Big data evaluation-based automobile production assembly management system and method

Abstract

The application discloses an automobile production assembly management system and method based on big data evaluation, relates to the technical field of automobile production, and performs influence analysis on each assembly part corresponding condition based on assembly part quality and assembly condition data during automobile production; performs abnormality analysis on assembly equipment corresponding to each assembly part based on historical operation condition data of the assembly equipment corresponding to each assembly part during automobile production; performs estimated assembly abnormality influence analysis based on usage condition data of each assembly part action position, influence analysis results of each assembly part corresponding condition and abnormality analysis results of the assembly equipment corresponding to each assembly part; and determines whether to maintain the automobile production assembly process according to the estimated assembly abnormality influence analysis results, so that the purpose of preventing troubles from occurring, improving automobile production quality and efficiency and reducing the whole life cycle cost of automobile production is achieved.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of automobile production, and in particular to an automobile production assembly management system and method based on big data evaluation. BACKGROUND

[0002] With the rapid development of the automobile industry, the assembly precision and quality in the automobile manufacturing process have become key factors affecting the quality and safety of the vehicle. In automobile production, any slight error in the assembly process can lead to substandard vehicle performance, and even affect driving safety. Therefore, improving automobile assembly quality, monitoring assembly abnormalities in real time during the production process, and making predictions and maintenance of potential problems are technical problems that need to be solved in modern automobile manufacturing.

[0003] With the continuous development of information technology and automation technology, more and more automobile production enterprises have begun to use big data, Internet of Things, machine learning and other technologies to intelligently manage and optimize the assembly quality by means of the massive data generated during the production process. However, in the automobile assembly process, due to the influence of factors such as the quality of assembly parts, the running state of assembly equipment, and the use of part action positions, assembly abnormalities occur from time to time. Existing technologies focus on monitoring and diagnosis of a single link and cannot consider the comprehensive influence of assembly parts, assembly equipment, and assembly environment on the automobile production process. Existing technologies also have a lag in problem discovery, and existing quality detection is mostly post-detection or passive response. This method cannot predict potential and implicit assembly abnormal risks in advance during the assembly process or before the completion of assembly. Often, problems are exposed after a large number of vehicles are offline, or even after they have been sold to the market. At this time, the cost of recall or repair is extremely high, and it causes serious damage to brand reputation. In order to solve the problems raised in the background art, the application designs an automobile production assembly management system and method based on big data evaluation. SUMMARY

[0004] In view of the above technical deficiencies, the application provides an automobile production assembly management system and method based on big data evaluation.

[0005] To solve the above technical problems, the application adopts the following technical solutions: The application provides an automobile production assembly management method based on big data evaluation, which includes the following specific steps:

[0006] S1, obtaining assembly part quality and assembly condition data during automobile production, assembly equipment historical operation condition data corresponding to the assembly parts during automobile production, and each assembly part action position use condition data;

[0007] S2, performing influence analysis of each assembly part corresponding condition based on the assembly part quality and assembly condition data during automobile production;

[0008] S3, performing abnormality analysis of the assembly equipment corresponding to each assembly part based on historical operation data of the assembly equipment corresponding to the assembly part during automobile production;

[0009] S4, performing estimated assembly abnormality influence analysis based on the usage data of the position where each assembly part acts, the influence analysis result of the corresponding condition of each assembly part, and the abnormality analysis result of the assembly equipment corresponding to each assembly part;

[0010] S5, determining whether to maintain the automobile production assembly process according to the estimated assembly abnormality influence analysis result.

[0011] It should be noted that, as a preferred technical solution of the automobile production assembly management method based on big data evaluation, the specific steps of S1 are as follows:

[0012] S11, obtaining assembly part quality and assembly condition data during automobile production through a high-definition industrial camera, a vision system, an assembly bill of materials and a process route file, and a PLC record table, wherein the assembly part quality and assembly condition data during automobile production includes bolt surface burr distribution quantity data, bolt surface area data, assembly part volume data, assembly step number data, and assembly time required data;

[0013] S12, obtaining historical operation data of the assembly equipment corresponding to the assembly part during automobile production through an Internet of Things gateway log, a machine vision system feedback, and a force sensor built in a gripper, wherein the historical operation data of the assembly equipment corresponding to the assembly part during automobile production includes fault downtime data, average fault duration data, actual positioning accuracy data of the assembly equipment corresponding to each assembly part, and gripper clamping force data in the historical operation cycle of the assembly equipment;

[0014] S13, obtaining usage data of the position where each assembly part acts from historical vehicle state information data, wherein the usage data of the position where each assembly part acts includes daily average usage times data and average usage intensity data of the position where each assembly part acts.

[0015] It should be noted that, as a preferred technical solution of the automobile production assembly management method based on big data evaluation, S2 includes the following specific steps:

[0016] S21, obtaining quality influence analysis results of each assembly part from bolt surface burr distribution quantity data and bolt surface area data;

[0017] S22, obtaining assembly difficulty analysis results of each assembly part from assembly step number data and assembly time required data;

[0018] S23, obtain the assembly part quality influence analysis result and the assembly difficulty analysis result, add the assembly part quality influence analysis result and the assembly difficulty analysis result after weighting to obtain the corresponding situation influence analysis result of each assembly part. It should be noted that the assembly part quality influence analysis result and the assembly difficulty analysis result are weighted and fused, the key information of two different dimensions is fused together, the aggregation effect is generated, and a global and unified view of the part risk in the automobile production assembly process is formed.

