A traction motor for electric tricycle and a detection method thereof

Abstract

The present invention relates to the field of motor technology, and in particular to a traction motor for an electric tricycle and a detection method thereof. In terms of motor structure, various components are rationally arranged inside the motor, and the stability and efficiency of the motor operation are ensured through designs such as liquid cooling channels and rolling bearings. In terms of detection, the stability coefficient is obtained through index pairing and a dynamic change curve is constructed, which is compared with the preset curve to calibrate the motor as a qualified or unqualified product. For unqualified products, the actual working parameters are further analyzed to find abnormal parameters, and finally the abnormal type is predicted through the constructed working state abnormal type prediction model. On the whole, the traction motor and the detection method thereof disclosed in the present invention are helpful to improve the motor quality control level, accurately evaluate the motor performance, reduce the rate of unqualified products, and provide a basis for motor fault diagnosis and repair, thereby improving the reliability and safety of electric tricycles.

Description

Technical Field

[0001] The invention relates to the technical field of motors, in particular to a traction motor for an electric tricycle and a detection method thereof. Background Art

[0002] In the electric tricycle industry, the traction motor is one of the core components of the vehicle, and its performance stability and reliability have a vital impact on the overall performance of the vehicle. With the continuous expansion of the electric tricycle market and the continuous advancement of technology, the requirements for traction motors are getting higher and higher. The traction motor needs to maintain efficient and stable operation in various complex working environments, which requires strict control of its design, manufacturing and testing links. Traditional traction motor detection methods often rely on manual detection and simple instrument testing, which are not only inefficient, but also difficult to fully and accurately evaluate the performance of the motor. In addition, due to the diversity and complexity of the working environment of electric tricycles, the traction motor exhibits different performance characteristics under different working parameter combinations, which makes it difficult for traditional detection methods to meet the high standards of traction motor performance detection in the modern electric tricycle industry. Summary of the invention

[0003] The present invention overcomes the deficiencies of the prior art and provides a traction motor for an electric tricycle and a detection method thereof.

[0004] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is:

[0005] The first aspect of the present invention discloses a traction motor for an electric tricycle, comprising a housing and a base, wherein the housing is fixedly mounted on the base;

[0006] A motor rotor and a motor stator are arranged inside the housing, the motor rotor and the motor stator are interlocked with each other, an output shaft is matched and connected at the axis of the motor rotor, and the bottom end of the output shaft extends to the outside of the base, a card interface is arranged on the output shaft extending to the outside of the base, and the output shaft is connected to the base through a rolling bearing;

[0007] The upper and lower ends of the motor rotor are both provided with rotor end plates, a motor encoder and a controller power board are provided above the motor rotor, a capacitor is provided between the motor encoder and the controller power board, and a heat sink is provided below the capacitor.

[0008] More specifically, a water pipe connecting seat is provided on the side of the shell, and a water inlet and a water outlet are provided on the water pipe connecting seat. A liquid cooling channel is also provided inside the shell, and the water inlet end of the liquid cooling channel is connected to the water inlet, and the water outlet end of the liquid cooling channel is connected to the water outlet.

[0009] More specifically, the housing is sealed with a rubber sealing ring.

[0010] More specifically, a power line and a control line are also arranged on the shell.

[0011] A second aspect of the present invention discloses a method for detecting a traction motor for an electric tricycle, which is applied to a traction motor for an electric tricycle as described in any one of the first aspects, and comprises the following steps:

[0012] Constructing a stability coefficient index library of the traction motor, and performing a simulated working test on the traction motor, and collecting actual working parameter combinations of the traction motor at several preset time nodes based on the Internet of Things;

[0013] Importing the actual working parameter combinations collected at several preset time nodes into the stability coefficient index library for index matching, and obtaining the stability coefficients corresponding to the traction motors at several preset time nodes;

[0014] Constructing a stability coefficient dynamic change curve of the traction motor during a test period according to the stability coefficients corresponding to the traction motor at a number of preset time nodes; analyzing the overlap between the stability coefficient dynamic change curve and the preset curve;

[0015] If the overlap is greater than the preset overlap, the traction motor is calibrated as a first-class product; if the overlap is not greater than the preset overlap, the traction motor is calibrated as a second-class product; wherein the first-class product is a qualified product, and the second-class product is an unqualified product;

[0016] If the traction motor is calibrated as a Class II product, the actual working parameter combinations collected at several preset time nodes are subjected to feature classification processing to obtain various actual working parameters of the traction motor calibrated as a Class II product during the test period;

[0017] Performing feature analysis on various actual working parameters of the traction motor calibrated as a Class II product during the test period, and obtaining abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period;

[0018] A prediction model for the abnormal working state type of the traction motor is constructed, and the abnormal actual working parameters of the traction motor labeled as Class II during the test time period are imported into the prediction model for the abnormal working state type for prediction, so as to obtain the abnormal working state type of the traction motor labeled as Class II.

[0019] More specifically, a stability coefficient index library of the traction motor is constructed, specifically:

[0020] Obtain the stability coefficient of the traction motor under various preset working parameter combinations through the big data network;

[0021] Constructing a database space, counting the number of various preset working parameter combinations, and dividing the database space into a corresponding number of sub-libraries according to the number of various preset working parameter combinations;

[0022] And the stability coefficients under various preset working parameter combination conditions are etched on the metadata area of ​​the corresponding sub-library, and then various preset working parameter combinations are stored in the index structure area of ​​the corresponding sub-library to obtain a stability coefficient index library.

[0023] More specifically, the actual working parameter combinations collected at several preset time nodes are imported into the stability coefficient index library for index matching to obtain the stability coefficients corresponding to the traction motors at several preset time nodes, specifically:

[0024] Importing the actual working parameter combinations collected at each preset time node into the stability coefficient index library in sequence according to the timestamp;

[0025] Calculating the Jaccard similarity coefficient between the actual working parameter combination and the preset working parameter combination in each sub-library, and sorting the calculated Jaccard similarity coefficients to obtain the maximum Jaccard similarity coefficient;

[0026] Marking a sub-library of a preset working parameter combination corresponding to the maximum Jaccard similarity coefficient, and extracting a corresponding stability coefficient in a metadata area of ​​the marked sub-library to obtain a stability coefficient corresponding to the traction motor at a corresponding preset time node;

[0027] After all actual working parameter combinations collected at each preset time node are imported into the stability coefficient index library for index matching, the stability coefficient corresponding to the traction motor at each preset time node is obtained.

