Automobile sunroof sealing strip aging early warning method and system based on multi-source data

Abstract

The present application relates to the technical field of sealing strips, and provides an aging warning method and system for automobile sunroof sealing strips based on multi-source data. The method includes: reading the sunroof sealing strip code and the matching vehicle model; based on the code and vehicle model, analyzing the environmental impact on the aging of the sealing strip to obtain the environmental aging coefficient; according to the vehicle model, evaluating the degree of aging under mechanical action to obtain the mechanical aging coefficient; performing cracking risk analysis on the vehicle model, determining the key inspection area, and collecting real-time images; using the aging analysis network to evaluate the apparent aging state and obtain the apparent aging coefficient; through the preset weights, comprehensively calculating and analyzing the three coefficients to obtain the sealing strip aging warning information. The present application solves the technical problem that the existing method does not fully consider the impact of multiple factors on the automobile sunroof sealing strip, resulting in inaccurate aging warning of the automobile sunroof sealing strip, and realizes accurate aging warning based on multi-source data, improving the accuracy and timeliness of the warning.

Description

Technical Field

[0001] The present application relates to the technical field of automobile sunroof aging warning, specifically to the technical field of sealing strips, and in particular to an automobile sunroof sealing strip aging warning method and system based on multi-source data. Background Art

[0002] With the rapid development of the automobile industry and consumers' increasing attention to vehicle performance and comfort, the popularity of car sunroofs has increased year by year. As a key component to ensure the air tightness, water tightness and sound insulation of the vehicle, the quality and performance of the car sunroof sealing strip directly affect the vehicle's user experience. However, as the vehicle is used for a longer time, the sunroof sealing strip will inevitably age, such as hardening, cracking, and loss of elasticity, which will not only reduce the sealing performance, but may also cause problems such as water leakage and noise, seriously affecting the normal use of the vehicle. Existing aging warning methods often ignore the impact of multiple factors on the aging process of the sealing strip, resulting in inaccurate warning results and failure to timely reflect the actual aging status of the sealing strip. Summary of the invention

[0003] The present application provides a method and system for warning the aging of automobile sunroof sealing strips based on multi-source data, aiming to solve the technical problem that the existing methods fail to fully consider the impact of multiple factors on automobile sunroof sealing strips, resulting in inaccurate warning of automobile sunroof sealing strip aging.

[0004] In view of the above problems, the present application provides a method and system for early warning of aging of automobile sunroof sealing strips based on multi-source data.

[0005] In a first aspect disclosed in the present application, a method for aging warning of automobile sunroof sealing strips based on multi-source data is provided, the method comprising: reading a target coding identifier and a target adaptation vehicle model of a target sunroof sealing strip, wherein the target sunroof sealing strip is a sealing strip to be tested for aging installed on a target automobile sunroof; performing an environmental aging correlation analysis on the target sunroof sealing strip according to the target coding identifier and the target adaptation vehicle model to obtain an environmental aging coefficient; performing a mechanical aging correlation analysis on the target sunroof sealing strip according to the environmental aging coefficient and the target adaptation vehicle model to obtain a mechanical aging coefficient; performing a cracking region of interest analysis according to the target adaptation vehicle model to obtain a target region of interest set, performing image acquisition on the target sunroof sealing strip according to the target region of interest set to obtain a target real-time image set; pre-constructing an image aging analysis network, and performing an apparent aging correlation analysis on the target sunroof sealing strip based on the image aging analysis network and the target real-time image set to obtain an apparent aging coefficient; pre-setting a weight distribution, and performing a comprehensive calculation and analysis on the environmental aging coefficient, the mechanical aging coefficient and the apparent aging coefficient according to the preset weight distribution to obtain sealing strip aging warning information.

[0006] Another aspect disclosed in the present application provides an automobile sunroof sealing strip aging warning system based on multi-source data, the system comprising: a data reading module, the data reading module is used to read a target coding identifier and a target adaptation vehicle type of a target sunroof sealing strip, wherein the target sunroof sealing strip is a sealing strip to be tested for aging installed on a target automobile sunroof; an environmental correlation analysis module, the environmental correlation analysis module is used to perform an environmental aging correlation analysis on the target sunroof sealing strip according to the target coding identifier and the target adaptation vehicle type to obtain an environmental aging coefficient; a mechanical correlation analysis module, the mechanical correlation analysis module is used to perform a mechanical aging correlation analysis on the target sunroof sealing strip according to the environmental aging coefficient and the target adaptation vehicle type to obtain a mechanical aging coefficient. number; an image acquisition module, the image acquisition module is used to analyze the cracking interest area according to the target adaptation vehicle model, obtain a target interest area set, and collect images of the target sunroof sealing strip according to the target interest area set to obtain a target real-time image set; a performance correlation analysis module, the performance correlation analysis module is used for a pre-constructed image aging analysis network, and based on the image aging analysis network and the target real-time image set, the target sunroof sealing strip is subjected to apparent aging correlation analysis to obtain an apparent aging coefficient; a calculation and analysis module: the calculation and analysis module is used for preset weight allocation, and according to the preset weight allocation, the environmental aging coefficient, the mechanical aging coefficient and the apparent aging coefficient are comprehensively calculated and analyzed to obtain sealing strip aging warning information.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0008] The above-mentioned automobile sunroof sealing strip aging warning method based on multi-source data identifies the sealing strip installed on the target automobile sunroof, and reads the coding identification and the adapted vehicle model information. Subsequently, using this information, the sealing strip is subjected to environmental aging analysis, the degree of aging affected by environmental factors is evaluated, and the environmental aging coefficient is obtained. After that, according to the environmental aging coefficient and vehicle model information, the aging of the sealing strip in mechanical use is analyzed to obtain the mechanical aging coefficient. In order to more comprehensively understand the aging state of the sealing strip, based on the vehicle model information, the areas on the sealing strip that are most prone to cracking are determined, and real-time images of these areas are collected. These images are analyzed by a pre-constructed image analysis network to determine the apparent aging degree of the sealing strip, and the apparent aging coefficient is obtained. Then, the environmental aging coefficient, the mechanical aging coefficient and the apparent aging coefficient are comprehensively considered, and a comprehensive calculation is performed according to the preset weight distribution, so as to obtain the aging warning information of the sealing strip. This method can comprehensively and accurately evaluate the aging of the automobile sunroof sealing strip, provide timely warnings, and ensure the sealing performance of the sunroof.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. 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 creative work.

[0011] Figure 1 The figure is a flow chart of an automobile sunroof sealing strip aging early warning method based on multi-source data in one embodiment.

[0012] Figure 2 This is an architecture diagram of an automobile sunroof sealing strip aging warning system based on multi-source data in one embodiment.

[0013] Explanation of the accompanying drawings: data reading module 1, environmental correlation analysis module 2, mechanical correlation analysis module 3, image acquisition module 4, performance correlation analysis module 5, calculation analysis module 6. DETAILED DESCRIPTION

[0014] The embodiments of the present application provide a method and system for warning the aging of automobile sunroof sealing strips based on multi-source data, so as to solve the technical problem that the existing methods fail to fully consider the influence of various factors on automobile sunroof sealing strips, resulting in inaccurate warning of aging of automobile sunroof sealing strips.

