Dynamic balance deterioration prediction method and system of axial flux motor for electric tricycle

Abstract

The invention provides a dynamic balance deterioration prediction method and system of an axial flux motor for an electric tricycle, and relates to the technical field of dynamic balance deterioration prediction.The dynamic balance deterioration prediction method comprises the steps that a target electric tricycle is placed in a preset fixed foundation state to be subjected to dynamic balance detection, and a fixed foundation detection record is obtained; a first record of the first vehicle speed is extracted, a dynamic balance fusion discrimination model is introduced for analysis, and a first dynamic balance index is obtained; reading a predetermined pavement excitation state for dynamic balance detection to obtain a pavement excitation detection record; obtaining a pavement excitation coefficient; extracting a second record of the first vehicle speed to obtain a second dynamic balance index; comparing to obtain a first dynamic balance deterioration degree; obtaining a target dynamic balance degradation trend line according to the first mapping relation; and carrying out dynamic balance degradation prediction. The technical problem that the dynamic balance degradation prediction precision is low due to the fact that dynamic balance state analysis is limited to a fixed foundation detection mode and systematic analysis for dynamic working conditions is lacked in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of dynamic balance degradation prediction, and in particular to a method and system for predicting dynamic balance degradation of an axial flux motor for an electric tricycle. Background Art

[0002] The axial flux motor used in electric three-wheelers is the core component of the drive system, and its dynamic balance performance directly affects the operating stability of the entire vehicle. The axial flux motor is a compact and high-efficiency motor, but due to its design and operating characteristics, it also has some special dynamic balance problems. For example, during long-term use, due to mechanical wear, external excitation, vehicle speed changes and other factors, the dynamic balance performance of the motor may gradually deteriorate, resulting in increased vibration and decreased efficiency, and may even cause motor failure in severe cases.

[0003] Some existing technologies use sensors and vibration monitoring technology to detect the vibration condition of electric tricycles in order to identify potential dynamic balance problems in a timely manner. However, these methods usually rely on a single detection indicator, such as vibration acceleration, vibration frequency, etc., and lack a comprehensive analysis of the comprehensive dynamic balance state of the vehicle under different working conditions. It is difficult to accurately reflect the dynamically changing road conditions and motor performance, especially in the face of complex road conditions or long-term use. The dynamic changes of dynamic balance degradation are difficult to accurately predict. Summary of the invention

[0004] The present application provides a method and system for predicting dynamic balance degradation of an axial flux motor for an electric tricycle, aiming to solve the technical problem that the analysis of the dynamic balance state in the prior art is limited to a fixed basic detection mode and lacks a systematic analysis method for dynamic working conditions, resulting in low accuracy in dynamic balance degradation prediction.

[0005] The first aspect disclosed in the present application provides a method for predicting dynamic balance degradation of an axial flux motor for an electric three-wheeler, the method comprising: placing a target electric three-wheeler in a predetermined fixed base state for dynamic balance detection to obtain a fixed base detection record; extracting a first record at a first vehicle speed in the fixed base detection record, and introducing a dynamic balance fusion discrimination model to analyze the first record to obtain a first dynamic balance index; reading a predetermined road excitation state, and placing the target electric three-wheeler in the predetermined road excitation state for dynamic balance detection to obtain a road excitation detection record; extracting features from the road excitation detection record to obtain road excitation features, and The road surface excitation feature is subjected to variation weighted calculation to obtain a road surface excitation coefficient; a second record at the first vehicle speed in the road surface excitation detection record is extracted, and the second record is analyzed according to the dynamic balance fusion discrimination model to obtain a second dynamic balance index; the first dynamic balance index is compared with the second dynamic balance index to obtain a first dynamic balance degradation degree of the target electric three-wheeler at the first vehicle speed; a target dynamic balance degradation trend line is obtained according to a first mapping relationship between the first vehicle speed and the first dynamic balance degradation degree; and a dynamic balance degradation trend line is used to predict the target axial flux motor of the target electric three-wheeler.

[0006] According to a second aspect disclosed in the present application, a dynamic balance degradation prediction system for an axial flux motor for an electric three-wheeler is provided. The system is used for the above-mentioned dynamic balance degradation prediction method for an axial flux motor for an electric three-wheeler. The system comprises: a first dynamic balance detection module, used to place the target electric three-wheeler in a predetermined fixed basic state for dynamic balance detection, and obtain a fixed basic detection record; a first record analysis module, used to extract a first record at a first vehicle speed in the fixed basic detection record, and introduce a dynamic balance fusion discrimination model to analyze the first record, and obtain a first dynamic balance index; a second dynamic balance detection module, used to read a predetermined road excitation state, and place the target electric three-wheeler in the predetermined road excitation state for dynamic balance detection, and obtain a road excitation detection record; a variation weighted calculation module, used to perform characteristic weighting on the road excitation detection record; A road excitation feature is extracted from the road excitation feature, and a variation weighted calculation is performed on the road excitation feature to obtain a road excitation coefficient; a second record analysis module is used to extract a second record at the first vehicle speed in the road excitation detection record, and analyze the second record according to the dynamic balance fusion discrimination model to obtain a second dynamic balance index; a dynamic balance degradation degree acquisition module is used to compare the first dynamic balance index with the second dynamic balance index to obtain a first dynamic balance degradation degree of the target electric three-wheeler at the first vehicle speed; a degradation trend line acquisition module is used to obtain a target dynamic balance degradation trend line according to a first mapping relationship between the first vehicle speed and the first dynamic balance degradation degree; a dynamic balance degradation prediction module is used to predict the dynamic balance degradation of a target axial flux motor of the target electric three-wheeler according to the target dynamic balance degradation trend line.