[0019] It should be noted that, as the preferred technical solution of the automobile production assembly management method based on big data evaluation, the specific steps of S21 are: performing quality influence analysis of each assembly part according to the bolt surface burr distribution quantity data and bolt surface area data of each assembly part, wherein the quality influence analysis process of each assembly part is: dividing the bolt surface burr distribution quantity data of each assembly part by the bolt surface area data of each assembly part to obtain the burr density of each assembly part, and dividing the burr density of each assembly part by the reference burr density to obtain the quality influence analysis result of each assembly part; it should be noted that the burr density is obtained by dividing the burr quantity by the surface area, which successfully converts the part quality influence analysis standard from absolute quantity to defect concentration per unit area; this enables bolts of different sizes and different models to be compared and sorted on the same fair scale; and then it is normalized to obtain a dimensionless and relative quality influence analysis result, which can more accurately identify assembly parts with deeper hidden dangers.

[0020] It should be noted that, as a preferred technical solution of the automobile production assembly management method based on big data evaluation, the specific steps of S22 are: analyzing the assembly difficulty of each assembly part according to the volume data, assembly step number data and assembly time required data of each assembly part, wherein the assembly difficulty analysis process of each assembly part is: dividing the assembly step number data of each assembly part by the reference assembly step number to quantify the assembly process complexity of each assembly part; dividing the assembly time required data of each assembly part by the maximum value of the reference assembly part required time to quantify the assembly operation complexity of each assembly part; multiplying the assembly process complexity and the assembly operation complexity of each assembly part to obtain the assembly complexity of each assembly part; dividing the difference between the maximum volume of each assembly part and the volume data of each assembly part by the maximum volume required for assembly to quantify the assembly precision of each assembly part; weighting and adding the assembly complexity and assembly precision of each assembly part to obtain the assembly difficulty analysis result of each assembly part; it should be noted that the difficulty is not simply attributed to a single factor, but is disassembled and quantified from three core dimensions of process complexity, operation complexity and physical precision, avoiding the one-sidedness of a single index, and providing a three-dimensional and comprehensive difficulty view; by dividing by the reference value, all absolute values are successfully normalized into relative ratios, which enables the difficulty of different types of parts, different production lines and different types of parts to be compared and ranked fairly under the same standard scale; the assembly precision is obtained, which means that the smaller the volume of the automobile production assembly part, the higher the assembly precision, that is, small parts are usually more difficult to grasp, align and install; weighting and adding give the assembly difficulty analysis result great flexibility, and different production environments can give different weights to complexity and precision.

[0021] It should be noted that, as a preferred technical solution of the automobile production assembly management method based on big data evaluation, the specific steps of S3 are:

[0022] S31, obtaining assembly equipment fault abnormality analysis results corresponding to each assembly part from the fault downtime data and average fault duration data in the historical operation cycle of the assembly equipment corresponding to each assembly part;

[0023] S32, obtaining assembly equipment operation abnormality analysis results corresponding to each assembly part from the actual positioning accuracy data and gripper clamping force data of the assembly equipment corresponding to each assembly part;

[0024] S33, obtaining the assembly equipment failure abnormality analysis result and the operation abnormality analysis result corresponding to each assembly part, weighting and adding the assembly equipment failure abnormality analysis result and the operation abnormality analysis result corresponding to each assembly part to obtain the assembly equipment abnormality analysis result corresponding to each assembly part; it should be noted that the assembly equipment failure abnormality analysis result and the operation abnormality analysis result corresponding to each assembly part are comprehensively analyzed to construct a comprehensive assembly equipment health condition view; the failure abnormality reflects the hardware health degree of the assembly equipment, and the operation abnormality reflects the performance health degree of the assembly equipment; by combining the two, a result reflecting the reliability and performance of the assembly equipment is obtained, and the accuracy of the assembly equipment abnormality analysis result corresponding to each assembly part is improved.

[0025] It should be noted that, as a preferred technical solution of the automobile production assembly management method based on big data evaluation, the specific steps of S31 are: performing assembly equipment failure abnormality analysis on each assembly part according to the fault downtime data and the average fault duration data in the historical running period of the assembly equipment corresponding to each assembly part, wherein the assembly equipment failure abnormality analysis process corresponding to each assembly part is: integrating the fault downtime data in the historical running period of the assembly equipment corresponding to each assembly part over the historical running period of the assembly equipment, dividing the integral result by the product of the historical running period of the assembly equipment and the reference fault downtime, and quantifying the failure frequency intensity of the assembly equipment corresponding to each assembly part; dividing the average fault duration data in the historical running period of the assembly equipment corresponding to each assembly part by the reference fault allowable duration to quantify the failure impact degree of the assembly equipment corresponding to each assembly part; multiplying the failure frequency intensity and the failure impact degree of the assembly equipment corresponding to each assembly part to obtain the assembly equipment failure abnormality analysis result corresponding to each assembly part; it should be noted that the failure frequency intensity reflects the frequency and density of equipment failure, and the failure impact degree reflects the damage caused by each failure; by comprehensively analyzing the two, the deviation degree relative to the historical period and the reference standard is obtained, and the health status of the assembly equipment is accurately described.

