A deformation prediction method and system for injection molded gears of automobile wipers

Abstract

The present invention relates to the technical field of deformation prediction of injection molded gears, and specifically to a deformation prediction method and system for injection molded gears of automobile wipers, which obtain working environment data of injection molded gears of automobile wipers and perform preprocessing; based on the working environment data, a first deformation prediction model is constructed to analyze the force and deformation of the gears under complex working conditions, and obtain first deformation prediction data of the injection molded gears; a deformation prediction optimization model is constructed to analyze the working environment data and the first deformation prediction data, and obtain an injection molded gear deformation optimization prediction data group; high-precision three-dimensional scanning data of the injection molded gears is obtained, the injection molded gear deformation is extracted, and an injection molded gear image deformation data group is obtained; the injection molded gear image deformation data group and the injection molded gear deformation optimization prediction data group are multi-dimensionally fused to generate an injection molded gear deformation prediction result; and operation suggestions are given based on the injection molded gear deformation prediction result.

Description

Technical Field

[0001] The invention relates to the technical field of deformation prediction of injection molded gears, and in particular to a deformation prediction method and system for an injection molded gear of an automobile wiper. Background Art

[0002] With the rapid development of the automobile industry, automobile wipers are important safety components of vehicles, and their performance and reliability are crucial to the normal use of vehicles. As one of its core transmission components, the wiper injection molded gear is responsible for driving the wiper to swing. Therefore, the deformation of its tooth shape and tooth surface directly affects the working efficiency and service life of the wiper. Injection molded gears are usually produced using the injection molding process. Although this process can efficiently and economically manufacture gears, due to its molding process and material properties, injection molded gears are easily affected by environmental conditions such as temperature, humidity, pressure, workload, and long-term friction during use, resulting in deformation of tooth shape and tooth surface.

[0003] In the prior art, the prediction of deformation of injection molded gears mainly relies on static physical models. Although they can simulate the deformation of tooth profile under static working conditions well, the prediction accuracy is low under complex environmental conditions, such as temperature fluctuations, humidity changes, load instability and other dynamic changes. In addition, with the long-term use of wipers, the external pressure and friction on the gears continue to accumulate, and the tooth profile and tooth surface will gradually deform. It is difficult for the prior art to accurately predict these dynamic changes during long-term use. Due to the material and processing characteristics of injection molded gears, there are nonlinear and time-varying characteristics in the deformation process, and the existing linear model cannot accurately reflect its deformation trend.

[0004] Therefore, a deformation prediction method and system for injection molded gear of automobile wiper is proposed. Summary of the invention

[0005] The purpose of the present invention is to provide a deformation prediction method and system for an injection molded gear of an automobile wiper, to obtain working environment data of the injection molded gear of the automobile wiper, and to perform preprocessing; based on the working environment data, to construct a first deformation prediction model to analyze the force and deformation of the gear under complex working conditions, and to obtain first deformation prediction data of the injection molded gear; to construct a deformation prediction optimization model to analyze the working environment data and the first deformation prediction data, and to obtain an injection molded gear deformation optimization prediction data group; to obtain high-precision three-dimensional scanning data of the injection molded gear, to extract the deformation of the injection molded gear, and to obtain an injection molded gear image deformation data group; to multi-dimensionally fuse the injection molded gear image deformation data group with the injection molded gear deformation optimization prediction data group, and to generate an injection molded gear deformation prediction result; and to provide operation suggestions based on the injection molded gear deformation prediction result.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A deformation prediction method for an injection molded gear of an automobile wiper, comprising:

[0008] Obtaining working environment data of the injection molded gear of the automobile wiper and performing preprocessing; the working environment data includes temperature, humidity, load, friction, operating frequency and working cycle;

[0009] Based on the working environment data, a first deformation prediction model is constructed to simulate and analyze the force and deformation of the gear under complex working conditions, so as to obtain first deformation prediction data of the injection molded gear; the first deformation prediction data includes first deformation prediction data of the tooth shape and first deformation prediction data of the tooth surface;

[0010] Constructing a deformation prediction optimization model to analyze the working environment data, the tooth shape first deformation prediction data and the tooth surface first deformation prediction data to obtain an injection molded gear deformation optimization prediction data group;

[0011] Acquire high-precision three-dimensional scanning data of injection-molded gears, analyze the three-dimensional scanning data using an image processing algorithm, and extract deformation features of the injection-molded gears; obtain an injection-molded gear image deformation data set based on the deformation features of the injection-molded gears;

[0012] Performing multi-dimensional fusion on the injection molded gear image deformation data group and the injection molded gear deformation optimization prediction data group to generate an injection molded gear deformation prediction result;

[0013] Based on the injection molded gear deformation prediction results, operational suggestions are given.

