A method and system for operation and maintenance management of a vehicle frame production

Abstract

The application discloses a kind of operation and maintenance management method and system for frame production, it is related to production operation and maintenance technical field.A kind of operation and maintenance management system for frame production, including have: frame production information acquisition module and frame production operation and maintenance management module.The present application is by using internal component sensing information and external image information, fusion equipment operating state and historical process parameters, not only pay attention to the appearance quality of product, also pay attention to dynamic change in processing process, realized the unified analysis of product quality, process stability and equipment health state, effectively improved the precision and real-time of middleware state identification;By introducing fine hierarchical mechanism, avoid the problem of excessive repair or misjudgment in traditional operation and maintenance management;Especially for the re-division and analysis of critical frame middleware, can more accurately identify the intermediate state product that is on the edge of failure but can be repaired, realize on-demand processing, dynamic scheduling, avoid resource waste.

Description

TECHNICAL FIELD

[0001] The present application relates to the production operation and maintenance technical field, and particularly relates to a production operation and maintenance management method and system for a vehicle frame. BACKGROUND

[0002] In the traditional vehicle frame production process, quality detection is mostly dependent on manual inspection or single sensor monitoring, which has problems such as response lag, inaccurate abnormality identification, rough defect positioning, and the like, and it is difficult to achieve fine quality control and predictive operation and maintenance of equipment. In the prior art, although image recognition or equipment state prediction means are introduced, they are mostly based on single-point data, lack comprehensive judgment ability for vehicle frame structure parts, equipment health state and process complexity, and cannot accurately identify critical states or provide executable control suggestions.

[0003] Therefore, a production operation and maintenance management method for a vehicle frame is needed, which can integrate internal sensing information and external image information, realize re-division of abnormal intermediate parts through a prediction model, and make multi-dimensional judgments based on vehicle frame structure parts and risk trends, so as to not only identify critical states, but also repair through an operation and maintenance strategy closed loop, thereby significantly improving the intelligentization, flexibility and traceable operation and maintenance capability of the vehicle frame production line. SUMMARY

[0004] The present application aims to provide a production operation and maintenance management method and system for a vehicle frame, which can manage the defect states that are easily ignored in vehicle frame production, and build an intelligent and traceable vehicle frame production line.

[0005] A production operation and maintenance management method for a vehicle frame, comprising the following steps:

[0006] Obtaining production process information and corresponding production equipment information of vehicle frame intermediate parts; the production process information includes internal component sensing information and external component image information; detecting based on the external component image information to obtain vehicle frame production operation and maintenance information; identifying a vehicle frame part label of the vehicle frame intermediate part;

[0007] Adding a vehicle frame operation and maintenance label to the vehicle frame intermediate part based on the vehicle frame production operation and maintenance information and the vehicle frame part label; the vehicle frame operation and maintenance label includes a vehicle frame intermediate part to be operated and maintained, an abnormal vehicle frame intermediate part and a normal vehicle frame intermediate part;

[0008] For the abnormal vehicle frame intermediate part, production prediction is performed based on the internal component sensing information and the production equipment information to obtain vehicle frame production prediction information; the abnormal vehicle frame intermediate part is re-divided based on the vehicle frame production prediction information to obtain a critical vehicle frame intermediate part or a vehicle frame intermediate part to be operated and maintained; the vehicle frame production prediction information marked as the critical vehicle frame intermediate part is analyzed to obtain production equipment operation and maintenance information;

[0009] The abnormal operation of the to-be-maintained frame middleware is performed on the to-be-maintained frame middleware, and the next production operation is continued to be performed on the normal frame middleware.

[0010] As a preferred technical solution of the present application, the specific steps of adding the frame maintenance label to the frame production middleware based on the frame production and maintenance information and the frame part label include:

[0011] The frame production comparison result is obtained by comparing the frame production and maintenance information and the preset frame part maintenance standard;

[0012] The frame production deviation factor is obtained by identifying the frame production deviation according to the frame production comparison result and the frame part label;

[0013] The frame production middleware is marked as a to-be-maintained frame middleware, an abnormal frame middleware and a normal frame middleware based on the frame production deviation factor.

[0014] As a preferred technical solution of the present application, the specific steps of production prediction based on internal component sensing information and production equipment information include:

[0015] The internal component sensing information and the production equipment information are aligned in time sequence to obtain a middleware processing information fusion sequence; the middleware processing information fusion sequence is processed by using a sliding window of indefinite length to obtain a plurality of continuous sliding middleware processing information pieces;

[0016] The change trend, the coefficient of variation and the perturbation disturbance ratio of the sliding middleware processing information piece are extracted;

[0017] The historical data about frame production is obtained; the equipment health factor and the process complexity index are determined based on the historical data;

[0018] The change trend, the coefficient of variation and the perturbation disturbance ratio are integrated with the equipment health factor and the process complexity index to obtain a frame processing fusion vector; the production prediction information is obtained by performing production prediction based on the frame processing fusion vector;

[0019] The production reliable prediction value, the production abnormal risk trend and the frame production abnormal prediction part are included in the frame production prediction information.