[0007] One or more technical solutions provided in this application have at least the following beneficial effects: By placing the electric three-wheeler in a predetermined fixed basic state for dynamic balancing detection, a fixed basic detection record is obtained, which is the benchmark data for subsequent analysis. This process ensures that all detection and evaluation are carried out under known standard conditions, providing an accurate reference for subsequent comparison and analysis; by introducing a dynamic balancing fusion discrimination model, the data in the fixed basic detection record is analyzed to obtain the first dynamic balancing index, which quantifies the dynamic balancing performance of the electric three-wheeler in the basic state, providing a quantitative indicator for subsequent performance analysis; dynamic balancing detection is carried out under a predetermined road surface excitation state to obtain a road surface excitation detection record. This step simulates the different excitation conditions that the electric three-wheeler will encounter during actual driving, and can evaluate the dynamic balancing performance of the electric three-wheeler in actual road conditions; the road surface excitation coefficient is obtained by extracting features from the road surface excitation detection record and performing variation weighted calculation. This process comprehensively considers the influence of road surface conditions on dynamic balancing performance, and can more accurately evaluate the influencing factors encountered by the vehicle during actual driving; at the first vehicle speed The second record is extracted and analyzed through the same dynamic balance fusion discriminant model to obtain the second dynamic balance index. This process reflects the dynamic balance change of the electric three-wheeler at the same vehicle speed under the influence of road excitation, and provides a basis for the subsequent calculation of the degree of degradation. The first dynamic balance index is compared with the second dynamic balance index to calculate the first dynamic balance degradation degree. This degradation degree can quantify the degree of dynamic balance loss of the electric three-wheeler in actual driving due to road excitation, and provides an evaluation basis for the health status of the vehicle. By establishing a mapping relationship between the first vehicle speed and the dynamic balance degradation degree, the target dynamic balance degradation trend line is obtained. This trend line can reflect the relationship between vehicle speed and dynamic balance degradation, and provides data support for further dynamic balance prediction, and can more accurately predict the future dynamic balance performance of the vehicle. Based on the target dynamic balance degradation trend line, the dynamic balance degradation of the axial flux motor of the electric three-wheeler is predicted. This prediction realizes the early warning of the dynamic balance degradation that may occur in the future use of the electric three-wheeler, thereby providing a scientific basis for the maintenance and maintenance of the vehicle.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic flow chart of a method for predicting dynamic balance degradation of an axial flux motor for an electric tricycle provided in an embodiment of the present application.

[0010] Figure 2 Schematic diagram of the structure of the dynamic balance degradation prediction system for the axial flux motor for an electric three-wheeler provided in an embodiment of the present application.

[0011] Explanation of the reference numerals: first dynamic balance detection module 10 , first record analysis module 20 , second dynamic balance detection module 30 , variation weighted calculation module 40 , second record analysis module 50 , dynamic balance degradation degree acquisition module 60 , degradation trend line acquisition module 70 , dynamic balance degradation prediction module 80 . DETAILED DESCRIPTION

[0012] The embodiments of the present application provide a method and system for predicting dynamic balance degradation of an axial flux motor for an electric tricycle, thereby solving the technical problem that the analysis of the dynamic balance state in the prior art is limited to a fixed basic detection mode and lacks a systematic analysis method for dynamic working conditions, resulting in low accuracy in dynamic balance degradation prediction.

[0013] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced below in conjunction with the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0014] Embodiment 1, as Figure 1 As shown, the embodiment of the present application provides a method for predicting dynamic balance degradation of an axial flux motor for an electric tricycle, the method comprising: The target electric tricycle is placed in a predetermined fixed foundation state for dynamic balance detection to obtain a fixed foundation detection record.

[0015] The target electric tricycle is placed in a predetermined fixed base state, which means that the wheels do not touch the ground, or the wheels are fixed by special supporting equipment (such as brackets) to ensure that the body is in a stationary state. At this time, the electric tricycle will not be affected by external road excitation and can reflect the dynamic balance performance of the motor and its system alone. In this state, a comprehensive test is performed through dynamic balancing detection equipment, such as accelerometers, vibration sensors, current sensors, etc., and all relevant data are recorded to form a fixed base detection record, which provides a basis for subsequent analysis.

[0016] A first record at a first vehicle speed in the fixed basic detection record is extracted, and a dynamic balance fusion discrimination model is introduced to analyze the first record to obtain a first dynamic balance index.

[0017] In the fixed basic detection process, including detection records at different vehicle speeds, the first record at the first vehicle speed is extracted at this time. The first vehicle speed is any one of the different vehicle speeds. This random extraction facilitates subsequent traversal analysis of all vehicle speeds.

[0018] The dynamic balance fusion discrimination model is introduced. The dynamic balance fusion discrimination model is a comprehensive analysis model that can combine multiple different types of data, such as vibration data, electrical data, and torque data, to make a comprehensive judgment and obtain an evaluation of the dynamic balance state. The extracted first recorded data is input into the dynamic balance fusion discrimination model. Based on these data, the model calculates the first dynamic balance index. This index reflects the dynamic balance performance of the electric tricycle at the vehicle speed. The higher the value, the better the dynamic balance state of the vehicle. Conversely, it may indicate the potential risk of imbalance or damage.

[0019] The predetermined road surface excitation state is read, and the target electric three-wheeler is placed in the predetermined road surface excitation state for dynamic balance detection to obtain a road surface excitation detection record.

[0020] Read the preset road excitation state. The road excitation state refers to the road characteristics and environmental factors that the electric three-wheeler will encounter during operation. For example, the unevenness, potholes, fluctuations, and ruggedness of the road surface can all be regarded as road excitations, which affect the dynamic balance state of the three-wheeler. The preset road excitation state refers to the pre-set road conditions, such as the degree of unevenness of a specific road section obtained by measurement or calculation.

[0021] In order to simulate the operation of electric tricycles on actual roads, it is necessary to place the electric tricycle on a road with a predetermined road excitation state. The road excitation can be controlled by a road vibration table, a simulated vibration test device, etc. When the electric tricycle runs under this specific road state, it will be exposed to different vibrations, torques, accelerations, etc. In the predetermined road excitation state, the dynamic balance test of the target electric tricycle is performed, including measuring the vibration of the wheels, body, frame and motor, while monitoring the electrical signals such as the current and voltage of the motor and the mechanical response, and obtaining the test data of the electric tricycle under road excitation through sensors to generate road excitation test records.