[0028] More specifically, the overlap between the dynamic change curve of the stability coefficient and the preset curve is analyzed, specifically:

[0029] Acquiring working performance requirement information of the traction motor, and determining a standard stability coefficient corresponding to the traction motor at a plurality of preset time nodes according to the working performance requirement information;

[0030] Draw a preset curve according to the standard stability coefficient corresponding to the traction motor at a number of preset time nodes;

[0031] Introducing a Euclidean distance algorithm, and calculating a Euclidean distance value between the stability coefficient dynamic change curve and a preset curve based on the Euclidean distance algorithm;

[0032] The degree of overlap between the stability coefficient dynamic change curve and a preset curve is determined according to the Euclidean distance value.

[0033] More specifically, various actual working parameters of the traction motor calibrated as a Class II product during the test period are subjected to characteristic analysis and processing to obtain abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period, specifically:

[0034] Constructing a dynamic variation curve diagram of various actual working parameters of the traction motor calibrated as a second-class product during the test period according to various actual working parameters of the traction motor calibrated as a second-class product during the test period;

[0035] Obtaining standard range values ​​of various actual working parameters of the traction motor during the test period, and determining upper and lower limits in the corresponding dynamic change curve diagrams according to the standard range values ​​of the various actual working parameters, thereby obtaining the limit areas of each dynamic change curve diagram;

[0036] Analyze the total length of the line segments in the boundary area and the total length outside the boundary area in each dynamic change curve graph respectively, divide the total length of the line segments in the boundary area and the total length outside the boundary area in each dynamic change curve graph, and obtain the offset coefficient of the curve in each dynamic change curve graph;

[0037] Compare the offset coefficient of each curve in the dynamic change curve diagram with the preset coefficient threshold value respectively;

[0038] If the offset coefficient of a curve in a certain dynamic change curve diagram is greater than a preset coefficient threshold, the actual working parameter corresponding to the dynamic change curve diagram is calibrated as an abnormal actual working parameter;

[0039] If the offset coefficient of a curve in a certain dynamic change curve graph is not greater than a preset coefficient threshold, the actual working parameter corresponding to the dynamic change curve graph is calibrated as a normal actual working parameter.

[0040] More specifically, a prediction model for the abnormal working state type of the traction motor is constructed, and the abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period are imported into the prediction model for the abnormal working state type to make a prediction, and the abnormal working state type of the traction motor calibrated as a Class II product is obtained, specifically:

[0041] Obtain various abnormal working state types of the traction motor through the big data network, and obtain abnormal characteristic parameters corresponding to various abnormal working state types, so as to obtain various abnormal working state types and their corresponding abnormal characteristic parameters;

[0042] Based on the grey correlation analysis method, various abnormal working state types and their corresponding abnormal characteristic parameters are correlated and analyzed to obtain the correlation text between various abnormal working state types and their corresponding abnormal characteristic parameters;

[0043] Introducing a hidden Markov random field, taking the abnormal working state type as a condition node, and taking the abnormal characteristic parameter as a variable node;

[0044] According to the association text between various working state abnormality types and their corresponding abnormal characteristic parameters, the condition node is associated with the variable node to obtain a topological structure diagram between various working state abnormality types and their corresponding abnormal characteristic parameters;

[0045] Building a prediction model based on a graph network, embedding the topological structure graph into the prediction model, and obtaining a prediction model for abnormal working state types of the traction motor;

[0046] Acquire abnormal actual working parameters of the traction motor calibrated as a Class II product during a test period, and import the abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period into the working state abnormality type prediction model;

[0047] The abnormal working state type of the traction motor calibrated as a second-class product is obtained through prediction, and the abnormal working state type of the traction motor calibrated as a second-class product is sent to a preset terminal for display.

[0048] The present invention solves the technical defects existing in the background technology, and has the following beneficial effects: in terms of motor structure, various components such as motor rotor, stator, encoder, controller power board, etc. are rationally arranged inside, and the stability and efficiency of motor operation are guaranteed through designs such as liquid cooling channels and rolling bearings. In terms of detection, a stability coefficient index library is constructed, and the stability coefficient is obtained through index pairing and a dynamic change curve is constructed. The motor is calibrated as a qualified or unqualified product by comparing it with the preset curve. For unqualified products, the actual working parameters are further analyzed to find abnormal parameters, and finally the abnormal type is predicted by the constructed working state abnormal type prediction model. On the whole, the traction motor and its detection method disclosed in the present invention are helpful to improve the motor quality control level, accurately evaluate the motor performance, reduce the unqualified rate, and provide a basis for motor fault diagnosis and repair, thereby improving the reliability and safety of electric tricycles. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0050] Figure 1 is a first three-dimensional structural schematic diagram of the traction motor;

[0051] Figure 2 is a second three-dimensional structural schematic diagram of the traction motor;

[0052] Figure 3 This is a schematic diagram of the structure of the traction motor from a bottom view;

[0053] Figure 4 for Figure 3 Schematic diagram of the cross section at BB;

[0054] The description of the accompanying drawings is as follows: 101, housing; 102, base; 103, motor rotor; 104, motor stator; 105, output shaft; 106, card interface; 107, rolling bearing; 108, rotor end plate; 109, motor encoder; 201, controller power board; 202, capacitor; 203, heat sink; 204, water pipe connection seat; 205, water inlet; 206, water outlet; 207, liquid cooling channel; 208, rubber sealing ring; 209, power cord; 301, control line. DETAILED DESCRIPTION

[0055] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. The preferred embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thoroughly understood.

[0056] like Figure 1 , Figure 2 As shown, the first aspect of the present invention discloses a traction motor for an electric tricycle, comprising a housing 101 and a base 102, wherein the housing is fixedly mounted on the base.

[0057] like Figure 2 , Figure 3 , Figure 4As shown, a motor rotor 103 and a motor stator 104 are arranged inside the shell, and the motor rotor and the motor stator are interlocked with each other. An output shaft 105 is connected to the axis of the motor rotor, and the bottom end of the output shaft extends to the outside of the base. A card interface 106 is arranged on the output shaft extending to the outside of the base, and the output shaft is connected to the base through a rolling bearing 107.