[0020] As a preferred technical solution of the present application, the specific steps of re-dividing the abnormal frame middleware based on the frame production prediction information include:

[0021] The production reliable level area information is obtained by interval mapping based on the production reliable prediction value in the frame production prediction information;

[0022] The production abnormal risk trend in the frame production prediction information is judged, if the production abnormal risk trend is in the trend enhancement interval but does not exceed the threshold value of the serious abnormal interval, critical frame weighted information is obtained, otherwise, the to-be-maintained frame weighted information is obtained.

[0023] Based on the frame production abnormal prediction part in the frame production prediction information, critical judgment is performed:

[0024] Step S1: identifying the importance level of the frame production abnormal prediction part; based on the importance level, the production reliable level area information and the critical frame weighted information or the to-be-maintained frame weighted information, comprehensive judgment is performed to obtain first frame intermediate part judgment information;

[0025] Step S2: based on the frame production prediction information and the pre-trained enhanced fuzzy judgment model, judgment is performed to obtain second frame intermediate part judgment information;

[0026] According to the first frame intermediate part judgment information and the second frame intermediate part judgment information, it is determined that the abnormal frame intermediate part belongs to the critical frame intermediate part or the to-be-maintained frame intermediate part.

[0027] As a preferred technical solution of the present application, the frame production prediction information marked as a critical frame intermediate part is analyzed to obtain specific steps of production equipment maintenance information, including:

[0028] Based on the production abnormal risk trend and the frame production abnormal prediction part, reverse judgment is performed, and similar known internal faults of the equipment are matched; if the matching is successful, the production equipment maintenance information is generated based on the known internal faults of the equipment; if the matching fails, the corresponding production equipment is adjusted based on the corresponding production equipment information to obtain the production equipment maintenance information.

[0029] After adjusting the corresponding production equipment by using the production equipment maintenance information, the critical frame intermediate part is reprocessed until a normal frame intermediate part is obtained.

[0030] As a preferred technical solution of the present application, based on the external component image information, detection is performed to obtain specific steps of frame production and maintenance information, including: using the trained frame defect detection and recognition model to perform feature recognition on the external component image information to obtain the frame production and maintenance information; the frame defect detection and recognition model is trained based on the CNN model.

[0031] A maintenance management system for frame production includes:

[0032] The frame production information acquisition module comprises a data acquisition unit; the data acquisition unit is used for acquiring production process information and corresponding production equipment information of frame production intermediate parts; the production process information comprises internal component sensing information and external component image information; frame production operation and maintenance information is obtained by detecting based on the external component image information; a frame part label of the frame production intermediate part is identified;

[0033] The frame production operation and maintenance management module comprises an operation and maintenance identification unit and an operation and maintenance judgment unit; the operation and maintenance identification unit is used for adding a frame operation and maintenance label to the frame production intermediate part based on the frame production operation and maintenance information and the frame part label; the frame operation and maintenance label comprises a to-be-operated frame intermediate part, an abnormal frame intermediate part and a normal frame intermediate part; the operation and maintenance judgment unit is used for performing production prediction based on the internal component sensing information and the production equipment information for the abnormal frame intermediate part, obtaining frame production prediction information; the abnormal frame intermediate part is re-divided based on the frame production prediction information, obtaining a critical frame intermediate part or a to-be-operated frame intermediate part; the frame production prediction information of the critical frame intermediate part is analyzed, obtaining production equipment operation and maintenance information; the to-be-operated frame intermediate part is executed for operation and maintenance abnormal operation; for the normal frame intermediate part, the next step production operation is continued to be executed on the normal frame intermediate part.

[0034] The present application has the following advantages:

[0035] 1、The present application uses internal component sensing information and external image information to fuse equipment running state and historical process parameters, not only pays attention to the appearance quality of the product, but also pays attention to the dynamic change in the processing process, realizes unified analysis of product quality, process stability and equipment health state, effectively improves the precision and real-time performance of intermediate part state recognition; by introducing a refined hierarchical mechanism, the problems of excessive maintenance or misjudgment in traditional operation and maintenance management are avoided; especially for the re-division and analysis of the critical frame intermediate part, the intermediate state product that is on the edge of failure but can be repaired can be more accurately identified, on-demand processing and dynamic scheduling are realized, and resource waste is avoided.

[0036] 2、The present application can make early judgment on production trend before defects appear by constructing a fusion feature vector and a trained prediction model, realize early identification of abnormal trend; especially the introduction of equipment health factor and process complexity index makes the model have better context understanding ability, improves the explainability and adaptability of prediction; after identifying the critical state, the equipment failure mode can be automatically matched, or the equipment parameters can be analyzed and adjusted in reverse, automatic repair or fine tuning is completed, and the intermediate part is reprocessed and verified, forming a complete processing control closed loop, effectively reducing the scrap rate and improving the stable operation ability of the production line. BRIEF DESCRIPTION OF DRAWINGS

[0037] Figure 1A structural schematic diagram of an operation and maintenance management system for vehicle frame production adopted by an embodiment of the present application. DETAILED DESCRIPTION

[0038] In order for those skilled in the art to better understand the technical solutions in the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0039] Embodiment 1, an operation and maintenance management method for vehicle frame production, comprising the following steps:

