Real-time monitoring and fault diagnosis method and system for trailer production line

Abstract

The application discloses a real-time monitoring and fault diagnosis method and system for a trailer production line, and relates to the technical field of production line monitoring.The real-time monitoring and fault diagnosis method comprises a production line diagnosis method and a workpiece diagnosis method.The workpiece diagnosis method comprises real-time monitoring and data acquisition: real-time detection of workpieces in a workpiece set by a detection device to obtain real-time data of the workpieces, the workpiece set comprising initial workpieces and combined workpieces.A virtual model of the combined workpieces is obtained according to a virtual model of the initial workpieces and structural change data, diagnosis data of the initial workpieces is reflected on the virtual model of the combined workpieces and is analyzed through global simulation, and simulation fault diagnosis of the combined workpieces is realized, that is, early warning of faults is realized before the faults occur, and the situation that the production efficiency of the trailer production line is affected by faults occurring after combination is reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of production line monitoring, in particular to a real-time monitoring and fault diagnosis method and system for a trailer production line. BACKGROUND

[0002] As an important transportation tool, the production efficiency and quality requirements of trailers are also increasingly high. However, the equipment on the trailer production line is complex and diverse, and the production process is complicated, which is prone to various faults and problems. Therefore, real-time monitoring of trailers on the trailer production line is needed to detect faults and ensure production efficiency and quality.

[0003] For example, patent publication "CN110146316A" entitled "Fault diagnosis method and system for automobile production line function cabinet". The fault diagnosis system of the invention detects the real-time related parameters of the function cabinet in real time through the acquisition unit when the production line is running, compares them with the preset warning value, judges the specific problems existing in the function cabinet according to the experience data recorded in the storage unit when the real-time related parameters exceed the warning value, and pushes the display unit to facilitate the operator to quickly repair without manual troubleshooting. It is beneficial to improve the efficiency of automobile welding.

[0004] The above method uses the way of real-time related parameters exceeding the warning value to realize early warning in order to facilitate subsequent repair. However, for the trailer structure on the trailer production line, the method detects each workpiece in the trailer structure, but the workpieces are connected to each other, and some problems may only occur when the workpieces are combined or after the combination, i.e. early warning is not possible. Therefore, a real-time monitoring and fault diagnosis method and system for a trailer production line are invented. SUMMARY

[0005] The purpose of the present application is to provide a real-time monitoring and fault diagnosis method and system for a trailer production line to solve the problems raised in the background art.

[0006] To achieve the above purpose, the present application provides the following technical solution: the real-time monitoring and fault diagnosis method includes a production line diagnosis method and a workpiece diagnosis method;

[0007] The workpiece diagnosis method includes:

[0008] Real-time monitoring and data acquisition: the detection equipment detects the workpiece set in real time to obtain real-time data of the workpiece, the workpiece set includes initial workpieces and combined workpieces, the combined workpieces are considered as a plurality of initial workpieces combined with each other, the combination of the combined workpieces is obtained, and the structure data of the initial workpieces is obtained;

[0009] Initial workpiece virtual model: a virtual model of the initial workpiece is established according to the real-time data of the initial workpiece;

[0010] Initial fault diagnosis: real-time data of the initial workpiece are subjected to fault diagnosis to obtain diagnosis data of the initial workpiece, and the diagnosis data of the initial workpiece are mapped onto the virtual model of the initial workpiece;

[0011] Establishment and application of workpiece correlation model: a workpiece correlation model is established, the correlation degree between each initial workpiece in the combined workpiece is obtained according to the workpiece correlation model, and structural change data are obtained based on the correlation degree and the combination of the combined workpiece;

[0012] Virtual model of combined workpiece: a virtual model of the combined workpiece is simulated and obtained according to the virtual model of the initial workpiece and the structural change data;

[0013] Virtual model analysis and diagnosis: diagnosis data of the combined workpiece are obtained based on the virtual model of the combined workpiece, the diagnosis data of the initial workpiece and the correlation degree, the diagnosis data of the initial workpiece are reflected on the virtual model of the combined workpiece, global simulation analysis is performed on the virtual model of the combined workpiece, influence data of the diagnosis data on the virtual model of the combined workpiece are simulated, and simulation total diagnosis data are obtained;

[0014] Early warning: simulation fault data are obtained by analyzing the simulation total diagnosis data, and early warning is performed according to the simulation fault data;

[0015] Calibration: real-time data of the combined workpiece are subjected to fault diagnosis, the real-time data of the combined workpiece, fault diagnosis structure of the combined workpiece and simulation total diagnosis data of the corresponding virtual model of the combined workpiece are compared, and comparison data used for modifying parameters in virtual model analysis and diagnosis are obtained.

