CN117150274B - Quality detection method for press fitting of plug - Google Patents
Quality detection method for press fitting of plug Download PDFInfo
- Publication number
- CN117150274B CN117150274B CN202311421194.8A CN202311421194A CN117150274B CN 117150274 B CN117150274 B CN 117150274B CN 202311421194 A CN202311421194 A CN 202311421194A CN 117150274 B CN117150274 B CN 117150274B
- Authority
- CN
- China
- Prior art keywords
- press
- plug
- vibration
- pressure
- fitting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 23
- 238000000034 method Methods 0.000 claims abstract description 84
- 230000008569 process Effects 0.000 claims abstract description 74
- 238000012544 monitoring process Methods 0.000 claims abstract description 24
- 238000004458 analytical method Methods 0.000 claims abstract description 21
- 238000007637 random forest analysis Methods 0.000 claims abstract description 13
- 230000000087 stabilizing effect Effects 0.000 claims description 18
- 230000008859 change Effects 0.000 claims description 17
- 238000011156 evaluation Methods 0.000 claims description 14
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000007689 inspection Methods 0.000 claims description 10
- 230000004927 fusion Effects 0.000 claims description 9
- 238000004364 calculation method Methods 0.000 claims description 5
- 239000003086 colorant Substances 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000012935 Averaging Methods 0.000 claims description 2
- 238000012858 packaging process Methods 0.000 claims description 2
- 230000000875 corresponding effect Effects 0.000 description 10
- 238000003908 quality control method Methods 0.000 description 7
- 238000012549 training Methods 0.000 description 6
- 238000004088 simulation Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000001303 quality assessment method Methods 0.000 description 3
- 238000007789 sealing Methods 0.000 description 3
- 230000006641 stabilisation Effects 0.000 description 3
- 238000011105 stabilization Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/253—Fusion techniques of extracted features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- Medical Informatics (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Automatic Assembly (AREA)
Abstract
The invention relates to the technical field of press-fit quality detection, in particular to a quality detection method for press-fit of a plug. And extracting the plug pressure point characteristic parameters and the plug vibration characteristic parameters. And (5) utilizing a random forest model to fuse characteristic parameters, and evaluating the press-fitting quality of the plug in real time. Calculating a press-fit stability index through analysis of the pressure stability interval and the plug amplitude; calculating the vibration index of the press-fitting process through analysis of the pressure fluctuation range and the plug vibration frequency; and the press-fitting efficiency index is calculated by analyzing the maximum pressure point position and the vibration duration. And (3) according to a preset weight value, carrying out weighted average on the three indexes to obtain a plug press-fitting quality index Q so as to evaluate the plug press-fitting quality in real time. The high efficiency and the accuracy of end cap pressure equipment quality monitoring have been realized.
Description
Technical Field
The invention relates to the technical field of press-fit quality detection, in particular to a quality detection method for plug press-fit.
Background
A plug is a common mechanical component that is commonly used to close an opening of a pipe, container, or other device to prevent leakage of liquid, gas, or other substances. The press-fitting quality of the plug is an important factor for determining the performance and safety of the plug. If the press fit is not in place or under pressure, leakage, pressure loss or other problems may result, thereby affecting the proper operation and safety of the system. Conversely, if the pressure is too high, damage to the plug or the pipe may result, and even serious safety accidents may result.
Existing plug presses are typically implemented by mechanical press-fitting equipment, which is typically operated by an operator, or by a simple automated control system to control the press-fitting process. In the press-fitting process, conventional quality control methods rely mainly on manual inspection or simple sensor monitoring. If the operator would determine if the plug is press-fit in place by inspection or manual inspection, or the press-fit process would be monitored by simple pressure and position sensors. However, the conventional quality control method lacks comprehensive analysis capability on press-fitting process data, is difficult to meet the requirement of high-efficiency production, and is difficult to accurately evaluate the press-fitting quality of the plug, so that quality problems may not be found and treated in time.
Disclosure of Invention
In order to solve the problems, the invention provides a quality detection method for press fitting of plugs.