[0015] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0016] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.

[0017] Embodiment 1

[0018] like Figure 1As shown, the present application provides an automobile sunroof sealing strip aging early warning method based on multi-source data, the method comprising:

[0019] The target coding identification and the target adaptation vehicle type of the target sunroof sealing strip are read, wherein the target sunroof sealing strip is a sealing strip to be tested for aging installed on the sunroof of a target vehicle.

[0020] In the automotive industry, the sunroof is an important component for improving the lighting and ventilation in the car. Its sealing performance is crucial to ensure driving comfort and safety. However, as the car ages, the sunroof seals often suffer varying degrees of damage due to environmental factors, mechanical wear and material aging, resulting in a decrease in sealing performance. This may not only lead to problems such as water leakage and increased noise in the car, but may also have a negative impact on the overall performance of the vehicle.

[0021] In an embodiment of the present application, in order to accurately evaluate the aging condition of the sealing strip, the system terminal first determines the target sunroof sealing strip, which refers to identifying the sealing strip installed on the car sunroof that needs to be tested for aging. Subsequently, the system terminal reads the target coding identification of the sealing strip, which is unique to each sealing strip and is used to identify the specific sealing strip model and specifications. At the same time, the system terminal also reads the target compatible vehicle model, because the sunroof sealing strips of different vehicle models may differ in design, material and use environment. By obtaining this information, the aging condition of the sealing strip can be analyzed more accurately, thereby improving the accuracy of the aging warning.

[0022] According to the target coding identifier and the target compatible vehicle model, an environmental aging correlation analysis is performed on the target sunroof sealing strip to obtain an environmental aging coefficient.

[0023] In one embodiment, in order to accurately evaluate the aging degree of the target sunroof sealing strip, the system terminal must not only consider the material properties of the sealing strip itself, but also deeply analyze the impact of the environment in which it is located on its aging process. By reading the target coding identification and the target adaptation vehicle model, a large amount of sample information can be obtained. This sample information contains information such as the car, environment, and aging status. Subsequently, based on this sample information, the system terminal constructs an environmental aging analysis model, which is used to perform environmental aging correlation analysis in combination with the target environmental information set to generate an environmental aging coefficient. This coefficient quantifies the contribution of environmental factors to the aging of the sealing strip, thereby more accurately evaluating the aging status of the sealing strip.

[0024] Furthermore, the present application provides a method for performing an environmental aging correlation analysis on the target sunroof sealing strip according to the target coding identifier and the target compatible vehicle model to obtain an environmental aging coefficient, including:

[0025] Vehicle matching is performed with the target coding identifier and the target adapted vehicle model as constraints to obtain a sample vehicle set, a sample initial environment information set and a sample aging information set; replacement node information of the target sunroof sealing strip is interactively obtained, and data alignment is performed on the sample initial environment information set according to the replacement node information to obtain a sample optimized environment information set.

[0026] Preferably, the system terminal selects a set of vehicles that match each other from the historical database based on the obtained target coding identifier and the target adaptation model, and obtains a sample car set, in which the car model and the sealing strip used are the same as the target adaptation model and the target coding identifier. Subsequently, the system terminal obtains the initial environmental information of the sealing strips installed on these vehicles, such as humidity, temperature, etc., as well as the corresponding aging information, and forms a sample initial environmental information set and a sample aging information set. After that, the system terminal interacts with the car terminal to obtain the replacement node information about the target sunroof sealing strip. This information includes the time point and reason for the replacement of the sealing strip. Then, the system terminal uses the replacement node information as a benchmark to align the sample initial environmental information set in time. For each sample car, select the environmental information data from the replacement node to the current time. In this way, the environmental information of all sample cars will take the replacement node as the starting point and have the same comparison benchmark, thereby obtaining a more accurate sample optimization environmental information set. The purpose of this process is to more accurately analyze the performance changes of the sunroof sealing strip, etc., so as to provide a scientific basis for subsequent aging warnings.

[0027] The sample optimization environment information set and the sample aging information set are used to pre-build an environment aging analysis model; and data of the target vehicle is called with the replacement node information as a constraint to obtain a target environment information set.

[0028] Preferably, when evaluating the aging of the automobile sunroof sealing strip, the system terminal first constructs an environmental aging analysis model for evaluating the degree to which the automobile sunroof sealing strip is affected by environmental factors. In order to construct this model, the system terminal uses the sample optimization environment information set and the sample aging information set to respectively construct the environment information visualization layer, the environment matching analysis layer and the aging information call layer, and cascades these three layers to construct an environmental aging analysis model. Subsequently, the system terminal uses the replacement node information as a constraint to call the target car for data and obtain the target environment information set. This information set is also the alignment data obtained based on the replacement node information, and contains various environmental factor data encountered by the target car during use, such as humidity, temperature, etc. In general, this process involves multiple steps such as data collection, processing, and model construction, aiming to provide scientific and accurate evaluation results for the aging of the automobile sunroof sealing strip through precise data analysis and model prediction.

[0029] The environmental aging analysis model and the target environmental information set are used to perform environmental aging correlation analysis on the target sunroof sealing strip to obtain the environmental aging coefficient.

[0030] Preferably, after the environmental aging analysis model is constructed, the system terminal performs environmental aging association analysis based on the environmental aging analysis model and the obtained target environmental information set. First, the system terminal inputs the target environmental information set into the environmental aging analysis model. The environmental information visualization layer inside the model converts these environmental data into the target environmental space based on these environmental data and the learned knowledge. Subsequently, the obtained target environmental space is combined with multiple sample optimized environmental spaces respectively, and the combined results are transmitted to the internal environmental matching analysis layer for spatial projection similarity calculation to obtain the sample similarity of each combined result. After that, the system terminal selects the largest one from these sample similarities, and selects an aging information corresponding to it from the sample aging information set according to the screening result, and then uses the same method as above to align the data according to the replacement node information to generate the environmental aging coefficient. This coefficient reflects the aging degree and speed of the target sunroof sealing strip under specific environmental conditions. In summary, environmental aging association analysis is the process of aging prediction and evaluation of the target sunroof sealing strip using the environmental aging analysis model and the target environmental information set. Through analysis, a quantitative environmental aging coefficient can be obtained, which can reflect the aging degree and speed of the sealing strip under specific conditions, providing a scientific basis for vehicle maintenance.

[0031] Furthermore, the present application provides a method for pre-building an environmental aging analysis model using the sample optimization environmental information set and the sample aging information set, and the method further includes:

[0032] The sample optimization environment information set includes multiple sample optimization environment information, wherein the multiple sample optimization environment information is associated and mapped with multiple sample cars in the sample car set, and each sample optimization environment information in the multiple sample optimization environment information sets includes a sample environment humidity sequence, a sample environment temperature sequence and a sample radiation intensity sequence; using the environment humidity, environment temperature and radiation intensity as spatial dimensions, constructing a standard environment space, and synchronizing the standard environment space to a pre-constructed environment information visualization layer; synchronizing the multiple sample optimization environment information to the standard environment space to obtain multiple sample optimization environment spaces.

