A life detection method, storage medium and execution device for a fixture for a motor housing

Abstract

The present invention proposes a life detection method, storage medium and execution device for a fixture for a motor housing. The method integrates the continuous expansion force characteristic value and response time parameter collected by the force sensor, and accurately extracts key features to evaluate the life of the fixture through SVM model classification prediction and high-dimensional feature space technology. The PSO algorithm is used to optimize the expansion force configuration, combined with the preset expansion force threshold and abnormal parameter identification, to achieve high-precision prediction of the fixture life. At the same time, the present invention constructs a regularized elastic net fixture SOH model, and uses a continuous scaling prediction method to generate a predictive relationship between the life parameter model and the expansion force configuration, further improving the accuracy of the prediction. The method improves the detection efficiency and automation level through automated data collection and processing, and provides scientific decision-making support for the maintenance and management of the fixture. The present invention not only enhances the reliability and safety of the fixture, but also reduces production costs and risks. It has broad application prospects and important practical value.

Description

Technical Field

[0001] The present invention relates to the technical field of sunroof motor production equipment, and in particular to a life detection method, storage medium and execution equipment for a jig for a motor housing. Background Art

[0002] Motor housing jigs play a vital role in industrial production, and their lifespan is directly related to production efficiency and product quality. However, traditional jig life detection methods mostly rely on empirical judgment or regular replacement, which lacks scientificity and accuracy. Therefore, how to develop an efficient and accurate jig life detection method has become an urgent problem to be solved.

[0003] like Figure 3-Figure 4 As shown, the detection device mainly includes a fixture body, a pressure sensor, a display, a highly integrated control processor and a driving member arranged at the bottom of the fixture body. The fixture body includes a contact arm and a movable body. In actual use, the driving member supports the movable body of the fixture body from bottom to top and drives it to move upward, thereby extending the contact arm to support the shell of the sunroof motor. Summary of the invention

[0004] 1. Technical issues to be resolved

[0005] To solve the above problems, the present invention proposes a life detection method, storage medium and execution device for a fixture for a motor housing, aiming to solve the problem that traditional fixture life detection methods in the prior art mostly rely on experience judgment or regular replacement and lack scientificity and accuracy.

[0006] (II) Technical solution

[0007] A life detection method for a jig for a motor housing of the present invention comprises:

[0008] Collecting a continuous expansion force characteristic value of the fixture based on a force sensor, preprocessing the continuous expansion force characteristic value to obtain a first characteristic data set, and performing classification prediction on the first characteristic data set through a SVM model;

[0009] Acquire the response time parameters of the fixture multiple times, convert the response time parameters into time vectors and map them to high-dimensional space using high-dimensional feature space technology to obtain an optimal separation hyperplane, and extract periodic features in the response time parameters to form a second feature data set;

[0010] Based on the optimization algorithm, the first feature data set and the second feature data set are integrated to find the optimal expansion force configuration, the expansion force threshold is preset and the abnormal parameters in the first feature data set are extracted to obtain the correlation between the life of the fixture and the first feature data set;

[0011] A jig life parameter model is constructed based on the correlation, a prediction relationship between the jig life parameter model and the expansion force configuration is generated, and a jig life prediction is obtained based on the correlation and the prediction relationship.

[0012] In the present invention, the step of preprocessing the continuous expansion force characteristic value includes:

[0013] Remove noise and outliers from data to improve data quality;

[0014] Processing the continuous expansion force characteristic value based on normalization to eliminate the dimensional influence between different characteristics;

[0015] Extracting the continuous expansion force characteristic values ​​that meet the requirements to form the first characteristic data set;

[0016] The SVM model is trained through cross-validation, and the trained SVM model is used to perform classification prediction on the first feature data set to obtain a classification result.

[0017] In the present invention, the response time parameters include a first reaction time when the fixture is started and in contact with the force sensor, a second reaction time when the driving device is in action and in contact with the fixture, a preset contact frequency, a contact time and a test cycle, and the response time parameters are collected based on the contact frequency and the contact time during the test cycle;

[0018] The time vector is mapped to a high-dimensional space based on a radial basis function kernel, and the response time parameters of different categories are separated to obtain an optimal separation hyperplane.

[0019] In the present invention, the optimization algorithm uses the PSO algorithm to determine the number of particles, which is the sum of the total data volume of the first feature data set and the second feature data set, and updates the speed of each particle according to the following formula: ,in express The particle in The speed in the weft, express The particle in The position on the latitude, and is a random number between 0 and 1, used to increase the randomness of the search. is the individual extreme value of the particle, For optimal expansion force configuration, is the inertia weight, is the first feature data set, is the second feature data set;

[0020] According to the updated particle velocity, based on the relationship: Update the position of each particle;

[0021] Check whether the preset expansion force threshold is reached. If the termination condition is met, the algorithm ends and the optimal expansion force configuration is output.

