores are greater than 85%. The higher the preset threshold, the more similar the test data between the sampled bicycles that represent a consistent trend, and the more consistent the quality.

[0092] Specifically, in step S3, the process of determining the braking state of the test group in response to the test characterization trend includes:

[0093] In response to the determination result representing a consistent trend, the test data of the corresponding test group is determined and compared with the initial test data, so as to determine the braking state of the corresponding test group based on the comparison result; it is understood that the method for obtaining the test data is the same as the method for determining the initial test data.

[0094] Alternatively, in response to the judgment results that characterize different trends, the braking state of the test group is determined to be an abnormal state.

[0095] Specifically, in step S3, determining the braking state of the corresponding test group based on the comparison between the experimental test data and the initial test data includes:

[0096] In response to the fact that the test data and the initial test data meet similarity conditions, the braking state of the corresponding test group is determined to be normal.

[0097] In response to the fact that the test data and the initial test data do not meet the similarity condition, the braking state of the corresponding test group is determined to be an abnormal state.

[0098] The similarity conditions include the ratio of the test braking time to the initial braking time being within a preset range, the test audio characteristics being the same as the initial audio characteristics, and the test vibration curve being similar to the initial vibration curve.

[0099] In practice, the preset range is [0.9, 1.1].

[0100] In practice, if the similarity between the test audio features and the initial audio features exceeds 90%, the test audio features are considered to be the same as the initial audio features.

[0101] Understandably, based on the results of the abnormal condition assessment, combined with the energy distribution and frequency peaks of the test audio characteristics, it is possible to determine which part (brake cable, brake pad, brake caliper) has loose fixing screws.

[0102] Specifically, in step S3, the vibration range is determined based on the initial vibration curve, and the test vibration curve is determined to be similar to the initial vibration curve based on the result that the test vibration curve is within the vibration range.

[0103] Understandably, the initial vibration interval is determined based on the maximum and minimum points of the initial vibration curve. The initial vibration interval is then expanded by 1 to 1.2 times to form a vibration interval. If all the values ​​of the vibration curve are within the vibration interval, the test vibration curve is considered similar to the initial vibration curve.

[0104] In practice, the vibration interval is 1.1 times the initial vibration interval.

[0105] Specifically, in step S4, based on the determination that the braking status of each test group is in a normal state, the first maintenance time of this batch of assembled bicycles is determined as the standard time.

[0106] Understandably, the standard time is usually around 6 months, meaning that the braking system needs to be calibrated and adjusted six months after the start of use to avoid loosening of the fixing screws due to insufficient tightness of the machine assembly during assembly and the break-in between parts during use, which would affect the braking performance.

[0107] Specifically, in step S4, based on the determination results of abnormal braking conditions in each test group, it is determined that the first maintenance time of this batch of assembled bicycles is less than the standard time, combined with the test cycle.

[0108] In practice, if only the third test group has an abnormal braking condition, the first maintenance period is usually 2 months; if the second test group has an abnormal braking condition, the first maintenance period is usually 3 months; if the first test group has an abnormal braking condition, all bicycles in that batch will have their screws tightened manually and the assembly robot's parameters for assembling the fixing screws will be adjusted to make the adjusted fixing screws tighter.

[0109] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0110] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for testing bicycle braking performance based on multivariate data, characterized in that, include: Step S1: Sample the assembled bicycles from the same batch and perform brake tests on all sampled bicycles to determine initial test data based on the brake test results. The initial test data includes initial braking time, initial audio characteristics, and initial vibration curve. Step S2: Divide the sampled bicycles into several test groups, conduct periodic cyclic tests on each test group, and conduct brake tests after the test. The number of test groups is at least 3. Step S3, determining the test characterization trend of the test group based on the brake test results of each sampled bicycle in a single test group, and determining the braking state of the test group in response to the test characterization trend, including: Determine the test data for the corresponding test group and compare it with the initial test data, so as to determine the braking state of the corresponding test group based on the comparison results; Alternatively, determine that the braking status of this test group is abnormal; The test characterization trend includes a consistent characterization trend and a inconsistent characterization trend; the test data includes test braking time, test audio characteristics, and test vibration curve; and the braking state includes a normal state and an abnormal state. Step S4: Determine whether to combine the test cycle to determine the first maintenance time for this batch of assembled bicycles based on the braking status of the test group.