[0016] Further, the production line diagnosis method comprises:

[0017] Data acquisition: historical production data of equipment in the trailer production line are obtained;

[0018] Real-time monitoring: real-time monitoring of the equipment in the trailer production line and real-time monitoring of the trailer structure are performed to obtain equipment real-time data and structure real-time data;

[0019] Division of equipment in the production line: the equipment in the production line is divided into partial real-time data and total real-time data according to the total attribute of the equipment in the production line;

[0020] Establishment and application of production line correlation model: a production line correlation model is established, and the production line correlation model establishes the correlation degree between the equipment in the production line according to the historical production data;

[0021] Establishment and application of prediction model: a prediction model is established, equipment real-time data are predicted through the prediction model, and equipment prediction data are obtained;

[0022] Application of continuous data: using time series analysis technology and combining the degree of association, the continuity of the sub-real-time data and the total real-time data is deeply mined, the trend in the production process and the continuity between the production lines are identified, and the continuous data between the sub-real-time data and the total real-time data is obtained;

[0023] Establish and apply threshold value: based on historical production data, corresponding equipment threshold value is established, whether the continuous data and the equipment prediction data will exceed the corresponding threshold value is judged, and the threshold value is exceeded, and a warning is given.

[0024] Further, the structural change data includes physical connection, mechanical property and vibration connection, the physical connection includes the connection mode between the initial workpieces in the combined workpiece, the mechanical property includes stress distribution and stiffness change, and the vibration connection includes resonance frequency and amplitude change.

[0025] Further, the method for obtaining the degree of association between the devices on the production line comprises: analyzing the historical production data of the devices in the trailer production line, judging whether there is an association between the devices, establishing the association between the devices, establishing a production line association model for reflecting the interaction and dependence relationship between the devices, analyzing the degree of mutual influence between the devices through the production line association model, and the degree of mutual influence is represented by a quantitative index.

[0026] Further, the method for obtaining the degree of association between the devices on the production line comprises: analyzing the historical production data of the devices in the trailer production line, judging whether there is an association between the devices, establishing the association between the devices, establishing a production line association model for reflecting the interaction and dependence relationship between the devices, analyzing the degree of mutual influence between the devices through the production line association model, and the degree of mutual influence is represented by a quantitative index.

[0027] Further, according to the initial workpieces in the workpiece set, the length data in the structure data is divided into long-span workpieces and non-long-span workpieces, and the fault diagnosis method for the workpiece data of the long-span workpiece comprises segmented monitoring, the segmented monitoring comprises dividing the long-span workpiece into a plurality of shorter paragraphs, and the vibration monitoring and data acquisition are performed on each paragraph respectively, and the segmented data and the position data of each paragraph are obtained.

[0028] Further, the fault diagnosis method for the long-span workpiece comprises:

[0029] The segmented data is established and corresponding to the position on the long-span workpiece according to the position data, the position is calibrated after the corresponding position, the segmented data is time-aligned, and the overall data is obtained;

[0030] Map the overall data to the virtual model, and personnel assist in analysis;

[0031] Signal processing is performed on the overall data, and the processed overall data is processed to obtain characteristic information of the fault;

[0032] The mode recognition is performed on the characteristic information of the fault, the fault type is judged, the specific position of the corresponding fault occurrence is determined by combining the position data and the fault characteristics, and the positioning algorithm is adopted.

[0033] A real-time monitoring and fault diagnosis system of a trailer production line adopts the real-time monitoring and fault diagnosis method of the trailer production line.

[0034] Compared with the prior art, the beneficial effects of the present application are:

[0035] The real-time monitoring and fault diagnosis method and system of the trailer production line, through the establishment of a virtual model, preliminary fault diagnosis and a correlation model, first analyzes the initial workpiece for faults before combining the workpieces, then establishes a virtual model of the initial workpiece, obtains a virtual model of the combined workpiece according to the virtual model of the initial workpiece and the structural change data, reflects the diagnosis data of the initial workpiece on the virtual model of the combined workpiece and analyzes through global simulation, realizes the simulation fault diagnosis of the combined workpiece, that is, realizes the early warning of the fault before the fault occurs, reduces the situation that the fault after combination will affect the production efficiency of the trailer production line.