[0022] In the present invention, the abnormal parameter is a continuous expansion force characteristic value that deviates from the expansion force threshold by about 10%-20%.

[0023] In the present invention, an elastic net fixture SOH model is constructed in the optimal separation hyperplane, and the elastic net fixture SOH model is regularized to obtain a regularized model.

[0024] In the present invention, based on the regularized model and using a continuous scaling prediction method, a prediction relationship between the fixture life parameter model and the expansion force configuration is generated, and a fixture life prediction is obtained based on the correlation and the prediction relationship.

[0025] Another computer-readable storage medium of the present invention stores a computer program thereon, which, when executed by a processor, implements the steps of the life detection method of a jig for a motor housing described in the above technical solution.

[0026] Another execution device of the present invention includes an execution device body and a controller, and is characterized in that the controller includes a processor and a computer program stored in a memory and executable on the processor, and when the processor executes the program, the steps of the life detection method of the fixture for the motor housing described in the above technical solution are implemented.

[0027] (III) Beneficial effects

[0028] Compared with the prior art, the present invention has the following beneficial effects:

[0029] (1) In the present invention, the continuous expansion force characteristic value of the fixture is accurately collected by force sensors, and noise and outliers are removed through preprocessing to ensure the accuracy and reliability of the data. The first feature data set is classified and predicted using the SVM model, which effectively identifies the different states of the fixture and provides a solid foundation for subsequent analysis. The response time parameters are mapped through high-dimensional feature space technology, and periodic features are extracted, which further enriches the data set and improves the comprehensiveness and accuracy of the detection.

[0030] (2) The present invention uses the PSO algorithm to fuse the first feature data set and the second feature data set to find the optimal expansion force configuration. By presetting the expansion force threshold and extracting abnormal parameters, the present invention can accurately identify the expansion force characteristics that have an important impact on the life of the fixture, providing a scientific basis for optimizing fixture design and usage conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] 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, other drawings can be obtained based on these drawings without paying creative work.

[0032] Figure 1 It is a logical structure diagram of the fixture life detection method;

[0033] Figure 2 is a schematic diagram of the training logic structure of the first feature data set;

[0034] Figure 3 It is a schematic diagram of the overall structure of the detection device in the background technology;

[0035] Figure 4 It is a schematic diagram of the exploded structure of the fixture body in the background technology;

[0036] Figure 5 Schematic diagram of the framework structure of the execution equipment.

[0037] 1. Processor, 2. Memory, 3. Communication interface, 4. Communication bus, 5. Fixture body, 51. Touch arm, 52. Movable body, 6. Pressure sensor, 7. Display, 8. Highly integrated control processor, 9. Driving part. DETAILED DESCRIPTION

[0038] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are part of the embodiments of the present invention, rather than 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. Example

[0039] like Figure 1-Figure 2 A life detection method for a motor housing fixture is shown, which mainly includes the following steps:

[0040] S100, collecting continuous expansion force characteristic values ​​of the fixture based on a force sensor, preprocessing the continuous expansion force characteristic values ​​to obtain a first characteristic data set, and performing classification and prediction on the first characteristic data set through a SVM model.

[0041] First, select the appropriate force sensor type according to the design and usage scenario of the fixture. For measuring expansion force, sensors that can accurately reflect changes in tension or pressure are usually selected, such as strain gauge force sensors, piezoresistive force sensors, or capacitive force sensors. The sensor's range, accuracy, response speed and other parameters need to be determined according to actual needs.

[0042] The force sensors are installed at the key parts of the jig, which should be able to directly reflect the changes in the expansion force of the jig during operation. In this example, the jig extends into the motor housing of the sunroof motor and extends the contact arm to abut against the inner wall of the housing, and the corresponding force sensors need to be set on the displacement path of each contact arm to facilitate the subsequent parameter collection.

[0043] Connect the force sensor to a data acquisition system (such as a data recorder, microcontroller or computer) via a data cable or wirelessly. Configure the sampling frequency of the data acquisition system to ensure that all important force changes during the expansion of the fixture can be captured. Start the data acquisition system and let the force sensor continuously record the expansion force data of the fixture during use, that is, present the specific force in the form of data. This includes force value changes under different operating conditions (such as different workpiece sizes, materials, clamping strength, etc.). The steps of preprocessing the continuous expansion force characteristic value include:

[0044] S101. Remove noise and outliers from data to improve data quality;

[0045] The original data may contain noise and invalid information, and needs to be denoised through filtering algorithms (such as low-pass filtering, median filtering, etc.) to improve data quality.