[0014] Preferably, the first deformation prediction model includes a tooth shape deformation prediction layer, a tooth surface deformation prediction layer, a correlation analysis layer and a result output layer;

[0015] The tooth profile deformation prediction layer obtains preliminary tooth profile deformation data by analyzing the working environment data;

[0016] The tooth surface deformation prediction layer analyzes the working environment data to obtain preliminary tooth surface deformation data;

[0017] The correlation analysis layer analyzes the preliminary tooth shape deformation data, preliminary tooth surface deformation data, temperature and humidity through correlation analysis to obtain the tooth shape first deformation prediction data and the tooth surface first deformation prediction data;

[0018] The result output layer obtains the first deformation prediction data of the injection gear by weighted sum processing of the first deformation prediction data of the tooth shape and the first deformation prediction data of the tooth surface.

[0019] Preferably, the calculation formula of the first deformation prediction data is:

[0020] ;

[0021] in, is the first deformation prediction data, is the first deformation prediction data of the tooth shape, is the first deformation prediction data of the tooth surface, is a multi-input multi-output weighted function, For standard injection gear data, For working environment data.

[0022] Preferably, the deformation prediction optimization model includes a data input layer, a model training layer and an optimized data output layer;

[0023] The data input layer inputs the first deformation prediction data into the deep neural network;

[0024] The model training layer performs preliminary training on the deep neural network by using historical working environment data to obtain a feature vector between the first deformation prediction data and the gear deformation; optimizes and updates the deep neural network by collecting working environment data in real time; the optimization and update is performed by calculating errors and adjusting weights and errors in the network;

[0025] The optimized data output layer obtains an injection molded gear deformation optimization prediction data set by continuously optimizing and updating the deep neural network.

[0026] Preferably, the three-dimensional scanning data acquisition process is:

[0027] Scan the surface of the injection-molded gear multiple times with a laser beam to obtain multiple three-dimensional point cloud data of the injection-molded gear;

[0028] De-noising and down-sampling the three-dimensional point cloud data; obtaining three-dimensional point cloud data of a complete gear by registering multiple three-dimensional point cloud data;

[0029] Analyzing the three-dimensional point cloud data and standard injection molded gear data to generate deformation characteristics of the injection molded gear; the deformation characteristics of the injection molded gear include deformation amplitude, local wear area and shape change trend;

[0030] The deformation characteristics of the injection molded gear are measured and calculated to obtain an injection molded gear image deformation data set.

[0031] Preferably, the injection molded gear deformation prediction result generation process includes:

[0032] Performing dimensionality reduction on the injection molded gear image deformation data group and the injection molded gear deformation optimization prediction data group through principal component analysis to generate low-dimensional space data;

[0033] The data in the low-dimensional spatial data set are multi-dimensionally fused to obtain the injection molded gear deformation prediction result.

[0034] A deformation prediction system for an injection molded gear of an automobile wiper, comprising:

[0035] A data acquisition unit, used to acquire working environment data of the injection molded gear of the automobile wiper and perform preprocessing; the working environment data includes temperature, humidity, load, friction, operating frequency and working cycle;

[0036] A first deformation prediction data acquisition unit is used to construct a first deformation prediction model based on the working environment data to simulate and analyze the force and deformation of the gear under complex working conditions, so as to obtain first deformation prediction data of the injection molded gear; the first deformation prediction data includes first deformation prediction data of the tooth shape and first deformation prediction data of the tooth surface;

[0037] A prediction value optimization unit is used to construct a deformation prediction optimization model to analyze the working environment data, the tooth shape first deformation prediction data and the tooth surface first deformation prediction data to obtain an injection molded gear deformation optimization prediction data group;

[0038] An image data acquisition unit is used to acquire high-precision three-dimensional scanning data of injection-molded gears, analyze the three-dimensional scanning data using an image processing algorithm, and extract deformation features of the injection-molded gears; and obtain an injection-molded gear image deformation data group based on the deformation features of the injection-molded gears;

[0039] The deformation result prediction unit is used to perform multi-dimensional fusion on the injection molded gear image deformation data group and the injection molded gear deformation optimization prediction data group to generate an injection molded gear deformation prediction result; and provide operation suggestions based on the injection molded gear deformation prediction result.