[0036] At the same time, through real-time monitoring and data acquisition, the equipment and workpieces on the trailer production line can be comprehensively and real-time monitored, ensuring the stability and reliability of the production process, using the real-time data of the initial workpiece to simulate the subsequent combined workpiece, simulating the influence of the diagnosis data on the virtual model of the combined workpiece, discovering potential fault points in advance, and the combination of the production line diagnosis method and the workpiece diagnosis method makes the fault diagnosis more comprehensive and accurate, and the fault in production can be discovered in time.

[0037] By segmenting the long-span workpiece, segmenting the long-span workpiece, and then time-aligning the segmented data, this method can more accurately obtain the vibration data and running state of the workpiece, improve the accuracy and efficiency of fault diagnosis, and combined with the virtual model of the long-span workpiece and personnel assisted analysis, the fault point can be located more quickly. BRIEF DESCRIPTION OF DRAWINGS

[0038] Figure 1 is a schematic diagram of the workpiece diagnosis method of the present application;

[0039] Figure 2 is a flowchart of the workpiece diagnosis method of the present application;

[0040] Figure 3A flow chart of the production line diagnosis method of the present application. DETAILED DESCRIPTION

[0041] The technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0042] As shown in Figure 1 - Figure 3 The present application provides a technical solution: a real-time monitoring and fault diagnosis method including a production line diagnosis method and a workpiece diagnosis method.

[0043] The workpiece diagnosis method includes:

[0044] Real-time monitoring and data acquisition: real-time detection of workpieces in a workpiece set by a detection device to obtain real-time data of the workpieces, the workpiece set including initial workpieces and combined workpieces, the combined workpieces being considered as being composed of a plurality of initial workpieces, the combination condition of the combined workpieces being obtained, and the structure data of the initial workpieces being obtained;

[0045] Initial workpiece virtual model: a virtual model of the initial workpieces is established according to the real-time data of the initial workpieces;

[0046] Preliminary fault diagnosis: fault diagnosis is performed on the real-time data of the initial workpieces to obtain diagnosis data of the initial workpieces, and the diagnosis data of the initial workpieces is mapped onto the virtual model of the initial workpieces;

[0047] Establishment and application of workpiece correlation model: a workpiece correlation model is established, the correlation degree between each initial workpiece in the combined workpieces is obtained according to the workpiece correlation model, and structure change data is obtained based on the correlation degree and the combination condition of the combined workpieces;

[0048] Combined workpiece virtual model: a virtual model of the combined workpieces is simulated and obtained according to the virtual model of the initial workpieces and the structure change data;

[0049] Virtual model analysis and diagnosis: diagnosis data of the combined workpieces is obtained based on the virtual model of the combined workpieces, the diagnosis data of the initial workpieces and the correlation degree, the diagnosis data of the initial workpieces is reflected on the virtual model of the combined workpieces, global simulation analysis is performed on the virtual model of the combined workpieces, influence data of the diagnosis data on the virtual model of the combined workpieces is simulated, and simulation total diagnosis data is obtained;

[0050] Early warning: simulation fault data is obtained by analyzing the simulation total diagnosis data, and early warning is performed according to the simulation fault data;

[0051] Calibration: fault diagnosis is performed on the real-time data of the combined workpiece, the real-time data of the combined workpiece, the fault diagnosis structure of the combined workpiece and the simulation total diagnosis data of the corresponding combined workpiece virtual model are compared, and comparison data for modifying parameters in virtual model analysis diagnosis are obtained.