[0046] S102, processing the continuous expansion force characteristic value based on normalization to eliminate the dimensional influence between different characteristics;

[0047] S103, extracting the continuous expansion force characteristic values ​​that meet the requirements to form the first characteristic data set;

[0048] Extract eigenvalues ​​from the preprocessed data, which can represent the key characteristics of the fixture expansion force. For example, statistical features such as maximum value, minimum value, average value, standard deviation, peak time, and frequency domain features obtained by Fourier transform can be extracted.

[0049] S104: Train the SVM model through cross-validation, and use the trained SVM model to perform classification prediction on the first feature data set to obtain a classification result.

[0050] Select support vector machine (SVM) as the classifier because SVM performs well in dealing with high-dimensional data and nonlinear problems. According to the characteristics of the feature data set, select a suitable kernel function (such as linear kernel, Gaussian kernel, etc.). Divide the first feature data set into a training set and a test set, usually with a ratio of 70% training set and 30% test set, or adjust it appropriately according to the size of the data. Use the training set data to train the SVM model and optimize the model performance by adjusting the model parameters (such as penalty parameter C, kernel function parameters, etc.).

[0051] Finally, the new fixture expansion force feature value is input into the trained SVM model for real-time classification prediction. Based on the prediction results, decision support is provided to operators or automated control systems, such as issuing alarms, adjusting process parameters, optimizing fixture design, etc.

[0052] S200, repeatedly acquiring the response time parameters of the fixture, converting the response time parameters into time vectors and mapping them to high-dimensional space using high-dimensional feature space technology to obtain an optimal separation hyperplane, and extracting periodic features in the response time parameters to form a second feature data set.

[0053] First, in order to obtain the response time parameters of the fixture, we divide the response time parameters into the first reaction time when the fixture is started and contacts the force sensor, and the second reaction time when the driving device is in action and contacts the fixture. Time data is collected through a small time recording device or response device, such as a digital timer, a contact timer, or by recording the starting time point of the action mechanism and the time point of contacting the contact arm driving element to indirectly obtain the response time parameters.

[0054] Specifically, the contact frequency, contact time and test cycle are preset, and the response time parameter is collected based on the contact frequency and contact time during the test cycle. For example, within 30 minutes, the contact frequency and contact time of the fixture are limited to collect the first reaction time when the fixture is started and contacts the force sensor, and the second reaction time when the drive device is in action and contacts the drive device.

[0055] Then, the time vector is mapped to a high-dimensional space based on a radial basis function kernel, and the response time parameters of different categories are separated to obtain an optimal separation hyperplane.

[0056] It needs to be explained that the preset contact frequency, contact time and test cycle are sorted in chronological order, and then the minimum dimension of space reconstruction is determined by the false neighbor method. Specifically, by examining the real neighbor points in the phase space and obtaining which are false neighbor points in the phase space, when there are no false neighbor points, it can be considered that the geometric structure is completely opened. When two non-adjacent points in the high-dimensional attractor become adjacent points when projected onto the low-dimensional trajectory, such neighbor points are false. The proportion of false nearest neighbor points is calculated, and the dimension is gradually increased to obtain the minimum dimension. Then, according to the minimum dimension, the characterization data sorted in time is reconstructed, that is, the fuzzy entropy of the preset contact frequency, contact time and test cycle is reconstructed and mapped to the phase space to obtain the second feature data set.

[0057] S300, based on the optimization algorithm, the first feature data set and the second feature data set are integrated to find the optimal expansion force configuration, the expansion force threshold is preset and the abnormal parameters in the first feature data set are extracted to obtain the correlation between the life of the fixture and the first feature data set.

[0058] Specifically, in this embodiment, the PSO algorithm is used to determine the number of particles, which is the sum of the total data volume of the first feature data set and the second feature data set, and the speed of each particle is updated according to the following formula: ,in express The particle in The speed in the weft, express The particle in The position on the latitude, and is a random number between 0 and 1, used to increase the randomness of the search. is the individual extreme value of the particle, For optimal expansion force configuration, is the inertia weight, is the first feature data set, is the second feature data set;

[0059] According to the updated particle velocity, based on the relationship: Update the position of each particle. Check whether the preset expansion force threshold is reached. In this embodiment, the abnormal parameter is a continuous expansion force characteristic value that deviates from the expansion force threshold by about 15%.