[0033] Optionally, the sample optimization environment information set obtained by the system terminal contains multiple sample optimization environment information. These sample optimization environment information are associated with the sample car set, that is, each sample car corresponds to a sample optimization environment information, including a sample environment humidity sequence, a sample environment temperature sequence, and a sample radiation intensity sequence. These sequences record the changes in environment humidity, temperature, and radiation intensity at different time points. Subsequently, the system terminal clarifies the main dimensions of the standard environment space, namely, environment humidity, environment temperature, and radiation intensity, based on the number of categories of actual data. These three dimensions represent the main characteristics of the environment and are very important for analyzing the aging process of the sunroof sealing strip. Afterwards, according to business needs, define appropriate ranges and scales for each dimension. For example, the environment humidity ranges from 0% (completely dry) to 100% (saturated), and the environment temperature may be -40°C (severe cold) to +60°C (hot), etc. Then, a three-dimensional modeling tool is used to construct a standard environment space based on the defined dimension range and scale. This space is a three-dimensional coordinate system in which each axis represents a dimension. Further, the system terminal synchronizes the constructed standard environment space to the environment information visualization layer. This environmental information visualization layer is pre-built and used to convert the input data set into an environmental space. Specifically, the system terminal first virtualizes a three-dimensional environmental space, using temperature as the X-axis, humidity as the Y-axis, and radiation intensity as the Z-axis of this space. This three-dimensional environmental space is used to map the data set. Subsequently, the system terminal integrates the pre-written environmental data mapping algorithm with the three-dimensional environmental space to construct the environmental information visualization layer. The pre-written environmental data mapping algorithm can convert the environmental information set input into the environmental information visualization layer, that is, map the temperature, humidity, and radiation intensity at a moment into the three-dimensional environmental space to form coordinate points representing the environmental conditions at that moment.

[0034] After completing the synchronization of the standard environment space, the system terminal synchronizes multiple sample optimization environment information to the standard environment space. Since the standard environment space has been synchronized to the environment information visualization layer, the synchronization process will also be reflected on the visualization layer. By mapping these sample optimization environment information in the three-dimensional environment space through the environment information visualization layer, the system terminal can obtain multiple sample optimization environment spaces, each space corresponding to a sample optimization environment information. The display of these spaces on the environment information visualization layer can help the system terminal more intuitively compare and analyze the performance of different sample cars in different environments.

[0035] Pre-construct a spatial similarity analyzer, and synchronize the multiple sample optimization environment spaces and the spatial similarity analyzer to the environment matching analysis layer; pre-construct an aging information call layer, and map and store the sample optimization environment information set and the sample aging information set to the aging information call layer; cascade the environment information imaging layer, the environment matching analysis layer and the aging information call layer to generate the environment aging analysis model.

[0036] Optionally, after constructing the environmental information visualization layer, the system terminal begins to construct the basic framework of the environmental matching analysis layer, including an input layer, a hidden layer, and an output layer. Subsequently, the system terminal collects a large amount of three-dimensional data including humidity, temperature, and radiation intensity, as well as the corresponding similarity with the sample optimization environment space. After that, the system terminal hides the one-dimensional data in the three-dimensional data, that is, hides the radiation data by spatial projection, hides the dimension data of humidity from the three-dimensional data including humidity, temperature, and radiation intensity, and generates temperature-radiation intensity two-dimensional data. Then, feature extraction and data division are performed on the obtained temperature-radiation intensity two-dimensional data to obtain humidity hidden training data and humidity hidden test data, and the humidity hidden analysis branch is initialized using a neural network, including determining the number of neurons, activation function, loss function, etc. Then, the system terminal uses random numbers to initialize the weights and biases of the initial humidity hidden analysis branch, and then uses the humidity hidden training data to iteratively train this initial humidity hidden analysis branch. In each iteration, the forward propagation input data is used to generate the prediction output through the initial humidity hidden analysis branch, and the loss function value is calculated using the determined loss function, and then the back propagation algorithm is used to calculate the gradient and update the model parameters. After the training is completed, the system terminal uses the humidity hidden test data to evaluate the performance of the trained initial humidity hidden analysis branch and calculates relevant performance indicators, such as accuracy. Finally, the model is tuned according to the evaluation results. This includes adjusting the network structure, learning rate, number of hidden layer units, etc. If the adjustment is completed and the performance on the humidity test data is good, the system terminal outputs this initial humidity hidden analysis branch and generates a humidity hidden analysis branch.

[0037] Furthermore, the system terminal uses the same method to construct the temperature hidden analysis branch and the radiation hidden analysis branch respectively. After these hidden analysis branches are constructed, the system terminal begins to construct the comprehensive similarity calculation layer, which has built-in weight allocation rules and weighted summation algorithms. The weight allocation rules are formulated by the system terminal in advance based on historical experience and professional knowledge. For example, through the analysis of professional knowledge, the system terminal knows that the sealing strip is used to prevent moisture intrusion, so the system terminal will assign a larger weight to humidity. The weighted summation algorithm is constructed based on the weighted summation formula and is used to fuse the similarities generated by the three branches. Subsequently, the system terminal connects the humidity hidden analysis branch, the temperature hidden analysis branch, and the radiation hidden analysis branch in parallel to ensure that they can process data in parallel and generate their own feature representations. The three branches connected in parallel are uniformly connected to the comprehensive similarity calculation layer to form a spatial similarity analyzer. After that, the system terminal synchronizes the constructed spatial similarity analyzer to the basic framework of the environment matching analysis layer to form the final environment matching analysis layer.

[0038] After building the environment matching analysis layer, the system terminal builds the aging information call layer based on the extreme value matching algorithm and the information extraction algorithm, and maps and stores the sample optimization environment information set and the sample aging information set in the aging information call layer. Among them, the extreme value matching algorithm and the information extraction algorithm are pre-written. The extreme value matching algorithm compares multiple similarities and obtains the largest similarity. The information extraction algorithm traces back to the corresponding sample optimization environment information set based on this similarity, and extracts the corresponding aging information from the sample aging information set based on the sample car identification in the sample optimization environment information set.

[0039] Finally, the system terminal cascades the environmental information visualization layer, environmental matching analysis layer, and aging information call layer, that is, connects them according to the control logic of the data flow to ensure that the data can be transmitted and processed between the layers according to the predetermined path and order, thereby generating a complete environmental aging analysis model. The workflow of this model is: the input environmental data is converted into a three-dimensional environmental space through the environmental information visualization layer, and then the similarity with the optimized environmental space of each sample is calculated through the environmental matching analysis layer, and then the aging information call layer provides the aging information with the highest similarity based on the calculated similarity. Through such a model design, the system terminal can realize the rapid analysis of environmental aging conditions and provide a scientific basis for the early warning of sealing strip aging.