In order to achieve the above purpose, the invention adopts the following technical scheme:
a quality detection method for press fitting of plugs comprises the following steps:
collecting position data, pressure data and vibration data in the plug press-fitting process in real time, and drawing a pressure position curve and a vibration curve in the press-fitting time period;
extracting and analyzing the pressure position curve to obtain plug pressure point characteristic parameters, wherein the pressure point characteristic parameters comprise a pressure stabilizing interval, a pressure fluctuation range and a maximum pressure point position;
extracting and analyzing the vibration curve to obtain plug vibration characteristic parameters, wherein the vibration characteristic parameters comprise plug vibration frequency, plug amplitude and vibration duration;
carrying out fusion analysis on the pressure point characteristic parameters and the vibration characteristic parameters based on a random forest model, evaluating the press-fitting quality of the plug in real time and sending out warning prompt;
the fusion analysis specifically comprises the following steps:
taking the pressure stabilizing interval and plug amplitude as input characteristics, and calculating to obtain a press-fit stability index S;
calculating a vibration index V in the press-fitting process by taking a pressure fluctuation range and a plug vibration frequency as input characteristics;
calculating a press-fitting efficiency index E by taking the position of the maximum pressure point and the vibration duration as input characteristics;
and carrying out weighted average on the press-fit stability index S, the vibration influence index V and the press-fit efficiency index E according to preset weight values, and calculating to obtain a plug press-fit quality index Q.
Further, the real-time collecting of the position data, the time data, the pressure data and the vibration data in the plug press-fitting process specifically comprises the following steps:
the position change of the plug is monitored in real time through a multi-dimensional camera, and position data are obtained;
monitoring pressure change in the pressure packaging process in real time through a pressure sensor to obtain pressure data;
vibration information in the press fitting process is monitored and recorded in real time through a vibration sensor, and vibration data are obtained.
Further, the drawing of the pressure position curve and the vibration curve in the press-fitting time period specifically includes the following steps:
the acquired position data are sorted according to the time sequence, and time sequence position data are generated;
the acquired pressure data are sorted according to the time sequence, and time sequence pressure data are generated;
aligning the time axis of the time-series position data with the time-series pressure data, and drawing a pressure position curve based on the time sequence;
and arranging the acquired vibration data according to a time sequence, and drawing a vibration curve of the time sequence.
Further, the extracting and analyzing the pressure position curve specifically comprises the following steps:
determining key points in the pressure position curve through an extreme point detection algorithm, wherein the key points comprise a pressure starting point, a pressure stabilizing point, a maximum pressure point and corresponding position information;
calculating a pressure stabilizing section in the plug press-fitting process according to the determined key points;
calculating a pressure fluctuation range, wherein the fluctuation range is determined by the difference between the maximum value and the minimum value of the pressure;
and determining the position of the maximum pressure point according to the position information of the key point.
Further, the extracting and analyzing the vibration curve specifically includes the following steps:
performing frequency domain analysis on the vibration data through Fourier transformation to determine plug vibration frequency in the plug press-fitting process;
a peak detection algorithm is applied in a time domain to identify the peak value and the valley value of the vibration curve, and plug amplitude is further calculated;
the duration of the vibration is determined by comparing the mean value of the vibration amplitude within the window with a preset vibration amplitude threshold.
Further, the step of calculating the plug press-fit quality index Q by weighted averaging the press-fit stability index S, the vibration influence index V and the press-fit efficiency index E according to preset weight values specifically includes the following steps:
setting weight values of all indexes according to actual requirements and experience data in the press-fitting process;
the obtained press-fit stability index S, vibration influence index V and press-fit efficiency index E are weighted according to corresponding weight values, and a specific calculation formula of the plug press-fit quality index Q is as follows:
，
wherein,for press-fit stability index weight, +.>For vibration influencing index weight, +.>Is the index weight of the press-fitting efficiency.
Further, the real-time evaluation of the press-fitting quality of the plug and the sending of the warning prompt comprise the following steps:
comparing the calculated quality index Q of the press fitting of the plug with a preset quality threshold; if Q is higher than a preset threshold, judging the press-fit quality to be qualified; if the Q is lower than a preset threshold, judging that the press-fit quality is unqualified.
