Motor rotating speed detection method, motor rotating speed detection system and motor device

By performing data augmentation and feature parameter contribution evaluation on the motor speed detection model, and combining optimization algorithms to adjust the motor speed, the problem of detection result deviation in the existing technology is solved, and accurate detection and optimization of motor speed are achieved.

CN118962438APending Publication Date: 2024-11-15杨梦雨
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411022609.9
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-29
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

Existing technologies consider too few factors in motor speed detection, leading to deviations in detection results and making it difficult to achieve accurate motor speed adjustment.

Method used

By acquiring sample data of the motor at different speeds, performing preprocessing and data augmentation, a motor speed detection model is established. The contribution of feature parameters is evaluated using a decision tree model, multiple speed detection sub-models are trained, and the motor speed is adjusted to achieve the optimal state through optimization algorithms.

Benefits of technology

It improves the accuracy and adaptability of motor speed detection, enabling more precise adjustment of motor speed and optimization of motor performance.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118962438A_ABST
    Figure CN118962438A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of motors, in particular to a motor rotating speed detection method and system and a motor device. Comprising the following steps of: firstly, calculating a change difference value between acquired data and standard motor rotating speed data in a preprocessing process, and evaluating contribution degree of each characteristic parameter to motor rotating speed change, thereby improving the accuracy of motor rotating speed detection; secondly, a motor rotating speed detection model is provided, samples are clustered in the training process of the model, and a plurality of rotating speed detection sub-models are obtained according to the clustering result; the method has an accurate detection effect on data sub-models with similar feature distribution; and finally, providing a motor rotating speed optimization method, adjusting the motor rotating speed by setting a motor rotating speed optimization target and constraint conditions, obtaining various performances according to a mathematical model between the motor rotating speed and the performances of the target automobile, and optimizing characteristic parameters about the motor rotating speed through an optimization algorithm to obtain the motor rotating speed. Therefore, the optimal motor adjustment rotating speed is obtained.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of motors, and in particular to a motor speed detection method, a detection system, and a motor device. Background Art

[0002] A motor is a device that converts electrical energy into mechanical energy and is widely used in a variety of industrial, commercial and household equipment. It operates mainly through the principle of electromagnetic induction and can be divided into DC motors and AC motors according to their working principles and structural characteristics.

[0003] The running state of the motor refers to the various performances and characteristics of the motor during the working process, including: current, temperature, vibration, speed and other parameters and indicators; the motor speed is one of the key parameters in the detection of the running state of the motor, and its stability and accuracy directly affect the performance and running rate of the equipment. The speed of the motor depends on factors such as power supply frequency, voltage, load and design specifications. If the speed does not meet certain requirements during the operation of the motor, it will lead to reduced efficiency, reduced output power, overload and overheating. Therefore, it is necessary to detect the speed of the motor during its operation so as to adjust the speed of the motor in time.

[0004] Electric motors are widely used in modern industry and modern life, especially in the new energy automobile industry, where they play a key role. Electric motors not only transmit driving force in modern automobiles, but also play a vital role in the performance, energy efficiency and environmental protection of automobiles. With the development of electric technology, the application of electric motors in the automotive industry will continue to expand and deepen. The motor speed not only directly affects the power output and energy efficiency of the vehicle, but also plays a key role in the vehicle's driving control, stability, comfort and overall energy efficiency. Therefore, in the design and operation of electric vehicles, the detection and optimization of motor speed is an important technical challenge and research field.

[0005] There are many methods in the prior art to study the speed of automobile motors; among them, including: such as the Chinese patent with the prior art application number CN202110569953.X discloses a motor speed redundancy detection method, circuit and motor controller, the detection method obtains the phase current signal of the motor, filters and performs zero-crossing comparison processing on the phase current signal, and determines the first speed of the motor; obtains the line voltage signal of the motor, filters and performs zero-crossing comparison processing on the line voltage signal, and determines the second speed of the motor; according to the first speed and the second speed, determines the actual speed of the motor; and also includes the Chinese patent with the prior art application number CN202010748726.9 discloses an automobile motor working speed detection device and a detection method thereof, the method can realize the vibration of the detection motor caused by the simulation of automobile operation, increase the accuracy of the speed test, and secondly can realize the speed test of the detection motor under two different test environments of load and no-load, so that the range of the test results is wider. The above-mentioned prior art realizes the speed detection of the motor, but there are few factors to consider about the motor speed during the detection process. The direct detection method is used, which may lead to deviation in the speed detection, and is not conducive to adjusting the speed of the automobile motor.

[0006] To this end, the present invention provides a motor speed detection method, a detection system, and a motor device. Summary of the invention

[0007] The present invention proposes a motor speed detection method, a detection system, and a motor device for detecting and adjusting the speed of a motor of an automobile; wherein, the content of the present invention mainly includes: first, the present invention obtains sample data of a motor of an automobile at different speeds, and pre-processes the obtained data; wherein, in order to make accurate detection of the speed of the motor subsequently, a data enhancement method is proposed in the pre-processing process, the method calculates the change difference between the obtained data and the standard motor speed data, evaluates the contribution of each characteristic parameter to the change of the motor speed according to the obtained change difference, and weights each characteristic parameter according to the contribution, so as to illustrate the different influence degrees in the motor speed detection; secondly, the present invention proposes a motor speed detection model, the model is clustered by the speed distribution characteristics of the motors in the training set; and multiple speed detection sub-models are trained according to the clustering results, and different sub-models have accurate detection effects on data with similar characteristic distributions; finally, the present invention proposes a motor speed optimization method for automobiles, the method adjusts the motor speed by setting the motor speed optimization target and the motor speed constraint condition, and optimizes the characteristic parameters about the motor speed by using an optimization algorithm through a mathematical model between the motor speed and the performance of the target automobile, so as to obtain the optimal motor adjustment speed.