[0040] Further, the present application provides performing environmental aging correlation analysis on the target sunroof sealing strip using the environmental aging analysis model and the target environmental information set to obtain the environmental aging coefficient, and the method further includes:

[0041] The target environment information set is abstracted into a target environment space through the environment information visualization layer of the environmental aging analysis model; a spatial projection combination is performed on the target environment space and the first sample optimized environment space to obtain a first environment humidity hidden group, a first environment temperature hidden group and a first radiation intensity hidden group, wherein the first sample optimized environment space is any one of the multiple sample optimized environment spaces; the first environment humidity hidden group, the first environment temperature hidden group and the first radiation intensity hidden group are synchronized to the spatial similarity analyzer of the environment matching analysis layer to perform spatial projection similarity calculation to obtain a first sample similarity, wherein the spatial similarity analyzer includes a humidity hidden analysis branch, a temperature hidden analysis branch, a radiation hidden analysis branch and a comprehensive similarity calculation layer.

[0042] Optionally, the system terminal inputs the target environment information set into the environmental aging analysis model. The target environment information set first converts the redundant original environment data into a three-dimensional environment space that is easier to understand and analyze through the processing of the environment information visualization layer, thereby obtaining the target environment space. Subsequently, the system terminal randomly extracts one from the generated multiple sample optimized environment spaces as the first sample optimized environment space. After that, three blank objects are created, namely the first environment humidity hidden object, the first environment temperature hidden object and the first radiation intensity hidden object. After creating the three blank objects, the system terminal maps the humidity, temperature and radiation intensity in the target environment space and the first sample optimized environment space to the corresponding hidden objects, that is, the humidity data in the target environment space and the first sample optimized environment space are mapped to the first environment humidity hidden object to form the first environment humidity hidden group. By analogy, the system terminal can obtain the first environment humidity hidden group, the first environment temperature hidden group and the first radiation intensity hidden group. Then, the system terminal synchronizes the obtained first environment humidity hidden group, first environment temperature hidden group and first radiation intensity hidden group to the environment matching analysis layer, and the spatial similarity analyzer inside the environment matching analysis layer inputs the obtained first environment humidity hidden group, first environment temperature hidden group and first radiation intensity hidden group into the corresponding hidden analysis branches, namely, the humidity hidden analysis branch, the temperature hidden analysis branch and the radiation hidden analysis branch, respectively, to calculate the similarity. Finally, the spatial similarity analyzer passes the similarity output by the three hidden analysis branches to the comprehensive similarity calculation layer to fuse the similarity and generate the first sample similarity.

[0043] By analogy, multiple sample similarities of the multiple sample optimization environment spaces are calculated based on the environment matching analysis layer; a maximum call is performed at the aging information calling layer according to the multiple sample similarities, and target aging information is extracted from the sample aging information set based on the call result; data alignment is performed on the target aging information according to the replacement node information to obtain the environmental aging coefficient.

[0044] Optionally, after completing the similarity calculation between the target environment space and the first sample optimized environment space, the system terminal performs the same operation on the target environment space and all other sample optimized environment spaces. The purpose of this is to comprehensively evaluate the similarity between the target environment space and all known optimized environment conditions. Specifically, the system terminal repeatedly performs steps such as spatial projection combination and model analysis to obtain multiple sample similarities, each of which represents the similarity between the target environment space and the corresponding sample optimized environment space. Subsequently, in the aging information call layer, the system terminal performs a maximum call based on these sample similarity scores, that is, the obtained multiple sample similarities are arranged in descending order, and the first sample similarity value is extracted from this as the maximum call result. Once the sample optimized environment space with the highest similarity is determined, the aging information call layer traces back based on this maximum call result, that is, consults the log record to understand the sample optimized environment space on which this maximum call result is based. After that, the aging information call layer extracts the corresponding target aging information from the sample aging information set according to the car identification of the sample optimized environment space, and outputs it. Then, the system terminal uses the replacement node information to align the target aging information, and then extracts the point where the fault occurs from the aligned data, and uses the value corresponding to this point as the environmental aging coefficient. This is because the target aging information is a smooth curve. When the performance of the sealing strip drops to a certain level, a breakpoint will appear on the curve. This breakpoint corresponds to the time point when the sealing strip ages and needs to be replaced, so the value corresponding to this point is the environmental aging coefficient. Through data alignment, it can be ensured that the environmental aging coefficient can accurately reflect the aging of the sealing strip under the target environmental conditions, thereby providing strong support for subsequent early warnings.

[0045] According to the environmental aging coefficient and the target compatible vehicle model, a mechanical aging correlation analysis is performed on the target sunroof sealing strip to obtain a mechanical aging coefficient.

[0046] In one embodiment, in order to obtain the mechanical aging coefficient, the system terminal performs correlation analysis between the environmental aging coefficient and the target adaptation vehicle model. Specifically, the system terminal uses the environmental aging coefficient as a constraint, screens out a sample environmental aging coefficient set that meets this constraint, as well as a corresponding sample sunroof usage frequency set and a sample mechanical aging coefficient set, and then extracts the sunroof usage frequency of the target adaptation vehicle model. Subsequently, a mechanical aging analysis model is constructed using these sample data, and the extracted sunroof usage frequency is input into this model to calculate a mechanical aging coefficient, which reflects the speed and degree of mechanical aging of the sealing strip due to the mechanical movement of the sunroof in a specific vehicle model and environment.

[0047] Further, the present application provides a method for performing a mechanical aging correlation analysis on the target sunroof sealing strip according to the environmental aging coefficient and the target adapted vehicle model to obtain a mechanical aging coefficient, and the method further includes:

[0048] Taking the environmental aging coefficient as a constraint, a sample environmental aging coefficient set, a sample skylight usage frequency set and a sample mechanical aging coefficient set are interactively obtained; and a mechanical aging analysis model is constructed using the sample environmental aging coefficient set, the sample skylight usage frequency set and the sample mechanical aging coefficient set.

[0049] Preferably, after obtaining the environmental aging coefficient, the system terminal interacts with the historical database to extract a large number of historical environmental aging coefficients, as well as the corresponding historical skylight usage frequency and historical mechanical aging coefficient. Subsequently, the system terminal uses the environmental aging coefficient as a constraint to retrieve the historical environmental aging coefficient, screen out the historical environmental aging coefficient that meets the constraint, and organize the historical environmental aging coefficient that meets the constraint into a sample environmental aging coefficient set. After that, the historical skylight usage frequency and historical mechanical aging coefficient corresponding to these historical environmental aging coefficients are sorted to generate a sample skylight usage frequency set and a sample mechanical aging coefficient set. After collecting enough sample data, the system terminal uses these data to build a mechanical aging analysis model. Specifically, the system terminal first designs the architecture of the neural network, including an input layer, a hidden layer, and an output layer. The number of hidden layers and the number of neurons in each layer are determined according to the complexity of the problem. Subsequently, the sample environmental aging coefficient set, the sample skylight usage frequency set, and the sample mechanical aging coefficient set are divided into a training sample set, a verification sample set, and a test sample set. The training sample set is used to train the neural network model, and the network weights are updated by the back propagation algorithm and the gradient descent algorithm. During the training process, the validation sample set is used to monitor the performance of the model in order to perform early stopping or adjust hyperparameters such as the learning rate. After the training is completed, the system terminal uses the test sample set to evaluate the performance of the trained neural network model. Evaluation indicators include accuracy, mean square error, etc. If the evaluation results do not meet the expected requirements, the system terminal adjusts the learning rate, batch size, number of iterations, etc. to improve the model. The model training and evaluation process is repeated until satisfactory model performance is obtained. Once the model training is completed and meets the performance requirements, the system terminal outputs the neural network model to generate a mechanical aging analysis model.