[0040] Obtaining production process information and corresponding production equipment information of vehicle frame intermediate parts; the production process information contains internal component sensing information and external component image information; detecting based on the external component image information to obtain vehicle frame production operation and maintenance information; identifying a vehicle frame part label of the vehicle frame intermediate part;

[0041] The main role of identifying the vehicle frame part label of the vehicle frame intermediate part is to realize the accuracy of defect positioning and processing, the structuring of quality analysis, the individualization of operation and maintenance strategies, and the automation of production decision-making; in the production process with complex vehicle frame structure and various processes, the processing requirements, tolerance standards, and quality control points of different parts are obviously different, and simply identifying that the intermediate part has a problem is far from enough, and the specific structure part where the problem occurs must be further identified;

[0042] The production equipment information refers to a collection of various static attributes, dynamic states, running history, and process configuration information related to the equipment involved in the processing in the vehicle frame production process, which reflects the technical capability of the equipment itself, the current running situation, and the influence on the production process; specifically, the specific process parameters used by the equipment when performing different processing tasks, such as the current and voltage of the welding machine, the speed and angle of the cutting equipment, and the clamping pressure of the assembly equipment, etc., these parameters are direct variables that affect product quality.

[0043] The operation and maintenance management method for vehicle frame production involves multiple data collection and analysis links, the internal component sensing information refers to the data collected by various sensors embedded in the vehicle frame tooling fixture, processing equipment or intermediate part body in real time during the vehicle frame production process, these information usually includes parameters such as temperature, pressure, torque, vibration, current, voltage, etc. generated during processing, to reflect the state change of the internal structure during processing; the acquisition method of these sensing information depends on the Internet of Things sensors installed at key positions, through real-time collection and data uploading operation, to realize continuous monitoring of the internal running state during processing.

[0044] The external component image information refers to the appearance images of the intermediate frame in the frame obtained by the industrial vision system. These images can cover key parts such as welds, connection points, and shell curved surfaces, and are used to detect whether there are external visible defects such as cracks, deformation, burns, incomplete welding, etc. Image acquisition usually uses multi-angle industrial cameras, which automatically take pictures after the frame completes a certain key process, and cooperate with the light source system to improve the imaging clarity. The obtained image information is then input into the frame defect detection and recognition model for recognition and analysis.

[0045] The specific steps of detecting based on the external component image information to obtain the frame production and operation information include:

[0046] The trained frame defect detection and recognition model is used for feature recognition of the external component image information to obtain the frame production and operation information. The frame defect detection and recognition model is trained based on a CNN model.

[0047] The frame production and operation information refers to a data set obtained based on real-time monitoring and analysis of the state of the intermediate frame (semi-finished frame) during frame production, which is used to guide subsequent operation judgment and decision-making. For example, whether there are visual visible defects (from external image recognition) such as cracks, deformation, and virtual welding, and whether there are stress abnormalities, temperature abnormalities, and current deviations (from sensor data) in the internal parameters.

[0048] The training of the frame defect detection and identification model first relies on the construction of a high-quality data set; in order to ensure that the model can accurately identify various external defects that may occur in the frame, a large number of frame image samples from the actual production line need to be collected, which should cover different process stages, different part areas, and contain multiple types of defects, such as welding cracks, virtual welding, burns, deformation, foreign matter adhesion, etc.; after the image collection is completed, experienced quality inspection engineers need to label the images to clearly indicate the location, range, and defect type of the defects, forming a structured image and label pair, which is used as the basic data for model training; the model is built based on a convolutional neural network architecture, usually selecting a deep network structure with strong feature extraction capability as the basis, such as a network structure based on residual connection or a lightweight feature pyramid structure, in order to better extract details such as welds and edges; during model training, pre-processed image data is input into the model along with the labeled defect labels, and parameter optimization is performed through a back propagation algorithm; during training, a cross-entropy loss function or a multi-task loss function is used to jointly optimize the accuracy and positioning accuracy of target detection, while a positive and negative sample balancing strategy is introduced to prevent the model from producing excessive bias towards large background areas. A validation set is set during training to monitor the model's performance on non-training samples to prevent overfitting; when the model performs stably on the validation set and reaches the expected accuracy standard, the training is stopped and the model weights are saved, obtaining the final frame defect detection and identification model that can be used in production environment.

[0049] Based on the frame production operation information and the frame part label, a frame operation label is added to the frame production intermediate part; the frame operation label includes a frame intermediate part to be operated, an abnormal frame intermediate part and a normal frame intermediate part;

[0050] The specific steps of adding a frame operation label to a frame production intermediate part based on frame production operation information and frame part labels include:

[0051] Based on the frame production operation information and the preset frame part operation standard, a frame production comparison result is obtained;

[0052] The frame operation standard is a quality threshold system for different structural parts established according to historical production data, quality specifications, process requirements and experience knowledge, such as allowable range of weld width, maximum stress value at connection, vibration signal fluctuation limit, etc.; when comparing, the system will compare the external image detection result and the internal sensor data with these standards one by one, judge whether each index exceeds the normal tolerance range, and output the comparison result in a structured form to provide a basis for subsequent deviation analysis.

