CN118537329A - Method and system for detecting assembly quality of vehicle chassis - Google Patents

Abstract

The present invention discloses a method and system for detecting the assembly quality of a vehicle chassis, which belongs to the field of quality detection, wherein the method comprises: numbering the components of a target vehicle chassis, dividing the components into modules, and establishing a standard module; constructing a network diagram of the target vehicle chassis; assigning weights to the edges in the network diagram, and constructing a comprehensive evaluation function; establishing a standard zero point, taking the standard zero point as a reference, activating a laser scanning device to perform measurements of each module, and generating a detection data set; establishing an additional data set; inputting the detection data set and the additional data set into the comprehensive evaluation function to perform assembly quality evaluation; and performing assembly management of the target vehicle chassis according to the assembly quality evaluation result. The present application solves the technical problems of one-sided assembly quality detection and inaccurate evaluation of vehicle chassis in the prior art, and achieves the technical effect of comprehensive quality detection and accurate evaluation of the assembly quality of vehicle chassis.

Description

Assembly quality detection method and system for vehicle chassis

Technical Field

The present invention relates to the field of quality inspection, and in particular to an assembly quality inspection method and system for a vehicle chassis.

Background Art

In the process of vehicle manufacturing, the vehicle chassis is an important assembly that supports the vehicle body and passengers, and its assembly quality directly affects the safety, reliability and service life of the vehicle. In the prior art, the assembly quality is evaluated by measuring and testing each module of the chassis separately. For example, the geometric dimensions of the frame are tested, and the alignment status of the suspension system is measured. Although this modular detection method can find some quality problems, it lacks a systematic analysis of the mutual influence and connection relationship between the modules of the chassis, and it is difficult to comprehensively evaluate the overall assembly quality of the chassis assembly. In addition, most of the existing assembly quality detection methods focus on several main performance indicators such as geometric dimensions and alignment status, and do not pay enough attention to other important indicators such as connection strength and functional coordination, and fail to fully reflect the overall assembly quality level of the vehicle chassis. It can be seen that the vehicle chassis assembly quality detection method in the prior art is still insufficient in terms of systematicity and comprehensiveness, resulting in one-sided vehicle chassis assembly quality detection and inaccurate evaluation.

Summary of the invention

The present application provides a method and system for detecting the assembly quality of a vehicle chassis, aiming to solve the technical problem in the prior art that the vehicle chassis assembly quality detection method lacks systematic and comprehensive evaluation, resulting in one-sided vehicle chassis assembly quality detection and inaccurate evaluation.

In view of the above problems, the present application provides a method and system for detecting the assembly quality of a vehicle chassis.

The first aspect disclosed in the present application provides an assembly quality detection method for a vehicle chassis, the method comprising: numbering the components of a target vehicle chassis, dividing the components into modules based on the component numbering results, and establishing a standard module; obtaining the module interface parameters of the standard module, and taking the modules of the target vehicle chassis as nodes in a graph, performing mutual influence and connection analysis of the nodes based on the module interface parameters, and constructing a network graph of the target vehicle chassis, wherein each node includes at least one associated node, and the associated node is an influence node; assigning weights to the edges in the network graph of the target vehicle chassis, the weights representing the degree of influence of one node on another node, calculating the overall influence of each node based on the path analysis in graph theory, and constructing a comprehensive evaluation function through the overall influence; establishing a standard zero point, using the standard zero point as a reference, activating a laser scanning device to perform each module measurement, and generating a detection data set, wherein the execution of each module measurement includes geometric dimension measurement and alignment detection; performing connection strength and functional coordination tests by combining sensors, and establishing an additional data set; inputting the detection data set and the additional data set into the comprehensive evaluation function to perform assembly quality evaluation; and performing assembly management of the target vehicle chassis according to the assembly quality evaluation result.

Another aspect disclosed in the present application provides an assembly quality inspection system for a vehicle chassis, the system comprising: a standard module establishment component for numbering the components of a target vehicle chassis, and performing module division based on the component numbering results to establish a standard module; a network diagram construction component for obtaining module interface parameters of the standard module, and using the modules of the target vehicle chassis as nodes in the diagram, performing mutual influence and connection analysis of the nodes based on the module interface parameters, and constructing a network diagram of the target vehicle chassis, wherein each node contains at least one associated node, and the associated node is an influence node; an evaluation function construction component for assigning weights to the edges in the network diagram of the target vehicle chassis, and the weights represent the influence of one node on another node. The overall impact of each node is calculated based on the path analysis in graph theory, and a comprehensive evaluation function is constructed through the overall impact; a module measurement execution component is used to establish a standard zero point, and with the standard zero point as the benchmark, the laser scanning device is activated to perform each module measurement and generate a detection data set, wherein the execution of each module measurement includes geometric dimension measurement and alignment detection; an additional data acquisition component is used to perform connection strength and functional coordination tests through joint sensors and establish additional data sets; an assembly quality evaluation component is used to perform assembly quality evaluation based on the detection data set and the additional data set input into the comprehensive evaluation function; a chassis assembly management component is used to perform assembly management of the target vehicle chassis according to the assembly quality evaluation results.