[0008] In view of this, an object of the present invention is to provide a motor speed detection method, a detection system, and a motor device to overcome the problems in the prior art.

[0009] In a first aspect, an embodiment of the present application provides a method for detecting a motor speed, comprising:

[0010] Acquire data samples of the motor of the target vehicle at different speeds; wherein the data samples include: motor speed data, motor power data, current data, voltage data, motor vibration data and motor temperature data;

[0011] Further, the data samples are marked according to actual needs to construct motor data; wherein the rotation speed of the motor is used as an output target in the motor data;

[0012] Further, the motor data is preprocessed to obtain standard motor data;

[0013] wherein, the motor data is cleaned to obtain first motor data; wherein the data cleaning includes: missing value filling, outlier processing and noise filtering;

[0014] Normalizing the first motor data to obtain second motor data;

[0015] Performing data enhancement on the second motor data to obtain third motor data;

[0016] The data enhancement includes:

[0017] A second motor data matrix is ​​established according to the second motor data; wherein the second motor data matrix is ​​expressed as: Among them, F NM Represented as the Mth characteristic parameter in the Nth second motor data; v N It is represented by the motor speed value in the Nth second television data;

[0018] Further, the change difference of each data in the second motor data matrix is ​​calculated; wherein the change difference is obtained by subtracting the data in the second motor data matrix from the standard motor speed parameter vector; the standard motor speed parameter vector is recorded as: (s1, s2, ..., s M ,s v ), where s M It is represented as the Mth parameter value in the standard motor speed parameter vector; s v Represented as the motor speed value in the standard motor speed parameter vector;

[0019] Further, a decision tree model is used to evaluate the contribution of each characteristic parameter to the motor speed change difference according to the change difference;

[0020] Among them, the change difference is expressed as e NM The motor parameter change difference represented by the Mth characteristic parameter of the Nth data in the second motor data matrix and the Mth parameter value in the standard motor speed parameter vector; Nv The motor speed change difference represented by the motor speed value of the Nth data in the second motor data matrix and the motor speed value in the standard motor speed parameter vector;

[0021] Furthermore, the second motor data is weighted according to the contribution degree to obtain the third motor data.

[0022] Furthermore, the third motor data is processed in a distributed manner to obtain the standard motor data.

[0023] Further, the motor standard data is divided into a training set and a test set according to a certain ratio;

[0024] Furthermore, a motor speed detection model is built;

[0025] Further, the training set is input into the motor speed detection model until the model training stop condition is met, thereby obtaining an initial motor speed detection model;

[0026] The training process of the motor speed detection model includes:

[0027] Inputting the training set into a clustering layer to obtain a plurality of clusters; wherein the clustering layer performs clustering according to the motor speed distribution characteristics in the training set;

[0028] Further, each data sample is assigned to a corresponding cluster according to the clustering result;

[0029] Further, each of the clusters is input into a corresponding speed detection sub-model; wherein the speed detection sub-model is parameter updated according to the loss function;

[0030] The loss function is expressed as: Wherein, ω represents the weight vector of the speed detection sub-model; b represents the bias term of the speed detection sub-model; X represents the training set; C represents the penalty parameter of the speed detection sub-model; x i represents the i-th training data in the cluster corresponding to the speed detection sub-model, and x i ∈X;

[0031] The constraints set for the loss function are: Among them, y irepresents the motor speed corresponding to the i-th training data; σ represents the output variable of the speed detection sub-model; ξ i Represented as the boundary width corresponding to the i-th training data;

[0032] Further, iterating the training process of the motor speed detection model, and determining whether the model training stop condition is met;

[0033] When the model training stop condition is met, the model training is completed to obtain the initial motor speed detection model; when the model training stop condition is not met, the model training continues; wherein the model training stop condition is set to the loss of the motor speed detection model reaching convergence.

[0034] Further, the test set is input into the initial motor speed detection model for testing, and the initial motor speed detection model is optimized according to the test result to obtain an optimal motor speed detection model;

[0035] Further, the motor speed of the target vehicle is detected using the optimal motor speed detection model to obtain a motor speed detection result;

[0036] Further, the speed is optimized according to the motor speed detection result to obtain the optimal speed of the motor of the target vehicle;

[0037] The process of speed optimization includes:

[0038] Setting a motor speed optimization target and motor speed constraint conditions for the target vehicle;

[0039] Wherein, the motor speed optimization target is expressed as: The motor speed constraint condition is expressed as:

[0040] Further, a mathematical model is used to obtain corresponding power output, power consumption and operating status data according to the motor of the target vehicle at different speeds;

[0041] Further, an optimization algorithm is used to optimize the characteristic parameters of the motor, and the motor speed detection model is used to perform speed detection;

[0042] Furthermore, when the motor speed optimization target is reached, the optimal motor speed is obtained.