[0053] According to the frame production comparison result and the frame part label, a frame production deviation factor is obtained;

[0054] After obtaining the frame production comparison result, in combination with the identified frame part label, the system will perform fine deviation identification on the production quality state of the current intermediate part. This process not only judges whether a certain index is out of range, but also comprehensively considers the position of the deviation, the deviation amplitude, the process step, and the historical abnormal distribution; for example, the same temperature anomaly appears in the bottom pipe and the adapter pipe, and the impact may be different. The system combines the comparison result and the part label, calculates the frame production deviation factor reflecting the deviation degree and impact strength, and clearly indicates the severity and impact range of the problem.

[0055] Based on the frame production deviation factor, the frame production intermediate part is marked as a to-be-maintained frame intermediate part, an abnormal frame intermediate part, and a normal frame intermediate part.

[0056] After obtaining the frame production deviation factor, the system will divide the intermediate part based on its numerical range, trend judgment, and context process condition; if the deviation factor is low, the fluctuation is stable, and there is no obvious development trend, the intermediate part will be marked as a normal frame intermediate part; if the deviation factor is in the medium interval, it may be controllable in the short term but has uncertain risks, and will be marked as a to-be-maintained frame intermediate part for further observation or reprocessing; and if the deviation factor exceeds the serious threshold, or a significant deviation occurs in the key part, it will be directly marked as an abnormal frame intermediate part, which needs to be reworked or intervened.

[0057] For the abnormal frame intermediate part, production prediction is performed based on internal component sensing information and production equipment information to obtain frame production prediction information; based on the frame production prediction information, the abnormal frame intermediate part is further divided to obtain a critical frame intermediate part or a to-be-maintained frame intermediate part; the production equipment maintenance information is obtained by analyzing the frame production prediction information marked as a critical frame intermediate part.

[0058] The specific steps of production prediction based on internal component sensing information and production equipment information include:

[0059] Align the internal component sensing information and production equipment information in time sequence to obtain intermediate part processing information fusion sequence; process the intermediate part processing information fusion sequence using a sliding window of indefinite length to obtain a plurality of continuous sliding intermediate part processing information pieces;

[0060] The two types of information are aligned according to a unified time axis to form an intermediate part processing information fusion sequence covering the entire processing period. The sequence integrates sensor values such as temperature, vibration, and current during processing, as well as equipment operating status, processing rhythm, load changes, and other content, and constructs a dynamic process data set reflecting the real processing state of the frame intermediate part in a specific time period, laying a foundation for subsequent feature extraction and trend analysis.

[0061] After generating the intermediate part processing information fusion sequence, the system performs slicing processing on the sequence using a sliding window of indefinite length. The length of the sliding window can be flexibly set according to different process stages or sensor data fluctuation characteristics to ensure that each information slice covers a key change period. By sliding the window, the system can obtain a series of continuous intermediate part processing information slices, each of which represents a local working condition change interval. This process helps to capture subtle but trend-relevant changes in the processing process, while avoiding misjudgments caused by data dilution or noise interference, enhancing the sensitivity and stability of local prediction.

[0062] Extracting the change trend, coefficient of variation, and perturbation disturbance ratio of the sliding intermediate part processing information slice;

[0063] After obtaining the sliding intermediate part processing information slice, feature extraction is performed on each slice, focusing on calculating the change trend, coefficient of variation, and perturbation disturbance ratio. The change trend is used to determine whether the processing parameters in the slice are tending to be stable, increasing, or decreasing, reflecting the evolution direction of the process. The coefficient of variation is used to measure the relative amplitude of data fluctuations within the slice, evaluating the stability of the processing process. The perturbation disturbance ratio focuses on the degree of mutation between the current slice and the previous slice, which is an important indicator for identifying sudden abnormalities. Through the extraction of these features, the system can construct a multi-dimensional dynamic description of the current intermediate part processing state.

[0064] Obtaining historical data related to frame production; determining equipment health factor and process complexity index based on historical data;

[0065] The equipment health factor is a comprehensive indicator reflecting the current operating condition, performance stability, and fault risk level of the equipment. It measures whether a device is in a healthy and safe processing state. Its value is usually a standardized score, with a range of 0 to 100. The higher the value, the healthier the equipment, and the lower the value, the worse the equipment operating state and the higher the risk. The process complexity index measures the degree of dependence and sensitivity of a frame processing procedure on equipment, operating precision, parameter control requirements, etc. during execution. It reflects the difficulty and risk level of the current process of the intermediate part in quality control. It is also a standardized score, and the higher the value, the more complex the process and the higher the requirement on the system.

[0066] By calling the frame production history database, historical production data related to the current middleware process node, device type and material batch are obtained; based on these historical data, the system conducts a comprehensive evaluation of the equipment currently participating in processing, generating an equipment health factor reflecting factors such as its running state, failure frequency, maintenance record and usage time length; at the same time, according to the process route complexity, processing precision requirement, control parameter fluctuation tolerance and other indicators, a process complexity index is generated; these two types of factors together reflect the background conditions of the current production environment affecting the frame middleware, which helps to improve the model's ability to judge boundary states and critical risks.