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

The target vehicle chassis is numbered and module division is performed based on the result of the part numbering to establish a standard module. Through the part numbering and module division, the complex structure of the vehicle chassis is decomposed into several standard modules, laying the foundation for the subsequent systematic analysis; the module interface parameters of the standard module are obtained, and the modules of the target vehicle chassis are used as nodes in the graph. The mutual influence and connection analysis of each node is performed based on the module interface parameters to construct the target vehicle chassis network graph, in which each node contains at least one associated node, and the associated node is an influencing node. By analyzing the module interface parameters, the mutual influence and connection relationship between the modules of the chassis are revealed, and the target vehicle chassis network graph is constructed to achieve a systematic analysis of the chassis assembly quality; weights are assigned to the edges in the target vehicle chassis network graph, and the weights represent the degree of influence of one node on another node. The overall influence of each node is calculated based on the path analysis in graph theory, and a comprehensive evaluation function is constructed through the overall influence to provide a tool for quantitatively evaluating the assembly quality; a standard zero point is established, and the laser scanning device is activated to perform the measurement of each module based on the standard zero point to generate a detection data set. Among them, the execution of each module measurement includes geometric dimension measurement and alignment detection, and the use of laser scanning technology to automatically collect the geometric dimension and alignment status data of each module of the chassis to obtain an objective and accurate detection data set; the connection strength and functional coordination test is carried out by joint sensors, and an additional data set is established to supplement the connection strength and functional coordination data reflecting the assembly quality and enrich the evaluation dimension; based on the detection data set and the additional data set input into the comprehensive evaluation function, the assembly quality evaluation is performed, and quantitative analysis and calculation are output to output accurate and comprehensive assembly quality evaluation results; according to the assembly quality evaluation results, the assembly management of the target vehicle chassis is carried out, and the systematic and comprehensive assembly quality evaluation results are fed back in time to guide the optimization of the chassis assembly process and quality management, and improve the product quality and production efficiency. The technical solution solves the technical problem that the vehicle chassis assembly quality detection method lacks systematic and comprehensive evaluation in the prior art, resulting in one-sided and inaccurate evaluation of the vehicle chassis assembly quality. By combining the measurement data of the laser scanning device and the additional data of the joint sensor, the technical effect of comprehensive quality detection and accurate evaluation of the vehicle chassis assembly quality is achieved.

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

FIG1 is a schematic flow chart of a method for detecting the assembly quality of a vehicle chassis according to an embodiment of the present application;

FIG2 is a schematic structural diagram of an assembly quality inspection system for a vehicle chassis provided in an embodiment of the present application.

Explanation of the reference numerals: standard module establishment component 11, network diagram construction component 12, evaluation function construction component 13, module measurement execution component 14, additional data acquisition component 15, assembly quality evaluation component 16, chassis assembly management component 17.

DETAILED DESCRIPTION

The overall idea of the technical solution provided by this application is as follows:

The embodiments of the present application provide a method and system for detecting the assembly quality of a vehicle chassis.

First, the vehicle chassis is systematically divided into modules and standard modules are established. Secondly, the mutual influence and connection relationship between the modules are analyzed to construct the network diagram of the target vehicle chassis. Then, weights are assigned to the edges of the network diagram of the target vehicle chassis to construct a comprehensive evaluation function. Subsequently, a laser scanning device and a joint sensor are used to collect geometric dimensions, alignment status, connection strength and functional coordination data reflecting the assembly quality in multiple dimensions to obtain a detection data set and an additional data set. Afterwards, the obtained detection data set and additional data set are input into the comprehensive evaluation function for quantitative analysis to obtain accurate and comprehensive assembly quality evaluation results, and the evaluation results are fed back and applied to the assembly management of the target vehicle chassis. Compared with the prior art, the present application more systematically and comprehensively quantitatively evaluates the overall assembly quality of the vehicle chassis from multiple dimensions. The evaluation results are objective and accurate, effectively reflecting the assembly quality level, and providing a reliable basis for the assembly management of the vehicle chassis.

After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically described below in conjunction with the drawings in the specification.

Embodiment 1

As shown in FIG1 , an embodiment of the present application provides a method for detecting the assembly quality of a vehicle chassis, the method comprising:

S1: Number the target vehicle chassis, divide the modules based on the part numbering results, and establish standard modules.

Specifically, the target vehicle chassis refers to a specific vehicle chassis that needs to be inspected for assembly quality. It integrates multiple systems such as wheels, suspension, transmission, steering, etc., which affect the driving performance and safety of the car. The target vehicle chassis refers to the chassis of a specific vehicle model, such as the chassis corresponding to different types of vehicles such as sedans, SUVs, and trucks. The chassis of different models will differ in structural layout, size, component composition, etc., so it is necessary to carry out assembly quality inspection on the specific target vehicle chassis.

First, the target vehicle chassis is numbered to clearly identify and distinguish the various components on the chassis, which is convenient for subsequent module division and quality inspection. Next, the vehicle chassis is divided into modules based on the results of the component numbering. Module division is to classify components with similar functions, adjacent positions or close assembly relationships into the same module, and each module can contain one or more components. Through module division, the complex vehicle chassis is simplified into several relatively independent and interrelated modules, which is conducive to analyzing the interface relationship and mutual influence between modules. After the module division is completed, a standard module is established. The standard module refers to the unified module division scheme and technical parameters of each module for the target vehicle chassis, such as the module's geometric dimensions, material properties, connection methods, etc. The establishment of the standard module provides a basis for judgment and benchmark for subsequent assembly quality inspection.

Through the establishment of part numbering, module division and standard modules, the foundation is laid for the assembly quality inspection of the vehicle chassis, which helps to identify the key components of the chassis, clarify the internal structure, and facilitate subsequent inspection and analysis work.

S2: Obtain the module interface parameters of the standard module, and use the modules of the target vehicle chassis as nodes in the graph, perform mutual influence and connection analysis of each node based on the module interface parameters, and construct a network graph of the target vehicle chassis, wherein each node contains at least one associated node, and the associated node is an influencing node.

Specifically, first, the module interface parameters of the standard module are obtained. The module interface parameters describe the connection characteristics between adjacent modules, such as connection position, connection method (bolt connection, welding, riveting, etc.), interface size, etc. By obtaining the module interface parameters of the standard module, the requirements and limitations on the spatial position and assembly relationship between modules can be mastered. Then, each module of the target vehicle chassis is used as a node of the network diagram. The node represents the component module of the chassis and is the basic unit of the target vehicle chassis network diagram. Constructing the target vehicle chassis network diagram with modules as nodes can intuitively express the overall structure and internal composition of the chassis.

Next, based on the module interface parameters, each node is analyzed for mutual influence and connection. The mutual position and connection relationship between the modules determined by the interface parameters are used to determine whether there is mutual influence between the modules. On this basis, the nodes of the two modules with influence association are connected with a line (i.e., the edge in the target vehicle chassis network diagram). The direction of the line indicates the direction of influence, and the assembly quality of the starting module will affect the end module. By constructing the target vehicle chassis network diagram, in the network diagram, each node (module) has an influence association with at least one other node (module), forming an associated node. The associated node directly or indirectly affects the assembly quality of the current node.