[0026] It should be noted that as the preferred technical solution of the automobile production assembly management method based on big data evaluation, the specific steps of S32 are: according to the actual positioning accuracy data and the actual clamping force data of each assembly part corresponding to the assembly equipment, the abnormality analysis of the assembly equipment of each assembly part is carried out, wherein the abnormality analysis process of the assembly equipment of each assembly part is: the difference between the actual positioning accuracy data of each assembly part corresponding to the assembly equipment and the standard positioning accuracy is divided by the reference positioning accuracy difference, and the actual positioning ability attenuation degree of each assembly part corresponding to the assembly equipment is quantified; the actual clamping force data of each assembly part corresponding to the assembly equipment is divided by the clamping force allowed difference, and the clamping jaw running wear of each assembly part corresponding to the assembly equipment is quantified; the actual positioning ability attenuation degree and the clamping jaw running wear of each assembly part corresponding to the assembly equipment are multiplied, and the abnormality analysis result of the assembly equipment of each assembly part is obtained in this way; it should be noted that through the continuous and quantitative analysis of the performance attenuation degree of the assembly equipment, the running trend of the assembly equipment is grasped; by calculating the positioning accuracy attenuation and the clamping force attenuation separately and then multiplying them, the cause of the performance degradation of the assembly equipment can be accurately located, and the future problems of the assembly equipment can be predicted from the slight performance changes.

[0027] It should be noted that as the preferred technical solution of the automobile production assembly management method based on big data evaluation, the specific steps of S4 are: according to the daily use frequency data, the average use intensity data, the corresponding situation influence analysis result and the abnormality analysis result of each assembly part corresponding to the assembly equipment, the estimated assembly abnormality influence analysis is carried out, wherein the estimated assembly abnormality influence analysis process is: the daily use frequency data of each assembly part corresponding to the assembly equipment is multiplied by the average use intensity, and then divided by the product of the reference use intensity data and the daily use frequency, the use intensity of each assembly part corresponding position is quantified, the corresponding situation influence analysis result and the abnormality analysis result of each assembly part corresponding to the assembly equipment are weighted and added, and the result is multiplied by the use intensity of each assembly part corresponding position to obtain the estimated assembly abnormality influence analysis result of each assembly part; the estimated assembly abnormality influence analysis result of each assembly part is summed and averaged to obtain the estimated assembly abnormality influence analysis result; it should be noted that instead of analyzing a certain link in isolation, the results of all previous steps: the risk of the part itself, the state abnormality of the equipment and the severity of the vehicle use are integrated, thereby realizing the cross-target from analysis to prediction, dynamically associating manufacturing attributes and use attributes, and realizing the result of foreseeing and avoiding problems.

[0028] It should be noted that as the preferred technical solution of the automobile production assembly management method based on big data evaluation, the specific steps of S5 are: comparing the estimated assembly abnormality influence analysis result with the set estimated assembly abnormality influence analysis result threshold, if the estimated assembly abnormality influence analysis result is greater than or equal to the set estimated assembly abnormality influence analysis result threshold, it is judged that the automobile production assembly process needs to be maintained; if the estimated assembly abnormality influence analysis result is less than the set estimated assembly abnormality influence analysis result threshold, it is judged that the automobile production assembly process does not need to be maintained; it should be noted that by comparing the estimated assembly abnormality influence analysis result with its threshold and directly outputting the decision judgment, the decision threshold is reduced, the automation and intelligentization of decision-making is realized, the human intervention is reduced, and the response speed is improved.

[0029] The automobile production assembly management system based on big data evaluation is realized based on the above-mentioned automobile production assembly management method based on big data evaluation, and specifically includes an automobile production assembly situation acquisition module, an assembly part corresponding situation influence analysis module, an assembly equipment abnormality analysis module, an estimated assembly abnormality influence analysis module, and an estimated assembly maintenance judgment module. The automobile production assembly situation acquisition module is used to acquire assembly part quality and assembly situation data during automobile production, assembly equipment historical operation situation data corresponding to assembly parts during automobile production, and each assembly part action position use situation data;

[0030] The assembly part corresponding situation influence analysis module is used to analyze the influence of each assembly part corresponding situation based on the assembly part quality and assembly situation data during automobile production;

[0031] The assembly equipment abnormality analysis module is used to analyze the abnormality of each assembly equipment corresponding to the assembly part based on the assembly equipment historical operation situation data corresponding to the assembly part during automobile production;

[0032] The estimated assembly abnormality influence analysis module is used to analyze the influence of the estimated assembly abnormality based on the each assembly part action position use situation data, the each assembly part corresponding situation influence analysis result, and the each assembly part corresponding assembly equipment abnormality analysis result;

[0033] The estimated assembly maintenance judgment module is used to judge whether to maintain the automobile production assembly process according to the estimated assembly abnormality influence analysis result.

[0034] Compared with the prior art, the application has the beneficial effects that: the application obtains the assembly part quality and assembly condition data during automobile production, the assembly equipment historical operation condition data corresponding to the assembly part during automobile production, and the use condition data of the action position of each assembly part; the corresponding condition influence analysis of each assembly part is carried out based on the assembly part quality and assembly condition data during automobile production; the abnormality analysis of the assembly equipment corresponding to each assembly part is carried out based on the assembly equipment historical operation condition data corresponding to the assembly part during automobile production; the influence analysis of the estimated assembly abnormality is carried out based on the use condition data of the action position of each assembly part, the corresponding condition influence analysis result of each assembly part, and the abnormality analysis result of the assembly equipment corresponding to each assembly part; and whether to maintain the automobile production assembly process is judged according to the influence analysis result of the estimated assembly abnormality, so that the purpose of preventing trouble from happening, improving the automobile production quality and efficiency, and reducing the whole life cycle cost of automobile production is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 It is a whole flow schematic diagram of the automobile production assembly management method based on big data evaluation of the application.