[0052] The production line diagnosis method comprises:

[0053] Data acquisition: obtaining historical production data of the equipment in the trailer production line;

[0054] Real-time monitoring: real-time monitoring of the equipment in the trailer production line and real-time monitoring of the trailer structure are performed to obtain equipment real-time data and structure real-time data;

[0055] Equipment division in the production line: the equipment in the production line is divided into sub-real-time data and total real-time data according to the total attribute of the equipment in the production line;

[0056] Establishing and applying a production line correlation model: a production line correlation model is established, and the production line correlation model establishes the correlation degree between the equipment in the production line according to the historical production data;

[0057] Establishing and applying a prediction model: a prediction model is established, and the equipment real-time data is predicted through the prediction model to obtain equipment prediction data;

[0058] Application of continuous data: using time series analysis technology and combining the correlation degree, the continuity of the sub-real-time data and the total real-time data is deeply mined to identify the trend in the production process and the continuity between the production lines, and the continuous data between the sub-real-time data and the total real-time data is obtained;

[0059] Establishing and applying a threshold value: based on the historical production data, the corresponding equipment threshold value is established, and it is judged whether the continuous data and the equipment prediction data will exceed the corresponding threshold value, and if the threshold value is exceeded, a warning will be given.

[0060] The structure change data includes physical connection, mechanical property and vibration connection, the physical connection includes the connection mode between the initial workpieces in the combined workpiece, the mechanical property includes stress distribution and stiffness change, and the vibration connection includes resonance frequency and amplitude change.

[0061] The method for obtaining the correlation degree between the equipment in the production line comprises: analyzing the historical production data of the equipment in the trailer production line, judging whether there is a correlation between the equipment, establishing the correlation between the equipment, establishing a production line correlation model for reflecting the interaction and dependency relationship between the equipment, and analyzing the mutual influence degree between the equipment through the production line correlation model, and the mutual influence degree is represented by a quantitative index.

[0062] The method for obtaining the correlation degree between each workpiece in the trailer structure comprises: analyzing initial workpieces in a combined structure, obtaining the connection mode between each initial workpiece in the combined workpiece, using the structure data and real-time running data of the workpiece to establish a workpiece correlation model capable of reflecting the physical connection, functional dependence and interaction between the workpieces, analyzing the mutual influence and correlation degree between the workpieces through the workpiece correlation model, and representing the mutual influence and correlation degree between the workpieces by quantitative indexes.

[0063] According to the initial workpieces in the workpiece set, the initial workpieces are divided into long-span workpieces and non-long-span workpieces according to the length data in the structure data, and the fault diagnosis method for the workpiece data of the long-span workpiece comprises segmented monitoring, which comprises dividing the long-span workpiece into a plurality of shorter paragraphs, and respectively monitoring and collecting data of each paragraph to obtain segmented data of each paragraph and position data of the paragraph.

[0064] The fault diagnosis method for the long-span workpiece comprises:

[0065] The segmented data is corresponded to the position on the long-span workpiece according to the position data, the position is calibrated after the corresponding, the segmented data is time-aligned to obtain overall data;

[0066] The overall data is mapped into a virtual model, and personnel perform auxiliary analysis;

[0067] The overall data is processed, and the processed overall data is processed to obtain characteristic information of the fault;

[0068] The characteristic information of the fault is subjected to pattern recognition to determine the fault type, and the specific position of the corresponding fault is determined by combining the position data and the fault characteristics and using a positioning algorithm.

[0069] A real-time monitoring and fault diagnosis system for a trailer production line adopts the real-time monitoring and fault diagnosis method for the trailer production line.

[0070] The present application performs real-time detection on the equipment in the production line, performs real-time detection on the trailer structure on the production line, predicts the running state of the equipment in the production line, and then detects the equipment and predicts and diagnoses the fault, diagnoses the fault of the trailer structure on the production line, diagnoses and analyzes the workpiece to obtain the fault condition of the workpiece, and the real-time data of the initial workpiece comprises real-time size detection of the initial workpiece and detection of the initial workpiece by a detection device for subsequent fault diagnosis.

[0071] At the same time, real-time data of the combined workpiece is obtained, and since the combined workpiece has been actually manufactured, data difference between the combined workpiece and the virtual model of the combined workpiece is analyzed, the data difference including size difference and difference between actual diagnostic data and simulation total diagnostic data, etc., which can optimize each parameter in the virtual model analysis and diagnosis.