[0022] The road surface excitation detection record is subjected to feature extraction to obtain a road surface excitation feature, and the road surface excitation feature is subjected to variation weighted calculation to obtain a road surface excitation coefficient.

[0023] The road surface excitation features, such as road surface roughness, crankshaft speed, vibration signal, etc., are extracted from the data obtained from the road surface excitation detection records. In the processing process, in order to better reflect the different effects of different road surface excitations, a variation weighting strategy is adopted, which means that different road surface excitation features will be given different weights according to their importance and degree of change. For example, for certain specific road surface excitation features (such as vibrations with higher frequencies), a higher weight can be given because it has a greater impact on the dynamic balance state. The extracted road surface excitation features and the variation weighted results are combined to obtain the road surface excitation coefficient through weighted calculation. This coefficient is used to characterize the degree of impact on the dynamic balance performance of the electric three-wheeler under specific road surface excitation conditions. A higher road surface excitation coefficient means that the wheels, body and other parts will be greatly affected, and there is a risk of imbalance.

[0024] A second record at the first vehicle speed is extracted from the road surface excitation detection record, and the second record is analyzed according to the dynamic balance fusion discrimination model to obtain a second dynamic balance index.

[0025] A second record corresponding to the first vehicle speed is extracted from the road excitation detection record. This record contains vibration, torque, electrical signal and other data of the electric three-wheel under a predetermined road excitation state when the vehicle speed is the first vehicle speed. The extracted second record is input into a dynamic balance fusion discrimination model. The model calculates a second dynamic balance index based on the input data. This index reflects the dynamic balance performance of the electric three-wheel at the first vehicle speed under the action of road excitation.

[0026] The first dynamic balance index and the second dynamic balance index are compared to obtain a first dynamic balance degradation degree of the target electric three-wheeler at the first vehicle speed.

[0027] The first dynamic balance index is compared with the second dynamic balance index to evaluate the degree of dynamic balance degradation of the electric three-wheeler at the first vehicle speed. Specifically, the difference between the first dynamic balance index and the second dynamic balance index is calculated and then the difference is divided by the first dynamic balance index to calculate the first dynamic balance degradation degree. This value reflects the difference between the dynamic balance performance of the electric three-wheeler under the influence of road excitation and that in the fixed basic state. The larger the ratio, the more serious the dynamic balance degradation, and vice versa, it indicates that the dynamic balance state of the electric three-wheeler is maintained well.

[0028] A target dynamic balance degradation trend line is obtained according to a first mapping relationship between the first vehicle speed and the first dynamic balance degradation degree.

[0029] A first mapping relationship between a first vehicle speed and a first dynamic balance degradation degree is established. The mapping relationship indicates the relationship between the dynamic balance degradation degree and the vehicle speed at a specific first vehicle speed. In order to obtain a more comprehensive trend line, it is necessary to expand this mapping relationship to multiple vehicle speeds. For each vehicle speed, the corresponding dynamic balance degradation degree is obtained through corresponding dynamic balance detection and analysis. The relationship between these vehicle speeds and dynamic balance degradation degrees forms multiple mapping relationships.

[0030] These data points are connected to form a complete trend line. In this process, the vehicle speed is used as the horizontal coordinate and the dynamic balance degradation degree is used as the vertical coordinate. Each vehicle speed and its corresponding dynamic balance degradation degree will form a data point in the coordinate system. All data points are connected to finally obtain the dynamic balance degradation trend line. This trend line shows how the dynamic balance degradation degree of the electric tricycle changes within the entire vehicle speed range. This trend line will provide a key basis for further prediction of dynamic balance degradation.

[0031] The dynamic balance degradation of the target axial flux motor of the target electric three-wheeler is predicted according to the target dynamic balance degradation trend line.

[0032] Electric tricycles use axial flux motors, which means that the dynamic balance degradation trend line can not only be used to evaluate the dynamic balance of the entire vehicle, but also be used specifically for motor performance evaluation. By analyzing the trend line, it is possible to predict the dynamic balance degradation of the electric tricycle axial flux motor in future use. For example, as the use time increases, the motor may accumulate vibration and imbalance, causing its dynamic balance performance to gradually decline. By using trend lines to predict dynamic balance degradation, it is possible to predict the degree of dynamic balance degradation of the motor in different future usage scenarios, thereby providing early warning or taking necessary maintenance measures.

[0033] Furthermore, extracting a first record at a first vehicle speed from the fixed basic detection record, and introducing a dynamic balance fusion discrimination model to analyze the first record to obtain a first dynamic balance index, including: Extract a first electrical data record from the first record, wherein the first electrical data record includes a first current and a first voltage; extract a first operating data record from the first record, wherein the first operating data record includes a first overturning moment and a first internal bending moment; extract a first vibration timing signal from the first record, and extract first vibration characteristic data of the first vibration timing signal, wherein the first vibration characteristic data includes a first vibration acceleration, a first vibration velocity, and a first vibration displacement; input the first current, the first voltage, the first overturning moment, the first internal bending moment, the first vibration acceleration, the first vibration velocity, and the first vibration displacement into the dynamic balance fusion discrimination model to obtain the first dynamic balance index.

[0034] Extract the first electrical data record in the first record. The first electrical data record includes the working parameters of the motor, including the first current and the first voltage. The first current is the current value of the motor during operation, which reflects the load condition, power consumption and overall electrical state of the electric tricycle. The change of the current indicates the working load of the motor, which in turn affects the dynamic balance. The first voltage is the input voltage of the motor. Voltage fluctuation or instability may mean that there is a problem in the electrical system, which indirectly affects the dynamic balance of the vehicle. These electrical data are collected through sensors, including current sensors, voltage sensors, etc.