2. The bicycle braking performance testing method based on multivariate data according to claim 1, characterized in that, In step S1, the process of determining the initial test data based on the brake test results includes: Step S11: Determine the initial braking time based on the average braking time of each assembled bicycle; Step S12: Draw a spectrum diagram based on the brake audio information of a single assembled bicycle to determine the energy distribution and frequency peak of the spectrum diagram of the assembled bicycle. Determine the initial audio characteristics based on the average value of the energy distribution and frequency peak of the spectrum diagrams of each assembled bicycle. The initial audio characteristics include the initial energy distribution and the initial frequency peak. Step S13: Determine the initial vibration curve based on the same spectral characteristics of the brake vibration curves of each assembled bicycle.

3. The bicycle braking performance testing method based on multivariate data according to claim 1, characterized in that, In step S1, the brake test process includes: Step A1: Install sensor groups on each of the sampled bicycles; Step A2: Fix the sampling bicycles to the roller test stand, drive each sampling bicycle to accelerate to the standard speed and maintain the standard time. Step A3: Control the brake lever to the preset lever travel and obtain braking duration, brake audio information and brake vibration curve.

4. The bicycle braking performance testing method based on multivariate data according to claim 1, characterized in that, In step S2, the cycle period is different for each of the test groups.

5. The bicycle braking performance testing method based on multivariate data according to claim 1, characterized in that, In step S3, the process of determining the test characterization trend of a test group based on the brake test results of each sampled bicycle in a single test group includes: Step S31: Determine the duration fluctuation parameter based on the ratio of the average deviation to the average value of the test braking time of each sampled bicycle. Step S32: Based on the calculation results of the duration fluctuation parameter, determine whether to determine the test characterization trend of the test group according to the test audio characteristics and test vibration curve, wherein: In response to the duration fluctuation parameter being greater than the preset fluctuation parameter, the test characterization trend of this experimental group is determined to be a characterization inconsistency trend; Step S33: In response to the duration fluctuation parameter being less than or equal to the preset fluctuation parameter, the feature similarity and curve similarity are determined by the machine learning model based on the test audio features and test vibration curves of each sampled bicycle. Step S34: Determine the test characterization trend of the experimental group based on the feature similarity and curve similarity, wherein: When both feature similarity and curve similarity are greater than or equal to the corresponding preset thresholds, the test characterization trend of this experimental group is determined to be a characterization consistency trend. Conversely, the test characterization trend of this experimental group is determined to be a characterization inconsistency trend.

6. The bicycle braking performance testing method based on multivariate data according to claim 1, characterized in that, In step S3, the process of determining the braking state of the test group in response to the test characterization trend includes: In response to the determination result representing a consistent trend, the test data of the corresponding test group are determined and compared with the initial test data, so as to determine the braking state of the corresponding test group based on the comparison result; Alternatively, in response to the judgment results that characterize different trends, the braking state of the test group is determined to be an abnormal state.

7. The bicycle braking performance testing method based on multivariate data according to claim 6, characterized in that, In step S3, the braking state of the corresponding test group is determined based on the comparison between the experimental test data and the initial test data, including: In response to the fact that the test data and the initial test data meet similarity conditions, the braking state of the corresponding test group is determined to be normal. In response to the fact that the test data and the initial test data do not meet the similarity condition, the braking state of the corresponding test group is determined to be an abnormal state. The similarity conditions include the ratio of the test braking time to the initial braking time being within a preset range, the test audio characteristics being the same as the initial audio characteristics, and the test vibration curve being similar to the initial vibration curve.

8. The bicycle braking performance testing method based on multivariate data according to claim 7, characterized in that, In step S3, the vibration range is determined based on the initial vibration curve, and the test vibration curve is determined to be similar to the initial vibration curve based on the result that the test vibration curve is within the vibration range.

9. The bicycle braking performance testing method based on multivariate data according to claim 1, characterized in that, In step S4, based on the determination that the braking status of each test group is normal, the first maintenance time of this batch of assembled bicycles is determined as the standard time.

10. The bicycle braking performance testing method based on multivariate data according to claim 9, characterized in that, In step S4, based on the determination results of abnormal braking conditions in each test group, it is determined that the first maintenance time of this batch of assembled bicycles is less than the standard time, combined with the test cycle.

Images