[0040] Preferably, the deformation prediction optimization model includes a data input layer, a model training layer and an optimized data output layer;

[0041] The data input layer inputs the first deformation prediction data into the deep neural network;

[0042] The model training layer performs preliminary training on the deep neural network by using historical working environment data to obtain a feature vector between the first deformation prediction data and the gear deformation; optimizes and updates the deep neural network by collecting working environment data in real time; the optimization and update is performed by calculating errors and adjusting weights and errors in the network;

[0043] The optimization data output layer obtains an injection molded gear deformation optimization prediction data set by continuously optimizing and updating the deep neural network;

[0044] The three-dimensional scanning data acquisition process is as follows:

[0045] Scan the surface of the injection-molded gear multiple times with a laser beam to obtain multiple three-dimensional point cloud data of the injection-molded gear;

[0046] De-noising and down-sampling the three-dimensional point cloud data; obtaining three-dimensional point cloud data of a complete gear by registering multiple three-dimensional point cloud data;

[0047] Analyzing the three-dimensional point cloud data and standard injection molded gear data to generate injection molded gear deformation characteristics; the injection molded gear deformation characteristics include deformation amplitude, local wear area and shape change trend;

[0048] Measuring and calculating the deformation characteristics of the injection molded gear to obtain an injection molded gear image deformation data set;

[0049] The injection molded gear deformation prediction result generation process includes:

[0050] Performing dimensionality reduction on the injection molded gear image deformation data set and the injection molded gear deformation optimization prediction data set by principal component analysis to generate a low-dimensional space data set;

[0051] The data in the low-dimensional spatial data set are multi-dimensionally fused to obtain the injection molded gear deformation prediction result.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] 1. The present invention analyzes the working environment data by constructing a first deformation prediction model to obtain the first deformation prediction data of the injection molded gear; wherein the first deformation prediction data includes the first deformation prediction data of the tooth shape and the first deformation prediction data of the tooth surface; by constructing a deformation prediction optimization model combined with a deep neural network and historical working environment data, the first deformation prediction data is optimized to obtain an injection molded gear deformation optimization prediction data group; by continuously training and optimizing the optimized prediction data group, the accuracy and reliability of deformation prediction are improved.

[0054] 2. The present invention obtains 3D point cloud data of the complete gear by scanning the injection molded gear multiple times and aligning multiple 3D point cloud data, thereby extracting the deformation characteristics of the gear surface under different working conditions; the deformation characteristics include deformation amplitude, local wear area and shape change trend, which can reflect the deformation of the gear surface under various environmental factors such as friction and load; by analyzing the deformation characteristics, the injection molded gear image deformation data is obtained, thereby improving the accuracy of gear deformation detection.

[0055] 3. The present invention uses multi-dimensional data fusion technology to effectively integrate the image deformation data set of injection molded gears with the optimized prediction data set, thereby generating more accurate deformation prediction results. At the same time, the fused prediction results comprehensively consider the working environment, deformation characteristics, historical data and real-time monitoring information. Through the dimensionality reduction processing and multi-dimensional fusion of the image deformation data set of injection molded gears and the optimized prediction data set, the complexity of the data is simplified, the key information in the data is retained, and the accuracy and efficiency of deformation prediction are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A schematic flow chart of a deformation prediction method for an injection molded gear of an automobile wiper provided by the present invention;

[0057] Figure 2 A structural schematic diagram of a deformation prediction system for an automobile wiper injection gear provided by the present invention;

[0058] Figure 3 It is a schematic diagram of the overall structure of the injection molded gear deformation prediction of the present invention. DETAILED DESCRIPTION

[0059] 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.