[0035] Extract the first operation data record in the first record. The operation data is usually related to the physical movement and mechanical state of the vehicle, including the first overturning moment and the first internal bending moment. The first overturning moment refers to the moment acting on the electric tricycle, which may cause the vehicle to tilt or overturn. The moment may be caused by an unbalanced load, a design problem of the vehicle itself, or an uneven road surface. By monitoring the overturning moment, the stability problem of the vehicle can be identified, which in turn affects the dynamic balance assessment; the first internal bending moment refers to the moment when the vehicle frame or body bends due to the internal force it receives during operation. The existence of the internal bending moment usually means that the vehicle has stress concentration during operation, which may be caused by imbalance or loose components. These torque data can be obtained through torque sensors, accelerometers, gyroscopes and other devices, which are usually installed on key components of the vehicle, such as the frame, wheels or motors, so as to monitor and record the changes in torque in real time.

[0036] Extract the first vibration time series signal in the first record. The first vibration time series signal records the vibration data generated by the vehicle during movement, which is usually collected by an accelerometer or a vibration sensor. It is time series data of vibration and can reflect the vibration state and intensity of the vehicle at different time points. Extract the key information of the vibration from the vibration time series signal as the first vibration feature data. These vibration features are usually extracted from the vibration time series signal through Fourier transform or wavelet transform, which can convert the time domain signal into frequency domain features so as to analyze the frequency components of the vibration and their corresponding vibration intensities.

[0037] The first vibration characteristic data includes a first vibration acceleration, a first vibration velocity and a first vibration displacement, wherein the first vibration acceleration reflects the intensity and frequency of the vibration, and the acceleration generally indicates the speed change of an object in unit time, and the larger the acceleration, the larger the vibration amplitude, and the vibration acceleration is very important to the dynamic balance of the vehicle, especially in high-frequency vibrations; the first vibration velocity indicates the speed change of the vibration, and reflects the dynamic characteristics of the vibration. The vibration velocity can be used to evaluate the propagation characteristics of the vibration and whether there is a resonance problem caused by imbalance; the first vibration displacement indicates the displacement of the object caused by the vibration, and the displacement can be used to evaluate the impact of the vibration on the structural part of the vehicle, such as the offset of the frame or wheels.

[0038] The extracted feature data of different types are input into the dynamic balance fusion discrimination model. The dynamic balance fusion discrimination model is a multi-input model that can perform comprehensive analysis based on the input values ​​of different features. By analyzing the above data, the model calculates the first dynamic balance index, which can quantify the dynamic balance status of the vehicle at the first vehicle speed.

[0039] Further, the first current, the first voltage, the first overturning moment, the first internal bending moment, the first vibration acceleration, the first vibration velocity, and the first vibration displacement are input into the dynamic balance fusion discrimination model to obtain the first dynamic balance index, including: Obtain a preliminary discriminant layer in the dynamic balance fusion discriminant model; analyze the first current and the first voltage through the first discriminant model in the preliminary discriminant layer to obtain a first discriminant index; analyze the first overturning moment and the first internal bending moment through the second discriminant model in the preliminary discriminant layer to obtain a second discriminant index; analyze the first vibration acceleration, the first vibration velocity and the first vibration displacement through the third discriminant model in the preliminary discriminant layer to obtain a third discriminant index; obtain a meta discriminant layer in the dynamic balance fusion discriminant model; analyze the first discriminant index, the second discriminant index and the third discriminant index through the meta discriminant layer to obtain the first dynamic balance index.

[0040] The initial discriminant layer is a key component in the dynamic balance fusion discriminant model. Its function is to perform preliminary analysis and classification of multiple input feature data, evaluate the impact of different types of sensor data on the dynamic balance state, and calculate the corresponding preliminary discriminant index. This layer includes multiple discriminant models, each of which is responsible for analyzing a specific type of input data.

[0041] The first discriminant model in the initial discriminant layer specifically processes the two input data of the first current and the first voltage. Current and voltage are key electrical parameters of the electric three-wheeler, which are usually closely related to the motor load, power output and operating status of the vehicle.

[0042] The first discriminant model is used to perform correlation analysis on the first current and the first voltage. The changes in current and voltage reflect the working condition of the motor. If the current and voltage fluctuate greatly or are unstable at a certain vehicle speed, it indicates that there is a problem with the motor system of the electric three-wheeler, which in turn affects the dynamic balance of the vehicle. For example, if the motor is unbalanced, it may cause abnormal fluctuations in current and voltage. The first discriminant model performs analysis based on the characteristic values ​​of current and voltage, such as amplitude, frequency, fluctuation, etc. By comparing historical data, normal working modes and abnormal modes, the model can determine the impact of current current and voltage data on dynamic balance and calculate the first discriminant index. This index quantifies the degree of influence of the electrical system of the electric three-wheeler on the dynamic balance at the first vehicle speed. For example, if the current and voltage fluctuate greatly, the first discriminant index is low, which means that the motor system may be unbalanced; if the current and voltage are stable, the first discriminant index is high, indicating that the electrical system is normal and the dynamic balance may be good.

[0043] The second discriminant model in the initial discriminant layer is specifically used to analyze parameters related to vehicle mechanics and stability. The first overturning moment and the first internal bending moment are input into the model for analysis. These data describe the mechanical behavior of the vehicle due to external excitation or imbalance during operation.

[0044] The second discriminant model analyzes the characteristics of the first overturning moment and the first internal bending moment to evaluate their impact on the dynamic balance of the vehicle. For example, if the overturning moment is too large, it means that the vehicle may face a large imbalance, affecting the overall stability. The model combines historical data to analyze the changing trend and fluctuation range of the moment to calculate the second discriminant index. The second discriminant index is a quantitative value that represents the impact of the vehicle's mechanical behavior on the dynamic balance. If the index is high, it means that the moment data indicates that the vehicle's mechanical state may be abnormal, affecting the vehicle's dynamic balance; if the index is low, it means that the vehicle performs well in terms of stability.