[0072] Wherein, the continuity refers to whether the production is continuous, the threshold of the continuous data refers to whether the data of the production line can be continuously produced according to the continuous data, the combination condition in the combined workpiece is known data, the combination condition refers to the mutual connection of each initial structure, etc., the combination condition of the combined workpiece includes obtaining the initial workpiece data of each initial workpiece in the combined workpiece, obtaining the mutual connection mode of each initial workpiece, the trailer total structure is regarded as a plurality of combined workpieces which are combined with each other, in actual application, the mutual connection requirement in the trailer total structure is not high, and the allowable error range is large, therefore, it is not described in the application, the sub-total attribute of the equipment in the production line refers to whether the equipment is installed on the sub-production line or the total production line, the running state and the value of the equipment in the production line are detected, whether the running state of the equipment in the production line is normal is judged to ensure that the equipment in the production line can normally run and judge, the continuous data of the sub-production line and the total production line in production, wherein, the continuous data includes the workpiece feeding speed on the sub-production line, the combination time of the workpiece on the sub-production line, the moving speed of the workpiece on the total production line, the combination time of the workpiece on the total production line, and the time cooperation on the sub-production line and the total production line, the time cooperation includes the production continuity between the sub-production line and the total production line.

[0073] On the trailer production line, not only the device for detecting the equipment in the production line exists, but also a plurality of detection devices for detecting the workpiece exist, the workpiece refers to each structural part of the trailer, which is uniformly represented by the workpiece, the initial workpiece is not necessarily only a single structure inside, but refers to the workpiece when it is not installed with other workpieces on the production line, on the trailer production line, each initial workpiece is combined with each other to obtain a combined workpiece, and the combined workpieces are connected with each other to obtain a trailer, and the fault diagnosis and analysis of the initial workpiece, since the initial workpiece is not associated with other initial workpieces at this time, the workpiece can be directly detected and fault diagnosed and analyzed, and the analysis result of the initial workpiece is relatively reliable at this time.

[0074] The structural data of the workpiece has been recorded in the database, a virtual workpiece is simulated through the structural data of the workpiece, real-time data about the long-span workpiece and the combined workpiece of long span is input into the corresponding virtual workpiece, the reaction of the real-time data in the virtual workpiece in the computer is simulated through the real-time data, virtual simulation data is obtained, personnel can watch through the display screen, and the fault diagnosis method is assisted by personnel. When the long-span initial workpiece is detected, it is inconvenient to directly diagnose and analyze through the real-time data because of the long span, for example, the long-span bearing and the suspension system may be more susceptible to stress concentration and fatigue damage, and the long-span workpiece is inconvenient to detect due to the long span during real-time detection, and the analysis of the real-time data is more inconvenient due to the span problem. There may be a complex relationship between different parts of the long-span workpiece, so the fault diagnosis method of the long-span workpiece is adopted.

[0075] The long-span workpiece is divided into several shorter paragraphs, the length of each paragraph should be determined according to the specific working condition and monitoring requirement, vibration monitoring is performed on each paragraph, vibration signals are collected in real time through sensors and other equipment, data collection should ensure comprehensiveness and accuracy to reflect the true state of the workpiece, segmented data of each paragraph is obtained through monitoring equipment, including vibration frequency, amplitude and other key parameters, and position data of each paragraph is recorded for subsequent data processing and fault positioning. The signal processing includes filtering, denoising and smoothing.

[0076] The segmented data is established according to the position data to correspond to the position on the long-span workpiece, the monitoring point position is calibrated to ensure the accuracy and consistency of the data, time alignment processing is performed to eliminate errors caused by different data collection times, thereby obtaining overall data, and the overall data is mapped to the virtual model, which helps personnel to more intuitively understand the state of the workpiece and possible problems. Personnel combine virtual model and actual data to assist in analysis, preliminarily judge whether the workpiece has a fault and the possible position of the fault, and perform signal processing on the overall data, such as filtering and denoising, to improve the quality of the data. Feature information of the fault is extracted through the signal processing method, such as frequency change and amplitude anomaly, the extracted feature information is classified and recognized by using pattern recognition technology, the fault type is judged according to the recognition result, such as crack, wear and looseness, the specific position of the fault is determined by combining the position data and the fault feature information through positioning algorithm. The long-span workpiece is monitored by segmentation, and the segmented data is time-aligned, which can more accurately obtain the vibration data and running state of the workpiece, improve the accuracy and efficiency of fault diagnosis, and quickly locate the fault point by combining the virtual model of the long-span workpiece and personnel assisted analysis.