[0043] In a second aspect, the present invention provides a motor speed detection system, the system comprising: a sensor and data acquisition unit, a motor data receiving unit, a data processing unit, a motor speed detection unit and a motor speed adjustment unit;

[0044] The sensor and data acquisition unit is used for obtaining various data of the target vehicle through current sensor, voltage sensor, vibration sensor and temperature sensor;

[0045] The motor data receiving unit is used to receive the collected data;

[0046] The data processing unit is used to pre-process the received data; wherein the data processing unit includes:

[0047] Performing data cleaning on the motor data to obtain first motor data; wherein the data cleaning includes: missing value filling, outlier value processing and noise filtering;

[0048] Normalizing the first motor data to obtain second motor data;

[0049] Performing data enhancement on the second motor data to obtain third motor data; wherein the data enhancement includes:

[0050] A second motor data matrix is ​​established according to the second motor data; wherein the second motor data matrix is ​​expressed as: Among them, F NM Represented as the Mth characteristic parameter in the Nth second motor data; v N It is represented by the motor speed value in the Nth second television data;

[0051] Further, the change difference of each data in the second motor data matrix is ​​calculated; wherein the change difference is obtained by subtracting the data in the second motor data matrix from the standard motor speed parameter vector; the standard motor speed parameter vector is recorded as: (s1, s2, ..., s M ,s v ), where s M It is represented as the Mth parameter value in the standard motor speed parameter vector; s v Represented as the motor speed value in the standard motor speed parameter vector;

[0052] Further, a decision tree model is used to evaluate the contribution of each characteristic parameter to the motor speed change difference according to the change difference;

[0053] Among them, the change difference is expressed as e NM The motor parameter change difference represented by the Mth characteristic parameter of the Nth data in the second motor data matrix and the Mth parameter value in the standard motor speed parameter vector; Nv The motor speed change difference represented by the motor speed value of the Nth data in the second motor data matrix and the motor speed value in the standard motor speed parameter vector;

[0054] Furthermore, the second motor data is weighted according to the contribution degree to obtain the third motor data.

[0055] Furthermore, the third motor data is processed in a distributed manner to obtain standard motor data.

[0056] The motor speed detection unit is used to detect the motor speed of the target vehicle; the motor speed detection unit adopts a motor speed detection model;

[0057] The training process of the motor speed detection model includes:

[0058] Inputting the training set into the clustering layer to obtain a plurality of clusters; wherein the clustering layer performs clustering according to the motor speed distribution characteristics in the training set;

[0059] Further, each data sample is assigned to a corresponding cluster according to the clustering result;

[0060] Further, each of the clusters is input into a corresponding speed detection sub-model; wherein the speed detection sub-model is parameter updated according to the loss function;

[0061] The loss function is expressed as: Wherein, ω represents the weight vector of the speed detection sub-model; b represents the bias term of the speed detection sub-model; X represents the training set; C represents the penalty parameter of the speed detection sub-model; x i represents the i-th training data in the cluster corresponding to the speed detection sub-model, and x i ∈X;

[0062] The constraints set for the loss function are: Among them, y i represents the motor speed corresponding to the i-th training data; σ represents the output variable of the speed detection sub-model; ξ i Represented as the boundary width corresponding to the i-th training data;

[0063] Further, iterating the training process of the motor speed detection model, and determining whether the model training stop condition is met;

[0064] Further, when the model training stop condition is met, the model training is terminated to obtain an initial motor speed detection model; when the model training stop condition is not met, the model training continues; wherein the model training stop condition is set to the loss of the motor speed detection model reaching convergence;

[0065] The motor speed adjustment unit is used to optimally adjust the motor speed.

[0066] In a third aspect, the present invention provides a motor device; wherein the device comprises: a sensing module, an acquisition module, a data processing module, a speed detection module and a speed adjustment module; and the device can be implemented by any step of a motor speed detection method; wherein the speed detection module comprises:

[0067] Obtain the current motor speed related characteristic parameters of the target vehicle;

[0068] Further, the characteristic parameters related to the current motor speed are input into the clustering layer to obtain a clustering result;

[0069] Further, a corresponding speed detection sub-model is selected according to the clustering result; wherein the speed detection sub-model uses a loss function to update parameters;

[0070] The loss function is expressed as: Wherein, ω represents the weight vector of the speed detection sub-model; b represents the bias term of the speed detection sub-model; X represents the training set; C represents the penalty parameter of the speed detection sub-model; x i represents the i-th training data in the cluster corresponding to the speed detection sub-model, and x i ∈X;

[0071] The constraints set for the loss function are: Among them, y i represents the motor speed corresponding to the i-th training data; σ represents the output variable of the speed detection sub-model; ξ i Represented as the boundary width corresponding to the i-th training data;

[0072] Furthermore, the motor speed of the target vehicle is output according to the speed detection sub-model.

[0073] Furthermore, the speed adjustment is performed by a speed adjustment module of the motor device; wherein the speed adjustment module comprises:

[0074] Setting a motor speed optimization target and motor speed constraint conditions for the target vehicle;

[0075] Wherein, the motor speed optimization target is expressed as: The motor speed constraint condition is expressed as:

[0076] Further, a mathematical model is used to obtain corresponding power output, power consumption and operating status data according to the motor of the target vehicle at different speeds;

[0077] Further, an optimization algorithm is used to optimize the characteristic parameters of the motor, and the motor speed detection model is used to perform speed detection;

[0078] Furthermore, when the motor speed optimization target is reached, the optimal motor speed is obtained.

[0079] Compared with the prior art, the present invention has the following beneficial effects:

[0080] 1. The present invention proposes a data enhancement method, which calculates the change difference between the acquired data and the standard motor speed data, evaluates the contribution of each characteristic parameter to the change of the motor speed based on the obtained change difference, and weights each characteristic parameter by the contribution to illustrate the different degrees of influence in the motor speed detection; this method can improve the accuracy of subsequent motor speed detection, which is conducive to optimizing the adjustment of the automobile motor speed.