[0067] The change trend, coefficient of variation and perturbation ratio are integrated with the equipment health factor and the process complexity index to obtain a frame processing fusion vector; based on the frame processing fusion vector, production prediction is performed to obtain frame production prediction information;

[0068] The change trend, coefficient of variation and perturbation ratio extracted from the processing information piece are multi-dimensionally integrated with the equipment health factor and the process complexity index to form a complete frame processing fusion vector. This fusion vector not only contains real-time processing dynamic characteristics, but also integrates equipment and process background factors, and has higher expression ability and prediction value. The fusion vector is then input into the trained production prediction model. Through learning from historical labeled data, the model can output the processing quality state and risk trend of the current frame middleware.

[0069] The training set of the production prediction model comes from the frame historical production data, including the sensing information (such as temperature, current, vibration) collected during the middleware processing process, the equipment running state (such as running time, fault record), the process parameters and the corresponding final quality judgment results (qualified or specific defect type). These data are extracted through the manufacturing execution system and the quality management system and then standardized. The model training uses a deep neural network-based architecture, taking the fused middleware processing features as input and the quality state as label output. The system adjusts the model parameters through repeated iterations to make the prediction results consistent with the actual labels as much as possible. Cross-validation is used in the training process to evaluate the model's performance on unseen samples, ensuring its good generalization ability. The training termination conditions include that the validation set performance no longer improves over multiple training rounds, the model reaches the preset accuracy requirement, or the training round number reaches the set upper limit. The model parameters with the best validation effect are finally retained for middleware state prediction and operation and maintenance judgment in actual frame production. CNN model or BP neural network model can be used as the basic model of the production prediction model.

[0070] The frame production prediction information includes production reliable prediction value, production abnormal risk trend and frame production abnormal prediction part; intelligent prediction is performed based on the frame processing fusion vector to generate frame production prediction information; the information mainly includes three aspects: first, the production reliable prediction value indicates whether the intermediate part has acceptable processing stability under the current process environment; second, the production abnormal risk trend is used to evaluate whether the intermediate part is in a potential deterioration trend, especially in a critical situation close to the failure edge; third, the frame production abnormal prediction part is the structure position most likely to have an abnormality, and these prediction results will be directly used as an important basis for subsequent division of intermediate part state, optimization of equipment parameters and scheduling of maintenance strategy.

[0071] The specific steps of re-dividing the abnormal frame intermediate part based on the frame production prediction information include:

[0072] Interval mapping is performed on the production reliable prediction value in the frame production prediction information to obtain production reliable level area information;

[0073] The production reliable prediction value in the frame production prediction information is interval mapped. The system classifies and processes the production reliable prediction value according to the preset scoring interval, for example, divides it into three level areas of high reliability, medium reliability and low reliability. The interval division is determined based on historical labeling data and quality stability analysis results, and can reflect the comprehensive stability degree of the current intermediate part under the production condition. After interval mapping, the system can obtain the production reliable level area information of the intermediate part, providing a basic reference for subsequent risk judgment.

[0074] The production abnormal risk trend in the frame production prediction information is judged. If the production abnormal risk trend is in the trend enhancement interval but has not exceeded the serious abnormality threshold, critical frame weighted information is obtained; otherwise, to-be-maintained frame weighted information is obtained;

[0075] The risk trend is curve information generated based on sensor data and equipment state changes, and is used to judge whether the state of the intermediate part is deteriorating. If the trend is in the obvious rising stage, that is, the state of the intermediate part is developing in the unstable direction, but has not exceeded the warning threshold of serious abnormality, the system will consider that the intermediate part has potential risks but can be recovered, and therefore generates critical frame weighted information. The information indicates that the intermediate part is in a controllable but dangerous boundary area; if the trend is obviously abnormal or close to the critical upper limit, the system will generate to-be-maintained frame weighted information, prompting that the intermediate part tends to need rework or maintenance.

[0076] Critical judgment is performed based on the frame production abnormal prediction part in the frame production prediction information:

[0077] Step S1: identifying the importance level of the abnormal frame production prediction part; based on the importance level, production reliability level area information and critical frame weighting information or to-be-maintained frame weighting information, comprehensive judgment is made to obtain first frame intermediate part judgment information;

[0078] The importance level is divided based on structural safety, load-bearing and product performance influence degree, for example, main weld and support connection point are high importance level parts, and auxiliary decorative structure is low importance level; after identification, the system will combine the importance level of the part, the aforementioned production reliability level area information and the generated critical frame or to-be-maintained frame weighting information, and make comprehensive logical judgment to obtain the first frame intermediate part judgment information. This information represents the preliminary classification suggestion for the intermediate part under the engineering rules and risk level standards.

[0079] Step S2: based on the frame production prediction information and the pre-trained enhanced fuzzy judgment model, judgment is made to obtain the second frame intermediate part judgment information; in order to enhance the flexibility and accuracy of judgment, the system also inputs the production prediction information of the frame intermediate part into a pre-trained enhanced fuzzy judgment model for auxiliary judgment; based on the principle of fuzzy logic reasoning, combined with fuzzy membership function and historical sample data, the model scores the fuzzy attribution of the input features through adaptive weight distribution, and outputs a critical state probability as the second frame intermediate part judgment information; the model is particularly sensitive to boundary states, and is suitable for processing fuzzy critical situations that cannot be judged by hard rules. The second frame intermediate part judgment information output by the model provides a supplementary judgment for the system.