Further, the real-time evaluation of the press-fitting quality of the plug and the sending of the warning prompt further comprise the following steps:
and continuously monitoring the change of the press-fit quality index Q, and if the Q has a descending trend and is close to a preset threshold value, sending out a warning prompt.
Further, the warning prompt includes:
depending on the implementation evaluation, the operator is alerted by an audible warning that prompt intervention or examination is required.
Further, the warning prompt further includes:
according to the real-time evaluation, various index anomalies are displayed in bright colors and icon changes on a monitoring interface and the cause of the problem is provided in a text form.
The invention has the beneficial effects that: the invention provides a comprehensive data base for evaluating and monitoring the press-fit quality by collecting the position data, the pressure data and the vibration data in the press-fit process of the plug in real time, and compared with the traditional quality control method, the vibration parameter plays a key role. In the plug press-fitting process, the vibration condition can directly reflect the stability and quality of the press-fitting process, including the contact condition between the plug and the pipeline, the stability of the plug and the operation accuracy of the press-fitting machine. By monitoring and analyzing the vibration data, possible quality problems including misplacement of the plug, press-fitting failure or excessive press-fitting can be effectively identified and prevented. According to the invention, the pressure position curve and the vibration curve in the press-fitting time period are drawn, the pressure position curve and the vibration curve are extracted and analyzed to obtain factors which possibly influence the press-fitting quality, such as plug pressure points and vibration characteristic parameters, and then the random forest model is combined to perform fusion analysis on the pressure point characteristic parameters and the vibration characteristic parameters, and weighted average is performed according to a preset weight value, so that the plug press-fitting quality index is calculated. Accurate and efficient plug press-fitting quality assessment is achieved through comprehensive assessment indexes and press-fitting quality assessment automation.
Drawings
Fig. 1 is a flow chart of the steps of the quality detection method for press fitting of plugs in the present invention.
FIG. 2 is a flow chart of steps for extraction analysis of a pressure location curve.
Fig. 3 is a flowchart of the steps for extraction analysis of vibration curves.
Detailed Description
Referring to fig. 1-3, the present invention relates to a quality detection method for press-fitting of a plug.
Specifically, the invention provides a quality detection method for press fitting of a plug, which comprises the following steps:
s1, collecting position data, pressure data and vibration data in the plug press-fitting process in real time, and drawing a pressure position curve and a vibration curve in a press-fitting time period; step S1 comprises the steps of:
s11, monitoring the position change of the plug in real time through a multi-dimensional camera to obtain position data;
specifically, the multidimensional camera should be installed at the key position of the plug press-fitting area, including the upper part, the side surface and other possible angles, so as to ensure that the image information of the plug and the press-fitting equipment can be comprehensively captured. When the press-fitting process starts, the multidimensional camera can capture image information of the plug and the press-fitting equipment in real time. The camera should be able to run at a high frame rate to ensure that every critical moment in the press-fit process can be captured. And extracting the position data of the plug from the captured image information through an image processing and analyzing technology. This may include the co-ordinate position of the plug, the angle of rotation, and the relative position with the pipe or other equipment, etc. And recording and transmitting the acquired position data to a system in real time. And the real-time performance and the accuracy of the position data are ensured.
S12, monitoring pressure change in the pressure assembly process in real time through a pressure sensor to obtain pressure data;
specifically, the pressure sensor is mounted on a press-fitting head or a press-fitting cylinder that is in contact with the plug, so as to ensure that pressure changes in the press-fitting process can be accurately captured. The sensor operates at a high sampling rate to ensure that pressure data at each critical moment in the press-fit process can be captured. Pressure data, including pressure values, time stamps, etc., are obtained in real time from the pressure sensor. And recording and transmitting the acquired pressure data to a system in real time, so that the real-time performance and accuracy of the position data are ensured.
And S13, monitoring and recording vibration information in the press fitting process in real time through a vibration sensor, and obtaining vibration data.
Specifically, vibration sensors are installed in the press ram to ensure accurate capture of vibration information during press-fitting. Vibration data including vibration amplitude, frequency, time stamp, etc. are obtained in real time from the vibration sensor. And recording and transmitting the acquired vibration data to a system in real time, so that the real-time performance and accuracy of the position data are ensured.