[0050] The target sunroof usage frequency is obtained by interacting with the target vehicle; the target sunroof usage frequency and the environmental aging coefficient are synchronized to the mechanical aging analysis model to perform mechanical aging correlation analysis to obtain the mechanical aging coefficient.

[0051] Preferably, the system terminal interacts with the target car to obtain the frequency of use of its sunroof. This interactive process is achieved through the vehicle-mounted system. The frequency of use of the sunroof is one of the important parameters for evaluating the degree of mechanical aging of the sunroof sealing strip, because it reflects the wear of the sunroof in daily use. Subsequently, the obtained target sunroof usage frequency and environmental aging coefficient are synchronized to the constructed mechanical aging analysis model. The mechanical aging analysis model performs a mechanical aging correlation analysis on the input target sunroof usage frequency and environmental aging coefficient based on the learned knowledge, and outputs a mechanical aging coefficient, which represents the mechanical aging state of the target sunroof sealing strip under the current environment and usage conditions. This coefficient is of great significance for the early warning of sealing strip aging.

[0052] An analysis of cracking regions of interest is performed according to the target adapted vehicle model to obtain a target region of interest set, and images of the target sunroof sealing strip are collected according to the target region of interest set to obtain a target real-time image set.

[0053] In one embodiment, the system terminal first determines in which areas of the sunroof sealing strip of the target adapted vehicle model are prone to cracking problems. This involves steps such as acquiring sample replacement cracking images, integrating sample replacement cracking images, and segmenting sample replacement cracking images. Through these steps, the system terminal can determine which areas belong to the target areas of interest. These areas may be the parts where the sealing strip contacts the sunroof frame, or areas that are greatly affected by the external environment. Once the target area of ​​interest set is determined, the system terminal can start image acquisition of these areas to obtain a target real-time image set, which contains real-time images of the target sunroof sealing strip in the area of ​​interest. These images will serve as an important basis for subsequent analysis and judgment of whether the sealing strip has cracking problems.

[0054] Further, the present application provides performing cracking interest region analysis according to the target adapted vehicle model to obtain a target interest region set, performing image acquisition on the target sunroof sealing strip according to the target interest region set to obtain a target real-time image set, and the method further includes:

[0055] The target adaptation vehicle model is used as a constraint to perform an online call to obtain multiple sample replacement crack images of multiple sample adaptation vehicles; an image alignment vector is preset, and the multiple sample replacement crack images are overlapped based on the image alignment vector to obtain a sample integrated crack image.

[0056] Preferably, the system terminal retrieves and obtains the replacement cracking images of multiple sample adapted cars related to the model by calling the online database online based on the specific parameters and characteristics of the target adapted model. These images represent the cracking of the sunroof sealing strip under different circumstances. Subsequently, in order to effectively integrate and analyze these images from different sample vehicles, the system terminal presets an image alignment vector. This vector is set based on historical experience and defines the feature points or areas that should be aligned between images to ensure that the images can be accurately aligned when overlapping. After that, the preset image alignment vector is used to overlap multiple sample replacement cracking images. This process involves operations such as image transformation, scaling, and rotation to ensure that the cracked parts in each image can accurately correspond and be superimposed together. Then, after overlapping processing, the system terminal obtains a sample integrated cracking image. This image integrates the cracking conditions of multiple sample vehicles, which can provide a more comprehensive and intuitive understanding and analysis for the system terminal.

[0057] A crack density threshold is preset, and a plurality of crack regions of interest are obtained from the sample integrated crack image according to the crack density threshold, wherein the plurality of crack regions of interest constitute the target region of interest set; and image acquisition is performed on the target sunroof sealing strip according to the target region of interest set to obtain the target real-time image set.

[0058] Preferably, the system terminal presets a crack density threshold value based on the historical data of cracking of the sunroof sealing strip and the expert experience. This threshold value is used to quantify the severity of cracking and help identify the area with more obvious cracking in the image. Subsequently, the system terminal uses this crack density threshold value to analyze the sample integrated cracking image, automatically detects the cracking area in the image, calculates the ratio of the number of images with cracking in each area to the number of all images, and obtains the crack density of each area. After that, the calculated crack density threshold value is compared with the crack density threshold, and the crack density greater than or equal to the crack density threshold is screened out, and the areas represented by these crack densities are segmented to divide multiple cracking interest areas to form a target interest area set. These areas are the parts of the image where the crack density exceeds the threshold and are also the focus of subsequent attention. Then, the image acquisition process of the target sunroof sealing strip is guided according to the target interest area set. The real-time image of the target sunroof sealing strip is obtained by a high-definition camera, and the parts corresponding to the crack interest area are focused on. In this way, the system terminal can ensure that the acquired image can cover all key areas on the sunroof sealing strip where cracking problems may occur. Finally, through the image acquisition process, the system terminal obtains a target real-time image set. This image set contains real-time images of all cracked areas of interest on the target sunroof seal, providing important data support for subsequent analysis and early warning.

[0059] A pre-constructed image aging analysis network, and performing an apparent aging correlation analysis on the target sunroof sealing strip based on the image aging analysis network and the target real-time image set to obtain an apparent aging coefficient;

[0060] In one embodiment, the system terminal collects a large number of sample aging and cracking images and sample aging coefficients, and segments these sample aging and cracking images according to multiple cracking interest regions. Subsequently, for the segmentation results of the same region, the system terminal constructs a regional aging analysis branch for the region. Similarly, the same operation is performed on the remaining regions to construct multiple regional aging analysis branches. Afterwards, these regional aging analysis branches are connected to the apparent aging weighted layer to construct an image aging analysis network. After completing the construction of the image aging analysis network, the system terminal inputs the target real-time image set into the network, performs apparent aging association analysis, and obtains the apparent aging coefficient. This coefficient is a quantitative indicator used to indicate the visual aging degree of the target sunroof sealing strip. The high or low coefficient directly reflects the aging state of the sealing strip. For example, a higher coefficient means that the sealing strip has shown obvious signs of aging, such as an increase in cracks.