[0058] The upper and lower ends of the motor rotor are both provided with rotor end plates 108, the motor encoder 109 and the controller power board 201 are provided above the motor rotor, a capacitor 202 is provided between the motor encoder and the controller power board, and a heat sink 203 is provided below the capacitor.

[0059] It should be noted that the controller power board 201 starts initialization after receiving power. The controller power board 201 works in conjunction with the motor encoder 109, which is used to detect information such as the position and speed of the motor rotor 103. The capacitor 202 plays a role in stabilizing the circuit voltage in this process, ensuring that the controller power board 201 can obtain a stable voltage supply during startup and operation.

[0060] A water pipe connection seat 204 is provided on the side of the shell, and a water inlet 205 and a water outlet 206 are provided on the water pipe connection seat. A liquid cooling channel 207 is also provided inside the shell, and the water inlet end of the liquid cooling channel is connected to the water inlet, and the water outlet end of the liquid cooling channel is connected to the water outlet.

[0061] The housing is sealed with a rubber sealing ring 208 .

[0062] The housing is also provided with a power line 209 and a control line 301 .

[0063] It should be noted that the controller power board 201 adjusts the frequency and magnitude of the current output to the motor stator 104 according to the speed information of the motor rotor 103 fed back by the motor encoder 109. The motor stator 104 generates a rotating magnetic field, interacts with the motor rotor 103, and causes the motor rotor 103 to rotate at a set speed. The rotation of the motor rotor 103 transmits power to the transmission system of the electric tricycle through the output shaft 105. During the operation of the motor, the liquid cooling channel 207 starts to work. The external coolant enters the liquid cooling channel 207 from the water inlet 205. The coolant flows in the liquid cooling channel 207, takes away the heat generated by the operation of the motor, and then flows out from the water outlet 206. In this way, the temperature of the motor can be effectively controlled to prevent the motor from overheating. When the electric tricycle stops running (for example, braking or turning off the power switch), the controller power board 201 stops outputting current to the motor stator 104. The motor rotor 103 gradually stops rotating under the action of inertia, and the output shaft 105 also stops power output.

[0064] In general, the housing is sealed with a rubber seal ring 208, which can effectively prevent dust, moisture and other impurities from entering the motor. This helps to protect the motor rotor 103, motor stator 104, motor encoder 109 and other precision components inside the motor, and prolong the service life of the motor.

[0065] Inside the housing, the motor rotor 103, the motor stator 104, the motor encoder 109, the controller power board 201, the capacitor 202 and the heat sink 203 are arranged compactly. This compact layout can reduce the overall size of the motor and facilitate installation in the limited space of the electric tricycle. For example, the motor encoder 109 is arranged adjacent to the controller power board 201, and the capacitor 202 is located between them. Such a layout is conducive to signal transmission and circuit integration, reduces line length, and reduces the possibility of electromagnetic interference.

[0066] Compared with traditional air cooling, the liquid cooling channel 207 has a higher cooling efficiency. When the electric tricycle runs for a long time or under high load, the motor generates a lot of heat. The liquid cooling system can quickly and effectively remove the heat and keep the motor within a suitable operating temperature range, thereby improving the working efficiency and reliability of the motor.

[0067] The cooperation between the motor encoder 109 and the controller power board 201 makes the control of the motor more precise. The motor encoder 109 can accurately feedback the operating state of the motor rotor 103 in real time, and the controller power board 201 can accurately adjust the operating parameters of the motor, such as speed and torque, according to the feedback information. This helps to improve the driving comfort and safety of the electric tricycle, for example, during climbing or deceleration, it can achieve smooth power output and speed adjustment.

[0068] The housing 101 is fixedly mounted on the base 102, and the bottom end of the output shaft 105 extends to the outside of the base 102 and is provided with a card interface 106. This structure is convenient for installing the motor on the transmission system of the electric tricycle, and the card interface 106 can be quickly connected with other transmission components. The output shaft 105 is connected to the base 102 through a rolling bearing 107, and the rolling bearing 107 can reduce the friction when the output shaft 105 rotates, improve the transmission efficiency, and also facilitate the positioning and support of the output shaft 105 during installation and maintenance. The power line 209 and the control line 301 of the motor are separately arranged on the housing, which is convenient for the inspection and maintenance of the line. If a circuit fault occurs, it can be relatively easy to find the problem and repair it. The water pipe connection seat 204 of the liquid cooling system is clearly provided with a water inlet 205 and a water outlet 206, which is relatively simple to operate when replacing the coolant or maintaining the liquid cooling system.

[0069] The present invention also discloses a detection method for a traction motor for an electric tricycle, which is applied to any one of the traction motors for an electric tricycle, and comprises the following steps:

[0070] Constructing a stability coefficient index library of the traction motor, and performing a simulated working test on the traction motor, and collecting actual working parameter combinations of the traction motor at several preset time nodes based on the Internet of Things;

[0071] Importing the actual working parameter combinations collected at several preset time nodes into the stability coefficient index library for index matching, and obtaining the stability coefficients corresponding to the traction motors at several preset time nodes;

[0072] Constructing a stability coefficient dynamic change curve of the traction motor during a test period according to the stability coefficients corresponding to the traction motor at a number of preset time nodes; analyzing the overlap between the stability coefficient dynamic change curve and the preset curve;

[0073] If the overlap is greater than the preset overlap, the traction motor is calibrated as a first-class product; if the overlap is not greater than the preset overlap, the traction motor is calibrated as a second-class product; wherein the first-class product is a qualified product, and the second-class product is an unqualified product;

[0074] If the traction motor is calibrated as a Class II product, the actual working parameter combinations collected at several preset time nodes are subjected to feature classification processing to obtain various actual working parameters of the traction motor calibrated as a Class II product during the test period;

[0075] Performing feature analysis on various actual working parameters of the traction motor calibrated as a Class II product during the test period, and obtaining abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period;

[0076] A prediction model for the abnormal working state type of the traction motor is constructed, and the abnormal actual working parameters of the traction motor labeled as Class II during the test time period are imported into the prediction model for the abnormal working state type for prediction, so as to obtain the abnormal working state type of the traction motor labeled as Class II.