[0036] Figure 2 It is a flow schematic diagram of the corresponding condition influence analysis result of each assembly part of the automobile production assembly management method based on big data evaluation of the application.

[0037] Figure 3 It is a whole framework schematic diagram of the automobile production assembly management system based on big data evaluation of the application. DETAILED DESCRIPTION

[0038] In order to better understand the application, various aspects of the application will be described in more detail with reference to the accompanying drawings.

[0039] In order to solve the technical problems proposed in the background art, the application provides a preferred embodiment:

[0040] The specific content of the embodiment is:

[0041] As shown in Figure 1 The automobile production assembly management method based on big data evaluation includes the following specific steps:

[0042] S1, obtaining the assembly part quality and assembly condition data during automobile production, the assembly equipment historical operation condition data corresponding to the assembly part during automobile production, and the use condition data of the action position of each assembly part;

[0043] In the embodiment, the specific steps of S1 are:

[0044] S11, the assembly part quality and assembly condition data during automobile production includes bolt surface burr distribution quantity data, bolt surface area data, each assembly part volume data, assembly step number data and assembly required time length data, wherein the assembly part quality and assembly condition data during automobile production is obtained in the following manner: under a high-definition industrial camera, a bolt sample is photographed from multiple angles, bolt surface burr distribution quantity data is obtained through an image processing algorithm (such as edge detection, feature extraction, deep learning image segmentation); the geometric size of the bolt is measured through a calibrated vision system, so as to calculate the bolt surface area data; each assembly part volume data and assembly step number data are obtained from the assembly bill of materials and process route file; the assembly required time length data is obtained from a PLC (programmable logic controller) record table;

[0045] S12, the assembly equipment historical operation condition data corresponding to the assembly part during automobile production includes fault downtime number data, average fault time length data in the historical operation cycle of the assembly equipment corresponding to each assembly part, actual positioning accuracy data of the assembly equipment corresponding to each assembly part and clamping force data of a clamping jaw; wherein the assembly equipment historical operation condition data corresponding to the assembly part during automobile production is obtained in the following manner: the fault downtime number data and the average fault time length data in the historical operation cycle of the assembly equipment corresponding to each assembly part are obtained through a thing gateway record log; the actual positioning accuracy data of the assembly equipment corresponding to each assembly part is obtained through a machine vision system feedback; the clamping force data of the clamping jaw is obtained from a built-in force sensor of the clamping jaw;

[0046] S13, the use condition data of each assembly part action position includes daily average use number data and average use intensity data of each assembly part action position, wherein the use condition data of each assembly part action position is obtained in the following manner: the daily average use number data and the average use intensity data of each assembly part action position are obtained from historical vehicle state information data.

[0047] S2, as shown in Figure 2 , based on the assembly part quality and assembly condition data during automobile production, a corresponding condition influence analysis of each assembly part is performed;

[0048] S21, the quality influence analysis result of each assembly part is obtained from the bolt surface burr distribution quantity data and the bolt surface area data;

[0049] In the embodiment, S21 comprises the following specific steps: performing quality influence analysis of each assembly part according to the bolt surface burr distribution quantity data and the bolt surface area data of each assembly part, wherein the quality influence analysis process of each assembly part is: dividing the bolt surface burr distribution quantity data of each assembly part by the bolt surface area data of each assembly part to obtain the burr density of each assembly part, and dividing the burr density of each assembly part by the reference burr density to obtain the quality influence analysis result of each assembly part; it should be noted that when the bolt has burrs, the defects will increase the friction of the bolt surface; because the bolt increases the friction, the phenomenon of false torque (i.e. the torque meets the standard, but the clamping force is insufficient) will occur; when the false torque occurs: the value on the torque dial meets the standard, but the value is incorrect, at this time the bolt has not been stretched to the predetermined extent, resulting in that the clamping force is far lower than the design value, which may cause the connection point to loosen, make abnormal sound and fail; the purpose of tightening the bolt is to obtain a large clamping force, which can tightly press the connected parts together to resist external separation force, and the applied torque is a means used to obtain the clamping force; insufficient clamping force means that the connected parts are not pressed tightly enough, under the continuous vibration, impact and alternating load in vehicle driving, the bolt connection will have microscopic relative sliding, which will cause the pre-tightening force to further attenuate, the loosening from microscopic to macroscopic, and finally the bolt completely loosens; the loosened connection point will produce knocking and friction between metal parts, which will make abnormal sound on bumpy road or under vibration working condition; insufficient clamping force will cause the external load to be entirely borne by the bolt itself instead of the friction between the connected parts, which is easy to cause the bolt to fatigue and break; the loosened connection will cause uneven load distribution, which not only damages the bolt, but also may damage the connected parts (such as installation holes of engine support, control arm and auxiliary frame), resulting in greater structural failure.