[0060] Embodiment 1

[0061] See also Figure 1 The present invention provides a deformation prediction method for an injection molded gear of an automobile wiper, and the technical solution is as follows:

[0062] Obtaining working environment data of the injection molded gear of the automobile wiper and performing preprocessing; the working environment data includes temperature, humidity, load, friction, operating frequency and working cycle;

[0063] Preprocessing includes data cleaning (removing outliers, such as obviously unreasonable data caused by sensor failure) and data normalization (converting data of different magnitudes to the same magnitude, for example, normalizing data of different ranges such as temperature and load to the [0,1] interval) to ensure data accuracy and consistency.

[0064] Based on the working environment data, a first deformation prediction model is constructed to simulate and analyze the force and deformation of the gear under complex working conditions, so as to obtain first deformation prediction data of the injection molded gear; the first deformation prediction data includes first deformation prediction data of the tooth shape and first deformation prediction data of the tooth surface;

[0065] Furthermore, the first deformation prediction model includes a tooth shape deformation prediction layer, a tooth surface deformation prediction layer, a correlation analysis layer and a result output layer;

[0066] The tooth profile deformation prediction layer obtains preliminary tooth profile deformation data by analyzing the working environment data;

[0067] The tooth surface deformation prediction layer analyzes the working environment data to obtain preliminary tooth surface deformation data;

[0068] The correlation analysis layer analyzes the preliminary tooth shape deformation data, preliminary tooth surface deformation data, temperature and humidity through correlation analysis to obtain the tooth shape first deformation prediction data and the tooth surface first deformation prediction data;

[0069] The result output layer obtains the first deformation prediction data of the injection gear by weighted sum processing of the first deformation prediction data of the tooth shape and the first deformation prediction data of the tooth surface.

[0070] In this embodiment, the constructed first deformation prediction model helps to analyze the influence of working environment data on the deformation of injection molded gears from multiple levels. The tooth shape deformation prediction layer and the tooth surface deformation prediction layer respectively perform preliminary analysis on the deformation of the tooth shape and tooth surface, the correlation analysis layer integrates multiple data through correlation analysis, and the result output layer obtains more comprehensive first deformation prediction data through weighted sum processing. This process can comprehensively consider the influence of multiple factors on the deformation of injection molded gears and effectively improve the accuracy of deformation prediction.

[0071] Furthermore, the calculation formula of the first deformation prediction data is:

[0072] ;

[0073] in, is the first deformation prediction data, is the first deformation prediction data of the tooth shape, is the first deformation prediction data of the tooth surface, is a multi-input multi-output weighted function, It is the standard injection gear data, which is determined by the drawing. For working environment data.

[0074] The weighted function of multiple input and multiple output The tooth shape and tooth surface weights are obtained by machine learning analysis of historical deformation data of injection molded gears.

[0075] The working environment data of the automobile wiper injection molded gear of the present invention is obtained by installing various sensors at key positions of the automobile wiper injection molded gear. The temperature sensor measures temperature, the humidity sensor measures humidity, the pressure sensor measures load, the friction sensor measures friction, the counter records the operating frequency, and the timer records the working cycle. Through the first deformation prediction model, the force and deformation of the injection molded gear under complex working conditions can be simulated and analyzed, providing basic data for subsequent optimization prediction, and improving the initial accuracy of the injection molded gear deformation prediction. For specific data, please refer to Table 1;

[0076] Table 1 Data analysis table of the first deformation prediction model

[0077]

[0078] Constructing a deformation prediction optimization model to analyze the working environment data, the tooth shape first deformation prediction data and the tooth surface first deformation prediction data to obtain an injection molded gear deformation optimization prediction data group;

[0079] Furthermore, the deformation prediction optimization model includes a data input layer, a model training layer and an optimized data output layer;

[0080] The data input layer inputs the first deformation prediction data into the deep neural network;

[0081] The model training layer performs preliminary training on the deep neural network by using historical working environment data to obtain a feature vector between the first deformation prediction data and the gear deformation; optimizes and updates the deep neural network by collecting working environment data in real time; the optimization and update is performed by calculating errors and adjusting weights and errors in the network; the historical working environment data is obtained through historical data of injection molded gears.