By utilizing the module interface parameters and constructing the target vehicle chassis network diagram in the form of nodes and edges, the internal correlation structure of the target vehicle chassis is vividly described, the mutual influence and constraint relationship between modules is revealed, and an intuitive analysis tool is provided for the subsequent improvement of assembly quality.

S3: Assign weights to the edges in the target vehicle chassis network graph, wherein the weights represent the degree of influence of one node on another node, calculate the overall influence on each node based on path analysis in graph theory, and construct a comprehensive evaluation function based on the overall influence.

Specifically, first, weights are assigned to the edges in the network graph of the target vehicle chassis. The weight represents the degree of influence of a node on another module node. The greater the degree of influence, the higher the weight value. The size of the weight can be quantitatively assigned according to factors such as the connection strength, force transmission path, and tolerance matching between modules. By assigning weights, the relative importance of different influence associations can be distinguished. Next, based on the path analysis method in graph theory, the overall influence of each node module is calculated. Through path analysis, all connection paths from one node to another are found, and the weight values on each path are accumulated to obtain the overall influence of the node, which reflects the degree of comprehensive influence of other modules on the assembly quality of a module. Finally, based on the overall influence of each module, a comprehensive evaluation function is constructed to quantitatively judge the overall assembly quality of the target vehicle chassis. The comprehensive evaluation function comprehensively considers the quality characteristics of each module and the mutual influence between modules, and obtains a quantitative index representing the assembly quality level of the entire chassis through weighted average or other mathematical methods.

Through weight distribution and path analysis, the influence transfer between modules in the network diagram is quantified, and on this basis, a comprehensive evaluation function is constructed to transform the qualitative network diagram into a quantitative mathematical model, providing a basis and evaluation criteria for subsequent assembly quality evaluation and optimization decisions.

S4: Establish a standard zero point, and use the standard zero point as a reference to activate the laser scanning device to perform measurement of each module and generate a detection data set, wherein the measurement of each module includes geometric dimension measurement and alignment detection.

Specifically, first, establish a standard zero point, which is a virtual origin of the coordinate system, as a reference for measuring the geometric dimensions and spatial positions of each module. For example, select a certain feature point on the chassis (such as a fixing bolt hole, a positioning pin hole, etc.) as the standard zero point, and establish a corresponding coordinate system (such as a rectangular coordinate system). The establishment of the standard zero point unifies the starting point and direction of the measurement, ensuring the consistency and comparability of data acquisition. Then, based on the standard zero point, activate the laser scanning device to measure each module. Laser scanning is a non-contact three-dimensional measurement technology that scans the surface of an object with a laser beam to obtain a large amount of point cloud data, and then reconstructs a three-dimensional digital model of the object. Among them, the laser scanning device scans each module of the target vehicle chassis one by one according to the preset scanning path and scanning parameters to obtain the surface morphology and spatial coordinate data of the module.

Two types of measurements are performed during the laser scanning process: geometric dimension measurement and alignment detection. Geometric dimension measurement refers to obtaining the length, width, height, aperture and other dimensional parameters of the module to determine whether it meets the design standards; alignment detection refers to measuring the relative position and posture between modules, such as parallelism, verticality, coaxiality, etc., to determine whether the assembly of the module meets the positioning requirements. By combining geometric dimension measurement and alignment detection, the processing quality and assembly quality of the module can be comprehensively evaluated. After the laser scanning device completes the measurement, a detection data set is generated. The detection data set contains information such as point cloud data, geometric dimension parameters and relative position relationship of each module, which is the basis for assembly quality evaluation.

Through laser scanning technology, the geometric parameters and spatial position data of each module of the target vehicle chassis are collected to establish a complete detection data set, converting the physical chassis into digital measurement data, providing basic input for subsequent comprehensive evaluation. At the same time, the introduction of geometric dimension measurement and alignment detection enhances the comprehensiveness and accuracy of assembly quality detection.

S5: Build additional datasets by testing the connection strength and functional coordination of the combined sensors.

Specifically, a joint sensor is used to test the connection strength and functional coordination of the module connection parts. The joint sensor is a composite sensor that integrates multiple sensor elements and can measure multiple physical quantity parameters at the same time. The joint sensor includes but is not limited to torque sensors, ultrasonic sensors, force sensors, etc., which are used to obtain the mechanical performance parameters of the module connection parts.

By testing the target vehicle chassis with joint sensors, additional data sets are established, including: the tightening torque size and change curve of fasteners (such as bolts, nuts, etc.); the strength, sealing and integrity of permanent connections such as welding and bonding; the force distribution and deformation of the connection parts under actual loads, etc. The additional data set reflects the force characteristics and mechanical properties of the module connection parts, which is the basis for judging the assembly quality. Together with the detection data set, the additional data set constitutes the complete input for the comprehensive evaluation of assembly quality. The additional data set supplements the mechanical performance data in addition to the geometric size and spatial position information, making the judgment of assembly quality more comprehensive and reliable.

By introducing the joint sensor, the mechanical performance parameters of the connection parts of the chassis module of the target vehicle were obtained, and an additional data set was established, which expanded the dimension of assembly quality inspection, broke through the limitations of geometric inspection, and revealed the factors affecting assembly quality from a mechanical perspective. The acquisition of additional data sets provides richer and more reliable inputs for the comprehensive evaluation function, improving the accuracy of assembly quality assessment.

S6: Based on the detection data set and the additional data set, the comprehensive evaluation function is input to perform assembly quality evaluation.

Specifically, the inspection dataset and the additional dataset are used as inputs to the comprehensive evaluation function to obtain the assembly quality evaluation result. Among them, the comprehensive evaluation function considers two aspects: one is the quality characteristics of the module itself, and the other is the mutual influence between modules.

The module's own quality characteristics, including dimensional accuracy, surface quality, material properties, etc., are obtained through the detection data set. The comprehensive evaluation function quantifies and scores the quality characteristics of each module to obtain the module's own quality score. The mutual influence between modules, including connection strength, matching accuracy, force transmission, etc., are obtained through additional data sets. The comprehensive evaluation function quantifies and scores the interaction influence data between modules to obtain the module interaction influence score.