S14, sorting the acquired position data according to a time sequence to generate time sequence position data;
specifically, the collected location data is sorted in time order according to the time stamps. The ordering should ensure continuity and consistency of the time series so that the subsequent analysis is more accurate. The ordered location data is consolidated into a time series format, with each data point including location information and a corresponding timestamp. And creating a simulation environment by using the collected position data, and generating position simulation images of the plugs at different time points in the simulation environment according to the time-series position data. Each simulation image shows the position state of the plug at a specific time point, and can clearly show the movement and the position change of the plug in the press-fitting process.
S15, sorting the acquired pressure data according to a time sequence to generate time sequence pressure data;
specifically, the collected pressure data is sorted in the order of the time stamps, generating a time-ordered sequence of pressure data. Then, it is checked whether the sorted data are correctly arranged in time order and whether there is a data missing or abnormal value. The chronological pressure data is saved to the system for later analysis and use.
S16, aligning time axes of the time series position data and the time series pressure data, and drawing a pressure position curve based on the time series;
specifically, the validation position data and the pressure data are first verified as being chronologically organized, and each data point has a corresponding time stamp. By comparing the time stamps of the position data and the pressure data, a perfect alignment of the two on the time axis is ensured. If the time stamps do not match exactly, the time axes of the two are aligned by interpolation or data synchronization methods. The time-axis aligned position data and pressure data are combined to form a new data set, wherein each data point contains a time stamp, a position value and a pressure value. Next, in the same graph, the position curve and the pressure curve are plotted with time as the horizontal axis and position and pressure as two independent vertical axes. In this way, the position and pressure values at each time point can be clearly shown on the graph. And finally, the drawn pressure position curve graph is saved as an image file and a data file, so that convenience is brought to subsequent analysis and evaluation.
S17, the acquired vibration data are sorted according to the time sequence, and a vibration curve of the time sequence is drawn.
Specifically, firstly, vibration data collected in real time through a vibration sensor is ordered according to time stamps, so that the time sequence of the data is ensured to be correct. The time stamps are then correlated with the corresponding vibration data values to construct a time series of vibration data sets. And drawing a vibration curve of the time sequence by taking the time stamp as a horizontal axis and the vibration data value as a vertical axis. The curve can clearly reflect the variation trend and characteristics of vibration in the plug press-fitting process, and provides an intuitive and accurate data basis for subsequent vibration characteristic extraction and analysis.
S2, extracting and analyzing the pressure position curve to obtain plug pressure point characteristic parameters;
the pressure point characteristic parameters comprise a pressure stabilizing section, a pressure fluctuation range and a maximum pressure point position;
step S2 comprises the steps of:
s21, determining key points in a pressure position curve through an extreme point detection algorithm, wherein the key points comprise a pressure starting point, a pressure stabilizing point, a maximum pressure point and corresponding position information;
specifically, the extreme point detection algorithm finds these key points mainly by identifying local maxima and local minima on the curve. Specifically, the algorithm first smoothes the pressure location curve to reduce the effects of noise, and then calculates the first and second derivatives of the curve. By analyzing the zero and sign changes of the first derivative and the second derivative, the extreme points on the curve are accurately located. After finding these extreme points, the pressure start point, the pressure stabilization point, and the maximum pressure point can be further determined. The pressure start point is usually the position where the pressure starts to change significantly, the pressure stabilization point is the position where the pressure change tends to be smooth, and the maximum pressure point is the position where the pressure reaches the maximum value.
S22, calculating a pressure stabilizing section in the plug press-fitting process according to the determined key points;
specifically, in a specific calculation process, the moment when the pressure starts to stabilize is initially determined according to the pressure change condition between the pressure starting point and the pressure stabilizing point. Then, by analyzing the pressure change trend after the maximum pressure point, the time at which the pressure ends to stabilize is determined, thereby determining the end point of the pressure stabilization section. By these two moments, the range of the pressure stabilizing section is obtained.