[0082] The optimized data output layer obtains an injection molded gear deformation optimization prediction data set by continuously optimizing and updating the deep neural network.

[0083] In this embodiment, the deformation prediction optimization model provided by the present invention uses a deep neural network and combines historical working environment data to deeply analyze and optimize the first deformation prediction data. By continuously learning data features and optimizing network weights through the model training layer, the optimized data output layer can obtain a more accurate injection molded gear deformation optimization prediction data set, which helps to further improve the accuracy and reliability of deformation prediction and reduce prediction errors. The present invention measures the error through the mean square error.

[0084] The first deformation prediction data is input into the deep neural network to construct a deformation prediction optimization model. The model training layer performs preliminary training on the deep neural network by using historical working environment data. For example, there are 100 sets of historical working environment data and their corresponding actual gear deformation data. The feature vector between the first deformation prediction data and the gear deformation is obtained through these data. In actual operation, the working environment data is collected in real time, and the deep neural network is optimized and updated. The optimization update is performed by calculating errors and adjusting weights and errors in the network. The optimized data output layer obtains the injection molding gear deformation optimization prediction data group by continuously optimizing and updating the deep neural network. For specific data, see Table 2;

[0085] Table 2 Comparison of injection molded gear deformation optimization prediction data

[0086]

[0087] By constructing a deformation prediction optimization model, the first deformation prediction data is optimized by combining the deep neural network and the historical working environment data. From the data comparison in Table 2, it can be seen that the error between the predicted data and the actual deformation data after optimization is significantly reduced, which improves the accuracy and reliability of deformation prediction.

[0088] Acquire high-precision three-dimensional scanning data of injection-molded gears, analyze the three-dimensional scanning data using an image processing algorithm, and extract deformation features of the injection-molded gears; obtain an injection-molded gear image deformation data set based on the deformation features of the injection-molded gears;

[0089] Furthermore, the three-dimensional scanning data acquisition process is:

[0090] Scan the surface of the injection-molded gear multiple times with a laser beam to obtain multiple three-dimensional point cloud data of the injection-molded gear;

[0091] De-noising and down-sampling the three-dimensional point cloud data; obtaining three-dimensional point cloud data of a complete gear by registering multiple three-dimensional point cloud data;

[0092] Analyzing the three-dimensional point cloud data and standard injection molded gear data to generate deformation characteristics of the injection molded gear; the deformation characteristics of the injection molded gear include deformation amplitude, local wear area and shape change trend;

[0093] The deformation characteristics of the injection molded gear are measured and calculated to obtain an injection molded gear image deformation data set, wherein the injection molded gear image deformation data set includes deformation data of shape, size, angle, and position;

[0094] Table 3 Comparison of deformation feature detection accuracy

[0095]

[0096] In this embodiment, by acquiring and analyzing high-precision three-dimensional scanning data, the deformation characteristics of injection molded gears under different working conditions can be accurately extracted. These deformation characteristics include deformation amplitude, local wear area, and shape change trend. Compared with traditional detection methods, they can more comprehensively reflect the deformation of the gear surface, thereby significantly improving the accuracy of gear deformation detection. For details, refer to Table 3;

[0097] The data in Table 3 are derived from a large number of experimental tests on various detection technologies for injection molded gear deformation detection. During the experiment, traditional manual measurement, three-coordinate measuring machine, optical measurement technology and the high-precision three-dimensional scanning technology of the present invention were used for detection. Each sample was measured multiple times under the same environmental conditions, and the average value was taken to ensure the reliability of the data.

[0098] The injection molded gear image deformation data group and the injection molded gear deformation optimization prediction data group are multi-dimensionally fused to generate an injection molded gear deformation prediction result; and operation suggestions are given based on the injection molded gear deformation prediction result.

[0099] Furthermore, the injection molded gear deformation prediction result generation process includes:

[0100] Performing dimensionality reduction on the injection molded gear image deformation data set and the injection molded gear deformation optimization prediction data set by factor analysis to generate a low-dimensional space data set;

[0101] The data in the low-dimensional spatial data set are multi-dimensionally fused to obtain the injection molded gear deformation prediction result.