After obtaining the module's own quality score and the module's interaction score, the comprehensive evaluation function uses a weighted summation method to average the two scores according to the preset weights to obtain a comprehensive evaluation value of the assembly quality. The weight setting needs to comprehensively consider factors such as the design requirements, process level, and operating conditions of the chassis assembly, and is determined through expert experience, data analysis, and other methods.

The calculation result of the comprehensive evaluation function can be expressed in percentage or grade to intuitively reflect the quality of assembly. When the comprehensive evaluation value exceeds the preset threshold, the assembly quality is judged to be qualified; when the comprehensive evaluation value is lower than the preset threshold, the assembly quality is judged to be unqualified, and cause analysis and improvement are required.

By constructing a comprehensive evaluation function and integrating the two dimensions of module quality and module interaction, the assembly quality of the target vehicle chassis is quantitatively evaluated. The test data and additional data are fully utilized, qualitative problems are converted into quantitative analysis, and objective and accurate assembly quality evaluation results are obtained.

S7: Perform assembly management of the target vehicle chassis according to the assembly quality evaluation result.

Specifically, the assembly quality evaluation result reflecting the overall assembly quality of the chassis was obtained through the calculation of the comprehensive evaluation function, which can not only judge whether the assembly quality is qualified or not, but also reveal the weak links and potential risks in the assembly process.

After obtaining the assembly quality evaluation results, use the assembly quality evaluation results to implement targeted management of the assembly process and continuously improve the assembly quality level. For example, based on the assembly quality evaluation results, identify chassis with unqualified assembly quality or low scores, trace their assembly process, find out the causes of quality problems, such as material defects, process parameter deviations, personnel operating errors, etc., to provide a basis for subsequent improvements; for the problems found in quality tracing, combine the assembly process regulations and operating instructions to optimize the assembly process, adjust the process parameters, improve the working methods, and improve the assembly quality and efficiency of key processes; according to the assembly quality evaluation results, set the quality warning threshold, and when the evaluation results of multiple consecutive chassis are lower than the warning value, issue a warning in time and activate the emergency plan to prevent the expansion of quality problems.

By applying the assembly quality evaluation results to the improvement of the assembly process, a closed-loop management model driven by quality data is formed, which can quickly discover and solve assembly quality problems, continuously optimize the assembly process and work flow, and achieve continuous improvement and enhancement of assembly quality.

Furthermore, the comprehensive evaluation function is configured, and the formula is as follows:

Among them, Global Score is the comprehensive evaluation value, i represents any module, n is the total number of modules, _Pi is the self-score of module i, which is calculated by docking accuracy, j represents any module different from module i, _Iij is the inter-module influence score, which is calculated by connection strength and functional coordination, _Pj is the self-score of module j, _αi is the self-score weight of module i, and _βij is the influence score weight of module j on module i.

Global Score represents the comprehensive evaluation value of assembly quality, which is a quantitative index between 0 and 100. The higher the comprehensive evaluation value, the better the assembly quality. The comprehensive evaluation function consists of three parts. The first part Represents the weighted sum of the quality scores of all modules. Where i represents any module, n is the total number of modules, _Pi represents the quality score of module i, which is obtained by calculating the docking accuracy of module i, and αi _is the weight coefficient of module i's own score, which represents the importance of module i's own quality to the overall assembly quality. Part II It represents the weighted sum of the quality scores after considering the interaction between modules, where j represents any module different from module i. _{I ij} represents the impact score of module j on module i, which is calculated by the connection strength and functional coordination of the connection parts of the two modules. β _ij is the weight coefficient of the impact score of module j on module i, which represents the degree of influence of module j on the assembly quality of module i. P _j represents the quality score of module j itself. Part III is a normalization factor, which makes the comprehensive evaluation value range between 0 and 100. The comprehensive evaluation function comprehensively considers the assembly quality of each chassis module and the interaction between modules. Through the weighted summation method, a quantitative index reflecting the assembly quality of the entire chassis is obtained, which overcomes the subjectivity and experience dependence of traditional qualitative evaluation and provides an objective and accurate means of assembly quality evaluation.

Among them, the weight coefficients α _i and β _ij in the comprehensive evaluation function are reasonably set according to the specific chassis structure, assembly process and other factors. The value of the weight coefficient is obtained through expert experience, test data analysis, machine learning and other methods to reflect the actual contribution of different modules and influencing factors to assembly quality.

By configuring a comprehensive evaluation function, the process of chassis assembly quality evaluation is quantified, providing an efficient and accurate quality evaluation method that makes full use of the acquired test data and additional data, taking into account both the module's own quality and the impact between modules, and providing support for evaluating the assembly level of vehicle chassis.

Furthermore, the embodiment of the present application also includes:

The torque sensor in the combined sensor is called to perform torque monitoring of the tightening process through the torque sensor to construct a first monitoring data set; the ultrasonic sensor in the combined sensor is called to perform welding, bonding and integrity tests on each module to establish a second monitoring data set; the force sensor in the combined sensor is called to perform connection point load measurement based on the force sensor to establish a third monitoring data set; the connection strength is evaluated based on the first monitoring data set, the second monitoring data set and the third monitoring data set, and the connection strength evaluation result is used as an additional data set.

In a feasible implementation, the combined sensor includes a torque sensor, an ultrasonic sensor and a force sensor. First, the torque sensor in the combined sensor is used to monitor the torque of fasteners (such as bolts, nuts, etc.) during the assembly process. The torque sensor measures the tightening torque of the fasteners in real time to obtain the curve data of the torque changing with time. By analyzing the torque data, the installation quality of the fasteners is judged, such as whether the predetermined tightening torque is reached, whether there is abnormal torque fluctuation, etc. The torque monitoring data is constructed into a first monitoring data set as a basis for evaluating the connection strength of the fasteners. At the same time, the ultrasonic sensor in the combined sensor is used to perform non-destructive testing on the welding and bonding parts of each module. The ultrasonic sensor transmits and receives high-frequency sound waves, measures the propagation time, attenuation and other parameters of the sound waves in the measured object, and thus detects the internal quality of the weld and the bonding surface, such as whether there are defects such as pores, cracks, and unfused. The ultrasonic detection data is constructed into a second monitoring data set as a basis for evaluating the strength and integrity of the welding and bonding connection. At the same time, the force sensor in the combined sensor is used to measure the load distribution of the chassis module connection point. The force sensor measures mechanical parameters such as stress and strain at the connection point to obtain the distribution characteristics of the load under different working conditions. By analyzing the load data, it is evaluated whether the stress state of the connection point is uniform and reasonable, and whether there are risks such as stress concentration and overload. The load measurement data is constructed into a third monitoring data set as a basis for evaluating the rationality of the load distribution at the connection point.