[0080] According to the first frame intermediate part judgment information and the second frame intermediate part judgment information, it is determined whether the abnormal frame intermediate part belongs to the critical frame intermediate part or the to-be-maintained frame intermediate part;

[0081] The first frame intermediate part judgment information and the second frame intermediate part judgment information are cross-compared and logically integrated to form a final decision; if they are consistent, the intermediate part is directly classified into the corresponding category; if there is a difference, the system will adopt a set weight mechanism or introduce confidence analysis to fuse the two results; in this way, the abnormal frame intermediate part will be finally divided into critical frame intermediate part or to-be-maintained frame intermediate part, and used as the basis for subsequent processing.

[0082] The frame production prediction information marked as critical frame intermediate part is analyzed to obtain the specific steps of production equipment maintenance information, including:

[0083] Based on the production abnormal risk trend and the frame production abnormal prediction position, reverse judgment is performed to match similar known internal equipment failures. If a match is successful, production equipment operation and maintenance information is generated based on the known internal equipment failures. If a match fails, the corresponding production equipment information is used to adjust the production equipment, and production equipment operation and maintenance information is obtained.

[0084] In the case of processing a frame intermediate part marked as critical, the system first analyzes the frame production prediction information corresponding to the intermediate part, especially the production abnormal risk trend and the frame production abnormal prediction position. The core purpose of reverse judgment is to speculate the possible equipment reasons that cause the intermediate part to enter a critical state. The system compares the current risk trend characteristics and the predicted abnormal position with the equipment failure modes in the historical failure knowledge base to find whether there is a known internal equipment failure case similar to the characteristics, such as insufficient heat input caused by welding current drift, welding seam deviation caused by feeding mechanism jamming, etc.

[0085] After adjusting the corresponding production equipment using the production equipment operation and maintenance information, the critical frame intermediate part is reprocessed until a normal frame intermediate part is obtained.

[0086] If the system matches a device failure case highly consistent with the current risk characteristics in the historical failure library, it means that the critical state is likely caused by internal equipment problems. At this time, the system extracts the processing measures, maintenance suggestions, and fault component information associated with the matching failure, and generates complete production equipment operation and maintenance information. The operation and maintenance information includes the equipment number to be repaired, the recommended repair site, the recommended repair operation (such as replacing the welding gun, recalibrating the electronic control unit, calibrating the clamp pressure, etc.), and may also include a priority level to guide maintenance personnel to accurately locate and quickly process.

[0087] If no explicit historical failure case is matched in the failure knowledge base, it means that the current critical state may not belong to known equipment failure types. At this time, the system analyzes the production equipment information used when processing the intermediate part, extracts its process parameters, running time, loading mode, current load and motion curve, and compares the differences with the historical optimal equipment configuration to speculate the potential setting problems that may cause the critical state. The system will generate a set of optimization adjustment suggestions for the equipment based on this, as new production equipment operation and maintenance information output. Such adjustment suggestions may include welding current fine tuning, slightly slower feeding speed, clamp synchronization time optimization, etc., aiming to restore the equipment processing stability through soft regulation.

[0088] After the operation and maintenance information is generated, the system applies it to the adjustment operation of the corresponding device; the device adjustment can be automatically completed by the system, or can be manually fine-tuned on site according to the suggestion; after the adjustment is completed, the original intermediate part marked as a critical state is returned to the corresponding work station for reprocessing; through reprocessing by applying new device parameters and synchronous quality monitoring, the system can determine whether the intermediate part has been converted from a critical state to a normal state; if the detection result meets the standard, it is determined that the intermediate part has returned to a normal frame intermediate part, and the process ends; if there are still critical signs, the device fine-tuning and reprocessing process can be entered again until the intermediate part meets the quality requirements, realizing a closed-loop processing mode of precise repair and device self-correction.

[0089] For the intermediate part of the frame to be operated and maintained, the operation and maintenance exception operation is performed on the intermediate part of the frame to be operated and maintained; for the normal frame intermediate part, the next production operation is continued on the normal frame intermediate part.

[0090] When the frame intermediate part is determined by the system to be in a state to be operated and maintained, it means that the intermediate part has a relatively clear exception in the production process and has exceeded the process tolerance range, and cannot directly flow into the next process, but has not reached the severity of scrapping or returning to the factory; therefore, the system will immediately perform an operation and maintenance exception operation on it; first, the intermediate part is automatically separated from the main production rhythm and transferred to a specially designed exception processing station or buffer area; according to the defect type, defect position and related device running state of the intermediate part, the system calls the operation and maintenance processing scheme library to generate a special reprocessing strategy or repair instruction; for example, for the case of poor welding, the system will assign repositioning, local repair welding and secondary quality inspection; if it is a slight deformation, it will enter the correction process; the entire exception operation process will be recorded by the system, including repair time, responsible equipment, processing method and result feedback, as an important data source for equipment traceability and quality statistics; after the operation and maintenance processing is completed and verified, the intermediate part can be re-integrated into the production process.