[0077] It should be noted that a dynamic change curve of the stability coefficient is constructed according to the stability coefficient of each time node. This curve intuitively shows the change trend of the stability of the motor during the test period. Then it is compared with the preset curve determined according to the working performance requirements, and the performance of the motor is evaluated by calculating the overlap between the two (such as using the Euclidean distance algorithm). If the overlap is high, it means that the actual operating stability of the motor meets expectations; otherwise, it does not meet expectations. According to the comparison results of the overlap and the preset overlap, the traction motor is calibrated as a Class I product (qualified product) or a Class II product (unqualified product). This calibration method provides a clear standard for the quality assessment of the motor and helps to screen out motor products that meet the requirements. For the traction motor calibrated as a Class II product, the collected actual working parameter combinations are further subjected to feature classification processing to obtain various actual working parameters of the motor during the test period. Then, by analyzing the dynamic change curves of these actual working parameters (such as determining the boundary area of ​​the curve, calculating the offset coefficient, etc.), the abnormal actual working parameters are found. This helps to gain a deeper understanding of the specific reasons for the motor failure.

[0078] By building a stability coefficient index library, collecting actual working parameters, analyzing the dynamic change curve of the stability coefficient, and other steps, the performance of the traction motor can be comprehensively and accurately evaluated. Not only can qualified products be distinguished from unqualified products, but unqualified products can also be deeply analyzed to find abnormal actual working parameters and predict the type of abnormal working state. This helps to improve the quality control level of electric tricycle traction motors, reduce the outflow rate of unqualified products, and provide effective technical support for motor fault diagnosis and repair, thereby improving the overall reliability and safety of electric tricycles.

[0079] More specifically, a stability coefficient index library of the traction motor is constructed, specifically:

[0080] Obtain the stability coefficient of the traction motor under various preset working parameter combinations through the big data network;

[0081] Constructing a database space, counting the number of various preset working parameter combinations, and dividing the database space into a corresponding number of sub-libraries according to the number of various preset working parameter combinations;

[0082] And the stability coefficients under various preset working parameter combination conditions are etched on the metadata area of ​​the corresponding sub-library, and then various preset working parameter combinations are stored in the index structure area of ​​the corresponding sub-library to obtain a stability coefficient index library.

[0083] It should be noted that a large amount of data about traction motors under different working conditions can be collected through the big data network. The preset working parameter combination can cover multiple parameters such as the motor's input voltage, current, load, speed, ambient temperature, etc. By analyzing the motor's operating data under these different combinations, the corresponding stability coefficient is calculated. This stability coefficient reflects the stability of the motor under a specific working parameter combination, such as the fluctuation of the motor's output power, whether it is prone to overheating or vibration abnormality, etc. The database space is constructed to systematically manage data related to the traction motor. The number of various preset working parameter combinations is counted, and the corresponding number of sub-libraries are divided according to this number. Each sub-library is equivalent to a storage space specifically for a certain type of preset working parameter combination. This division method makes data management more orderly and facilitates subsequent data query, analysis and processing. For example, if there are 10 different preset working parameter combinations, 10 sub-libraries will be divided, each corresponding to a combination type. The stability coefficient is etched in the metadata area of ​​the corresponding sub-library. The metadata area is a description area of ​​the overall information of the sub-library. The stability coefficient is placed here to facilitate the rapid acquisition of the stability profile of the motor under the working parameter combination when querying the sub-library. Then various preset working parameter combinations are stored in the index structure area of ​​the corresponding sub-library. The index structure area helps to improve the efficiency of data query. When it is necessary to find data of a specific working parameter combination, the corresponding storage location can be quickly located through the index.

[0084] By building such a stability coefficient index library, efficient management and quick query of stability data of traction motors under different preset working parameter combinations can be achieved. The stability of the motor under various working conditions can be accurately grasped, providing strong data support for performance evaluation, fault diagnosis, and optimization design of traction motors, which will help improve the overall performance and reliability of traction motors for electric tricycles, and also provide a solid data foundation for subsequent detection methods (such as motor calibration based on stability coefficients, etc.).

[0085] More specifically, the actual working parameter combinations collected at several preset time nodes are imported into the stability coefficient index library for index matching to obtain the stability coefficients corresponding to the traction motors at several preset time nodes, specifically:

[0086] Importing the actual working parameter combinations collected at each preset time node into the stability coefficient index library in sequence according to the timestamp;

[0087] Calculating the Jaccard similarity coefficient between the actual working parameter combination and the preset working parameter combination in each sub-library, and sorting the calculated Jaccard similarity coefficients to obtain the maximum Jaccard similarity coefficient;

[0088] Marking a sub-library of a preset working parameter combination corresponding to the maximum Jaccard similarity coefficient, and extracting a corresponding stability coefficient in a metadata area of ​​the marked sub-library to obtain a stability coefficient corresponding to the traction motor at a corresponding preset time node;

[0089] After all actual working parameter combinations collected at each preset time node are imported into the stability coefficient index library for index matching, the stability coefficient corresponding to the traction motor at each preset time node is obtained.

[0090] It should be noted that the timestamp provides a sequential identifier for the actual working parameter combination collected. Importing the actual working parameter combination collected into the stability coefficient index library in sequence according to the timestamp is an orderly data processing method. For example, during the operation of the electric tricycle, the working state of the motor is different at different times. Importing data in chronological order can fully reflect the state change of the motor during the entire operation process. Sort the calculated Jaccard similarity coefficients to find the largest Jaccard similarity coefficient. The purpose of this step is to find the one that is most similar to the actual working parameter combination among many preset working parameter combinations. For example, if the actual working parameter combination includes parameters such as motor speed, load, and temperature, by calculating the similarity coefficient, the preset working parameter combination type closest to the current actual working state can be found. Since the metadata area of ​​each sub-library stores the stability coefficient under the corresponding preset working parameter combination, the stability coefficient is extracted from the metadata area of ​​the marked sub-library. This stability coefficient is the stability coefficient corresponding to the traction motor at the corresponding preset time node. This process achieves the purpose of finding a matching preset working parameter combination according to the actual working parameter combination and obtaining the corresponding stability coefficient. Repeat the above steps until all the actual working parameter combinations collected at each preset time node are imported into the stability coefficient index library for index matching. In this way, the stability coefficient corresponding to each preset time node of the traction motor can be obtained, so as to fully understand the stability of the traction motor at different times during the entire test period.

[0091] Through this index pairing method, the stability coefficient of the traction motor at each preset time node can be accurately obtained. This helps to accurately evaluate the stability state of the traction motor at different times, and provides key data basis for the subsequent construction of the stability coefficient dynamic change curve, analysis of motor performance, judgment of motor qualification, and fault diagnosis, thereby achieving comprehensive monitoring and evaluation of traction motor performance and improving the reliability and safety of electric tricycle traction motors.