[0081] 2. The present invention proposes a motor speed detection model, which clusters the training set during the training process and trains multiple corresponding speed detection sub-models according to the clustering results; during the training process, the parameters of the model are trained by minimizing the loss function, and restrictions on the training samples are set for the loss function; in this way, the training process of the motor speed detection model can be effectively optimized to make it more adaptable to actual application needs; thereby improving the performance of the motor speed detection model and providing accurate detection results for subsequent motor speed adjustment.

[0082] 3. The present invention proposes a method for adjusting the speed of an automobile motor. The method adjusts the motor speed by setting a motor speed optimization target and a motor speed constraint condition, obtains various performance values ​​through a mathematical model between the motor speed and the performance of the target automobile, and uses an optimization algorithm to optimize characteristic parameters related to the motor speed to obtain the optimal motor adjustment speed. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 A flow chart of a motor speed detection method provided by an embodiment of the present invention;

[0084] Figure 2 A structural diagram of a motor speed detection system provided by an embodiment of the present invention;

[0085] Figure 3 A structural diagram of a motor device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0086] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0087] As a device that converts electrical energy into mechanical energy, motors are widely used in modern industry and daily life, especially in the new energy vehicle industry.

[0088] During the operation of the motor, there are various performances and characteristics, including: current, temperature, vibration, speed and other parameters and indicators; in the detection of the motor's operating status, the motor speed is one of the key parameters, and its stability and accuracy directly affect the performance and operating rate of the equipment. The speed of the motor depends on factors such as power supply frequency, voltage, load and design specifications. If the speed does not meet certain requirements during the operation of the motor, it will lead to reduced efficiency, reduced output power, overload and overheating. Therefore, it is necessary to detect the speed of the motor during its operation so as to adjust the speed of the motor in time.

[0089] There are many studies on motor speed detection in the prior art, but there are some defects therein. In order to solve the defects in the prior art, the present invention proposes a motor speed detection method, a detection system, and a motor device. The following will describe in detail two embodiments.

[0090] Embodiment 1

[0091] In the embodiment of the present application, the method proposed in the present invention is combined with the system, and the method corresponds to each unit of the system.

[0092] First, the present invention proposes a motor speed detection method, such as Figure 1 As shown; the method includes the following steps: S10. Obtain data samples of the motor of the target vehicle at different speeds; S20. Mark the data samples according to actual needs to construct motor data; S30. Preprocess the motor data to obtain standard motor data; S40. Divide the motor standard data into training set and test set according to a certain ratio; S50. Build a motor speed detection model; S60. Train the motor speed detection model; S70. Test the motor speed detection model; S80. Detect the motor speed of the target vehicle; S90. Adjust the motor speed of the target vehicle. In the process of detecting the speed of the automobile motor, combine Figure 2 The system unit is implemented in Figure 2As shown, the system includes: a sensor and data acquisition unit, a motor data receiving unit, a data processing unit, a motor speed detection unit and a motor speed adjustment unit;

[0093] Furthermore, combined with Figure 1 and Figure 2 The specific process is as follows:

[0094] The sensor and data acquisition unit of the system obtain data samples of the motor of the target vehicle at different speeds, corresponding to Figure 1 Step S10; wherein the sensor and data acquisition unit are respectively arranged with a current sensor, a voltage sensor, a vibration sensor and a temperature sensor, for obtaining motor power data, current data, voltage data, motor vibration data and motor temperature data;

[0095] Further, the data acquired by the sensor and the data acquisition unit are received by the motor data receiving unit of the system;

[0096] Further, the acquired data samples are marked according to the above step S20; wherein the rotation speed of the motor is used as a mark and also as an output target;

[0097] Furthermore, the data processing unit of the system pre-processes the data samples to obtain standard motor data, corresponding to Figure 1 Step S30; wherein the pre-processing mentioned in the embodiment of the present application includes:

[0098] Performing data cleaning on the motor data to obtain first motor data; wherein the data cleaning includes: missing value filling, outlier processing and noise filtering;

[0099] Normalizing the first motor data to obtain second motor data;

[0100] Performing data enhancement on the second motor data to obtain third motor data;

[0101] The data enhancement includes:

[0102] A second motor data matrix is ​​established according to the second motor data; wherein the second motor data matrix is ​​expressed as: Among them, F NM Represented as the Mth characteristic parameter in the Nth second motor data; v N It is represented by the motor speed value in the Nth second television data;

[0103] Further, the change difference of each data in the second motor data matrix is ​​calculated; wherein the change difference is obtained by subtracting the data in the second motor data matrix from the standard motor speed parameter vector; the standard motor speed parameter vector is recorded as: (s1, s2, ..., s M ,s v ), where s M It is represented as the Mth parameter value in the standard motor speed parameter vector; s v Represented as the motor speed value in the standard motor speed parameter vector;

[0104] Further, a decision tree model is used to evaluate the contribution of each characteristic parameter to the motor speed change difference according to the change difference;

[0105] Among them, the change difference is expressed as e NM The motor parameter change difference represented by the Mth characteristic parameter of the Nth data in the second motor data matrix and the Mth parameter value in the standard motor speed parameter vector; Nv The motor speed change difference represented by the motor speed value of the Nth data in the second motor data matrix and the motor speed value in the standard motor speed parameter vector;

[0106] In the application implementation, the decision tree model is used to evaluate the contribution scores of each feature parameter; see Table 1:

[0107] Table 1. Contribution of motor speed characteristic parameters

[0108] Characteristic parameters Contribution Motor power 0.30 Current 0.25 Voltage 0.20 Motor vibration 0.15 Motor temperature 0.10

[0109] From Table 1, it can be concluded that the contribution of motor power is 0.30, the contribution of current is 0.25, the contribution of voltage is 0.20, the contribution of motor vibration is 0.15 and the contribution of motor temperature is 0.10.