Afterwards, the first, second and third monitoring data sets are combined to comprehensively evaluate the connection strength of the chassis module. The connection strength evaluation includes: the fastening quality of the fasteners, the integrity of the welding and bonding parts, the uniformity of the load distribution at the connection points, etc. By analyzing and comparing the three types of monitoring data, the connection strength evaluation results reflecting the connection strength are obtained as an additional data set for subsequent comprehensive evaluation of assembly quality.

Through the combined monitoring methods of torque sensors, ultrasonic sensors, force sensors, etc., the mechanical performance data of the chassis module connection parts were obtained, and the connection strength was evaluated based on these data. Finally, the connection strength evaluation results were used as an additional data set to provide data support for the comprehensive evaluation of assembly quality.

Furthermore, the embodiment of the present application also includes:

The displacement sensor in the joint sensor is called to perform relative position changes of each module in motion under a preset motion scheme to construct a fourth monitoring data set; the acceleration sensor and the vibration analysis sensor in the joint sensor are called to construct a fifth monitoring data set based on the acceleration sensor and the vibration analysis sensor; after time-series alignment of the fourth monitoring data set and the fifth monitoring data set, time series analysis is performed to generate time series analysis results; the module vibration data in the fifth monitoring data set is called to perform main frequency component analysis of the module vibration data, and vibration coordination results are generated according to the frequency resonance results; the functional coordination evaluation results are constructed through the time series analysis results and the vibration coordination results, and the functional evaluation results and the connection strength evaluation results are used as additional data sets.

In a preferred embodiment, the combined sensor further includes a displacement sensor, an acceleration sensor and a vibration analysis sensor. The displacement sensor in the combined sensor is used to measure the relative position change of the chassis module under a preset motion scheme (such as driving, steering, braking and other working conditions). The displacement sensor obtains a displacement-time curve reflecting the relative motion of the module by monitoring the relative linear displacement or angular displacement between the modules. By analyzing the displacement data, the motion coordination between the modules is evaluated, such as whether there are abnormal phenomena such as relative slippage and disconnection. The displacement monitoring data is constructed into a fourth monitoring data set. The acceleration sensor and vibration analysis sensor in the combined sensor are used to measure the acceleration and vibration response of the chassis module under a preset motion scheme. The acceleration sensor measures the magnitude and direction of the dynamic acceleration of the module to obtain an acceleration-time curve; the vibration analysis sensor measures the vibration frequency, amplitude and other parameters of the module to obtain a vibration spectrum diagram. By analyzing the acceleration and vibration data, the dynamic performance of the module is evaluated, such as whether there are abnormal impacts, resonance and other problems. The acceleration and vibration monitoring data are constructed into a fifth monitoring data set.

Then, the fourth monitoring data set and the fifth monitoring data set are time-series aligned so that the data of different sensors are synchronized on the time axis. On this basis, time series analysis methods such as correlation analysis and spectrum analysis are used to reveal the dynamic characteristics such as motion correlation and transfer characteristics between different modules, and generate time series analysis results. At the same time, the module vibration data in the fifth monitoring data set are analyzed separately, and the main frequency components of the vibration signal are extracted using spectrum analysis methods. By comparing the main frequencies of each module, it is determined whether there are frequency coupling, resonance and other phenomena. Frequency coupling indicates that there is vibration energy exchange between modules, which may cause resonance problems; frequency separation indicates that the module vibration is relatively independent and is not easy to resonate. According to the frequency resonance results, the vibration coordination evaluation between modules is performed to generate vibration coordination results. Subsequently, the functional coordination evaluation results between chassis modules are constructed by comprehensively considering the time series analysis results and the vibration coordination evaluation. The content of the functional coordination evaluation includes the motion synchronization, power transmission characteristics, vibration coupling degree and so on between modules. Finally, the functional coordination evaluation results are used together with the aforementioned connection strength evaluation results as additional data sets to provide more comprehensive and in-depth data support for the comprehensive evaluation of assembly quality.

Furthermore, the embodiment of the present application also includes:

An image acquisition unit is called to perform image acquisition of a target vehicle chassis under a preset light source to construct a multi-scale image set; contours of the multi-scale image set are annotated through edge computing, and a convolutional neural network is called to perform feature recognition on the multi-scale image set with contour annotations; authentication of the inspection data corresponding to the assembly quality evaluation result is performed according to the feature recognition result, and the assembly quality evaluation result is updated according to the authentication result.

In a feasible implementation, first, the target vehicle chassis is imaged at multiple scales using an image acquisition unit under preset light source conditions. The preset light source may be visible light, infrared light, etc., which is used to provide stable and uniform lighting conditions. The image acquisition unit may be an industrial camera, a linear array camera, etc., and chassis images at different scales (such as the entire vehicle, a part, details, etc.) are obtained by changing the imaging distance or the focal length of the lens. Images of different scales are constructed into a multi-scale image set as input data for subsequent visual analysis. Then, the multi-scale image set is contour labeled and feature recognized. For example, using edge computing technology, the contour extraction and annotation of the image is completed on the local processor of the image acquisition unit. The contour extraction algorithm (such as the Canny operator) can find the edge contour in the image and mark it as a closed curve. After that, the multi-scale image set with contour annotation is input into the pre-trained convolutional neural network to perform the feature recognition task. The convolutional neural network automatically extracts multi-level features (such as texture, shape, structure, etc.) of the image through operations such as convolution, pooling, and activation, and classifies or recognizes the image content based on these features. After that, the feature recognition results of the convolutional neural network are used to visually authenticate and update the assembly quality evaluation results. The chassis features identified by the convolutional neural network are compared with the standard template to determine whether the appearance quality of the chassis is qualified, such as whether there are defects such as deformation, damage, and foreign matter. The authentication results are compared with the obtained assembly quality evaluation results. If the two are consistent, the authentication is passed; if there are differences, the reasons need to be further analyzed and the assembly quality evaluation results updated if necessary. Through visual authentication, the accuracy of the assembly quality evaluation can be verified, and intuitive image basis can be provided to facilitate the tracing and improvement of quality problems.