[0092] More specifically, the overlap between the dynamic change curve of the stability coefficient and the preset curve is analyzed, specifically:

[0093] Acquiring working performance requirement information of the traction motor, and determining a standard stability coefficient corresponding to the traction motor at a plurality of preset time nodes according to the working performance requirement information;

[0094] Draw a preset curve according to the standard stability coefficient corresponding to the traction motor at a number of preset time nodes;

[0095] Introducing a Euclidean distance algorithm, and calculating a Euclidean distance value between the stability coefficient dynamic change curve and a preset curve based on the Euclidean distance algorithm;

[0096] The degree of overlap between the stability coefficient dynamic change curve and a preset curve is determined according to the Euclidean distance value.

[0097] It should be noted that the working performance requirement information of the traction motor must be clarified first, which is the basis for evaluating whether the motor is qualified. The working performance requirement information may include the stability standards that the motor should meet under different working conditions. According to these requirements, the standard stability coefficient corresponding to the traction motor at each preset time node is determined. For example, for the traction motor for electric tricycles, in different periods such as the starting stage, the uniform speed driving stage and the climbing stage, according to the design requirements of the vehicle, the motor should have corresponding stable operation standards, and these standards are reflected in the standard stability coefficient. With the standard stability coefficient corresponding to each preset time node, the preset curve can be drawn. This preset curve represents the change trend of the stability coefficient when the traction motor runs according to the ideal working performance requirements. It is a benchmark for comparison with the dynamic change curve of the stability coefficient obtained by actual testing. The Euclidean distance algorithm is introduced, which is a common method for calculating the distance between two points in multidimensional space. Here, the dynamic change curve of the stability coefficient and the preset curve are regarded as two curves in the space with time nodes as the dimension. The distance value between the two curves is calculated by the Euclidean distance algorithm. This distance value reflects the comprehensive situation of the difference in the stability coefficient of the two curves at each time node. The overlap between the dynamic change curve of the stability coefficient and the preset curve is determined based on the calculated Euclidean distance value. The overlap is an indicator to measure the degree of closeness between the actual operation and the ideal working performance requirements. If the Euclidean distance value is small, it means that the two curves are close and the overlap is high; otherwise, the overlap is low.

[0098] Through the above steps, the overlap between the dynamic change curve of the stability coefficient and the preset curve can be accurately analyzed. This overlap can intuitively reflect the degree of compliance between the performance of the traction motor in the actual operation process and the ideal working performance requirements, thereby providing a quantitative basis for judging whether the traction motor is qualified, and helping to accurately calibrate the traction motor as a qualified product (class I product) or a non-qualified product (class II product), thereby ensuring that the quality and performance of the traction motor of the electric tricycle meet the requirements.

[0099] More specifically, various actual working parameters of the traction motor calibrated as a Class II product during the test period are subjected to characteristic analysis and processing to obtain abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period, specifically:

[0100] Constructing a dynamic variation curve diagram of various actual working parameters of the traction motor calibrated as a second-class product during the test period according to various actual working parameters of the traction motor calibrated as a second-class product during the test period;

[0101] Obtaining standard range values ​​of various actual working parameters of the traction motor during the test period, and determining upper and lower limits in the corresponding dynamic change curve diagrams according to the standard range values ​​of the various actual working parameters, thereby obtaining the limit areas of each dynamic change curve diagram;

[0102] Analyze the total length of the line segments in the boundary area and the total length outside the boundary area in each dynamic change curve graph respectively, divide the total length of the line segments in the boundary area and the total length outside the boundary area in each dynamic change curve graph, and obtain the offset coefficient of the curve in each dynamic change curve graph;

[0103] Compare the offset coefficient of each curve in the dynamic change curve diagram with the preset coefficient threshold value respectively;

[0104] If the offset coefficient of a curve in a certain dynamic change curve diagram is greater than a preset coefficient threshold, the actual working parameter corresponding to the dynamic change curve diagram is calibrated as an abnormal actual working parameter;

[0105] If the offset coefficient of a curve in a certain dynamic change curve graph is not greater than a preset coefficient threshold, the actual working parameter corresponding to the dynamic change curve graph is calibrated as a normal actual working parameter.

[0106] It should be noted that for the traction motors calibrated as Class II products (unqualified products), a dynamic change curve chart is constructed based on various actual working parameters during the test period. These actual working parameters can be the speed, torque, temperature, current, etc. of the motor. By drawing a dynamic change curve chart, the change trend of these parameters during the test period can be intuitively displayed, providing a visual basis for subsequent analysis. The standard range values ​​of various actual working parameters of the traction motor during the test period are obtained. These standard range values ​​are set based on the requirements for normal operation of the motor. According to these standard range values, the upper and lower limits are determined in the corresponding dynamic change curve chart, thereby obtaining the limit area of ​​each dynamic change curve chart. This limit area represents the range of normal working parameters, within which the motor is considered to be operating normally.

[0107] Analyze the total length of the line segments in each dynamic change curve graph within and outside the boundary area. Divide the total length in the boundary area by the total length outside the boundary area to obtain the offset coefficient of the curve in each dynamic change curve graph. This offset coefficient reflects the degree to which the change curve of the actual working parameter deviates from the normal range (boundary area). If the offset coefficient is large, it means that the actual working parameter deviates from the normal range more seriously; if the offset coefficient is small, it means that the degree of deviation is small. Compare the offset coefficient of the curve in each dynamic change curve graph with the preset coefficient threshold. The preset coefficient threshold is a pre-set judgment standard used to distinguish between normal and abnormal working parameters. If the offset coefficient of a curve in a dynamic change curve graph is greater than the preset coefficient threshold, the actual working parameter corresponding to the dynamic change curve graph is calibrated as an abnormal actual working parameter; if the offset coefficient is not greater than the preset coefficient threshold, the corresponding actual working parameter is calibrated as a normal actual working parameter. This comparison and calibration method can accurately find out the abnormal actual working parameters that cause the traction motor to fail.

[0108] Through the above steps, the various actual working parameters of the traction motor calibrated as Class II during the test period can be analyzed in detail, so as to accurately obtain the abnormal actual working parameters. This helps to gain a deeper understanding of the reasons for the failure of the traction motor, provides key data support for the subsequent construction of a prediction model for the abnormal type of working state, fault diagnosis and improvement of motor performance, and improves the ability of traction motor quality control and fault analysis.