[0060] If the termination condition is met, the algorithm ends and the optimal expansion force configuration is output. It should be noted that the above optimization algorithm is not limited to the PSO algorithm, and the AdaBoost algorithm, RMSProp algorithm, etc. can also be used for data optimization to improve the overall data accuracy of the data set.

[0061] In particular, an elastic net fixture SOH model is constructed in the optimal separation hyperplane, and the elastic net fixture SOH model is regularized to obtain a regularized model. Based on the regularized model and a continuous scaling prediction method, a prediction relationship between the fixture life parameter model and the expansion force configuration is generated, and the fixture life prediction is obtained based on the correlation and the prediction relationship.

[0062] S400 , constructing a fixture life parameter model based on the correlation, generating a prediction relationship between the fixture life parameter model and the expansion force configuration, and obtaining a fixture life prediction based on the correlation and the prediction relationship.

[0063] The second feature data set can be classified in the optimal separating hyperplane, and then the elastic net fixture SOH model is constructed in the optimal separating hyperplane. In this embodiment, the elastic net fixture SOH model is a regularization process of linear regression, which can improve the prediction accuracy by automatically selecting variables and performing continuous scaling, so as to obtain the optimal SOH prediction of the fixture service life, that is, the health of the spring; the optimal expansion force configuration output by the PSO algorithm can dynamically judge the standard expansion force configuration that the fixture should achieve. The goal of regularization optimization is usually to improve the SOH prediction result based on ordinary least squares by reducing the parameter estimate to zero.

[0064] In summary, the life detection method of the jig for the motor housing in the above-mentioned embodiment of the present invention, by combining the first feature data and the second feature data, can not only update the life detection of the jig in real time, but also effectively improve the accuracy of the detection. By performing continuous scaling prediction on the elastic net jig SOH model constructed in the optimal separating hyperplane, the reduction of prediction accuracy can be avoided. The prediction generated by the correlation and prediction relationship can more accurately predict the life of the jig. Example

[0065] An embodiment of the present invention provides a computer-readable storage medium.

[0066] The computer-readable storage medium provided in the embodiment of the present invention stores a computer program, and when the computer program is executed by a processor, the steps of any of the above-mentioned methods for detecting the life of a fixture for a motor housing can be implemented.

[0067] The computer-readable storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0068] For an introduction to the computer-readable storage medium provided in an embodiment of the present invention, please refer to the above method embodiment, and the present invention will not be elaborated here.

[0069] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0070] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0071] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. Example

[0072] An embodiment of the present invention provides an execution device.

[0073] Please refer to Figure 5 , Figure 5 This is a schematic diagram of the structure of an execution device provided by the present invention, and the execution device may include:

[0074] Memory for storing computer programs;

[0075] The processor is used to implement the steps of any of the above-mentioned methods for detecting the life of a fixture for a motor housing when executing a computer program.

[0076] like Figure 5 , which is a schematic diagram of the composition structure of the execution device, the execution device may include: a processor 1, a memory 2, a communication interface 3 and a communication bus 4. The processor 1, the memory 2, and the communication interface 3 all communicate with each other through the communication bus 4.

[0077] In the embodiment of the present invention, the processor 1 may be a central processing unit (CPU), an application specific integrated circuit, a digital signal processor, a field programmable gate array or other programmable logic devices.

[0078] The processor 1 may call a program stored in the memory 2. Specifically, the processor 1 may execute the operations in the embodiment of the push button switch fault detection method.

[0079] The memory 2 is used to store one or more programs, which may include program codes, and the program codes include computer operation instructions. In the embodiment of the present invention, the memory 2 at least stores programs for implementing the following functions:

[0080] Collecting a continuous expansion force characteristic value of the fixture based on a force sensor, preprocessing the continuous expansion force characteristic value to obtain a first characteristic data set, and performing classification prediction on the first characteristic data set through a SVM model;

[0081] Acquire the response time parameters of the fixture multiple times, convert the response time parameters into time vectors and map them to high-dimensional space using high-dimensional feature space technology to obtain an optimal separation hyperplane, and extract periodic features in the response time parameters to form a second feature data set;

[0082] Based on the optimization algorithm, the first feature data set and the second feature data set are integrated to find the optimal expansion force configuration, the expansion force threshold is preset and the abnormal parameters in the first feature data set are extracted to obtain the correlation between the life of the fixture and the first feature data set;

[0083] A jig life parameter model is constructed based on the correlation, a prediction relationship between the jig life parameter model and the expansion force configuration is generated, and a jig life prediction is obtained based on the correlation and the prediction relationship.