[0102] In this embodiment, the multi-dimensional fusion technology effectively integrates the image deformation data set of the injection molded gear with the optimized prediction data set, performs dimensionality reduction processing through principal component analysis, simplifies the data complexity and retains key information, thereby generating more accurate deformation prediction results. This prediction result that comprehensively considers multiple factors can improve the accuracy and efficiency of deformation prediction.

[0103] The final tooth profile deformation prediction results are fed back to the gear manufacturing and maintenance system. Based on the prediction results, the system can provide relevant operational suggestions, such as adjusting the manufacturing process, replacing severely worn gears, etc. At the same time, when the prediction results exceed the set threshold, the system will issue an alarm to remind the operator to take timely measures to ensure the normal operation of the gear. Figure 3 .

[0104] The deformation prediction method and system of the injection molded gear of the automobile wiper of the present invention can realize the accurate deformation prediction of the injection molded gear under different working conditions by comprehensively analyzing multi-dimensional data, and significantly improve the reliability, service life and maintenance efficiency of the gear. By obtaining detailed working environment data (such as temperature, humidity, load, etc.) and combining with deep learning algorithms, the force and deformation of the gear under actual working conditions can be accurately simulated and predicted. First, through the joint use of the first deformation prediction model and the deformation prediction optimization model, the present invention can simulate and predict the deformation characteristics of the tooth shape and tooth surface under different working conditions in detail based on the working environment data. After continuous optimization and training of the deep neural network, the prediction accuracy of the model has been significantly improved, and the accuracy of the deformation prediction results has been effectively improved. Secondly, by using high-precision three-dimensional scanning technology and combining image processing algorithms to extract gear deformation features, the local wear, deformation amplitude and shape change of the gear can be accurately detected, and the meticulousness and comprehensiveness of the prediction results are further enhanced. Finally, through multi-dimensional data fusion, the image deformation data is combined with the optimized prediction data, and the principal component analysis is used for dimensionality reduction processing, so that the prediction results are more concise and easy to analyze, the error is reduced, and the accuracy of the decision is improved.

[0105] Embodiment 2

[0106] See also Figure 2 The present invention provides a deformation prediction system for an injection molded gear of an automobile wiper, and the technical solution is as follows:

[0107] A data acquisition unit, used to acquire working environment data of the injection molded gear of the automobile wiper and perform preprocessing; the working environment data includes temperature, humidity, load, friction, operating frequency and working cycle;

[0108] A first deformation prediction data acquisition unit is used to construct a first deformation prediction model based on the working environment data to simulate and analyze the force and deformation of the gear under complex working conditions, so as to obtain first deformation prediction data of the injection molded gear; the first deformation prediction data includes first deformation prediction data of the tooth shape and first deformation prediction data of the tooth surface;

[0109] A prediction value optimization unit is used to construct a deformation prediction optimization model to analyze the working environment data, the tooth shape first deformation prediction data and the tooth surface first deformation prediction data to obtain an injection molded gear deformation optimization prediction data group;

[0110] An image data acquisition unit is used to acquire high-precision three-dimensional scanning data of injection-molded gears, analyze the three-dimensional scanning data using an image processing algorithm, and extract deformation features of the injection-molded gears; and obtain an injection-molded gear image deformation data group based on the deformation features of the injection-molded gears;

[0111] The deformation result prediction unit is used to perform multi-dimensional fusion on the injection molded gear image deformation data group and the injection molded gear deformation optimization prediction data group to generate an injection molded gear deformation prediction result; and provide operation suggestions based on the injection molded gear deformation prediction result.

[0112] Furthermore, the injection molded gear deformation prediction result generation process includes:

[0113] Performing dimensionality reduction on the injection molded gear image deformation data set and the injection molded gear deformation optimization prediction data set by principal component analysis to generate a low-dimensional space data set;

[0114] The data in the low-dimensional spatial data set are multi-dimensionally fused to obtain the injection molded gear deformation prediction result.