[0045] The third discriminant model in the primary discriminant layer is responsible for analyzing data related to vehicle vibration, including vibration acceleration, vibration velocity and vibration displacement. Vibration analysis is crucial to dynamic balancing evaluation because the vibration state of the vehicle directly reflects its dynamic balancing performance.

[0046] The third discriminant model combines the data of the first vibration acceleration, the first vibration velocity and the first vibration displacement for analysis to evaluate their impact on the overall dynamic balance of the vehicle. Generally, larger or irregular vibration data will indicate that there is a problem with the dynamic balance and further repair is required. The output third discriminant index is a quantitative result that reflects the impact of vibration factors on the dynamic balance of the vehicle. A higher index usually means greater vibration, the dynamic balance may be impaired, and the vehicle may have an asymmetric or unstable vibration mode.

[0047] The meta-discrimination layer is an advanced layer in the dynamic balance fusion discrimination model. It is responsible for comprehensively analyzing the multiple discrimination indexes obtained in the initial discrimination layer. By integrating multiple discrimination indexes, it further optimizes the dynamic balance evaluation results. Through the integration of the meta-discrimination layer, the model can more accurately reflect the dynamic balance state of the electric three-wheeler, avoid the deviation caused by a single data source, and improve the accuracy of the evaluation.

[0048] The first discriminant index, the second discriminant index and the third discriminant index of the initial discriminant layer are taken as input and passed to the meta-discriminant layer. The meta-discriminant layer conducts a comprehensive analysis of these discriminant indexes, such as using weighted average and other methods, and optimizes them according to the importance of each discriminant index. Finally, the fused first dynamic balance index is calculated. This index is a comprehensive evaluation value that reflects the dynamic balance status of the electric three-wheeler in multiple dimensions. A higher value indicates that the dynamic balance performance of the electric three-wheeler is better, and a lower value indicates that there may be an imbalance problem, which affects the stability and safety of the vehicle.

[0049] Furthermore, the road surface excitation feature at least includes road surface roughness and crankshaft speed, wherein the road surface roughness refers to the excitation detection characteristic value of the predetermined road surface excitation state calculated by a predetermined road surface roughness function, and the expression of the predetermined road surface roughness function is as follows: ; in, The unit length The road surface roughness within The unit length The height of the road surface inside, is the average road height. is the length of the measuring section, refers to the amplitude of the predetermined road roughness function, and , is the attenuation factor, and .

[0050] Specifically, the expression of the predetermined road roughness function is as follows: ; This function Indicated in unit length The roughness of the road surface reflects the degree of change in the height of the road surface at that location. It means that the road surface is relatively uneven, and the vehicle will be strongly stimulated when driving, affecting the dynamic balance of the vehicle. The unit length The road surface height inside is the average height of the entire road surface. and The square of the difference reflects the impact of road height changes on vehicles. The greater the road roughness, the more significant the fluctuation of the road surface. It refers to the length of the measurement section, which is set in advance; amplitude Controls the overall strength of the roughness function and determines the magnitude of the calculated road roughness. The closer the value is to 1, the greater the road roughness and the more significant the impact. The closer the value is to 0, the smoother the road surface is and the less excitation the vehicle receives. Used to adjust the impact of road roughness on the vehicle. As the value increases, the impact of road roughness gradually weakens, and vice versa. The attenuation factor simulates the actual situation, where the impact of road roughness on different vehicles may be different under certain conditions. In general, this formula can be used to reflect the impact of road roughness on dynamic balance during vehicle driving by describing the calculation method of road roughness, combining different physical parameters and adjustment coefficients.

[0051] Furthermore, after extracting the features of the road surface excitation detection record to obtain the road surface excitation features, and performing variation weighted calculation on the road surface excitation features to obtain the road surface excitation coefficient, the method further includes: Extracting environmental excitation features from the road surface excitation detection record; performing weighted calculation on the environmental temperature and environmental humidity in the environmental excitation features to obtain an environmental feedback coefficient; and adjusting the road surface excitation coefficient using the environmental feedback coefficient as a weight.

[0052] Extract environmental excitation features from road excitation detection records. Environmental excitation features refer to external environmental factors that are independent of the road excitation state but may affect the vehicle's dynamic balance. They usually include ambient temperature, humidity, air pressure, etc. They may indirectly affect the dynamic balance by affecting the road condition or the performance of vehicle components. These environmental features are collected through external sensors, such as temperature and humidity sensors, and stored together with the road excitation data for subsequent analysis.

[0053] Variation weighting means assigning different weights according to the fluctuation or change degree of ambient temperature and humidity. Different environmental factors may have different effects on the dynamic balance of the vehicle. Therefore, they need to be weighted according to their changes during analysis. Specifically, changes in ambient temperature and humidity can affect the mechanical components or electrical systems of the vehicle and change its operating characteristics. Therefore, different weights need to be assigned to these changes. The weighted calculation adjusts the coefficient according to the assigned weight. The purpose is to derive the environmental feedback coefficient based on the actual changes in the environment, so as to reflect the comprehensive impact of the external environment on the dynamic balance.

[0054] The road excitation coefficient reflects the impact of the vehicle's dynamic balance under specific road conditions. It is calculated based on the excitation characteristic data measured under the road excitation state. The road excitation coefficient is adjusted according to the environmental feedback coefficient. For example, at higher temperatures, the battery and motor of the electric three-wheeler may be more susceptible to temperature changes, resulting in changes in vibration characteristics. Therefore, it is necessary to make corresponding weight adjustments to the road excitation coefficient through the environmental feedback coefficient. The adjusted road excitation coefficient can more accurately reflect the dynamic balance state of the electric three-wheeler under specific environmental conditions.

[0055] Further, comparing the first dynamic balance index with the second dynamic balance index to obtain a first dynamic balance degradation degree of the target electric three-wheeler at the first vehicle speed includes: The first dynamic balance index and the second dynamic balance index are compared to obtain a first index difference; a first ratio of the first index difference to the first dynamic balance index is taken as the first dynamic balance degradation degree.