[0109] More specifically, a prediction model for the abnormal working state type of the traction motor is constructed, and the abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period are imported into the prediction model for the abnormal working state type to make a prediction, and the abnormal working state type of the traction motor calibrated as a Class II product is obtained, specifically:

[0110] Obtain various abnormal working state types of the traction motor through the big data network, and obtain abnormal characteristic parameters corresponding to various abnormal working state types, so as to obtain various abnormal working state types and their corresponding abnormal characteristic parameters;

[0111] Based on the grey correlation analysis method, various abnormal working state types and their corresponding abnormal characteristic parameters are correlated and analyzed to obtain the correlation text between various abnormal working state types and their corresponding abnormal characteristic parameters;

[0112] Introducing a hidden Markov random field, taking the abnormal working state type as a condition node, and taking the abnormal characteristic parameter as a variable node;

[0113] According to the association text between various working state abnormality types and their corresponding abnormal characteristic parameters, the condition node is associated with the variable node to obtain a topological structure diagram between various working state abnormality types and their corresponding abnormal characteristic parameters;

[0114] Building a prediction model based on a graph network, embedding the topological structure graph into the prediction model, and obtaining a prediction model for abnormal working state types of the traction motor;

[0115] Acquire abnormal actual working parameters of the traction motor calibrated as a Class II product during a test period, and import the abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period into the working state abnormality type prediction model;

[0116] The abnormal working state type of the traction motor calibrated as a second-class product is obtained through prediction, and the abnormal working state type of the traction motor calibrated as a second-class product is sent to a preset terminal for display.

[0117] It should be noted that various types of abnormal working conditions of traction motors are collected through the big data network, and the corresponding abnormal characteristic parameters when each abnormal type occurs are obtained. This step lays the data foundation for the subsequent establishment of a prediction model. For example, the abnormal characteristic parameters corresponding to motor overheating may be excessive current and poor heat dissipation (manifested as abnormal temperature sensor readings, etc.); abnormal motor vibration may be related to rotor imbalance (manifested as characteristic parameters such as speed fluctuations). Based on the gray correlation analysis method, various types of abnormal working conditions and their corresponding abnormal characteristic parameters are analyzed for correlation. This analysis method can determine the degree of correlation between them, thereby obtaining a correlation text describing their relationship. For example, through analysis, it may be found that under a certain abnormal state of the motor, a certain abnormal characteristic parameter has a very high correlation with the abnormal state, which helps to accurately identify the key factors of the abnormality.

[0118] Hidden Markov random fields are introduced, and the abnormal working state type is set as a conditional node, and the abnormal characteristic parameters are set as variable nodes. According to the previously obtained association text, the conditional nodes are associated with the variable nodes to construct a topological structure diagram. This topological structure diagram intuitively shows the relationship structure between the abnormal working state type and the abnormal characteristic parameters, and is an important intermediate product for building a prediction model. The prediction model is built based on the graph network, and the topological structure diagram is embedded in it to obtain the abnormal working state type prediction model of the traction motor. The graph network provides an effective framework for processing complex relationship structures, so that the prediction model can accurately predict the abnormal working state type based on the input abnormal characteristic parameters.

[0119] The actual abnormal working parameters of the traction motor calibrated as Class II during the test period are obtained and imported into the constructed working state abnormality type prediction model for prediction. Through the calculation of the model, the abnormal working state type of the traction motor calibrated as Class II is obtained, and the result is sent to the preset terminal for display. This enables relevant personnel to intuitively understand the specific abnormal type of the motor, which is helpful for targeted maintenance and improvement.

[0120] The abnormal working state type prediction model constructed through the above steps can effectively use the information obtained from the big data network to accurately analyze the relationship between the abnormal working state type of the traction motor and the abnormal characteristic parameters. After the abnormal actual working parameters of the traction motor calibrated as a Class II product are imported into the model, its abnormal working state type can be accurately predicted and the results can be displayed on the preset terminal. This helps to quickly locate the fault type of the traction motor, improve maintenance efficiency, and reduce maintenance costs, thereby ensuring the normal operation of the traction motor of the electric tricycle and improving the reliability and safety of the entire electric tricycle.

[0121] In addition, the actual working parameter combinations collected at several preset time nodes are subjected to feature classification processing to obtain various actual working parameters of the traction motor calibrated as Class II during the test period, specifically:

[0122] Constructing a decision tree, initializing a root node of the decision tree, and importing actual working parameter combinations collected at a number of preset time nodes into the root node;

[0123] Based on the root node feature, the actual working parameter combination is divided to form different branches; each branch is divided again, and the division is performed recursively; wherein, in the division process, the actual working parameter combination is allocated to different sub-branches according to the value of each feature;

[0124] The division is stopped until there are no other actual working parameters in each obtained branch; a cosine similarity algorithm is introduced, and the cosine similarity between each branch is calculated based on the cosine similarity algorithm;

[0125] The branches whose cosine similarity is greater than a preset similarity threshold are merged to obtain several branch groups;

[0126] Calculate the silhouette coefficient of each branch group, and mark the branch group with a silhouette coefficient greater than a preset value as a branch group with a density that meets the standard; mark the branch group with a silhouette coefficient not greater than the preset value as a branch group with a density that does not meet the standard;

[0127] The actual working parameters in the branch group with substandard density are discretized again, and then the cosine similarity between the actual working parameters after discretization is recalculated, and then the merging process is re-performed;

[0128] When the contour coefficients of all branch groups are greater than the preset values, several branch groups with qualified density are obtained; each branch group with qualified density is cut in the decision tree and calibrated as various actual working parameters of the second-class traction motor during the test period.