[0110] Further, the second motor data is weighted according to the contribution to obtain the third motor data;

[0111] In the embodiment of the present application, a data enhancement method is proposed in the data preprocessing process. The method calculates the change difference between the acquired data and the standard motor speed data, evaluates the contribution of each characteristic parameter to the motor speed change based on the obtained change difference, and weights each characteristic parameter by the contribution to illustrate the different degrees of influence in the motor speed detection; this method can improve the accuracy of subsequent motor speed detection, which is conducive to optimizing the adjustment of the automobile motor speed.

[0112] Furthermore, the third motor data is processed in a distributed manner to obtain the standard motor data.

[0113] In the embodiment of the present application, a series of data processing methods are used to pre-process the sample data related to the motor speed, including: data cleaning, normalization, data enhancement and distribution processing; among them, data cleaning can help remove noise, outliers and inconsistencies in the data, thereby improving the quality and accuracy of the data; in motor speed detection, accurate input data is the basis for ensuring the effective operation of models and algorithms; by cleaning the data, errors and deviations can be reduced, so that subsequent motor speed detection models can be trained and predicted based on more reliable and consistent data; normalization processing improves the consistency and stability of the data, accelerates the training process of the model, improves the feature distribution and data analysis effect, and enhances the generalization ability of the model; data enhancement can effectively improve the utilization efficiency of data without increasing the cost of data acquisition, and can use a limited original data set to train a more robust and accurate motor speed detection model; distribution processing can adjust or optimize the data distribution, which may help to further improve the effect of data analysis or the performance of the model.

[0114] Furthermore, the motor standard data is divided into a training set and a test set according to a certain ratio, corresponding to Figure 1 In step S40, in the embodiment of the present application, the data set is divided into a training set and a test set in a ratio of 8:2;

[0115] Among them, the main implementation process of the motor speed detection model includes:

[0116] Build a motor speed detection model, corresponding to step S50;

[0117] Further, in the above step S60, the training set is input into the motor speed detection model until the model training stop condition is met, thereby obtaining an initial motor speed detection model;

[0118] The training process of the motor speed detection model includes:

[0119] Inputting the training set into the clustering layer to obtain a plurality of clusters; wherein the clustering layer performs clustering according to the motor speed distribution characteristics in the training set;

[0120] Further, each data sample is assigned to a corresponding cluster according to the clustering result;

[0121] Further, each of the clusters is input into a corresponding speed detection sub-model; wherein the speed detection sub-model is parameter updated according to the loss function; the speed detection sub-model can use a multi-layer perceptron as a basic model, which can effectively learn the nonlinear relationship between input features;

[0122] The loss function is expressed as: Wherein, ω represents the weight vector of the speed detection sub-model; b represents the bias term of the speed detection sub-model; X represents the training set; C represents the penalty parameter of the speed detection sub-model; x i represents the i-th training data in the cluster corresponding to the speed detection sub-model, and x i ∈X;

[0123] The constraints set for the loss function are: Among them, y i represents the motor speed corresponding to the i-th training data; σ represents the output variable of the speed detection sub-model; ξ i Represented as the boundary width corresponding to the i-th training data;

[0124] Further, iterating the training process of the motor speed detection model, and determining whether the model training stop condition is met;

[0125] Further, when the model training stop condition is met, the model training is terminated to obtain an initial motor speed detection model; when the model training stop condition is not met, the model training continues; wherein the model training stop condition is set to the loss of the motor speed detection model reaching convergence;

[0126] In an embodiment of the present application, a motor speed detection model is proposed. During the training process, the model clusters the training set and trains multiple corresponding speed detection sub-models according to the clustering results; during the training process, the parameters of the model are trained by minimizing the loss function, and restrictions on the training samples are set for the loss function; this can effectively optimize the training process of the motor speed detection model and make it more adaptable to actual application needs; thereby improving the performance of the motor speed detection model.

[0127] Further, according to step S70, the test set is input into the initial motor speed detection model for testing, and the initial motor speed detection model is optimized according to the test result to obtain an optimal motor speed detection model;

[0128] Further, the motor speed detection unit of the system performs motor speed detection, corresponding to the above step S80; wherein the motor speed detection unit uses a motor speed detection model for detection;

[0129] Furthermore, the motor speed adjustment unit of the system adjusts the motor speed of the target vehicle, corresponding to the above step S90; the specific adjustment process includes:

[0130] Setting a motor speed optimization target and motor speed constraint conditions for the target vehicle;

[0131] Wherein, the motor speed optimization target is expressed as: The motor speed constraint condition is expressed as:

[0132] Further, a mathematical model is used to obtain corresponding power output, power consumption and operating status data according to the motor of the target vehicle at different speeds;

[0133] The mathematical model is expressed as: motor =T(θ)*θ; where P motor It is expressed as output performance; T(θ) is expressed as a function of speed; θ is expressed as speed;

[0134] The speed can be expressed as: Wherein, MSDM() represents the speed detection model; Represented as motor speed characteristic parameter vector;

[0135] Furthermore, an optimization algorithm is used to optimize the motor speed characteristic parameters, and the motor speed detection model is used to perform speed detection; wherein the optimization algorithm can use an ant colony algorithm to update the motor speed characteristic parameters, and obtain the optimized motor speed based on the updated parameters.