S23, calculating a pressure fluctuation range, wherein the fluctuation range is determined by the difference value between the maximum value and the minimum value of the pressure;
specifically, the maximum pressure value and the minimum pressure value are first identified from the pressure data in the stable region, and then the difference between the maximum pressure value and the minimum pressure value is calculated, so that the obtained result is the pressure fluctuation range.
S24, determining the position of the maximum pressure point according to the position information of the key point;
specifically, by the application of the extreme point detection algorithm in the previous step S21, the key points in the pressure position curve have been identified. Each key point has corresponding position information, and the position information is based on position data in the plug press-fitting process collected in real time. The maximum pressure point is the point on the pressure position curve, where the pressure value reaches the maximum, and the position information directly reflects the position of the plug when the pressure reaches the maximum in the press mounting process.
S3, extracting and analyzing the vibration curve to obtain a plug vibration characteristic parameter;
the vibration characteristic parameters comprise plug vibration frequency, plug amplitude and vibration duration;
the step S3 specifically comprises the following steps:
s31, performing frequency domain analysis on vibration data through Fourier transformation to determine plug vibration frequency in the plug press-fitting process;
specifically, the acquired time-series vibration data is first input into a fourier transform algorithm. The fourier transform analyzes the data, converts the time domain vibration signal into a frequency domain signal, and then identifies the main frequency component, particularly the frequency with the strongest signal energy, which is the frequency of the choke plug. In general, the frequency with the strongest signal energy corresponds to the fundamental frequency of the signal and is also the main vibration frequency of the plug.
S32, a peak detection algorithm is applied in a time domain to identify a peak value and a valley value of the vibration curve, and plug amplitude is further calculated;
specifically, the time-series vibration curve acquired from step S17 is first taken as input data. On this time series of vibration curves, each peak and trough represents a period of vibration. By identifying these peaks and valleys, we can accurately calculate the amplitude of each cycle and thus the average amplitude of the plug throughout the press-fit process. Specifically, a sliding window based peak detection algorithm accurately identifies the locations of peaks and valleys on the vibration curve. The amplitude of each cycle, i.e. the difference between the peak and the adjacent valley, is calculated after the peak and valley are identified. And finally, calculating the average value of all the periodic amplitudes to obtain the average amplitude of the plug.
S33, determining the duration time of vibration by comparing the average value of the vibration amplitude in the window with a preset vibration amplitude threshold value;
specifically, a suitable window size is first selected, and the window size may be selected based on the characteristics of the actual press-fitting process and vibration data. In general, the window size should be large enough to capture the important features of the vibration and not so large as to mask the variations in the vibration. The mean value of the vibration amplitude within the window is then calculated at each location. This average is the arithmetic average of all vibration amplitude values within the window. The calculated vibration amplitude mean is then compared with a preset vibration amplitude threshold for each window position. If the mean value of the vibration amplitude is greater than or equal to the threshold value, it indicates that the vibration is more significant in this period of time. We will record all window positions where the mean value of the vibration amplitude is greater than or equal to the threshold value and determine the start and end times of the vibration from these positions. Eventually, the duration of the vibration will be equal to the end time of the vibration minus the start time.
S4, carrying out fusion analysis on the pressure point characteristic parameters and the vibration characteristic parameters based on a random forest model, evaluating the press-fitting quality of the plug in real time and sending out warning prompt;
the fusion analysis specifically comprises the following steps:
taking the pressure stabilizing interval and plug amplitude as input characteristics, and calculating to obtain a press-fit stability index S;
calculating a vibration index V in the press-fitting process by taking a pressure fluctuation range and a plug vibration frequency as input characteristics;
calculating a press-fitting efficiency index E by taking the position of the maximum pressure point and the vibration duration as input characteristics;
and carrying out weighted average on the press-fit stability index S, the vibration influence index V and the press-fit efficiency index E according to preset weight values, and calculating to obtain a plug press-fit quality index Q.