[0129] It should be noted that the decision tree is first constructed and the root node is initialized, which is the basic framework of the entire feature classification processing. The collected actual working parameter combinations are imported into the root node to prepare for the subsequent classification based on the decision tree. The structure of the decision tree helps to gradually analyze and classify complex actual working parameter combinations. The actual working parameter combinations are divided into different branches based on the root node features, and then each branch is divided in the same way. This is a recursive process. When dividing, the actual working parameter combinations are assigned to different sub-branches according to their values ​​under each feature, so that similar parameter combinations can be clustered together according to different parameter value ranges or states. When there are no other actual working parameters in each branch, the division is stopped, and the preliminary structure of the decision tree is formed. The branches whose cosine similarity is greater than the preset similarity threshold are merged to obtain several branch groups. Then the silhouette coefficient of each branch group is calculated. The silhouette coefficient is used to evaluate the density and separation of the branch group, so as to judge the quality of the branch group. For branch groups whose density does not meet the standard (the silhouette coefficient is not greater than the preset value), the actual working parameters belonging to them are re-discretized, and then the cosine similarity is recalculated and merged again. This step aims to optimize the structure of the branch group and improve its density and quality. Repeat the above operation until the contour coefficient of each branch group is greater than the preset value, and obtain several branch groups with qualified density. Finally, these branch groups with qualified density are cut in the decision tree to obtain various actual working parameters of the traction motor calibrated as Class II during the test period. These parameters have been strictly classified and processed, which helps to analyze the reasons for the motor failure in more depth. Through the above series of steps, the actual working parameter combination of the traction motor calibrated as Class II during the test period can be effectively classified. By using the recursive partitioning of the decision tree, the branch merging based on cosine similarity, the contour coefficient evaluation, and the reprocessing of the branch group with unqualified density, various actual working parameters that have been strictly classified are finally obtained. This helps to more accurately analyze the working status of the Class II traction motor, find out possible problems, and provide strong data support for further improving the motor performance or fault diagnosis.

[0130] In addition, in practical application, the detection method may further include the following steps:

[0131] Acquire image information of the traction motor at several times, and construct an actual assembly feature model diagram of the traction motor according to the image information; and acquire assembly drawing information of the traction motor, and construct a standard assembly feature model diagram of the traction motor according to the assembly drawing information;

[0132] Introducing a Euclidean distance algorithm, and calculating the overlap between the actual assembly feature model graph and the standard assembly feature model graph based on the Euclidean distance algorithm;

[0133] If the overlap is not greater than a preset overlap, a grid space is constructed, and the grid space is divided into a plurality of sub-grids, and the actual assembly feature model diagram and the standard assembly feature model diagram are imported into the grid space for registration;

[0134] Calculating the degree of intersection between the actual assembly feature model graph existing in each sub-grid and the model part of the standard assembly feature model graph;

[0135] Only the subgrids corresponding to the intersection degree not greater than the preset intersection degree are extracted; and the model parts corresponding to the subgrids corresponding to the intersection degree not greater than the preset intersection degree are marked to obtain the assembly offset area of ​​the traction motor;

[0136] The assembly offset area of ​​the traction motor is sent to a preset terminal for display.

[0137] It should be noted that in actual applications, the image information of the traction motor is first obtained and the actual assembly feature model diagram is constructed, and the standard assembly feature model diagram is constructed based on the assembly drawing information. These two model diagrams represent the actual assembly situation and the ideal assembly standard of the traction motor respectively. The actual assembly feature model diagram can reflect the actual state of the motor after actual assembly, including the actual situation such as the position and shape of each component; while the standard assembly feature model diagram is a model of the ideal assembly state formulated according to design requirements and specifications. The Euclidean distance algorithm is introduced to calculate the overlap between the actual assembly feature model diagram and the standard assembly feature model diagram. The Euclidean distance algorithm can quantify the similarity between the two model diagrams, and the overlap index can intuitively reflect the closeness between the actual assembly and the standard assembly. If the overlap is not high, it means that there is a problem with the actual assembly. When the overlap is not greater than the preset overlap, a grid space is constructed and divided into subgrids, and the two model diagrams are imported into it for registration. In each subgrid, the overlap between the model parts of the actual assembly feature model diagram and the standard assembly feature model diagram is calculated. This method of analysis in grid space refines the overall model into local sub-grids, which can more accurately locate areas where assembly differences may exist. Only sub-grids with a degree of intersection not greater than the preset degree of intersection are extracted, and the model parts corresponding to these sub-grids are the assembly offset areas. By marking and identifying these areas, it is possible to clearly identify the parts where the actual assembly differs significantly from the standard assembly. Finally, the assembly offset area is sent to the preset terminal for display, so that relevant personnel can intuitively see the areas where there are problems in the traction motor assembly, so as to conduct targeted inspections, adjustments or improvements.

[0138] Through the above steps, the assembly of the traction motor can be effectively detected and analyzed. By constructing a model diagram and calculating the overlap and intersection, the assembly offset area can be accurately located, and the results can be intuitively displayed on the preset terminal. This helps to improve the detection efficiency and accuracy of the traction motor assembly quality, timely discover problems in the assembly process, and facilitate adjustments and improvements, thereby ensuring the performance and reliability of the traction motor.

[0139] The above is based on the ideal embodiment of the present invention, and its description is relatively specific and detailed, but it cannot be understood as limiting the scope of the invention patent. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be based on the attached claims.