[0077] The positioning algorithm is usually based on the principle of pattern recognition or template matching, which compares and analyzes the known fault types and feature information with the real-time monitored data to determine the specific location of the fault. Combined with the position data and fault characteristics, it can accurately point out the fault point, providing important basis for subsequent maintenance and fault handling. Common positioning algorithms include determining the fault location by measuring signal strength and comparing the signal strength differences at different positions, calculating the fault location using the time difference of signal propagation at different positions, and identifying and predicting the fault location by training machine learning models.

[0078] Through virtual simulation data for fault diagnosis analysis, personnel can assist in analysis, through virtual workpieces in the computer simulation real-time data, since the simulation process can be displayed on the display screen, so that personnel can make auxiliary diagnosis analysis through the simulation picture in the display screen, through the vibration situation of the simulation picture, can directly understand the running state of the equipment, analyze the frequency, amplitude, phase and other characteristics of the vibration, these characteristics are often associated with a particular fault type, through the establishment of virtual model, preliminary fault diagnosis and correlation model, the initial workpiece is analyzed before the combination of workpieces, then the virtual model of the initial workpiece is established, the virtual model of the combined workpiece is obtained according to the virtual model and structure change data of the initial workpiece, the diagnosis data of the initial workpiece is reflected on the virtual model of the combined workpiece and analyzed through global simulation, realizing the simulation fault diagnosis of the combined workpiece, that is, realizing the early warning of the fault before the fault occurs, reducing the situation that the fault after combination will affect the production efficiency of the trailer production line.

[0079] The overall size of the combined workpiece may change when different initial workpieces are combined together due to the cooperation, connection or stacking between the workpieces. The mass distribution of the initial workpieces may change during the combination process, which in turn affects the overall mass distribution of the combined workpieces. After the workpieces are combined, the stiffness and strength of the combined workpieces may change due to mutual support, constraint or reinforcement. This change may make the combined workpieces exhibit different deformation resistance and load capacity when subjected to force. For combined workpieces that need to move, such as transmission components or motion mechanisms in mechanical systems, the combination method between workpieces will affect their motion characteristics. In some cases, the combination between workpieces will also affect the thermal characteristics of the combined workpieces, such as thermal conductivity, heat capacity or thermal expansion coefficient, etc.

[0080] Professional detection equipment (such as crack detector, stress analyzer, ultrasonic flaw detector) carries out in-depth detection and analysis on the preliminary structure and combined structure in the trailer, and the detection and analysis results of the preliminary structure are used to exclude some defective products, and the remaining qualified products will be put into the production line for production, but the initial workpieces put into the production line are only qualified in quality, which may not be found or considered as qualified in separate detection, but when these initial workpieces are combined into combined workpieces, these defects may interact with each other, resulting in a decline in overall performance or safety hazards. At the same time, when the initial workpieces are connected to each other, especially during the connection process such as welding, internal stress concentration may occur due to material deformation and uneven cooling, which may cause problems in the combined workpieces. Real-time monitoring and data acquisition can comprehensively and real-time monitor the equipment and workpieces on the trailer production line, ensure the stability and reliability of the production process, use real-time data of the initial workpieces to simulate the subsequent combined workpieces, simulate the influence of the diagnostic data on the virtual model of the combined workpieces, and find potential fault points in advance. The combination of production line diagnosis method and workpiece diagnosis method makes fault diagnosis more comprehensive and accurate, and can timely find faults in production.

[0081] The prediction model for predicting numerical trends already exists in the prior art, and the specific principle of the prediction model is not described here. Different specific devices, therefore, the prediction model adopted by the corresponding device will be different. The prediction model realizes the prediction of the numerical value, and the feature information extraction is one of the key steps of fault diagnosis. By extracting fault feature information from the processed overall data, strong support can be provided for subsequent fault recognition and classification. Feature information extraction is usually based on the time domain, frequency domain or time-frequency domain characteristics of the signal. By calculating and analyzing these characteristics, fault-related feature information such as amplitude, frequency, phase, energy, etc. can be extracted. Common feature extraction methods include: time domain feature extraction: extracting statistical indicators such as average value, variance, maximum value, minimum value, etc. of the signal, reflecting the overall waveform characteristics and amplitude change of the signal; frequency domain feature extraction: converting the signal from the time domain to the frequency domain through Fourier transform or wavelet transform, etc. to extract frequency spectrum characteristics such as frequency spectrum energy, spectral peak frequency, etc.; time-frequency domain feature extraction: combining the characteristics of time domain and frequency domain to extract time-frequency characteristics of the signal such as wavelet packet decomposition coefficient, IMF component of empirical mode decomposition, etc.