[0061] Furthermore, the present application provides a pre-constructed image aging analysis network, and based on the image aging analysis network and the target real-time image set, performs an apparent aging correlation analysis on the target sunroof sealing strip to obtain an apparent aging coefficient, and the method further includes:

[0062] The target adaptation vehicle model is used as a constraint to perform an online call to obtain a plurality of sample aging and cracking images and a plurality of sample aging coefficients; the plurality of sample aging and cracking images are divided using the plurality of cracking regions of interest as a constraint to obtain a plurality of groups of sample region of interest images; and a plurality of region aging analysis branches are constructed using the plurality of groups of sample region of interest images and the plurality of sample aging coefficients.

[0063] Optionally, the system terminal uses the target adaptation vehicle model as a constraint condition, and retrieves and obtains sample aging cracking images and corresponding sample aging coefficients of multiple sample adaptation vehicles related to the vehicle model by making an online call to the online database. These sample data provide the system terminal with rich reference information for building and training the aging analysis model. Subsequently, the system terminal uses the identified multiple cracking interest regions as constraints to divide the acquired multiple sample aging cracking images. This process actually extracts the part corresponding to the cracking interest region in each sample image to form multiple groups of sample interest region images. This allows the model to analyze and learn the aging characteristics of these specific areas more attentively and improve the accuracy of the analysis. Afterwards, the system terminal uses these multiple groups of sample interest region images and their corresponding sample aging coefficients to construct multiple regional aging analysis branches. These branches are important components of the aging analysis model, and they can perform independent analysis and prediction for each interest region. The training method is the same as the aforementioned training mechanical aging analysis model. By training these branches, the model can learn the aging characteristics of different regions and their relationship with the aging coefficient, thereby achieving a comprehensive assessment of the aging degree of the sunroof sealing strip.

[0064] The multiple regional aging analysis branches are connected in parallel, and the output ends of the multiple regional aging analysis branches are connected to a pre-constructed apparent aging weighted layer to complete the construction of the image aging analysis network, wherein the apparent aging weighted layer pre-stores a plurality of regional weights obtained by weighting the multiple cracking regions of interest; and an apparent aging correlation analysis is performed on the target sunroof sealing strip based on the image aging analysis network and the target real-time image set to obtain the apparent aging coefficient.

[0065] Optionally, the system terminal connects the constructed multiple regional aging analysis branches in parallel. These branches each focus on analyzing different cracking interest regions on the sunroof sealing strip and output their own evaluation results of the aging degree of the region. Subsequently, the system terminal connects the output ends of these branches with a pre-constructed apparent aging weighted layer to construct an image aging analysis network. This weighted layer is constructed based on a weighted summation algorithm, and pre-stores weight assignments for multiple cracking interest regions, i.e., multiple regional weights, during construction. These weights are set according to the importance of each region in the aging assessment of the sealing strip, and they can ensure that appropriate attention is given to different regions during analysis. Afterwards, the system terminal uses this network to perform apparent aging association analysis on the target sunroof sealing strip. The target real-time image set is input into the network, and each regional aging analysis branch in the network will process the corresponding interest region in the image respectively and output its own aging evaluation result. Then, these results are weighted by the apparent aging weighted layer, and the aging of each region is comprehensively considered to finally obtain an overall apparent aging coefficient.

[0066] A preset weight distribution is performed, and the environmental aging coefficient, the mechanical aging coefficient and the apparent aging coefficient are comprehensively calculated and analyzed according to the preset weight distribution to obtain sealing strip aging warning information.

[0067] In one embodiment, when evaluating the aging of the sunroof sealing strip, the system terminal comprehensively considers multiple factors, including environmental aging, mechanical aging, and apparent aging. In order to comprehensively consider these factors and obtain a comprehensive aging warning information, the system terminal presets a weight distribution. These preset weights are determined according to the relative importance of different aging factors in the aging process of the sealing strip, and are formulated according to expert advice. After obtaining the environmental aging coefficient, mechanical aging coefficient, and apparent aging coefficient, the system terminal multiplies each aging coefficient by its corresponding weight according to the preset weight distribution, and then adds the products to obtain a comprehensive aging evaluation value. Subsequently, the system terminal compares this comprehensive aging evaluation value with the set risk level to determine the current risk level. Afterwards, this comprehensive aging evaluation value and the corresponding risk level are sorted into sealing strip aging warning information and output. This sealing strip aging warning information combines the aging conditions of the three aspects of environment, mechanics, and appearance, and can more comprehensively reflect the actual aging degree of the sealing strip. In this way, the aging problem of the sealing strip can be discovered in time, and corresponding measures can be taken to ensure the normal use and safety performance of the sunroof.

[0068] In summary, the embodiments of the present application have at least the following technical effects:

[0069] The embodiment of the present application reads the coding identification and the compatible vehicle model of the target sunroof sealing strip, and then obtains the environmental aging coefficient and the mechanical aging coefficient through environmental aging correlation analysis and mechanical aging correlation analysis. Subsequently, the cracking region of interest analysis is performed to obtain the target region of interest set, and the target real-time image set is collected. After that, the apparent aging correlation analysis of the target sunroof sealing strip is performed using the pre-built image aging analysis network to obtain the apparent aging coefficient. Then, the sealing strip aging warning information is obtained by comprehensive calculation and analysis based on the preset weight distribution. These technical effects jointly solve the technical problem that the existing method does not fully consider the impact of multiple factors on the automobile sunroof sealing strip, resulting in inaccurate aging warning of the automobile sunroof sealing strip, and realizes accurate aging warning based on multi-source data, thereby improving the accuracy and timeliness of the warning.

[0070] Embodiment 2

[0071] Based on the same inventive concept as the automobile sunroof sealing strip aging early warning method based on multi-source data in the aforementioned embodiment, Figure 2 As shown, the present application provides an automobile sunroof sealing strip aging early warning system based on multi-source data, the system comprising:

[0072] Data reading module 1: the data reading module 1 is used to read the target coding identification and the target adaptation vehicle model of the target sunroof sealing strip, wherein the target sunroof sealing strip is a sealing strip to be installed on the sunroof of a target car to be tested for aging.

[0073] Environmental correlation analysis module 2: The environmental correlation analysis module 2 is used to perform environmental aging correlation analysis on the target sunroof sealing strip according to the target coding identifier and the target compatible vehicle model to obtain an environmental aging coefficient.

[0074] Mechanical correlation analysis module 3: The mechanical correlation analysis module 3 is used to perform mechanical aging correlation analysis on the target sunroof sealing strip according to the environmental aging coefficient and the target adaptation vehicle model to obtain the mechanical aging coefficient.

[0075] Image acquisition module 4: the image acquisition module 4 is used to analyze the cracking region of interest according to the target adaptation vehicle model, obtain a target region of interest set, and acquire images of the target sunroof sealing strip according to the target region of interest set to obtain a target real-time image set.

[0076] Performance correlation analysis module 5: The performance correlation analysis module 5 is used for pre-constructing an image aging analysis network, and performs an apparent aging correlation analysis on the target sunroof sealing strip based on the image aging analysis network and the target real-time image set to obtain an apparent aging coefficient.

[0077] Calculation and analysis module 6: The calculation and analysis module 6 is used to preset weight distribution, and perform comprehensive calculation and analysis on the environmental aging coefficient, mechanical aging coefficient and apparent aging coefficient according to the preset weight distribution to obtain sealing strip aging warning information.