Claims

1. A method for detecting a traction motor for an electric tricycle, characterized in that: The following steps are involved: Constructing a stability coefficient index library of the traction motor, and performing a simulated working test on the traction motor, and collecting actual working parameter combinations of the traction motor at several preset time nodes based on the Internet of Things; Importing the actual working parameter combinations collected at several preset time nodes into the stability coefficient index library for index matching, and obtaining the stability coefficients corresponding to the traction motors at several preset time nodes; Constructing a stability coefficient dynamic change curve of the traction motor during a test period according to the stability coefficients corresponding to the traction motor at a number of preset time nodes; analyzing the overlap between the stability coefficient dynamic change curve and the preset curve; If the overlap is greater than the preset overlap, the traction motor is calibrated as a first-class product; if the overlap is not greater than the preset overlap, the traction motor is calibrated as a second-class product; wherein the first-class product is a qualified product, and the second-class product is an unqualified product; If the traction motor is calibrated as a Class II product, the actual working parameter combinations collected at several preset time nodes are subjected to feature classification processing to obtain various actual working parameters of the traction motor calibrated as a Class II product during the test period; Performing feature analysis on various actual working parameters of the traction motor calibrated as a Class II product during the test period, and obtaining abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period; Constructing a prediction model for the abnormal working state type of the traction motor, importing the abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period into the prediction model for the abnormal working state type to perform prediction, and obtaining the abnormal working state type of the traction motor calibrated as a Class II product; Construct the stability coefficient index library of the traction motor, specifically: Obtain the stability coefficient of the traction motor under various preset working parameter combinations through the big data network; Constructing a database space, counting the number of various preset working parameter combinations, and dividing the database space into a corresponding number of sub-libraries according to the number of various preset working parameter combinations; and etching the stability coefficients under various preset working parameter combination conditions on the metadata area of ​​the corresponding sub-library, and then storing the various preset working parameter combinations on the index structure area of ​​the corresponding sub-library to obtain a stability coefficient index library; The actual working parameter combinations collected at several preset time nodes are imported into the stability coefficient index library for index matching to obtain the stability coefficients corresponding to the traction motors at several preset time nodes, specifically: Importing the actual working parameter combinations collected at each preset time node into the stability coefficient index library in sequence according to the timestamp; Calculating the Jaccard similarity coefficient between the actual working parameter combination and the preset working parameter combination in each sub-library, and sorting the calculated Jaccard similarity coefficients to obtain the maximum Jaccard similarity coefficient; Marking a sub-library of a preset working parameter combination corresponding to the maximum Jaccard similarity coefficient, and extracting a corresponding stability coefficient in a metadata area of ​​the marked sub-library to obtain a stability coefficient corresponding to the traction motor at a corresponding preset time node; After all actual working parameter combinations collected at each preset time node are imported into the stability coefficient index library for index matching, the stability coefficient corresponding to the traction motor at each preset time node is obtained; The overlap between the dynamic change curve of the stability coefficient and the preset curve is analyzed, specifically: Acquiring working performance requirement information of the traction motor, and determining a standard stability coefficient corresponding to the traction motor at a plurality of preset time nodes according to the working performance requirement information; Draw a preset curve according to the standard stability coefficient corresponding to the traction motor at a number of preset time nodes; Introducing a Euclidean distance algorithm, and calculating a Euclidean distance value between the stability coefficient dynamic change curve and a preset curve based on the Euclidean distance algorithm; Determining the degree of overlap between the stability coefficient dynamic change curve and a preset curve according to the Euclidean distance value; The various actual working parameters of the traction motor calibrated as a Class II product during the test period are subjected to characteristic analysis and processing to obtain the abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period, specifically: Constructing a dynamic variation curve diagram of various actual working parameters of the traction motor calibrated as a second-class product during the test period according to various actual working parameters of the traction motor calibrated as a second-class product during the test period; Obtaining standard range values ​​of various actual working parameters of the traction motor during the test period, and determining upper and lower limits in the corresponding dynamic change curve diagrams according to the standard range values ​​of the various actual working parameters, thereby obtaining the limit areas of each dynamic change curve diagram; Analyze the total length of the line segments in the boundary area and the total length outside the boundary area in each dynamic change curve graph respectively, divide the total length of the line segments in the boundary area and the total length outside the boundary area in each dynamic change curve graph, and obtain the offset coefficient of the curve in each dynamic change curve graph; Compare the offset coefficient of each curve in the dynamic change curve diagram with the preset coefficient threshold value respectively; If the offset coefficient of a curve in a certain dynamic change curve diagram is greater than a preset coefficient threshold, the actual working parameter corresponding to the dynamic change curve diagram is calibrated as an abnormal actual working parameter; If the offset coefficient of a curve in a certain dynamic change curve diagram is not greater than the preset coefficient threshold, the actual working parameter corresponding to the dynamic change curve diagram is calibrated as a normal actual working parameter; Construct a prediction model for abnormal working state types of traction motors, specifically: Obtain various abnormal working state types of the traction motor through the big data network, and obtain abnormal characteristic parameters corresponding to various abnormal working state types, so as to obtain various abnormal working state types and their corresponding abnormal characteristic parameters; Based on the grey correlation analysis method, various abnormal working state types and their corresponding abnormal characteristic parameters are correlated and analyzed to obtain the correlation text between various abnormal working state types and their corresponding abnormal characteristic parameters; Introducing a hidden Markov random field, taking the abnormal working state type as a condition node, and taking the abnormal characteristic parameter as a variable node; According to the association text between various working state abnormality types and their corresponding abnormal characteristic parameters, the condition node is associated with the variable node to obtain a topological structure diagram between various working state abnormality types and their corresponding abnormal characteristic parameters; A prediction model is constructed based on a graph network, and the topological structure diagram is embedded in the prediction model to obtain a prediction model for the abnormal type of the working state of the traction motor.

2. A method for detecting a traction motor for an electric tricycle according to claim 1, characterized in that: The abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period are imported into the abnormal working state type prediction model for prediction, and the abnormal working state type of the traction motor calibrated as a Class II product is obtained, specifically: Acquire abnormal actual working parameters of the traction motor calibrated as a Class II product during a test period, and import the abnormal actual working parameters of the traction motor calibrated as a Class II product during the test period into the working state abnormality type prediction model; The abnormal working state type of the traction motor calibrated as a second-class product is obtained through prediction, and the abnormal working state type of the traction motor calibrated as a second-class product is sent to a preset terminal for display.

3. The method for detecting a traction motor for an electric tricycle according to claim 1, characterized in that: The traction motor for an electric tricycle comprises a housing and a base, wherein the housing is fixedly mounted on the base; A motor rotor and a motor stator are arranged inside the housing, the motor rotor and the motor stator are interlocked with each other, an output shaft is matched and connected at the axis of the motor rotor, and the bottom end of the output shaft extends to the outside of the base, a card interface is arranged on the output shaft extending to the outside of the base, and the output shaft is connected to the base through a rolling bearing; The upper and lower ends of the motor rotor are both provided with rotor end plates, a motor encoder and a controller power board are provided above the motor rotor, a capacitor is provided between the motor encoder and the controller power board, and a heat sink is provided below the capacitor.

4. The method for detecting a traction motor for an electric tricycle according to claim 3, characterized in that: A water pipe connection seat is provided on the side of the shell, and a water inlet and a water outlet are provided on the water pipe connection seat. A liquid cooling channel is also provided inside the shell, and the water inlet end of the liquid cooling channel is connected to the water inlet, and the water outlet end of the liquid cooling channel is connected to the water outlet.

5. The method for detecting a traction motor for an electric tricycle according to claim 3, characterized in that: The housing is sealed by a rubber sealing ring.

6. The method for detecting a traction motor for an electric tricycle according to claim 3, characterized in that: The shell is also provided with a power line and a control line.

Images