[0136] Furthermore, when the motor speed optimization target is reached, the optimal motor speed is obtained.

[0137] In an embodiment of the present application, a method for adjusting the speed of an automobile motor is adopted. The method adjusts the motor speed by setting a motor speed optimization target and a motor speed constraint condition, obtains various performance values ​​through a mathematical model between the motor speed and the performance of the target automobile, and uses an optimization algorithm to optimize the characteristic parameters of the motor speed to obtain the optimal motor adjustment speed.

[0138] In the first embodiment, the system and method of the present invention are combined to realize the detection and adjustment of the motor speed of the target vehicle; the present invention mainly proceeds from the following aspects: first, the present invention obtains sample data of the motor of the vehicle at different speeds, and pre-processes the obtained data; wherein, in order to make accurate detection of the motor speed in the future, a data enhancement method is proposed in the pre-processing process, the method calculates the change difference between the obtained data and the standard motor speed data, and evaluates the contribution of each characteristic parameter to the change of the motor speed according to the obtained change difference, and weights each characteristic parameter by the contribution, so as to illustrate the different influence degrees in the motor speed detection; secondly, the present invention proposes a motor speed detection model, the model is clustered by the speed distribution characteristics of the motor in the training set; and multiple speed detection sub-models are trained according to the clustering results, and different sub-models have accurate detection effects on data with similar feature distributions; finally, the present invention proposes a motor speed optimization method for automobiles, the method adjusts the motor speed by setting the motor speed optimization target and the motor speed constraint condition, and obtains various performance values ​​through the mathematical model between the motor speed and the performance of the target vehicle, and optimizes the characteristic parameters of the motor speed by using the optimization algorithm to obtain the optimal motor adjustment speed.

[0139] Embodiment 2

[0140] In the first embodiment, the motor speed detection method is combined with the detection system to realize the speed detection and subsequent speed adjustment of the automobile motor; in the embodiment of the present application, a motor speed detection method will be described again, and the specific process is the same as the implementation process of the first embodiment, including the following:

[0141] Obtain data samples of the motor of the target vehicle at different speeds;

[0142] Further, the data samples are labeled according to actual needs to construct motor data;

[0143] Further, the motor data is preprocessed to obtain standard motor data;

[0144] Among them, the preprocessing process includes data enhancement operations; the specific data enhancement process is as follows:

[0145] A second motor data matrix is ​​established according to the second motor data; wherein the second motor data matrix is ​​expressed as: Among them, F NM Represented as the Mth characteristic parameter in the Nth second motor data; v N It is represented by the motor speed value in the Nth second television data;

[0146] Further, the change difference of each data in the second motor data matrix is ​​calculated; wherein the change difference is obtained by subtracting the data in the second motor data matrix from the standard motor speed parameter vector; the standard motor speed parameter vector is recorded as: (s1, s2, ..., s M ,s v ), where s M It is represented as the Mth parameter value in the standard motor speed parameter vector; s v Represented as the motor speed value in the standard motor speed parameter vector;

[0147] Further, a decision tree model is used to evaluate the contribution of each characteristic parameter to the motor speed change difference according to the change difference;

[0148] Among them, the change difference is expressed as e NM The motor parameter change difference represented by the Mth characteristic parameter of the Nth data in the second motor data matrix and the Mth parameter value in the standard motor speed parameter vector; Nv The motor speed change difference represented by the motor speed value of the Nth data in the second motor data matrix and the motor speed value in the standard motor speed parameter vector;

[0149] Furthermore, the second motor data is weighted according to the contribution degree to obtain the third motor data.

[0150] Further, the motor standard data is divided into a training set and a test set according to a certain ratio;

[0151] Furthermore, a motor speed detection model is built;

[0152] Further, the training set is input into the motor speed detection model until the model training stop condition is met, thereby obtaining an initial motor speed detection model;

[0153] Further, the test set is input into the initial motor speed detection model for testing, and the initial motor speed detection model is optimized according to the test result to obtain an optimal motor speed detection model;

[0154] Further, the motor speed of the target vehicle is detected using the optimal motor speed detection model to obtain a motor speed detection result;

[0155] Further, the speed is optimized according to the motor speed detection result to obtain the optimal speed of the motor of the target vehicle;

[0156] The process of speed optimization includes:

[0157] Setting a motor speed optimization target and motor speed constraint conditions for the target vehicle;

[0158] Wherein, the motor speed optimization target is expressed as: The motor speed constraint condition is expressed as:

[0159] Further, a mathematical model is used to obtain corresponding power output, power consumption and operating status data according to the motor of the target vehicle at different speeds;

[0160] Further, an optimization algorithm is used to optimize the characteristic parameters of the motor, and the motor speed detection model is used to perform speed detection;

[0161] Furthermore, when the motor speed optimization target is reached, the optimal motor speed is obtained.