[0078] Furthermore, the environment association analysis module 2 is also used to perform the following method:

[0079] Vehicle matching is performed with the target coding identifier and the target adapted vehicle model as constraints to obtain a sample vehicle set, a sample initial environmental information set and a sample aging information set; replacement node information of the target sunroof sealing strip is interactively obtained, and data alignment of the sample initial environmental information set is performed according to the replacement node information to obtain a sample optimized environmental information set; an environmental aging analysis model is pre-built using the sample optimized environmental information set and the sample aging information set; data is called for the target vehicle with the replacement node information as a constraint to obtain a target environmental information set; environmental aging association analysis is performed on the target sunroof sealing strip using the environmental aging analysis model and the target environmental information set to obtain the environmental aging coefficient.

[0080] Furthermore, the environment association analysis module 2 is also used to perform the following method:

[0081] The sample optimization environment information set includes multiple sample optimization environment information, wherein the multiple sample optimization environment information is associated and mapped with multiple sample cars in the sample car set, and each sample optimization environment information in the multiple sample optimization environment information sets includes a sample environment humidity sequence, a sample environment temperature sequence and a sample radiation intensity sequence; using the environment humidity, environment temperature and radiation intensity as spatial dimensions, constructing a standard environment space, and synchronizing the standard environment space to a pre-constructed environment information visualization layer; synchronizing the multiple sample optimization environment information to the standard environment space to obtain multiple sample optimization environment spaces; pre-constructing a spatial similarity analyzer, and synchronizing the multiple sample optimization environment spaces and the spatial similarity analyzer to an environment matching analysis layer; pre-constructing an aging information calling layer, and mapping and storing the sample optimization environment information set and the sample aging information set to the aging information calling layer; cascading the environment information visualization layer, the environment matching analysis layer and the aging information calling layer to generate the environment aging analysis model.

[0082] Furthermore, the environment association analysis module 2 is also used to perform the following method:

[0083] The target environment information set is abstracted into a target environment space through the environment information visualization layer of the environment aging analysis model; a spatial projection combination is performed on the target environment space and the first sample optimized environment space to obtain a first environment humidity hidden group, a first environment temperature hidden group and a first radiation intensity hidden group, wherein the first sample optimized environment space is any one of the multiple sample optimized environment spaces; the first environment humidity hidden group, the first environment temperature hidden group and the first radiation intensity hidden group are synchronized to the spatial similarity analyzer of the environment matching analysis layer to perform spatial projection similarity calculation to obtain a first sample similarity, wherein the spatial similarity analyzer includes a humidity hidden analysis branch, a temperature hidden analysis branch, a radiation hidden analysis branch and a comprehensive similarity calculation layer; and so on, multiple sample similarities of the multiple sample optimized environment spaces are calculated based on the environment matching analysis layer; a maximum call is performed in the aging information call layer according to the multiple sample similarities, and the target aging information is extracted and obtained in the sample aging information set based on the call result; data alignment is performed on the target aging information according to the replacement node information to obtain the environment aging coefficient.

[0084] Furthermore, the mechanical correlation analysis module 3 is also used to perform the following method:

[0085] Taking the environmental aging coefficient as a constraint, interactively obtain a sample environmental aging coefficient set, a sample sunroof usage frequency set and a sample mechanical aging coefficient set; use the sample environmental aging coefficient set, the sample sunroof usage frequency set and the sample mechanical aging coefficient set to build a mechanical aging analysis model; interactively obtain a target sunroof usage frequency for the target vehicle; synchronize the target sunroof usage frequency and the environmental aging coefficient to the mechanical aging analysis model to perform mechanical aging correlation analysis to obtain the mechanical aging coefficient.

[0086] Furthermore, the image acquisition module 4 is also used to execute the following method:

[0087] A network call is performed with the target adaptation vehicle model as a constraint to obtain multiple sample replacement crack images of multiple sample adaptation vehicles; an image alignment vector is preset, and the multiple sample replacement crack images are overlapped based on the image alignment vector to obtain a sample integrated crack image; a crack density threshold is preset, and multiple crack interest regions are obtained from the sample integrated crack image according to the crack density threshold, wherein the multiple crack interest regions constitute the target interest region set; and image acquisition is performed on the target sunroof sealing strip according to the target interest region set to obtain the target real-time image set.

[0088] Furthermore, the image acquisition module 4 is also used to execute the following method:

[0089] The target adaptive vehicle model is used as a constraint to perform a network call to obtain a plurality of sample aging and cracking images and a plurality of sample aging coefficients; the plurality of sample aging and cracking images are divided based on the plurality of cracking regions of interest as a constraint to obtain a plurality of groups of sample region of interest images; a plurality of region aging analysis branches are constructed using the plurality of groups of sample region of interest images and the plurality of sample aging coefficients; the plurality of region aging analysis branches are connected in parallel, and the output ends of the plurality of region aging analysis branches are connected to a pre-constructed apparent aging weighted layer to complete the construction of the image aging analysis network, wherein the apparent aging weighted layer pre-stores a plurality of region weights obtained by weighting the plurality of cracking regions of interest; an apparent aging association analysis is performed on the target sunroof sealing strip based on the image aging analysis network and the target real-time image set to obtain the apparent aging coefficient.

[0090] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0091] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0092] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. An early warning method for aging of automobile sunroof sealing strips based on multi-source data, characterized in that: The method comprises: Reading a target code identification and a target adaptation vehicle model of a target sunroof sealing strip, wherein the target sunroof sealing strip is a sealing strip to be tested for aging installed on a target vehicle sunroof; According to the target coding identifier and the target adapted vehicle model, an environmental aging correlation analysis is performed on the target sunroof sealing strip to obtain an environmental aging coefficient; According to the environmental aging coefficient and the target adaptation vehicle model, a mechanical aging correlation analysis is performed on the target sunroof sealing strip to obtain a mechanical aging coefficient; Perform cracking interest region analysis according to the target adaptation vehicle model to obtain a target interest region set, and perform image acquisition on the target sunroof sealing strip according to the target interest region set to obtain a target real-time image set; A pre-constructed image aging analysis network, and performing an apparent aging correlation analysis on the target sunroof sealing strip based on the image aging analysis network and the target real-time image set to obtain an apparent aging coefficient; A preset weight distribution is performed, and the environmental aging coefficient, the mechanical aging coefficient and the apparent aging coefficient are comprehensively calculated and analyzed according to the preset weight distribution to obtain sealing strip aging warning information.

2. The method according to claim 1, characterized in that According to the target coding identifier and the target compatible vehicle model, an environmental aging correlation analysis is performed on the target sunroof sealing strip to obtain an environmental aging coefficient, and the method further includes: Vehicle matching is performed with the target coding identifier and the target adapted vehicle model as constraints to obtain a sample vehicle set, a sample initial environment information set, and a sample aging information set; interactively obtaining replacement node information of the target sunroof sealing strip, and performing data alignment on the sample initial environment information set according to the replacement node information to obtain a sample optimized environment information set; Pre-constructing an environmental aging analysis model using the sample optimization environmental information set and the sample aging information set; Using the replacement node information as a constraint, data of the target vehicle is called to obtain a target environment information set; The environmental aging analysis model and the target environmental information set are used to perform environmental aging correlation analysis on the target sunroof sealing strip to obtain the environmental aging coefficient.