[0115] Through multi-dimensional data fusion technology, the image deformation data set of the injection molded gear and the optimization prediction data set are effectively integrated to generate more accurate deformation prediction results. Compared with the prediction of single data, the prediction accuracy of the present invention is higher. The data is shown in Table 4;

[0116] Table 4 Comparison of prediction accuracy and efficiency

[0117]

[0118] The present invention provides a deformation prediction system for injection molded gears of automobile wipers. By considering working environment data, gear stress conditions, deformation characteristics and image analysis, the deformation dynamic prediction of injection molded gears under complex working conditions is realized, and the performance evaluation and fault warning capabilities of gears are improved. Specifically, by obtaining detailed environmental data and constructing a multi-level prediction model, the stress and deformation conditions of gears under various working conditions are accurately simulated, providing strong data support for the optimization design and use of injection molded gears. Based on the combination of working environment data and high-precision three-dimensional scanning technology, the system can deeply analyze the deformation characteristics of the gear surface, such as deformation amplitude, local wear area and shape change trend, etc., to provide an accurate basis for the health status monitoring and maintenance of the gear. Secondly, through the continuous learning and adjustment of the deep neural network optimization model, combined with real-time data update, the gear deformation prediction results can be optimized in real time, improving the accuracy and reliability of the prediction. Finally, through multi-dimensional data fusion technology, the image deformation data is combined with the optimized prediction data to generate a low-dimensional space data set, ensuring that the deformation of the injection molded gear can be accurately predicted in a highly complex working environment, reducing the incidence of unexpected failures.

[0119] 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 deformation prediction method for an injection molded gear of an automobile wiper, characterized in that: include: Obtaining working environment data of the injection molded gear of the automobile wiper and performing preprocessing; the working environment data includes temperature, humidity, load, friction, operating frequency and working cycle; Based on the working environment data, a first deformation prediction model is constructed to simulate and analyze the force and deformation of the gear under complex working conditions, so as to obtain first deformation prediction data of the injection molded gear; The first deformation prediction data includes tooth shape first deformation prediction data and tooth surface first deformation prediction data; Constructing a deformation prediction optimization model to analyze the working environment data, the tooth shape first deformation prediction data and the tooth surface first deformation prediction data to obtain an injection molded gear deformation optimization prediction data group; Acquire high-precision three-dimensional scanning data of injection-molded gears, analyze the three-dimensional scanning data using an image processing algorithm, and extract deformation features of the injection-molded gears; obtain an injection-molded gear image deformation data set based on the deformation features of the injection-molded gears; Performing multi-dimensional fusion on the injection molded gear image deformation data group and the injection molded gear deformation optimization prediction data group to generate an injection molded gear deformation prediction result; Based on the injection molded gear deformation prediction results, operational suggestions are given.

2. The deformation prediction method of an injection molded gear of an automobile wiper according to claim 1, characterized in that: The first deformation prediction model includes a tooth shape deformation prediction layer, a tooth surface deformation prediction layer, a correlation analysis layer and a result output layer; The tooth profile deformation prediction layer obtains preliminary tooth profile deformation data by analyzing the working environment data; The tooth surface deformation prediction layer analyzes the working environment data to obtain preliminary tooth surface deformation data; The correlation analysis layer analyzes the preliminary tooth shape deformation data, preliminary tooth surface deformation data, temperature and humidity through correlation analysis to obtain the tooth shape first deformation prediction data and the tooth surface first deformation prediction data; The result output layer obtains the first deformation prediction data of the injection gear by weighted sum processing of the first deformation prediction data of the tooth shape and the first deformation prediction data of the tooth surface.

3. The deformation prediction method of an injection molded gear for a car wiper according to claim 1, characterized in that: The calculation formula for the first deformation prediction data is: ； in, is the first deformation prediction data, is the first deformation prediction data of the tooth shape, is the first deformation prediction data of the tooth surface, is a multi-input multi-output weighted function, For standard injection gear data, For working environment data.

4. The deformation prediction method of the injection molded gear of an automobile wiper according to claim 1, characterized in that: The deformation prediction optimization model includes a data input layer, a model training layer and an optimized data output layer; The data input layer inputs the first deformation prediction data into the deep neural network; The model training layer performs preliminary training on the deep neural network by using historical working environment data to obtain a feature vector between the first deformation prediction data and the gear deformation; optimizes and updates the deep neural network by collecting working environment data in real time; the optimization and update is performed by calculating errors and adjusting weights and errors in the network; The optimized data output layer obtains an injection molded gear deformation optimization prediction data set by continuously optimizing and updating the deep neural network.