Furthermore, the embodiment of the present application also includes:

The target vehicle chassis is subjected to secondary segmentation based on standard modules, and an acquisition scale is configured based on the secondary segmentation result, wherein the first segmentation in the secondary segmentation is an area segmentation based on the size of the standard module, and the second segmentation in the secondary segmentation is an area segmentation based on the connection position of the standard module; an image acquisition unit is called through the acquisition scale to execute image acquisition and construct a multi-scale image set.

In a feasible implementation, when constructing a multi-scale image set, first, the target vehicle chassis is subjected to secondary segmentation based on the standard module to generate two levels of segmentation regions, and the image acquisition scale is configured accordingly. The first segmentation is a regional segmentation based on the size of the standard module, which divides the chassis into several sub-regions corresponding to the size of the standard module. This segmentation method takes into account the modular composition characteristics of the chassis, ensuring that each segmentation region can cover the complete standard module, which is convenient for subsequent feature extraction and quality assessment. The second segmentation is a regional segmentation based on the connection position of the standard module. On the basis of the first segmentation, each standard module is further divided into several sub-regions containing key connection positions. This segmentation method focuses on the connection parts between modules, such as bolt holes, welds, flange surfaces, etc., to ensure higher resolution image acquisition of these quality-sensitive areas. According to the results of the secondary segmentation, the image acquisition scale adapted to the size of the segmentation region is configured, such as the medium scale corresponding to the standard module and the local high-magnification scale corresponding to the connection position.

Subsequently, according to the configured acquisition scale, the image acquisition unit is called to perform adaptive multi-scale image acquisition on the target vehicle chassis to construct a multi-scale image set. For each segmented area, the image acquisition unit automatically adjusts the imaging parameters (such as working distance, aperture, exposure time, etc.) according to its corresponding acquisition scale to obtain image data of the best quality. The acquisition process can realize scanning and imaging of each part of the chassis one by one through automated devices such as robotic arms, or one-time imaging of the entire chassis through an array camera. Through the adaptive adjustment of the acquisition scale, the global coverage and local details of the image can be taken into account, which can not only reflect the overall level of the vehicle assembly quality, but also can specifically evaluate the assembly quality of key parts. The image data obtained at different acquisition scales are constructed into a multi-scale image set to form hierarchical chassis quality characterization data.

Through secondary segmentation and adaptive multi-scale acquisition methods, the entire vehicle chassis is segmented into regions of interest at different levels and scales, and the optimal acquisition parameters are configured for each region to obtain more comprehensive, detailed and accurate chassis image data, providing high-quality input for subsequent assembly quality assessment and improving the information density and discrimination ability of image data.

Furthermore, the embodiment of the present application also includes:

After performing image acquisition, a sensitive area is constructed based on the pre-recognition result; an adaptive scale update is performed based on the sensitive area and the acquisition scale, additional image acquisition is performed based on the adaptive scale update result, and a multi-scale image set is constructed based on the original image acquisition result and the additional image acquisition result.

In a preferred embodiment, a sensitive area division and adaptive scale update mechanism based on pre-recognition results are introduced to achieve dynamic focusing and additional image acquisition on key parts of assembly quality, further improving the pertinence and richness of the multi-scale image set.

After image acquisition is completed, the acquired image data is pre-identified and processed to construct sensitive areas. Pre-identification processing can use rule-based, template-based or machine learning methods to quickly detect and locate feature areas in the image that are closely related to assembly quality, such as holes, gaps, connectors, etc. These feature areas are usually more sensitive to assembly deviations and assembly defects and require special attention and analysis. The feature areas obtained by pre-identification are constructed into sensitive areas of assembly quality to form a dynamically updated set of areas of interest, which provides direction and basis for subsequent image acquisition optimization.

Then, based on the constructed sensitive areas and the original acquisition scale, adaptive scale update and additional image acquisition are performed. For each sensitive area, the local image acquisition scale is dynamically adjusted according to its location, size, shape and other attributes, such as improving resolution, narrowing the field of view, changing the angle, etc., so that the acquisition parameters match the characteristics of the sensitive area and obtain clearer and more detailed local image data. After completing the adaptive update of the acquisition scale, the image acquisition unit is called to perform additional image acquisition for the sensitive area to obtain additional images focused on the key parts of the assembly quality. The original image acquisition results obtained by the acquisition and the additional image acquisition results obtained by the additional acquisition are constructed into a more comprehensive and hierarchical multi-scale image set to form multi-perspective and multi-granular characterization data for assembly quality.

Through sensitive area division and adaptive scale update mechanism, the weak links of assembly quality are discovered, and targeted image enhancement acquisition is implemented for these weak links, so as to obtain the visual basis of assembly defects more efficiently and accurately and improve the reliability of assembly quality assessment.

In summary, the assembly quality inspection method of a vehicle chassis provided in the embodiment of the present application has the following technical effects:

The target vehicle chassis is numbered, and the modules are divided based on the result of the part numbering. The standard modules are established to transform the complex chassis structure into standardized modules, laying the foundation for subsequent analysis and facilitating the systematic evaluation of assembly quality. The module interface parameters of the standard modules are obtained, and the modules of the target vehicle chassis are taken as nodes in the graph. The mutual influence and connection analysis of each node is performed based on the module interface parameters, which provides support for a comprehensive understanding of the internal connection of the chassis system. Weights are assigned to the edges in the network graph of the target vehicle chassis. The weights represent the degree of influence of one node on another node. The overall influence of each node is calculated based on the path analysis in graph theory. A comprehensive evaluation function is constructed through the overall influence, providing theoretical tools and analysis methods for quantitative evaluation of assembly quality. A standard zero point is established. Taking the standard zero point as the reference, the laser scanning device is activated to perform the measurement of each module and generate a detection data set. The measurement of each module includes geometric dimension measurement and alignment detection, which provides data support for assembly quality evaluation. The connection strength and functional coordination tests are performed by joint sensors to establish additional data sets, enrich the evaluation dimensions of assembly quality, and improve the comprehensiveness of the evaluation. Based on the test data set and additional data set input into the comprehensive evaluation function, the assembly quality evaluation is performed and the quantitative assembly quality evaluation results are output. The assembly management of the target vehicle chassis is carried out according to the assembly quality evaluation results, and the closed-loop feedback of the evaluation results to the assembly management is realized to improve the overall assembly quality of the vehicle chassis.