[0050] S22, obtaining the assembly difficulty analysis result of each assembly part from the volume data, the assembly step number data and the assembly required time length data of each assembly part;

[0051] In the embodiment, S22 comprises the following specific steps: performing assembly part assembly difficulty analysis according to the assembly part volume data, the assembly step number data and the assembly time length data required, wherein the assembly part assembly difficulty analysis process is: dividing the assembly part assembly step number data by the reference assembly step number to quantify the assembly part assembly process complexity; dividing the assembly part assembly time length data required by the maximum value of the reference assembly part time length required to quantify the assembly part assembly operation complexity; multiplying the assembly part assembly process complexity and the assembly operation complexity to obtain the assembly part assembly complexity; dividing the difference between the maximum value of the assembly part volume and the assembly part volume data by the maximum value of the assembly part volume required to quantify the assembly part assembly precision; weighting and adding the assembly part assembly complexity and the assembly precision to obtain the assembly part assembly difficulty analysis result; it should be noted that the step number is the most direct indicator of measuring the logical complexity and tediousness of an assembly task; the more steps, the more sequences need to be executed, and the error probability (such as missing and wrong sequence) increases; the operation complexity is reflected by the assembly part time length required, and the longer the time length required, the more precise and difficult the operation itself may be, or more adjustment and confirmation is required; therefore, an assembly part that requires many steps but each step is fast and a part that has few steps but each step is extremely time-consuming may have similar complexity; the part volume is a core physical parameter affecting ergonomics and operation accessibility; large-volume parts usually have problems such as grabbing, handling and positioning difficulties, and may require auxiliary equipment, but the operation precision requirement may be relatively low; and small-volume parts usually mean that grabbing is difficult, positioning is extremely high, more precise tools are required, and the stability of assembly is more challenging.

[0052] S23, obtaining the assembly part quality influence analysis result and the assembly difficulty analysis result, weighting and adding the assembly part quality influence analysis result and the assembly difficulty analysis result to obtain the corresponding situation influence analysis result of each assembly part, it should be noted that weighting means assigning weights according to the importance of different factors, which can more accurately reflect the actual situation; for example, the quality influence may be more important than the assembly difficulty, so the weight may be higher, which reflects the scientificity of the corresponding situation influence analysis result, and improves the accuracy and reliability of the prediction.

[0053] S3, performing assembly equipment abnormality analysis of each assembly part corresponding assembly equipment based on the historical operation data of the assembly equipment corresponding to the assembly part during automobile production;

[0054] S31, obtaining the assembly equipment fault abnormality analysis result of each assembly part corresponding assembly equipment from the fault downtime data and the average fault time length data in the historical operation cycle of the assembly equipment corresponding to the assembly part;

[0055] In the embodiment, the specific steps of S31 are: performing assembly equipment fault abnormality analysis of each assembly part according to the fault downtime data and the average fault duration data in the historical running period of the assembly equipment corresponding to each assembly part, wherein the process of the assembly equipment fault abnormality analysis of each assembly part is: integrating the fault downtime data in the historical running period of the assembly equipment corresponding to each assembly part over the historical running period of the assembly equipment, dividing the integral result by the product of the historical running period of the assembly equipment and the reference fault downtime, and quantifying the fault frequency intensity of the assembly equipment corresponding to each assembly part; dividing the average fault duration data in the historical running period of the assembly equipment corresponding to each assembly part by the reference fault allowable duration, and quantifying the fault influence degree of the assembly equipment corresponding to each assembly part; multiplying the fault frequency intensity and the fault influence degree of the assembly equipment corresponding to each assembly part to obtain the assembly equipment fault abnormality analysis result of each assembly part; it should be noted that the two types of data, the fault downtime and the average fault duration in the historical running period of the assembly equipment, are selected as the input of the fault abnormality analysis, which is from the two dimensions of reliability and maintainability, and accurately analyzes the core characteristics of the fault behavior of the assembly equipment; the fault downtime data is an index for describing the reliability and fault frequency of the assembly equipment, and the more frequent the fault occurs in a unit of time, the shorter the average fault-free time of the assembly equipment, and the worse the health condition; in the automobile production assembly process, the more the production process is interrupted, the greater the damage to the automobile production rhythm and plan; analyzing the average fault duration data is a key index for measuring the fault repair difficulty and the maintainability of the equipment, and the longer the time required for each fault repair of the assembly equipment, the more likely the fault itself is more serious (for example, a large core component needs to be replaced and only a restart can solve the problem), and the production loss and economic loss caused by each fault are greater; the integral of the fault frequency is the distribution density and trend of the fault on the time axis, and by dividing the reference fault downtime and the reference fault allowable duration, standardized evaluation can be realized to support fair comparison across devices; multiplying the fault frequency intensity and the fault influence degree means that a high-frequency, long-duration fault equipment will have its risk value amplified sharply.

[0056] S32, obtaining the assembly equipment operation abnormality analysis result of each assembly part from the actual positioning accuracy data and the gripper clamping force data of the assembly equipment corresponding to each assembly part;