[0091] For the part determined to be a normal frame intermediate part, the system will directly allow it to enter the next standardized production process without any intervention or correction operation; the system marks the processing quality state of the intermediate part as qualified, and updates its state code in the production traceability system, automatically transfers the intermediate part to the corresponding processing station of the next process according to the current process route and plan scheduling, such as from welding to polishing, from assembly to painting, etc.; during the transfer process, the system will synchronously carry the complete production data package of the intermediate part, including process parameters, detection records and equipment number, to ensure that the subsequent process can continue to perform seamless processing operation according to the current state, and finally realize stable and efficient production closed loop.

[0092] In this embodiment, a usable vehicle frame production operation and maintenance management method is given. For example, in the production process of the main weld connection area of the vehicle frame, the system first acquires the internal component sensing information and equipment operation data generated by the intermediate part in the welding stage; the sensor collects real-time temperature, current, voltage and vibration parameters, and at the same time collects the running time of the equipment, process setting value, clamp pressure and welding rhythm. Combined with the external image information obtained by the industrial vision system, the system detects that there is a slight discontinuous weld in the main weld area of the intermediate part through the defect recognition model trained based on the convolutional neural network; then the system identifies the defect in the main weld connection area according to the standardized vehicle frame structure model, and establishes a corresponding label with historical quality data.

[0093] The system further aligns the sensing data and the equipment operation state by time, constructs the intermediate part processing information fusion sequence, and applies an indefinite length sliding window to extract the change trend, volatility and disturbance index. At the same time, according to the process precision requirement and equipment maintenance condition corresponding to the process, the device health factor and the process complexity index are generated; the fused features are input into the trained vehicle frame production prediction model, and the system outputs the prediction results as a medium-low reliability score, an abnormal trend of upward trend, and an abnormal risk positioning of the weld connection area; according to these results, the system preliminarily marks the intermediate part as abnormal, and further executes the redivision process.

[0094] The system maps the production reliability score to the medium-low section, generates critical vehicle frame weighted information in combination with the abnormal trend which has not broken through the critical threshold; at the same time, the main weld connection area is identified as a high structure importance level area, and in combination with its critical state and risk upward trend, the system determines that the intermediate part is a critical vehicle frame intermediate part through rule reasoning and fuzzy judgment model double verification. The system then reversely searches the historical fault knowledge base, matches similar welding current decline type fault records, and automatically generates a recommended repair content, prompting to adjust the current control module and recalibrate the welding gun trajectory; after completing the equipment adjustment, the intermediate part is arranged into the repair process for local repair welding and re-detection, and finally qualified and restored to the normal intermediate part, and continues to flow to the subsequent polishing process.

[0095] Embodiment 2, a vehicle frame production operation and maintenance management system, as shown in Figure 1 The system comprises:

[0096] The vehicle frame production information acquisition module comprises a data acquisition unit; the data acquisition unit is used to acquire the production process information of the vehicle frame intermediate part and the corresponding production equipment information; the production process information comprises internal component sensing information and external component image information; the vehicle frame production operation and maintenance information is obtained by detecting based on the external component image information; the vehicle frame part label of the vehicle frame production intermediate part is identified;

[0097] The frame production operation and maintenance management module comprises an operation and maintenance identification unit and an operation and maintenance judgment unit; the operation and maintenance identification unit is used for adding a frame operation and maintenance label to a frame intermediate part based on frame production operation and maintenance information and a frame part label; the frame operation and maintenance label comprises an abnormal frame intermediate part, an abnormal frame intermediate part and a normal frame intermediate part; the operation and maintenance judgment unit is used for obtaining frame production prediction information by performing production prediction based on internal component sensing information and production equipment information for the abnormal frame intermediate part; the abnormal frame intermediate part is re-divided based on the frame production prediction information to obtain a critical frame intermediate part or a to-be-operated frame intermediate part; production equipment operation and maintenance information is obtained by analyzing frame production prediction information marked as a critical frame intermediate part; operation and maintenance abnormal operation is performed on the to-be-operated frame intermediate part for the to-be-operated frame intermediate part; and the next production operation is continued to be performed on the normal frame intermediate part for the normal frame intermediate part.

[0098] It should be understood that, for those skilled in the art, improvements or changes can be made according to the above description, and all these improvements and changes shall belong to the protection scope of the appended claims of the present application. The parts not described in detail in the specification belong to the prior art known to those skilled in the art.