[0056] The first dynamic balance index is compared with the second dynamic balance index. These two dynamic balance indexes reflect the dynamic balance state of the electric tricycle under different conditions. The first dynamic balance index represents the ideal balance state of the vehicle without external excitation, while the second dynamic balance index represents the actual balance state of the vehicle when it is excited by the road surface. The first index difference is the difference between the two indexes, which indicates the change of the dynamic balance state of the vehicle under different excitation conditions. It reflects the influence of external excitation on the dynamic balance of the vehicle. The larger the difference, the more significant the influence of external excitation on the dynamic balance.

[0057] The ratio between the first index difference and the first dynamic balance index is calculated. This ratio is the first dynamic balance degradation degree, which indicates the degree of dynamic balance degradation of the electric three-wheeler under specific vehicle speed and excitation conditions. When the ratio is large, it means that the dynamic balance performance of the vehicle is seriously degraded when it is stimulated by the road surface; when the ratio is small, it indicates that the dynamic balance of the vehicle is well maintained.

[0058] Furthermore, predicting the dynamic balance degradation of the target axial flux motor of the target electric three-wheeler according to the target dynamic balance degradation trend line includes: The target dynamic balance degradation trend line is subjected to polynomial regression processing to obtain a target fitting polynomial; a target speed is obtained, and a target predicted dynamic balance degradation degree is obtained in combination with the target polynomial.

[0059] The target dynamic balance degradation trend line is processed by polynomial regression. The purpose of this process is to obtain a mathematical model that can predict future dynamic balance degradation by fitting the data points of the target dynamic balance degradation trend line, that is, the target fitting polynomial. Specifically, polynomial regression is a statistical analysis method used to establish the relationship between the input variable (vehicle speed) and the output variable (dynamic balance degradation degree). Here, the regression model will fit a polynomial model based on the relationship between the vehicle speed and the dynamic balance degradation degree. For example, assuming that the relationship between the dynamic balance degradation degree and the vehicle speed is a quadratic curve, a quadratic polynomial can be used for regression to obtain the target fitting polynomial. Through polynomial regression processing, an accurate mathematical model is obtained. This model can better reflect the relationship between the vehicle speed and the dynamic balance degradation degree, providing a basis for subsequent prediction and analysis.

[0060] Obtain the target speed. The target speed refers to the vehicle speed at which the dynamic balance degradation of the electric three-wheeler is expected to be predicted in actual use or future applications. The target speed is substituted into the target fitting polynomial obtained by polynomial regression to calculate the corresponding target predicted dynamic balance degradation, which is the predicted value of the dynamic balance degradation of the electric three-wheeler at the target speed, and is used to predict the performance of the vehicle at a specific speed.

[0061] In summary, the method for predicting dynamic balance degradation of an axial flux motor for an electric tricycle provided in the embodiments of the present application has the following technical effects: By placing the electric three-wheeler in a predetermined fixed basic state for dynamic balancing detection, a fixed basic detection record is obtained, which is the benchmark data for subsequent analysis. This process ensures that all detection and evaluation are carried out under known standard conditions, providing an accurate reference for subsequent comparison and analysis; by introducing a dynamic balancing fusion discrimination model, the data in the fixed basic detection record is analyzed to obtain the first dynamic balancing index, which quantifies the dynamic balancing performance of the electric three-wheeler in the basic state, providing a quantitative indicator for subsequent performance analysis; dynamic balancing detection is carried out under a predetermined road surface excitation state to obtain a road surface excitation detection record. This step simulates the different excitation conditions that the electric three-wheeler will encounter during actual driving, and can evaluate the dynamic balancing performance of the electric three-wheeler in actual road conditions; the road surface excitation coefficient is obtained by extracting features from the road surface excitation detection record and performing variation weighted calculation. This process comprehensively considers the influence of road surface conditions on dynamic balancing performance, and can more accurately evaluate the influencing factors encountered by the vehicle during actual driving; at the first vehicle speed The second record is extracted and analyzed through the same dynamic balance fusion discriminant model to obtain the second dynamic balance index. This process reflects the dynamic balance change of the electric three-wheeler at the same vehicle speed under the influence of road excitation, and provides a basis for the subsequent calculation of the degree of degradation. The first dynamic balance index is compared with the second dynamic balance index to calculate the first dynamic balance degradation degree. This degradation degree can quantify the degree of dynamic balance loss of the electric three-wheeler in actual driving due to road excitation, and provides an evaluation basis for the health status of the vehicle. By establishing a mapping relationship between the first vehicle speed and the dynamic balance degradation degree, the target dynamic balance degradation trend line is obtained. This trend line can reflect the relationship between vehicle speed and dynamic balance degradation, and provides data support for further dynamic balance prediction, and can more accurately predict the future dynamic balance performance of the vehicle. Based on the target dynamic balance degradation trend line, the dynamic balance degradation of the axial flux motor of the electric three-wheeler is predicted. This prediction realizes the early warning of the dynamic balance degradation that may occur in the future use of the electric three-wheeler, thereby providing a scientific basis for the maintenance and maintenance of the vehicle.