[0057] In the embodiment, the specific steps of S32 are: performing assembly equipment operation abnormality analysis on each assembly part corresponding assembly equipment according to the actual positioning accuracy data and actual gripper clamping force data of each assembly part corresponding assembly equipment, wherein the assembly equipment operation abnormality analysis process of each assembly part corresponding assembly equipment is: dividing the difference between the actual positioning accuracy data of each assembly part corresponding assembly equipment and the standard positioning accuracy by the reference positioning accuracy difference to quantify the actual positioning ability attenuation degree of each assembly part corresponding assembly equipment; dividing the difference between the actual gripper clamping force data of each assembly part corresponding assembly equipment and the standard clamping force by the clamping force allowable difference to quantify the gripper running wear condition of each assembly part corresponding assembly equipment; and multiplying the actual positioning ability attenuation degree of each assembly part corresponding assembly equipment and the gripper running wear condition, so as to obtain the assembly equipment operation abnormality analysis result of each assembly part corresponding assembly equipment; it should be noted that the positioning accuracy and the gripper wear degree of the assembly equipment not only affect each other, but also usually form a negative feedback cycle that is continuously deteriorating; when the initial positioning deviates, non-ideal contact (such as oblique insertion) occurs between the gripper and the workpiece, resulting in abnormal stress and friction, at which time the abnormal wear condition of the gripper will be aggravated, the clamping center will be offset, the clamping force will be reduced or uneven, and the workpiece clamping posture will be changed; in order to achieve stable clamping, the system needs to compensate for the deviation, and the system positioning error will be further increased; by dividing the reference positioning accuracy difference and the clamping force allowable difference, the absolute tolerance requirements of different devices and different processes are normalized into dimensionless relative performance indicators, and standardized analysis is realized; by separately calculating the positioning accuracy attenuation and the clamping force attenuation and then multiplying them, the result has a preliminary equipment deterioration cause positioning ability; for example, if the positioning ability attenuation degree is high and the gripper wear condition is normal, the problem may be in the servo system, the mechanical transmission mechanism or the calibration system; if the gripper wear condition is high and the positioning ability attenuation degree is normal, the problem may be in the pneumatic system, the gripper body mechanism or the force sensor; if both are high, it indicates that the equipment is aging comprehensively.

[0058] S33, obtaining the assembly equipment fault abnormality analysis result and the operation abnormality analysis result of each assembly part corresponding assembly equipment, weighting and adding the assembly equipment fault abnormality analysis result and the operation abnormality analysis result of each assembly part corresponding assembly equipment to obtain the assembly equipment abnormality analysis result of each assembly part corresponding assembly equipment; it should be noted that by comprehensively analyzing the health status of the assembly equipment in two different aspects, namely the fault abnormality and the operation abnormality, the one-sidedness of a single perspective is avoided, and the analysis result can reflect the survival state and performance state of the assembly equipment at the same time.

[0059] S4, performing predicted assembly abnormality influence analysis based on the use condition data of each assembly part action position, the situation influence analysis result of each assembly part corresponding situation, and the assembly equipment abnormality analysis result of each assembly part corresponding assembly equipment;

[0060] In the embodiment, the specific steps of S4 are: performing estimated assembly abnormality influence analysis according to the daily use frequency data of each assembly part position, the average use intensity data, the corresponding situation influence analysis result of each assembly part, and the assembly equipment abnormality analysis result corresponding to each assembly part. The process of the estimated assembly abnormality influence analysis is: multiplying the daily use frequency data of each assembly part position by the average use intensity, dividing the product by the product of the reference use intensity data and the daily use frequency, quantifying the use intensity of the corresponding position of each assembly part, weighting and adding the corresponding situation influence analysis result of each assembly part and the assembly equipment abnormality analysis result corresponding to each assembly part, multiplying the result by the use intensity of the corresponding position of each assembly part, and obtaining the estimated assembly abnormality influence analysis result corresponding to each assembly part. The estimated assembly abnormality influence analysis result corresponding to each assembly part is summed and averaged to obtain the estimated assembly abnormality influence analysis result. It should be noted that the estimated assembly abnormality influence analysis is performed using the daily use frequency data of each assembly part position, the average use intensity data, the corresponding situation influence analysis result of each assembly part, and the assembly equipment abnormality analysis result corresponding to each assembly part, which realizes the goal of crossing from micro-analysis to macro-prediction, from static evaluation to dynamic risk insight. The use intensity analysis realizes personalized risk prediction, that is, the same assembly problem has different failure risks and failure speeds on vehicles in different use scenarios. A key part (i.e., high influence) is assembled by a high-abnormality device and installed in a high-load position, and the safety risk is extremely high. A non-key part (i.e., low influence) is assembled by a device in poor condition, but installed in a low-load position (i.e., low use intensity), and the safety risk is not necessarily high. Dynamically associating manufacturing attributes (i.e., part quality, device state) with use attributes (i.e., use intensity) can simulate the evolution process of potential assembly abnormalities under different use intensities, thereby intervening before problems occur.

[0061] S5, determining whether to maintain the automobile production assembly process according to the estimated assembly abnormality influence analysis result;

[0062] In the embodiment, the specific steps of S5 are: comparing the estimated assembly abnormality influence analysis result with the set estimated assembly abnormality influence analysis result threshold value. If the estimated assembly abnormality influence analysis result is greater than or equal to the set estimated assembly abnormality influence analysis result threshold value, it is determined that the automobile production assembly process needs to be maintained. If the estimated assembly abnormality influence analysis result is less than the set estimated assembly abnormality influence analysis result threshold value, it is determined that the automobile production assembly process does not need to be maintained.