5. The deformation prediction method of the injection molded gear of a car wiper according to claim 1, characterized in that: The three-dimensional scanning data acquisition process is as follows: Scan the surface of the injection-molded gear multiple times with a laser beam to obtain multiple three-dimensional point cloud data of the injection-molded gear; De-noising and down-sampling the three-dimensional point cloud data; obtaining three-dimensional point cloud data of a complete gear by registering multiple three-dimensional point cloud data; Analyzing the three-dimensional point cloud data and standard injection molded gear data to generate injection molded gear deformation features; The deformation characteristics of the injection molded gear include deformation amplitude, local wear area and shape change trend; The deformation characteristics of the injection molded gear are measured and calculated to obtain an injection molded gear image deformation data set.

6. The deformation prediction method of the injection molded gear of a car wiper according to claim 1, characterized in that: The injection molded gear deformation prediction result generation process includes: Performing dimensionality reduction on the injection molded gear image deformation data set and the injection molded gear deformation optimization prediction data set by principal component analysis to generate a low-dimensional space data set; The data in the low-dimensional spatial data set are multi-dimensionally fused to obtain the injection molded gear deformation prediction result.

7. A deformation prediction system for injection molded gears of automobile wipers, characterized in that: include: A data acquisition unit, used to acquire the working environment data of the injection molded gear of the automobile wiper and perform preprocessing; The working environment data includes temperature, humidity, load, friction, operating frequency and working cycle; A first deformation prediction data acquisition unit is used to construct a first deformation prediction model based on the working environment data to simulate and analyze the force and deformation of the gear under complex working conditions, so as to obtain first deformation prediction data of the injection molded gear; The first deformation prediction data includes tooth shape first deformation prediction data and tooth surface first deformation prediction data; A prediction value optimization unit is used to construct a deformation prediction optimization model to analyze the working environment data, the tooth shape first deformation prediction data and the tooth surface first deformation prediction data to obtain an injection molded gear deformation optimization prediction data group; An image data acquisition unit is used to acquire high-precision three-dimensional scanning data of injection-molded gears, analyze the three-dimensional scanning data using an image processing algorithm, and extract deformation features of the injection-molded gears; and obtain an injection-molded gear image deformation data group based on the deformation features of the injection-molded gears; The deformation result prediction unit is used to perform multi-dimensional fusion on the injection molded gear image deformation data group and the injection molded gear deformation optimization prediction data group to generate an injection molded gear deformation prediction result; and provide operation suggestions based on the injection molded gear deformation prediction result.

8. The deformation prediction system for injection molded gears of automobile wipers according to claim 7, characterized in that: The deformation prediction optimization model includes a data input layer, a model training layer and an optimized data output layer; The data input layer inputs the first deformation prediction data into the deep neural network; The model training layer performs preliminary training on the deep neural network by using historical working environment data to obtain a feature vector between the first deformation prediction data and the gear deformation; optimizes and updates the deep neural network by collecting working environment data in real time; the optimization and update is performed by calculating errors and adjusting weights and errors in the network; The optimization data output layer obtains an injection molded gear deformation optimization prediction data set by continuously optimizing and updating the deep neural network; The three-dimensional scanning data acquisition process is as follows: Scan the surface of the injection-molded gear multiple times with a laser beam to obtain multiple three-dimensional point cloud data of the injection-molded gear; De-noising and down-sampling the three-dimensional point cloud data; obtaining three-dimensional point cloud data of a complete gear by registering multiple three-dimensional point cloud data; Analyzing the three-dimensional point cloud data and standard injection molded gear data to generate injection molded gear deformation features; The deformation characteristics of the injection molded gear include deformation amplitude, local wear area and shape change trend; Measuring and calculating the deformation characteristics of the injection molded gear to obtain an injection molded gear image deformation data set; The injection molded gear deformation prediction result generation process includes: Performing dimensionality reduction on the injection molded gear image deformation data set and the injection molded gear deformation optimization prediction data set by principal component analysis to generate a low-dimensional space data set; The data in the low-dimensional spatial data set are multi-dimensionally fused to obtain the injection molded gear deformation prediction result.

Images