3. The method according to claim 2, characterized in that The sample optimization environment information set and the sample aging information set are used to pre-build an environment aging analysis model, and the method further includes: The sample optimization environment information set includes a plurality of sample optimization environment information, wherein the plurality of sample optimization environment information is associated and mapped with a plurality of sample cars in the sample car set, and each sample optimization environment information in the plurality of sample optimization environment information sets includes a sample environment humidity sequence, a sample environment temperature sequence and a sample radiation intensity sequence; Using environmental humidity, environmental temperature and radiation intensity as spatial dimensions, constructing a standard environmental space, and synchronizing the standard environmental space to a pre-constructed environmental information imaging layer; Synchronizing the plurality of sample optimization environment information to the standard environment space to obtain a plurality of sample optimization environment spaces; Pre-building a spatial similarity analyzer, and synchronizing the plurality of sample optimization environment spaces and the spatial similarity analyzer to an environment matching analysis layer; Pre-constructing an aging information calling layer, and mapping and storing the sample optimization environment information set and the sample aging information set to the aging information calling layer; The environmental information imaging layer, the environmental matching analysis layer and the aging information calling layer are cascaded to generate the environmental aging analysis model.

4. The method according to claim 3, characterized in that Using the environmental aging analysis model and the target environmental information set to perform environmental aging correlation analysis on the target sunroof sealing strip to obtain the environmental aging coefficient, the method further includes: The target environment information set is abstracted into a target environment space via the environment information visualization layer of the environment aging analysis model; Performing spatial projection combination on the target environment space and the first sample optimized environment space to obtain a first environment humidity hidden group, a first environment temperature hidden group and a first radiation intensity hidden group, wherein the first sample optimized environment space is any one of the multiple sample optimized environment spaces; Synchronizing the first environment humidity hidden group, the first environment temperature hidden group and the first radiation intensity hidden group to the spatial similarity analyzer of the environment matching analysis layer to perform spatial projection similarity calculation to obtain a first sample similarity, wherein the spatial similarity analyzer includes a humidity hidden analysis branch, a temperature hidden analysis branch, a radiation hidden analysis branch and a comprehensive similarity calculation layer; By analogy, multiple sample similarities of the multiple sample optimization environment spaces are obtained based on the environment matching analysis layer calculation; Performing maximum value calling according to the similarities of the multiple samples at the aging information calling layer, and extracting target aging information from the sample aging information set based on the calling result; The target aging information is data aligned according to the replacement node information to obtain the environmental aging coefficient.

5. The method according to claim 1, characterized in that Performing cracking interest region analysis according to the target adaptation vehicle model to obtain a target interest region set, performing image acquisition on the target sunroof sealing strip according to the target interest region set to obtain a target real-time image set, the method further comprising: Performing a network call with the target adaptation vehicle model as a constraint to obtain multiple sample replacement cracking images of multiple sample adaptation vehicles; Preset an image alignment vector, and overlap the plurality of sample replacement crack images based on the image alignment vector to obtain a sample integrated crack image; Presetting a crack density threshold, and obtaining a plurality of crack regions of interest from the sample integrated crack image according to the crack density threshold, wherein the plurality of crack regions of interest constitute the target region of interest set; The target sunroof sealing strip is imaged according to the target region of interest set to obtain the target real-time image set.

6. The method according to claim 5, characterized in that A pre-constructed image aging analysis network is used, and based on the image aging analysis network and the target real-time image set, an apparent aging correlation analysis is performed on the target sunroof sealing strip to obtain an apparent aging coefficient. The method further comprises: Performing an online call with the target adaptation vehicle model as a constraint to obtain multiple sample aging and cracking images and multiple sample aging coefficients; Dividing the plurality of sample aging and cracking images using the plurality of cracking regions of interest as constraints to obtain a plurality of groups of sample region of interest images; Constructing a plurality of regional aging analysis branches using the plurality of groups of sample region of interest images and the plurality of sample aging coefficients; The plurality of regional aging analysis branches are connected in parallel, and the output ends of the plurality of regional aging analysis branches are connected to a pre-constructed apparent aging weighted layer, so as to complete the construction of the image aging analysis network, wherein the apparent aging weighted layer pre-stores a plurality of regional weights obtained by weighting the plurality of cracking regions of interest; Based on the image aging analysis network and the target real-time image set, an apparent aging correlation analysis is performed on the target sunroof sealing strip to obtain the apparent aging coefficient.

7. The method according to claim 1, characterized in that According to the environmental aging coefficient and the target adaptation vehicle model, a mechanical aging correlation analysis is performed on the target sunroof sealing strip to obtain a mechanical aging coefficient. The method further includes: Taking the environmental aging coefficient as a constraint, interactively obtaining a sample environmental aging coefficient set, a sample skylight usage frequency set, and a sample mechanical aging coefficient set; A mechanical aging analysis model is constructed using the sample environmental aging coefficient set, the sample skylight usage frequency set and the sample mechanical aging coefficient set; Interacting with the target car to obtain a target sunroof usage frequency; The target skylight usage frequency and the environmental aging coefficient are synchronized to the mechanical aging analysis model to perform mechanical aging correlation analysis to obtain the mechanical aging coefficient.

8. The automobile sunroof sealing strip aging early warning system based on multi-source data is characterized by: The steps for implementing the automobile sunroof sealing strip aging early warning method based on multi-source data as described in any one of claims 1 to 7 include: Data reading module: reads the target code identification and target adaptation vehicle model of the target sunroof sealing strip, wherein the target sunroof sealing strip is the sealing strip to be tested for aging installed on the sunroof of the target vehicle; Environmental correlation analysis module: performing environmental aging correlation analysis on the target sunroof sealing strip according to the target coding identifier and the target adaptation vehicle model to obtain an environmental aging coefficient; Mechanical correlation analysis module: performing mechanical aging correlation analysis on the target sunroof sealing strip according to the environmental aging coefficient and the target adaptation vehicle model to obtain a mechanical aging coefficient; Image acquisition module: performing cracking interest region analysis according to the target adaptation vehicle model to obtain a target interest region set, and performing image acquisition on the target sunroof sealing strip according to the target interest region set to obtain a target real-time image set; Performance correlation analysis module: a pre-constructed image aging analysis network, and performing an apparent aging correlation analysis on the target sunroof sealing strip based on the image aging analysis network and the target real-time image set to obtain an apparent aging coefficient; Calculation and analysis module: preset weight distribution, and perform comprehensive calculation and analysis on the environmental aging coefficient, mechanical aging coefficient and apparent aging coefficient according to the preset weight distribution to obtain sealing strip aging warning information.

Images