[0082] Time series analysis is a statistical method specially used for processing data sequences that change over time, aiming to find the inherent patterns, trends, periodicities and other characteristics in the data, and make future predictions and decisions based on these characteristics, time series analysis arranges data in chronological order, analyzes the correlation and change law between data by using statistical methods, it relies on the continuity and correlation of data in time to analyze data, at the same time, through time series analysis, the change trend and periodic fluctuation of the workpiece transportation speed on each flow line are monitored, the consistency and synchronization of the speed are evaluated, so as to ensure the synchronization of workpiece production on the whole flow line.

[0083] Unbalance judgment, by analyzing the vibration data of the equipment, especially the high frequency components in the vibration spectrum, it can be judged whether the equipment exists unbalance problem, unbalance usually leads to periodic vibration of the equipment during operation, and the vibration amplitude will increase with the increase of speed, misalignment includes shaft misalignment and angle misalignment, by analyzing the deviation of the equipment's center line, vibration data and temperature data, etc., it can be judged whether the equipment exists asymmetry problem, equipment looseness usually leads to increase of vibration amplitude, broadening of vibration spectrum and appearance of non-periodic vibration components, by analyzing the vibration data and spectrum characteristics of the equipment, it can be judged whether the equipment exists looseness problem, wear is a common failure form during equipment operation, by analyzing the vibration data, oil analysis data and running time of the equipment, etc., it can be judged whether the equipment exists wear problem, the above not only can realize the judgment of the equipment, but also can realize the judgment of the initial workpiece in the trailer.

[0084] After issuing a warning, personnel can carry out maintenance or adjust equipment parameters, which is not described here, the types of equipment in the production line are different, so the threshold setting of the corresponding equipment will be different, the degree of mutual influence is represented by quantitative indicators, that is, the influence degree between equipment is represented by numerical value, the virtual model of the combined workpiece is obtained by simulating the virtual model of the initial workpiece and the structure change data, since the length of the combined workpiece is affected by the initial workpiece and different connection methods will directly affect the size data of the combined workpiece, therefore, the virtual model of the combined workpiece needs to combine the virtual model of the initial workpiece and the structure change data.

[0085] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, replacements and variations can be made to these embodiments without departing from the principles and spirits of the present application, the scope of the present application is defined by the appended embodiments and their equivalents.

Claims

1. A method for real-time monitoring and fault diagnosis of a trailer production line, characterized in that, The real-time monitoring and fault diagnosis method comprises a production line diagnosis method and a workpiece diagnosis method; The workpiece diagnosis method comprises: Real-time monitoring and data acquisition: real-time detection of workpieces in a workpiece set by a detection device to obtain real-time data of the workpieces, the workpiece set comprising initial workpieces and combined workpieces, the combined workpieces being considered as being formed by mutual combination of the initial workpieces, the combination condition of the combined workpieces being obtained, and the structure data of the initial workpieces being obtained; Initial workpiece virtual model: a virtual model of the initial workpieces is established according to the real-time data of the initial workpieces; Preliminary fault diagnosis: fault diagnosis is performed on the real-time data of the initial workpieces to obtain diagnosis data of the initial workpieces, and the diagnosis data of the initial workpieces is mapped onto the virtual model of the initial workpieces; Establishment and application of workpiece correlation model: a workpiece correlation model is established, the correlation degree between each initial workpiece in the combined workpiece is obtained according to the workpiece correlation model, and structure change data is obtained based on the correlation degree and the combination condition of the combined workpiece; Combined workpiece virtual model: a virtual model of the combined workpiece is simulated according to the virtual model of the initial workpieces and the structure change data; Virtual model analysis and diagnosis: diagnosis data of the combined workpiece is obtained based on the virtual model of the combined workpiece, the diagnosis data of the initial workpieces and the correlation degree, the diagnosis data of the initial workpieces is reflected on the virtual model of the combined workpiece, global simulation analysis is performed on the virtual model of the combined workpiece, influence data of the diagnosis data on the virtual model of the combined workpiece is simulated, and simulation total diagnosis data is obtained; Early warning: simulation fault data is obtained by analyzing the simulation total diagnosis data, and early warning is performed according to the simulation fault data; Calibration: fault diagnosis is performed on the real-time data of the combined workpiece to obtain fault diagnosis results of the combined workpiece, the fault diagnosis results of the combined workpiece are compared with the simulation total diagnosis data of the corresponding combined workpiece virtual model, and comparison data used for modifying parameters in the virtual model analysis and diagnosis are obtained.