[0062] Embodiment 2, based on the same inventive concept as the method for predicting dynamic balance degradation of an axial flux motor for an electric tricycle in the aforementioned embodiment, Figure 2 As shown, the embodiment of the present application provides a dynamic balance degradation prediction system for an axial flux motor for an electric three-wheeler, the system comprising: The first dynamic balance detection module 10 is used to place the target electric tricycle in a predetermined fixed basic state for dynamic balance detection to obtain a fixed basic detection record; the first record analysis module 20 is used to extract the first record at the first vehicle speed in the fixed basic detection record, and introduce a dynamic balance fusion discrimination model to analyze the first record to obtain a first dynamic balance index; the second dynamic balance detection module 30 is used to read a predetermined road surface excitation state, and place the target electric tricycle in the predetermined road surface excitation state for dynamic balance detection to obtain a road surface excitation detection record; the variation weighted calculation module 40 is used to extract features from the road surface excitation detection record to obtain road surface excitation features, and perform variation weighted calculation on the road surface excitation features to obtain road surface excitation features. excitation coefficient; a second record analysis module 50, used to extract the second record at the first vehicle speed in the road excitation detection record, and analyze the second record according to the dynamic balance fusion discrimination model to obtain a second dynamic balance index; a dynamic balance degradation degree acquisition module 60, used to compare the first dynamic balance index with the second dynamic balance index to obtain the first dynamic balance degradation degree of the target electric three-wheel at the first vehicle speed; a degradation trend line acquisition module 70, used to obtain a target dynamic balance degradation trend line according to a first mapping relationship between the first vehicle speed and the first dynamic balance degradation degree; a dynamic balance degradation prediction module 80, used to predict the dynamic balance degradation of the target axial flux motor of the target electric three-wheel according to the target dynamic balance degradation trend line.

[0063] Furthermore, the first record analysis module 20 includes: a first electrical data record extraction unit, used to extract a first electrical data record from the first record, wherein the first electrical data record includes a first current and a first voltage; a first operation data record extraction unit, used to extract a first operation data record from the first record, wherein the first operation data record includes a first overturning moment and a first internal bending moment; a first vibration timing signal extraction unit, used to extract a first vibration timing signal from the first record, and extract first vibration characteristic data of the first vibration timing signal, wherein the first vibration characteristic data includes a first vibration acceleration, a first vibration velocity and a first vibration displacement; a first dynamic balance index acquisition unit, used to input the first current, the first voltage, the first overturning moment, the first internal bending moment, the first vibration acceleration, the first vibration velocity and the first vibration displacement into the dynamic balance fusion discrimination model to obtain the first dynamic balance index.

[0064] Furthermore, the first dynamic balance index obtaining unit includes: An initial discrimination layer acquisition channel is used to obtain the initial discrimination layer in the dynamic balance fusion discrimination model; a first analysis channel is used to analyze the first current and the first voltage through the first discrimination model in the initial discrimination layer to obtain a first discrimination index; a second analysis channel is used to analyze the first overturning moment and the first internal bending moment through the second discrimination model in the initial discrimination layer to obtain a second discrimination index; a third analysis channel is used to analyze the first vibration acceleration, the first vibration velocity and the first vibration displacement through the third discrimination model in the initial discrimination layer to obtain a third discrimination index; a meta-discrimination layer acquisition channel is used to obtain the meta-discrimination layer in the dynamic balance fusion discrimination model; a fourth analysis channel is used to analyze the first discrimination index, the second discrimination index and the third discrimination index through the meta-discrimination layer to obtain the first dynamic balance index.

[0065] Furthermore, the road surface excitation feature at least includes road surface roughness and crankshaft speed, wherein the road surface roughness refers to the excitation detection characteristic value of the predetermined road surface excitation state calculated by a predetermined road surface roughness function, and the expression of the predetermined road surface roughness function is as follows: ; in, The unit length The road surface roughness within The unit length The road surface height inside is the average road height. is the length of the measuring section, refers to the amplitude of the predetermined road roughness function, and , is the attenuation factor, and .

[0066] Furthermore, the variation weighted calculation module 40 includes: An environmental excitation feature extraction unit is used to extract the environmental excitation features in the road surface excitation detection record; a variation weighted calculation unit is used to perform variation weighted calculation on the ambient temperature and ambient humidity in the environmental excitation features to obtain an environmental feedback coefficient; an adjustment unit is used to adjust the road surface excitation coefficient using the environmental feedback coefficient as a weight.

[0067] Furthermore, the dynamic balance degradation degree acquisition module 60 includes: The first index difference acquisition unit is used to compare the first dynamic balance index with the second dynamic balance index to obtain a first index difference; the first dynamic balance degradation degree acquisition unit is used to obtain a first ratio of the first index difference to the first dynamic balance index, which is recorded as the first dynamic balance degradation degree.

[0068] Furthermore, the dynamic balance degradation prediction module 80 includes: The polynomial regression processing unit is used to perform polynomial regression processing on the target dynamic balance degradation trend line to obtain a target fitting polynomial; the target predicted dynamic balance degradation degree acquisition unit is used to acquire a target speed and obtain a target predicted dynamic balance degradation degree in combination with the target polynomial.

[0069] Through the detailed description of the dynamic balance degradation prediction method of the axial flux motor for an electric three-wheeled vehicle in the foregoing description, those skilled in the art can clearly understand the dynamic balance degradation prediction system for the axial flux motor for an electric three-wheeled vehicle in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0070] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting dynamic balance degradation of an axial flux motor for an electric three-wheeled vehicle, characterized in that: The method comprises: The target electric tricycle is placed in a predetermined fixed foundation state for dynamic balancing detection to obtain a fixed foundation detection record; Extracting a first record at a first vehicle speed from the fixed basic detection record, and introducing a dynamic balance fusion discrimination model to analyze the first record to obtain a first dynamic balance index; Reading a predetermined road surface excitation state, and placing the target electric tricycle in the predetermined road surface excitation state for dynamic balance detection, to obtain a road surface excitation detection record; Extracting features from the road surface excitation detection records to obtain road surface excitation features, and performing variation weighted calculation on the road surface excitation features to obtain road surface excitation coefficients; Extracting a second record at the first vehicle speed from the road surface excitation detection record, and analyzing the second record according to the dynamic balance fusion discrimination model to obtain a second dynamic balance index; Comparing the first dynamic balance index with the second dynamic balance index to obtain a first dynamic balance degradation degree of the target electric three-wheeler at the first vehicle speed; obtaining a target dynamic balance degradation trend line according to a first mapping relationship between the first vehicle speed and the first dynamic balance degradation degree; The dynamic balance degradation of the target axial flux motor of the target electric three-wheeler is predicted according to the target dynamic balance degradation trend line.