Embodiment 2

Based on the same inventive concept as the assembly quality inspection method of a vehicle chassis in the aforementioned embodiment, as shown in FIG2 , the embodiment of the present application provides an assembly quality inspection system for a vehicle chassis, the system comprising:

A standard module establishment component 11 is used to number the target vehicle chassis, divide the modules based on the result of the numbering, and establish a standard module;

The network diagram construction component 12 is used to obtain the module interface parameters of the standard module, and take the modules of the target vehicle chassis as nodes in the diagram, perform mutual influence and connection analysis of the nodes based on the module interface parameters, and construct the network diagram of the target vehicle chassis, wherein each node includes at least one associated node, and the associated node is an influence node;

An evaluation function construction component 13 is used to assign weights to the edges in the target vehicle chassis network graph, wherein the weights represent the degree of influence of one node on another node, calculate the overall influence of each node based on the path analysis in graph theory, and construct a comprehensive evaluation function based on the overall influence;

The module determination execution component 14 is used to establish a standard zero point, and based on the standard zero point, activate the laser scanning device to perform each module determination and generate a detection data set, wherein the execution of each module determination includes geometric dimension measurement and alignment detection;

An additional data acquisition component 15 is used to perform connection strength and functional coordination tests by combining sensors to establish additional data sets;

An assembly quality evaluation component 16, configured to perform assembly quality evaluation based on the detection data set and the additional data set input into a comprehensive evaluation function;

The chassis assembly management component 17 is used to manage the assembly of the target vehicle chassis according to the assembly quality evaluation result.

Furthermore, the evaluation function construction component 13 includes the following execution steps:

Configure the comprehensive evaluation function, the formula is as follows:

Among them, Global Score is the comprehensive evaluation value, i represents any module, n is the total number of modules, _Pi is the self-score of module i, which is calculated by docking accuracy, j represents any module different from module i, _Iij is the inter-module influence score, which is calculated by connection strength and functional coordination, _Pj is the self-score of module j, _αi is the self-score weight of module i, and _βij is the influence score weight of module j on module i.

Furthermore, the additional data acquisition component 15 includes the following execution steps:

Invoking a torque sensor in the combined sensor, performing torque monitoring of a tightening process through the torque sensor, and constructing a first monitoring data set;

Invoking the ultrasonic sensor in the combined sensor to perform welding, bonding and integrity tests on each module to establish a second monitoring data set;

calling a force sensor in the combined sensor, performing connection point load measurement based on the force sensor, and establishing a third monitoring data set;

A connection strength evaluation is performed based on the first monitoring data set, the second monitoring data set, and the third monitoring data set, and the connection strength evaluation result is used as an additional data set.

Furthermore, the additional data acquisition component 15 also includes the following execution steps:

Calling the displacement sensor in the joint sensor to perform relative position changes of each module during movement under a preset movement scheme to construct a fourth monitoring data set;

Invoking an acceleration sensor and a vibration analysis sensor in the combined sensor, and constructing a fifth monitoring data set based on the acceleration sensor and the vibration analysis sensor;

After performing time series alignment on the fourth monitoring data set and the fifth monitoring data set, performing time series analysis to generate a time series analysis result;

Calling the module vibration data in the fifth monitoring data set, performing main frequency component analysis of the module vibration data, and generating a vibration coordination result according to the frequency resonance result;

The functional coordination evaluation results are constructed through the time series analysis results and the vibration coordination results, and the functional evaluation results and the connection strength evaluation results are used as additional data sets.

Furthermore, the chassis assembly management component 17 includes the following execution steps:

Calling the image acquisition unit to perform image acquisition of the target vehicle chassis under a preset light source to construct a multi-scale image set;

Annotating the contours of the multi-scale image set through edge computing, and calling a convolutional neural network to perform feature recognition on the multi-scale image set with contour annotations;

Execute the test data authentication corresponding to the assembly quality evaluation result according to the feature recognition result, and update the assembly quality evaluation result according to the authentication result.

Furthermore, the chassis assembly management component 17 also includes the following execution steps:

Performing secondary segmentation on the target vehicle chassis based on standard modules, configuring a collection scale based on the secondary segmentation result, wherein the first segmentation in the secondary segmentation is a region segmentation based on the size of the standard module, and the second segmentation in the secondary segmentation is a region segmentation based on the connection position of the standard module;

The image acquisition unit is called through the acquisition scale to perform image acquisition and construct a multi-scale image set.

Furthermore, the chassis assembly management component 17 also includes the following execution steps:

When image acquisition is performed, sensitive areas are constructed based on pre-recognition results;

An adaptive scale update is performed based on the sensitive area and the acquisition scale, additional image acquisition is performed according to the adaptive scale update result, and a multi-scale image set is constructed according to the original image acquisition result and the additional image acquisition result.

Any step of the method described above can be stored as a computer instruction or program in an unlimited computer memory, and can be called and recognized by an unlimited computer processor to implement any method in the embodiments of the present application, without any unnecessary restrictions.