[0063] According to the above implementation, the embodiment has the following advantages over the prior art: the embodiment obtains assembly part quality and assembly condition data during automobile production, assembly equipment historical operation condition data corresponding to the assembly parts during automobile production, and use condition data of the positions of the assembly parts; performs corresponding condition influence analysis of the assembly parts based on the assembly part quality and assembly condition data during automobile production; performs assembly equipment abnormality analysis of the assembly parts based on the assembly equipment historical operation condition data corresponding to the assembly parts during automobile production; performs estimated assembly abnormality influence analysis based on the use condition data of the positions of the assembly parts, the corresponding condition influence analysis results of the assembly parts, and the assembly equipment abnormality analysis results of the assembly parts; and determines whether to maintain the automobile production assembly process according to the estimated assembly abnormality influence analysis results, so as to prevent problems from occurring, improve automobile production quality and efficiency, and reduce the total life cycle cost of automobile production.

[0064] As shown in Figure 3 The embodiment also provides an automobile production assembly management system based on big data evaluation, which is implemented based on the above automobile production assembly management method based on big data evaluation, and specifically includes an automobile production assembly condition acquisition module, a corresponding condition influence analysis module of assembly parts, an assembly equipment abnormality analysis module, an estimated assembly abnormality influence analysis module, and an estimated assembly maintenance determination module. The automobile production assembly condition acquisition module is used to obtain assembly part quality and assembly condition data during automobile production, assembly equipment historical operation condition data corresponding to the assembly parts during automobile production, and use condition data of the positions of the assembly parts. The corresponding condition influence analysis module of assembly parts is used to perform corresponding condition influence analysis of the assembly parts based on the assembly part quality and assembly condition data during automobile production. The assembly equipment abnormality analysis module is used to perform assembly equipment abnormality analysis of the assembly parts based on the assembly equipment historical operation condition data corresponding to the assembly parts during automobile production. The estimated assembly abnormality influence analysis module is used to perform estimated assembly abnormality influence analysis based on the use condition data of the positions of the assembly parts, the corresponding condition influence analysis results of the assembly parts, and the assembly equipment abnormality analysis results of the assembly parts. The estimated assembly maintenance determination module is used to determine whether to maintain the automobile production assembly process according to the estimated assembly abnormality influence analysis results.

[0065] The specific steps of each unit module in the automobile production assembly management system based on big data evaluation of the present application for implementing the corresponding functions can refer to the steps in the above embodiment of the automobile production assembly management method based on big data evaluation, and will not be repeated here.

[0066] The above description is only the preferred embodiment of the present application and the explanation of the technical principles. It should be understood by those skilled in the art that the application scope of the present application is not limited to the technical solutions with the specific combination of the above technical features, and should also cover other technical solutions formed by combining the above technical features or their equivalent features without departing from the concept of the application. For example, the technical solutions formed by replacing the above features with the technical features with similar functions applied in the present application (but not limited to) with each other.

Claims

1. A method for evaluating automobile production assembly management based on big data, characterized by, The method comprises the following steps: S1, obtaining assembly part quality and assembly condition data during automobile production, assembly equipment historical operation data corresponding to the assembly part during automobile production, and use condition data of the position of each assembly part; S2, performing influence analysis on the corresponding condition of each assembly part based on the assembly part quality and assembly condition data during automobile production, wherein S2 comprises the following specific steps: S21, obtaining assembly part quality influence analysis results from bolt surface burr distribution quantity data and bolt surface area data of each assembly part; S22, obtaining assembly part assembly difficulty analysis results from volume data, assembly step number data, and assembly time length data of each assembly part; S23, obtaining assembly part quality influence analysis results and assembly difficulty analysis results, and adding the weighted assembly part quality influence analysis results and assembly difficulty analysis results to obtain corresponding condition influence analysis results of each assembly part; S3, performing abnormality analysis on the assembly equipment corresponding to each assembly part based on the assembly equipment historical operation data corresponding to the assembly part during automobile production, wherein the specific steps of S3 are as follows: S31, obtaining assembly equipment fault abnormality analysis results corresponding to each assembly part from fault downtime frequency data and average fault time length data in the historical operation cycle of the assembly equipment corresponding to each assembly part; S32, obtaining assembly equipment operation abnormality analysis results corresponding to each assembly part from actual positioning accuracy data and gripper clamping force data of the assembly equipment corresponding to each assembly part; S33, obtaining assembly equipment fault abnormality analysis results and operation abnormality analysis results corresponding to each assembly part, and adding the weighted assembly equipment fault abnormality analysis results and operation abnormality analysis results corresponding to each assembly part to obtain abnormality analysis results of the assembly equipment corresponding to each assembly part; S4, performing estimated assembly abnormality influence analysis based on use condition data of the position of each assembly part, corresponding condition influence analysis results of each assembly part, and abnormality analysis results of the assembly equipment corresponding to each assembly part; S5, determining whether to maintain the automobile production assembly process according to the estimated assembly abnormality influence analysis results.

2. The big data evaluation-based automobile production assembly management method according to claim 1, characterized by, The specific steps of S21 are as follows: performing quality influence analysis on each assembly part according to bolt surface burr distribution quantity data and bolt surface area data of each assembly part, wherein the quality influence analysis process of each assembly part is as follows: dividing bolt surface burr distribution quantity data of each assembly part by bolt surface area data of each assembly part to obtain burr density of each assembly part, and dividing the burr density of each assembly part by a reference burr density to obtain quality influence analysis results of each assembly part.