2. The real-time monitoring and fault diagnosis method of a trailer production line according to claim 1, characterized in that: The production line diagnosis method comprises: Data acquisition: historical production data of devices in the trailer production line is obtained; Real-time monitoring: real-time monitoring is performed on the devices in the trailer production line and real-time monitoring is performed on the trailer structure to obtain device real-time data and structure real-time data; Device division in the production line: the devices in the production line are divided into sub-real-time data and total real-time data according to the sub-total attribute of the devices in the production line; Establishment and application of production line correlation model: a production line correlation model is established, and the production line correlation model establishes the correlation degree between the devices in the production line according to the historical production data; Establishment and application of prediction model: a prediction model is established, and the device real-time data is predicted by the prediction model to obtain device prediction data; Application of continuous data: time series analysis technology is used in combination with the correlation degree to deeply mine the continuity of the sub-real-time data and the total real-time data, identify the trend in the production process and the continuity between the production lines, and obtain continuous data between the sub-real-time data and the total real-time data; Establishment and application of threshold value: corresponding device threshold values are established based on the historical production data, and it is judged whether the continuous data and the device prediction data will exceed the corresponding threshold values, and if the threshold values are exceeded, early warning will be performed.

3. The real-time monitoring and fault diagnosis method of a trailer production line according to claim 1, characterized in that: The structure change data includes physical connection, mechanical property and vibration connection, the physical connection includes the connection mode between the initial workpieces in the combined workpiece, the mechanical property includes stress distribution and stiffness change, and the vibration connection includes resonance frequency and amplitude change.

4. The real-time monitoring and fault diagnosis method of a trailer production line according to claim 1, characterized in that: The method for obtaining the correlation degree between devices on a production line includes: analyzing historical production data of devices in a trailer production line, judging whether there is a correlation between devices, establishing the correlation between devices, establishing a production line correlation model for reflecting the interaction and dependency relationship between devices, analyzing the mutual influence degree between devices through the production line correlation model, and representing the mutual influence degree by a quantitative index.

5. The method for real-time monitoring and fault diagnosis of a trailer production line according to claim 1, characterized in that: The method for obtaining the correlation degree between workpieces in a trailer structure includes: analyzing initial workpieces in a combined structure, obtaining the connection mode between each initial workpiece in a combined workpiece, establishing a workpiece correlation model capable of reflecting the physical connection, functional dependency and interaction between workpieces by using structure data and real-time running data of the workpieces, analyzing the mutual influence and correlation degree between workpieces through the workpiece correlation model, and representing the mutual influence and correlation degree between workpieces by a quantitative index.

6. The method for real-time monitoring and fault diagnosis of a trailer production line according to claim 1, characterized in that: According to the initial workpieces in the workpiece set, the initial workpieces are divided into long-span workpieces and non-long-span workpieces according to length data in the structure data, and the fault diagnosis method for the workpiece data of the long-span workpieces includes segmented monitoring, which includes dividing the long-span workpieces into a plurality of shorter paragraphs, respectively performing vibration detection and data acquisition on each paragraph, and obtaining segmented data and position data of each paragraph.

7. The method for real-time monitoring and fault diagnosis of a trailer production line according to claim 6, characterized in that: The fault diagnosis method for the long-span workpieces includes: The segmented data is corresponded to the position on the long-span workpiece according to the position data, the detection point position of the long-span workpiece is calibrated after the position is corresponded, the segmented data is time-aligned to obtain overall data; The overall data is mapped into a virtual model for personnel to assist in analysis; The overall data is processed, and the processed overall data is processed to obtain characteristic information of the fault; The characteristic information of the fault is subjected to pattern recognition to determine the fault type, and the specific position of the corresponding fault is determined by using a positioning algorithm in combination with the position data and the fault characteristics.

8. A real-time monitoring and fault diagnosis system for a trailer production line, characterized by: A trailer production line real-time monitoring and fault diagnosis method is adopted.

Images