Furthermore, the first or second mentioned above may not only represent an order relationship, but may also represent a specific concept, and/or refer to multiple elements that can be selected individually or in full. Obviously, those skilled in the art can make various changes and modifications 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 (8)

1. A method for detecting assembly quality of a chassis for a vehicle, the method comprising:

The chassis of the target vehicle is numbered, and module division is carried out based on the result of the component numbering, so that a standard module is established;

Obtaining module interface parameters of a standard module, taking each module of a chassis of a target vehicle as a node in a graph, and carrying out mutual influence and connection analysis of each node based on the module interface parameters to construct a network graph of the chassis of the target vehicle, wherein each node at least comprises an associated node which is an influence node;

Assigning weights to edges in the chassis network diagram of the target vehicle, wherein the weights represent the influence degree of one node on another node, calculating the overall influence of each node based on path analysis in graph theory, and constructing a comprehensive evaluation function through the overall influence;

Establishing a standard zero point, and activating a laser scanning device to execute each module measurement by taking the standard zero point as a reference to generate a detection data set, wherein the execution of each module measurement comprises geometric dimension measurement and alignment detection;

performing connection strength and functional coordination tests through a joint sensor, and establishing an additional data set;

based on the detection data set and the additional data set, inputting the detection data set and the additional data set into a comprehensive evaluation function, and executing assembly quality evaluation;

and carrying out assembly management on the chassis of the target vehicle according to the assembly quality evaluation result.

2. The method of claim 1, wherein said constructing a comprehensive evaluation function from said overall impact further comprises:

The comprehensive evaluation function is configured, and the formula is as follows:

Wherein GlobalScore is a comprehensive evaluation value, I represents any one module, n is the total number of modules, P _i is the self score of the module I, the result is obtained through calculation of docking accuracy, j represents any one module different from the module I, I _ij is the inter-module influence score, the result is obtained through calculation of connection strength and functional coordination, P _j is the self score of the module j, alpha _i is the self score weight of the module I, and beta _ij is the influence score weight of the module j to the module I.

3. The method of claim 1, wherein the establishing additional data sets by joint sensor connection strength and functional coordination testing further comprises:

Invoking a torque sensor in the joint sensor, and executing torque monitoring in a fastening process through the torque sensor to construct a first monitoring data set;

Invoking an ultrasonic sensor in the combined sensor, executing welding, bonding and integrity test of each module, and establishing a second monitoring data set;

Invoking a force sensor in the joint sensor, performing connection point load measurement based on the force sensor, and establishing a third monitoring data set;

And performing connection strength evaluation based on the first monitoring data set, the second monitoring data set and the third monitoring data set, and taking a connection strength evaluation result as an additional data set.

4. A method as claimed in claim 3, wherein the method further comprises:

invoking a displacement sensor in the combined sensor, and carrying out relative position change in the motion of each module under a preset motion scheme to construct a fourth monitoring data set;

Invoking an acceleration sensor and a vibration analysis sensor in the joint sensor, and constructing a fifth monitoring data set based on the acceleration sensor and the vibration analysis sensor;

Performing time sequence analysis after aligning the fourth monitoring data set and the fifth monitoring data set in time sequence to generate a time sequence analysis result;

Calling module vibration data in the fifth monitoring data set, executing main frequency component analysis of the module vibration data, and generating a vibration coordination result according to a frequency resonance result;

and constructing a functional coordination evaluation result through the time sequence analysis result and the vibration coordination result, and taking the functional evaluation result and the connection strength evaluation result as additional data sets.

5. The method of claim 1, wherein the performing assembly management of the chassis for the target vehicle according to the assembly quality evaluation result further comprises:

calling an image acquisition unit, and executing image acquisition of a chassis of a target vehicle under a preset light source to construct a multi-scale image set;

Marking the outline of the multi-scale image set through edge calculation, and calling a convolutional neural network to execute the multi-scale image set with the outline marking to perform feature recognition;

And executing detection data authentication corresponding to the assembly quality evaluation result according to the characteristic recognition result, and updating the assembly quality evaluation result according to the authentication result.

6. The method of claim 5, wherein the invoking the image acquisition unit performs image acquisition of the chassis of the target vehicle under a preset light source to construct a multi-scale image set, further comprising:

Performing secondary segmentation based on a standard module on the chassis of the target vehicle, and configuring an acquisition scale based on a secondary segmentation result, wherein the first segmentation in the secondary segmentation is region segmentation based on the size of the standard module, and the second segmentation in the secondary segmentation is region segmentation based on the connection position of the standard module;

and calling an image acquisition unit through the acquisition scale, executing image acquisition and constructing a multi-scale image set.

7. The method of claim 6, wherein invoking the image acquisition unit through the acquisition scale performs image acquisition to construct a multi-scale image set, further comprising:

After image acquisition is executed, a sensitive area is constructed based on a pre-recognition result;

And carrying out self-adaptive scale updating based on the sensitive area and the acquisition scale, carrying out additional image acquisition through a self-adaptive scale updating result, and constructing a multi-scale image set through an original image acquisition result and an additional image acquisition result.

8. A system for detecting the quality of assembly of a vehicle chassis, for implementing a method for detecting the quality of assembly of a vehicle chassis according to any one of claims 1 to 7, said system comprising:

The standard module building assembly is used for numbering the components of the chassis of the target vehicle, dividing the modules based on the component numbering result and building a standard module;

The network diagram construction assembly is used for acquiring module interface parameters of the standard modules, taking each module of the chassis of the target vehicle as a node in the diagram, and carrying out mutual influence and connection analysis of each node based on the module interface parameters to construct a network diagram of the chassis of the target vehicle, wherein each node at least comprises one associated node which is an influence node;

The evaluation function construction component is used for distributing weights to edges in the chassis network diagram of the target vehicle, the weights represent the influence degree of one node on the other node, the overall influence of each node is calculated based on path analysis in graph theory, and a comprehensive evaluation function is constructed through the overall influence;

The module measurement execution assembly is used for establishing a standard zero point, and activating the laser scanning device to execute each module measurement by taking the standard zero point as a reference to generate a detection data set, wherein the execution of each module measurement comprises geometric dimension measurement and alignment detection;

the additional data acquisition component is used for testing the connection strength and the functional coordination through the joint sensor and establishing an additional data set;

The assembly quality evaluation component is used for performing assembly quality evaluation based on the detection data set and the additional data set which are input into a comprehensive evaluation function;

and the chassis assembly management component is used for carrying out assembly management on the chassis of the target vehicle according to the assembly quality evaluation result.