3. The big data evaluation-based automobile production assembly management method according to claim 2, characterized by, The specific step of S22 is: performing assembly part assembly difficulty analysis according to the assembly part volume data, the assembly step number data and the assembly required time length data, wherein the assembly part assembly difficulty analysis process is: dividing the assembly part assembly step number data by the reference assembly step number to quantify the assembly part assembly process complexity; dividing the assembly part assembly required time length data by the reference assembly part required time length maximum value to quantify the assembly part assembly operation complexity; multiplying the assembly part assembly process complexity and the assembly operation complexity to obtain the assembly part assembly complexity; dividing the difference between the maximum volume of the assembly part and the assembly part volume data by the required assembly part volume maximum value to quantify the assembly part assembly fineness; and adding the assembly part assembly complexity and the assembly fineness after weighting to obtain the assembly part assembly difficulty analysis result.

4. The big data evaluation-based automobile production assembly management method according to claim 3, characterized by, The specific step of S31 is: performing assembly part corresponding assembly equipment fault abnormality analysis according to the fault downtime number data and the average fault time length data in the historical running cycle of the assembly part corresponding assembly equipment, wherein the assembly part corresponding assembly equipment fault abnormality analysis process is: integrating the fault downtime number data in the historical running cycle of the assembly part corresponding assembly equipment over the historical running cycle of the assembly equipment, dividing the integration result by the product of the historical running cycle of the assembly equipment and the reference fault downtime number, and quantifying the fault frequency intensity of the assembly part corresponding assembly equipment; dividing the average fault time length data in the historical running cycle of the assembly part corresponding assembly equipment by the reference fault allowed time length to quantify the fault influence degree of the assembly part corresponding assembly equipment; and multiplying the fault frequency intensity and the fault influence degree of the assembly part corresponding assembly equipment to obtain the assembly part corresponding assembly equipment fault abnormality analysis result.

5. The big data evaluation-based automobile production assembly management method according to claim 4, characterized by, The specific step of S32 is: performing assembly part corresponding assembly equipment operation abnormality analysis according to the actual positioning accuracy data and the actual gripper clamping force data of the assembly part corresponding assembly equipment, wherein the assembly part corresponding assembly equipment operation abnormality analysis process is: dividing the difference between the actual positioning accuracy data of the assembly part corresponding assembly equipment and the standard positioning accuracy by the reference positioning accuracy difference to quantify the actual positioning ability attenuation degree of the assembly part corresponding assembly equipment; dividing the difference between the actual gripper clamping force data of the assembly part corresponding assembly equipment and the standard clamping force by the clamping force allowed difference to quantify the gripper running wear condition of the assembly part corresponding assembly equipment; and multiplying the actual positioning ability attenuation degree and the gripper running wear condition of the assembly part corresponding assembly equipment to obtain the assembly part corresponding assembly equipment operation abnormality analysis result.

6. The big data evaluation-based automobile production assembly management method according to claim 5, wherein The specific steps of S4 are: performing estimated assembly abnormality influence analysis according to the daily use frequency data of each assembly part position, the average use intensity data, the corresponding situation influence analysis result of each assembly part, and the assembly equipment abnormality analysis result corresponding to each assembly part, wherein the estimated assembly abnormality influence analysis process is: multiplying the daily use frequency data of each assembly part position by the average use intensity, dividing the product by the product of the reference use intensity data and the daily use frequency, quantifying the use intensity of each assembly part corresponding position, weighting and adding the corresponding situation influence analysis result of each assembly part and the assembly equipment abnormality analysis result corresponding to each assembly part, multiplying the result by the use intensity of each assembly part corresponding position, and obtaining the estimated assembly abnormality influence analysis result corresponding to each assembly part; and performing summation and average operation on the estimated assembly abnormality influence analysis result corresponding to each assembly part to obtain the estimated assembly abnormality influence analysis result.

7. The big data evaluation-based automobile production assembly management method according to claim 6, characterized by, The specific steps of S5 are: comparing the estimated assembly abnormality influence analysis result with the set estimated assembly abnormality influence analysis result threshold value, if the estimated assembly abnormality influence analysis result is greater than or equal to the set estimated assembly abnormality influence analysis result threshold value, it is judged that the automobile production assembly process needs to be maintained; If the estimated assembly abnormality influence analysis result is less than the set estimated assembly abnormality influence analysis result threshold value, it is judged that the automobile production assembly process does not need to be maintained.

8. The automobile production assembly management system based on big data evaluation, which is implemented based on the automobile production assembly management method based on big data evaluation according to any one of claims 1-7, characterized in that, It specifically includes an automobile production assembly situation acquisition module, an assembly part corresponding situation influence analysis module, an assembly equipment abnormality analysis module, an estimated assembly abnormality influence analysis module, and an estimated assembly maintenance judgment module, the automobile production assembly situation acquisition module is used to acquire the assembly part quality and assembly situation data during automobile production, the assembly equipment historical operation situation data corresponding to the assembly part during automobile production, and the use situation data of each assembly part position; The assembly part corresponding situation influence analysis module is used to perform corresponding situation influence analysis of each assembly part based on the assembly part quality and assembly situation data during automobile production; The assembly equipment abnormality analysis module is used to perform assembly equipment abnormality analysis of each assembly part corresponding to the assembly part based on the assembly equipment historical operation situation data corresponding to the assembly part during automobile production; The estimated assembly abnormality influence analysis module is used to perform estimated assembly abnormality influence analysis based on the use situation data of each assembly part position, the corresponding situation influence analysis result of each assembly part, and the assembly equipment abnormality analysis result corresponding to each assembly part; The estimated assembly maintenance judgment module is used to judge whether to maintain the automobile production assembly process according to the estimated assembly abnormality influence analysis result.

Images