2. The method for predicting dynamic balance degradation of an axial flux motor for an electric three-wheeled vehicle according to claim 1, characterized in that: Extracting a first record at a first vehicle speed from the fixed basic detection record, and introducing a dynamic balance fusion discrimination model to analyze the first record to obtain a first dynamic balance index, including: extracting a first electrical data record from the first record, wherein the first electrical data record includes a first current and a first voltage; Extracting a first operation data record from the first record, wherein the first operation data record includes a first overturning moment and a first internal bending moment; Extracting a first vibration time sequence signal from the first record, and extracting first vibration characteristic data of the first vibration time sequence signal, wherein the first vibration characteristic data includes a first vibration acceleration, a first vibration velocity, and a first vibration displacement; The first current, the first voltage, the first overturning moment, the first internal bending moment, the first vibration acceleration, the first vibration velocity, and the first vibration displacement are input into the dynamic balance fusion discrimination model to obtain the first dynamic balance index.

3. The method for predicting dynamic balance degradation of an axial flux motor for an electric three-wheeled vehicle according to claim 2, characterized in that: Inputting the first current, the first voltage, the first overturning moment, the first internal bending moment, the first vibration acceleration, the first vibration velocity, and the first vibration displacement into the dynamic balance fusion discrimination model to obtain the first dynamic balance index includes: Obtaining the initial discrimination layer in the dynamic balance fusion discrimination model; Analyze the first current and the first voltage by using the first discriminant model in the initial discriminant layer to obtain a first discriminant index; Analyze the first overturning moment and the first internal bending moment by using the second discriminant model in the initial discriminant layer to obtain a second discriminant index; Analyzing the first vibration acceleration, the first vibration velocity, and the first vibration displacement by using a third discriminant model in the initial discriminant layer to obtain a third discriminant index; Obtaining a meta-discriminant layer in the dynamic balance fusion discriminant model; The first discriminant index, the second discriminant index and the third discriminant index are analyzed by the meta-discriminant layer to obtain the first dynamic balance index.

4. The method for predicting dynamic balance degradation of an axial flux motor for an electric three-wheeled vehicle according to claim 1, characterized in that: The road surface excitation feature at least includes road surface roughness and crankshaft speed, wherein the road surface roughness refers to the excitation detection characteristic value of the predetermined road surface excitation state calculated by a predetermined road surface roughness function, and the expression of the predetermined road surface roughness function is as follows: ; in, The unit length The road surface roughness within The unit length The height of the road surface inside, is the average road height. is the length of the measuring section, refers to the amplitude of the predetermined road roughness function, and , is the attenuation factor, and .

5. The method for predicting dynamic balance degradation of an axial flux motor for an electric three-wheeled vehicle according to claim 4, characterized in that: After extracting features from the road surface excitation detection record to obtain road surface excitation features, and performing variation weighted calculation on the road surface excitation features to obtain road surface excitation coefficients, the method further includes: Extracting environmental excitation features from the road surface excitation detection record; Performing a weighted calculation on the ambient temperature and ambient humidity in the environmental excitation characteristics to obtain an environmental feedback coefficient; The road surface excitation coefficient is adjusted using the environmental feedback coefficient as a weight.

6. The method for predicting dynamic balance degradation of an axial flux motor for an electric three-wheeled vehicle according to claim 1, characterized in that: Comparing the first dynamic balance index with the second dynamic balance index to obtain a first dynamic balance degradation degree of the target electric three-wheeler at the first vehicle speed includes: Comparing the first dynamic balance index with the second dynamic balance index to obtain a first index difference; A first ratio of the first index difference to the first dynamic balance index is taken as the first dynamic balance degradation degree.

7. The method for predicting dynamic balance degradation of an axial flux motor for an electric three-wheeled vehicle according to claim 1, characterized in that: Predicting the dynamic balance degradation of the target axial flux motor of the target electric three-wheeler according to the target dynamic balance degradation trend line includes: Performing polynomial regression processing on the target dynamic balance degradation trend line to obtain a target fitting polynomial; The target speed is obtained, and the target predicted dynamic balance degradation degree is obtained in combination with the target polynomial.

8. A dynamic balance degradation prediction system for an axial flux motor for an electric three-wheeled vehicle, characterized in that: A method for predicting dynamic balance degradation of an axial flux motor for an electric three-wheeled vehicle according to any one of claims 1 to 7, the system comprising: A first dynamic balance detection module is used to place the target electric tricycle in a predetermined fixed base state for dynamic balance detection to obtain a fixed base detection record; A first record analysis module is used to extract a first record at a first vehicle speed from the fixed basic detection record, and introduce a dynamic balance fusion discrimination model to analyze the first record to obtain a first dynamic balance index; A second dynamic balance detection module is used to read a predetermined road surface excitation state, and place the target electric tricycle in the predetermined road surface excitation state to perform dynamic balance detection, and obtain a road surface excitation detection record; A variation weighted calculation module, used for extracting features from the road surface excitation detection record to obtain road surface excitation features, and performing variation weighted calculation on the road surface excitation features to obtain road surface excitation coefficients; A second record analysis module is used to extract a second record at the first vehicle speed from the road excitation detection record, and analyze the second record according to the dynamic balance fusion discrimination model to obtain a second dynamic balance index; a dynamic balance degradation degree acquisition module, configured to compare the first dynamic balance index with the second dynamic balance index to obtain a first dynamic balance degradation degree of the target electric three-wheeler at the first vehicle speed; a degradation trend line acquisition module, configured to obtain a target dynamic balance degradation trend line according to a first mapping relationship between the first vehicle speed and the first dynamic balance degradation degree; A dynamic balance degradation prediction module is used to predict the dynamic balance degradation of the target axial flux motor of the target electric three-wheeler according to the target dynamic balance degradation trend line.

Images