[0162] For the detection method in the embodiments of the present application, see Figure 3 , the embodiment of the present application further proposes a motor device, the motor device comprising: a sensing module, a collection module, a data processing module, a speed detection module and a speed adjustment module; wherein the motor device can be implemented by the motor speed detection method proposed above;

[0163] Wherein, the speed detection module includes:

[0164] Obtain the current motor speed related characteristic parameters of the target vehicle;

[0165] Further, the characteristic parameters related to the current motor speed are input into the clustering layer to obtain a clustering result;

[0166] Further, a corresponding speed detection sub-model is selected according to the clustering result; wherein the speed detection sub-model uses a loss function to update parameters;

[0167] The loss function is expressed as: Wherein, ω represents the weight vector of the speed detection sub-model; b represents the bias term of the speed detection sub-model; X represents the training set; C represents the penalty parameter of the speed detection sub-model; x i represents the i-th training data in the cluster corresponding to the speed detection sub-model, and x i ∈X;

[0168] The constraints set for the loss function are: Among them, y i represents the motor speed corresponding to the i-th training data; σ represents the output variable of the speed detection sub-model; ξ i Represented as the boundary width corresponding to the i-th training data;

[0169] Furthermore, the motor speed of the target vehicle is output according to the speed detection sub-model.

[0170] Furthermore, the speed adjustment is performed by a speed adjustment module of the motor device; wherein the speed adjustment module comprises:

[0171] Setting a motor speed optimization target and motor speed constraint conditions for the target vehicle;

[0172] Wherein, the motor speed optimization target is expressed as: The motor speed constraint condition is expressed as:

[0173] Further, a mathematical model is used to obtain corresponding power output, power consumption and operating status data according to the motor of the target vehicle at different speeds;

[0174] Further, an optimization algorithm is used to optimize the characteristic parameters of the motor, and the motor speed detection model is used to perform speed detection;

[0175] Furthermore, when the motor speed optimization target is reached, the optimal motor speed is obtained.

[0176] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for detecting motor speed, It is characterized in that include: Obtain data samples of the motor of the target vehicle at different speeds; Marking the data samples according to actual needs to construct motor data; Preprocessing the motor data to obtain standard motor data; dividing the motor standard data into a training set and a test set according to a certain ratio; A motor speed detection model is constructed; the training set is input into the motor speed detection model until the model training stop condition is met to obtain an initial motor speed detection model; the test set is input into the initial motor speed detection model for testing, and the initial motor speed detection model is optimized according to the test result to obtain an optimal motor speed detection model; the motor speed of the target vehicle is detected using the optimal motor speed detection model to obtain a motor speed detection result; the speed is optimized according to the motor speed detection result to obtain the optimal motor speed of the target vehicle; wherein the speed optimization process includes: Setting a motor speed optimization target and motor speed constraint conditions for the target vehicle; Wherein, the motor speed optimization target is expressed as: The motor speed constraint condition is expressed as: Using a mathematical model to obtain corresponding power output, power consumption and operating status data according to the motor of the target vehicle at different speeds; The motor characteristic parameters are optimized by using an optimization algorithm, and the motor speed detection model is used to perform speed detection; When the motor speed optimization target is reached, the optimal motor speed is obtained.

2. A motor speed detection method according to claim 1, It is characterized in that The pre-processing comprises: Performing data cleaning on the motor data to obtain first motor data; wherein the data cleaning includes: missing value filling, outlier processing and noise filtering; Normalizing the first motor data to obtain second motor data; Performing data enhancement on the second motor data to obtain third motor data; The third motor data is processed in a distributed manner to obtain the standard motor data.

3. A motor speed detection method according to claim 2, It is characterized in that The data enhancement includes: establishing a second motor data matrix according to the second motor data; wherein the second motor data matrix is ​​expressed as: Among them, F NM Represented as the Mth characteristic parameter in the Nth second motor data; v N It is represented by the motor speed value in the Nth second television data; Calculate the change difference of each data in the second motor data matrix; wherein the change difference is obtained by subtracting the data in the second motor data matrix from the standard motor speed parameter vector; the standard motor speed parameter vector is recorded as: (s 1 ,s 2 ,…,s M ,s v ), where s M It is represented as the Mth parameter value in the standard motor speed parameter vector; s v Represented as the motor speed value in the standard motor speed parameter vector; According to the change difference, a decision tree model is used to evaluate the contribution of each characteristic parameter to the motor speed change difference; Among them, the change difference is expressed as e NM The motor parameter change difference represented by the Mth characteristic parameter of the Nth data in the second motor data matrix and the Mth parameter value in the standard motor speed parameter vector; Nv The motor speed change difference represented by the motor speed value of the Nth data in the second motor data matrix and the motor speed value in the standard motor speed parameter vector; The second motor data is weighted according to the contribution degree to obtain the third motor data.

4. A motor speed detection method according to claim 1, It is characterized in that The training process of the motor speed detection model includes: Inputting the training set into a clustering layer to obtain a plurality of clusters; wherein the clustering layer performs clustering according to the motor speed distribution characteristics in the training set; Allocating each data sample to the corresponding cluster according to the clustering result; Input each of the clusters into a corresponding speed detection sub-model; wherein the speed detection sub-model is parameter updated according to the loss function; The loss function is expressed as: Wherein, ω represents the weight vector of the speed detection sub-model; b represents the bias term of the speed detection sub-model; X represents the training set; C represents the penalty parameter of the speed detection sub-model; x i represents the i-th training data in the cluster corresponding to the speed detection sub-model, and x i ∈X; The constraints set for the loss function are: Among them, y i represents the motor speed corresponding to the i-th training data; σ represents the output variable of the speed detection sub-model; ξ i Represented as the boundary width corresponding to the i-th training data; Iterate the training process of the motor speed detection model, and determine whether the model training stop condition is met; When the model training stop condition is met, the model training is completed to obtain the initial motor speed detection model; when the model training stop condition is not met, the model training continues; wherein the model training stop condition is set to the loss of the motor speed detection model reaching convergence.