Claims

1. A method for operation and maintenance management of a vehicle frame production, characterized by, The method comprises the following steps: Obtain production process information and corresponding production equipment information of a frame production intermediate part; The production process information includes internal component sensing information and external component image information; Detect based on the external component image information to obtain frame production operation and maintenance information; identify the frame part label of the frame production intermediate part; Add frame operation and maintenance labels to the frame production intermediate part based on the frame production operation and maintenance information and the frame part label; the frame operation and maintenance label includes a frame intermediate part to be maintained, an abnormal frame intermediate part, and a normal frame intermediate part; For the abnormal frame intermediate part, production prediction is performed based on the internal component sensing information and the production equipment information to obtain frame production prediction information; the abnormal frame intermediate part is re-divided based on the frame production prediction information to obtain a critical frame intermediate part or a frame intermediate part to be maintained; analyze the frame production prediction information marked as the critical frame intermediate part to obtain production equipment operation and maintenance information; For the frame intermediate part to be maintained, perform operation and maintenance abnormal operation on the frame intermediate part to be maintained; for the normal frame intermediate part, continue to perform the next production operation on the normal frame intermediate part; The specific steps of production prediction based on internal component sensing information and production equipment information include: Align the internal component sensing information and the production equipment information in time sequence to obtain intermediate part processing information fusion sequence; process the intermediate part processing information fusion sequence using a sliding window of indefinite length to obtain a plurality of continuous sliding intermediate part processing information pieces; Extract the change trend, coefficient of variation, and perturbation disturbance ratio of the sliding intermediate part processing information piece; Obtain historical data related to frame production; determine the equipment health factor and the process complexity index based on the historical data; Integrate the change trend, coefficient of variation, and perturbation disturbance ratio with the equipment health factor and the process complexity index to obtain a frame processing fusion vector; perform production prediction based on the frame processing fusion vector to obtain frame production prediction information; The frame production prediction information includes production reliable prediction value, production abnormal risk trend, and frame production abnormal prediction part; The specific steps of re-dividing the abnormal frame intermediate part based on the frame production prediction information include: Interval mapping based on the production reliable prediction value in the frame production prediction information to obtain production reliable level area information; Judge the production abnormal risk trend in the frame production prediction information; if the production abnormal risk trend is in the trend enhancement interval but does not exceed the serious abnormality threshold, obtain critical frame weighting information; otherwise, obtain frame intermediate part to be maintained weighting information; Based on the frame production abnormal prediction part in the frame production prediction information, critical judgment is performed: Step S1: Identify the importance level of the frame production abnormal prediction part; based on the importance level, the production reliable level area information, and the critical frame weighting information or the frame intermediate part to be maintained weighting information, comprehensive judgment is performed to obtain first frame intermediate part judgment information; Step S2: Based on the frame production prediction information and the pre-trained enhanced fuzzy judgment model, judgment is performed to obtain second frame intermediate part judgment information; Determine whether the abnormal frame intermediate part belongs to the critical frame intermediate part or the maintenance-to-be frame intermediate part according to the first frame intermediate part determination information and the second frame intermediate part determination information.

2. The operation and maintenance management method for vehicle frame production according to claim 1, characterized in that, The specific steps of adding the frame maintenance label to the frame production intermediate part based on the frame production and maintenance information and the frame part label include: Comparing the frame production and maintenance information with the preset frame part maintenance standard to obtain a frame production comparison result; According to the frame production comparison result and the frame part label, a frame production deviation is identified to obtain a frame production deviation factor; Based on the frame production deviation factor, the frame production intermediate part is marked as a maintenance-to-be frame intermediate part, an abnormal frame intermediate part, and a normal frame intermediate part.

3. The operation and maintenance management method for vehicle frame production according to claim 2, characterized in that, The specific steps of analyzing the frame production prediction information marked as the critical frame intermediate part to obtain the production equipment maintenance information include: Based on the production abnormal risk trend and the frame production abnormal prediction part, a reverse judgment is performed to match similar known internal faults of the equipment; if the matching is successful, the production equipment maintenance information is generated based on the known internal faults of the equipment; if the matching fails, the corresponding production equipment is adjusted based on the corresponding production equipment information to obtain the production equipment maintenance information; After adjusting the corresponding production equipment using the production equipment maintenance information, the critical frame intermediate part is reprocessed until a normal frame intermediate part is obtained.

4. The operation and maintenance management method for vehicle frame production according to claim 3, characterized in that, The specific steps of obtaining the frame production and maintenance information based on the external component image information include: using a trained frame defect detection and identification model to perform feature recognition on the external component image information to obtain the frame production and maintenance information; the frame defect detection and identification model is trained based on a CNN model.

5. An operation and maintenance management system for vehicle frame production, characterized by, The system applies the operation and maintenance management method for frame production in any one of claims 1-4, comprising: The frame production information acquisition module includes a data acquisition unit; the data acquisition unit is used to acquire the production process information of the frame production intermediate part and the corresponding production equipment information; the production process information includes internal component sensing information and external component image information; the frame production and maintenance information is obtained based on the detection of the external component image information; the frame part label of the frame production intermediate part is identified; The frame production and maintenance management module includes an operation and maintenance identification unit and an operation and maintenance judgment unit; the operation and maintenance identification unit is used to add the frame maintenance label to the frame production intermediate part based on the frame production and maintenance information and the frame part label; the frame maintenance label includes the maintenance-to-be frame intermediate part, the abnormal frame intermediate part, and the normal frame intermediate part; the operation and maintenance judgment unit is used to perform production prediction based on the internal component sensing information and the production equipment information for the abnormal frame intermediate part to obtain frame production prediction information; the abnormal frame intermediate part is re-divided based on the frame production prediction information to obtain the critical frame intermediate part or the maintenance-to-be frame intermediate part; the production equipment maintenance information is obtained by analyzing the frame production prediction information marked as the critical frame intermediate part; the maintenance-to-be frame intermediate part is executed for the maintenance-to-be frame intermediate part; the next production operation is continued for the normal frame intermediate part.

Images