[0084] In a possible implementation, the memory 2 may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required for at least one function, etc.; the data storage area may store data created during use.

[0085] In addition, the memory 2 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device or other volatile solid-state storage device.

[0086] The communication interface 3 may be an interface of a communication module, and is used to connect to other devices or systems.

[0087] Of course, it should be noted that Figure 5 The structure shown does not constitute a limitation on the execution device in the embodiment of the present invention. In actual applications, the execution device may include Figure 5 More or fewer components than shown, or combinations of certain components.

[0088] The above-described embodiments are merely descriptions of preferred implementations of the present invention, and are not intended to limit the concept and scope of the present invention. Without departing from the design concept of the present invention, various modifications and improvements made by ordinary persons in the art to the technical solution of the present invention should fall within the protection scope of the present invention, and the technical contents for which protection is sought in the present invention have been fully recorded in the claims.

Claims

1. A method for detecting the life of a fixture for a motor housing, characterized in that: The detection method comprises: collecting a continuous expansion force characteristic value of the fixture based on a force sensor, preprocessing the continuous expansion force characteristic value to obtain a first feature data set, and performing classification prediction on the first feature data set using a SVM model; Acquiring the response time parameters of the fixture multiple times, converting the response time parameters into time vectors and mapping them to high-dimensional space using high-dimensional feature space technology to obtain an optimal separating hyperplane, and extracting periodic features from the response time parameters to form a second feature data set; fusing the first feature data set and the second feature data set based on an optimization algorithm to find an optimal expansion force configuration, presetting an expansion force threshold and extracting abnormal parameters in the first feature data set to obtain a correlation between the life of the fixture and the first feature data set; Constructing a fixture life parameter model based on the correlation, generating a prediction relationship between the fixture life parameter model and the expansion force configuration, and obtaining a fixture life prediction based on the correlation and the prediction relationship; The optimization algorithm uses the PSO algorithm to determine the number of particles, which is the sum of the total data volume of the first feature data set and the second feature data set, and updates the speed of each particle according to the following formula: ,in express The particle in The speed in the weft, express The particle in The position on the latitude, and is a random number between 0 and 1, used to increase the randomness of the search. is the individual extreme value of the particle, For optimal expansion force configuration, is the inertia weight, is the first feature data set, is the second feature data set; According to the updated particle velocity, based on the relationship: Update the position of each particle; Check whether the preset expansion force threshold is reached. If the termination condition is met, the algorithm ends and the optimal expansion force configuration is output.

2. The life detection method of a motor housing jig according to claim 1, characterized in that: The step of pre-processing the continuous expansion force characteristic value includes: Remove noise and outliers from data to improve data quality; Normalizing the sustained expansion force characteristic value to eliminate dimensional effects between different characteristics; Extracting the continuous expansion force characteristic values ​​that meet the requirements to form the first characteristic data set; The SVM model is trained through cross-validation, and the trained SVM model is used to perform classification prediction on the first feature data set to obtain a classification result.

3. The life detection method of a motor housing jig according to claim 2, characterized in that: The response time parameters include a first reaction time when the fixture contacts the force sensor when the fixture is started, a second reaction time when the driving device is in action and when the driving device contacts the fixture, a preset contact frequency, a contact time, and a test period, and the response time parameters are collected based on the contact frequency and the contact time during the test period; The time vector is mapped to a high-dimensional space based on a radial basis function kernel, and the response time parameters of different categories are separated to obtain an optimal separation hyperplane.

4. The life detection method of a jig for a motor housing according to claim 3, characterized in that: The abnormal parameter is a continuous expansion force characteristic value that deviates from the expansion force threshold by about 10%-20%.

5. The life detection method of a jig for a motor housing according to claim 4, characterized in that: An elastic net jig SOH model is constructed in the optimal separating hyperplane, and the elastic net jig SOH model is regularized to obtain a regularized model.

6. The life detection method of a jig for a motor housing according to claim 5, characterized in that: Based on the regularized model and using a continuous scaling prediction method, a prediction relationship between the fixture life parameter model and the expansion force configuration is generated, and a fixture life prediction is obtained based on the correlation and the prediction relationship.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the life detection method of the jig for the motor housing according to any one of claims 1 to 6 are implemented.

8. An execution device, comprising an execution device body and a controller, characterized in that: The controller includes a processor and a computer program stored in a memory and executable on the processor. When the processor executes the program, the steps of the method for detecting the life of a jig for a motor housing according to any one of claims 1 to 6 are implemented.

Images