5. A motor speed detection system, It is characterized in that The system comprises: a sensor and data acquisition unit, a motor data receiving unit, a data processing unit, a motor speed detection unit and a motor speed adjustment unit; Among them, the sensor and data acquisition unit is used for current sensor, voltage sensor, vibration sensor and temperature sensor to obtain various data of the target vehicle; the motor data receiving unit is used to receive the collected data; the data processing unit is used to pre-process the received data; the motor speed detection unit is used to detect the motor speed of the target vehicle; the motor speed adjustment unit is used to optimally adjust the motor speed.

6. A motor speed detection system according to claim 5, It is characterized in that The data processing unit comprises: Performing data cleaning on the motor data to obtain first motor data; wherein the data cleaning includes: missing value filling, outlier value processing and noise filtering; Normalizing the first motor data to obtain second motor data; Performing data enhancement on the second motor data to obtain third motor data; The third motor data is processed in a distributed manner to obtain standard motor data.

7. A motor speed detection system according to claim 6, It is characterized in that The data enhancement includes: A second motor data matrix is ​​established according to the second motor data; wherein the second motor data matrix is ​​expressed as: Among them, F NM Represented as the Mth characteristic parameter in the Nth second motor data; v N It is represented by the motor speed value in the Nth second television data; Calculate the change difference of each data in the second motor data matrix; wherein the change difference is obtained by subtracting the data in the second motor data matrix from the standard motor speed parameter vector; the standard motor speed parameter vector is recorded as: (s 1 ,s 2 ,…,s M ,s v ), where s M It is represented as the Mth parameter value in the standard motor speed parameter vector; s v Represented as the motor speed value in the standard motor speed parameter vector; According to the change difference, a decision tree model is used to evaluate the contribution of each characteristic parameter to the motor speed change difference; Among them, the change difference is expressed as e NM The motor parameter change difference represented by the Mth characteristic parameter of the Nth data in the second motor data matrix and the Mth parameter value in the standard motor speed parameter vector; Nv The motor speed change difference represented by the motor speed value of the Nth data in the second motor data matrix and the motor speed value in the standard motor speed parameter vector; The second motor data is weighted according to the contribution degree to obtain the third motor data.

8. A motor speed detection system according to claim 5, It is characterized in that The motor speed detection unit adopts a motor speed detection model; wherein the training process of the motor speed detection model includes: Inputting the training set into the clustering layer to obtain a plurality of clusters; wherein the clustering layer performs clustering according to the motor speed distribution characteristics in the training set; Allocating each data sample to the corresponding cluster according to the clustering result; Input each of the clusters into a corresponding speed detection sub-model; wherein the speed detection sub-model is parameter updated according to the loss function; The loss function is expressed as: Wherein, ω represents the weight vector of the speed detection sub-model; b represents the bias term of the speed detection sub-model; X represents the training set; C represents the penalty parameter of the speed detection sub-model; x i represents the i-th training data in the cluster corresponding to the speed detection sub-model, and x i ∈X; The constraints set for the loss function are: Among them, y i represents the motor speed corresponding to the i-th training data; σ represents the relaxation variable of the speed detection sub-model; ξ i Represented as the boundary width corresponding to the i-th training data; Iterate the training process of the motor speed detection model and determine whether the model training stop condition is met; When the model training stop condition is met, the model training is completed to obtain the initial motor speed detection model; when the model training stop condition is not met, the model training continues; wherein the model training stop condition is set to the loss of the motor speed detection model reaching convergence.

9. A motor device, It is characterized in that include: Sensing module, acquisition module, data processing module, speed detection module and speed adjustment module; The speed detection module includes: obtaining characteristic parameters related to the current motor speed of the target vehicle; inputting the characteristic parameters related to the current motor speed into the clustering layer to obtain clustering results; selecting a corresponding speed detection sub-model according to the clustering results; wherein the speed detection sub-model uses a loss function to update parameters; The loss function is expressed as: Wherein, ω represents the weight vector of the speed detection sub-model; b represents the bias term of the speed detection sub-model; X represents the training set; C represents the penalty parameter of the speed detection sub-model; x i represents the i-th training data in the cluster corresponding to the speed detection sub-model, and x i ∈X; The constraints set for the loss function are: Among them, y i represents the motor speed corresponding to the i-th training data; σ represents the output variable of the speed detection sub-model; ξ i Represented as the boundary width corresponding to the i-th training data; The motor speed of the target vehicle is output according to the speed detection sub-model.

10. The motor device according to claim 9, It is characterized in that The speed adjustment module comprises: Setting a motor speed optimization target and motor speed constraint conditions for the target vehicle; Wherein, the motor speed optimization target is expressed as: The motor speed constraint condition is expressed as: Using a mathematical model to obtain corresponding power output, power consumption and operating status data according to the motor of the target vehicle at different speeds; The motor characteristic parameters are optimized by using an optimization algorithm, and the motor speed detection model is used to perform speed detection; When the motor speed optimization target is reached, the optimal motor speed is obtained.

Citation Information

Cited By

  • Low-power-consumption rotating speed testing device, electronic equipment and storage medium

    CN120102921A