Specifically, to calculate the press-fit stability index S, two input features, namely, a pressure stability interval and a plug amplitude, are selected, because they can effectively reflect the stability and vibration conditions of the press-fit process. The pressure stabilizing section represents a section length in which a pressure change during press-fitting is small, and in general, a longer pressure stabilizing section means that the press-fitting process is stable, and a shorter pressure stabilizing section may mean that an unstable factor exists in the press-fitting process. Meanwhile, the vibration amplitude of the plug is obtained by analyzing vibration data in the press-fitting process, the vibration condition of the plug in the press-fitting process can be reflected, and the press-fitting quality can be influenced due to the fact that the vibration amplitude is large. The two characteristics are fused together, so that a comprehensive index which can reflect the stability and vibration condition of the press mounting process can be obtained.
The random forest model is an integrated learning model and is composed of a plurality of decision trees, can process nonlinear data relations and has good generalization capability and fault tolerance capability. By taking the pressure stability interval and the plug amplitude as input features, the random forest model can learn the relationship between the two features and the press-fit stability. The process of model training is based on historical press-fit data, which includes pressure stability intervals and plug amplitudes for each press-fit process, and corresponding press-fit quality assessment results. After model training is finished, the pressure stabilizing interval and plug amplitude data collected in real time can be input into a random forest model, the model can output a press-fit stability index S, and the index can reflect the stability of the current press-fit process in real time. Therefore, through real-time monitoring and analysis, the possible press-fit quality problem can be found in time, and an important information basis is provided for subsequent quality control and warning prompt.
The vibration index V is calculated based on two key input features, namely the pressure fluctuation range and the plug vibration frequency. The choice of these two features is of clear interest. The pressure fluctuation range can quantify the amplitude of pressure change in the press-fitting process, and directly reflects the pressure stability in the press-fitting process. A large pressure fluctuation range may mean that an unstable factor exists in the press-fitting process, and the press-fitting quality may be affected. The vibration frequency of the plug is obtained by analyzing vibration data through a frequency domain, reflects the vibration frequency of the plug in the press-fitting process, and is an important parameter for evaluating the vibration condition in the press-fitting process. Vibration conditions are important factors influencing the press-fit quality, and excessive vibration may cause unstable press-fit and further influence the press-fit quality. The random forest model is again utilized to achieve a fusion analysis of these two features. The random forest model has the advantages that the random forest model can process nonlinear relations and high-dimensional characteristics and has strong generalization capability. Through the training process, the model can capture the association relation between the pressure fluctuation range and the plug vibration frequency and the vibration index V. The training process is based on historical press-fitting data, and the data comprise the pressure fluctuation range, the plug vibration frequency and the corresponding press-fitting quality evaluation result of each press-fitting process. After model training is finished, the pressure fluctuation range and plug vibration frequency data collected in real time can be utilized to be input into a random forest model, so that the vibration index V of the press fitting process is calculated in real time, the vibration index V can provide important information for real-time evaluation and monitoring of the press fitting quality, the vibration index V helps to find possible vibration problems in the press fitting process in real time, and a valuable basis is provided for subsequent quality control and warning prompt.
The press-fitting efficiency index E is calculated by combining two input features of a maximum pressure point position and a vibration duration, and the maximum pressure point position is an important parameter reflecting the mechanical characteristics of the press-fitting process and can help identify a mechanical peak value in the press-fitting process, so that the mechanical efficiency of the press-fitting is evaluated. The duration of vibration is an important parameter reflecting the stability of the press-fitting process, which can help identify the dynamic stability during the press-fitting process, thereby evaluating the dynamic efficiency of the press-fitting. Through training random forest model, can catch the incidence relation between maximum pressure point position and vibration duration and the pressure equipment efficiency index E, this pressure equipment efficiency index E can provide important information for real-time evaluation and control pressure equipment quality, helps finding the pressure equipment efficiency problem that probably exists in real time, provides the basis for follow-up quality control and warning suggestion.
The calculation process of the plug press-fitting quality index Q comprises the following steps:
setting weight values of all indexes according to actual requirements and experience data in the press-fitting process;
the obtained press-fit stability index S, vibration influence index V and press-fit efficiency index E are weighted according to corresponding weight values, and a specific calculation formula of the plug press-fit quality index Q is as follows:
，
wherein,for press-fit stability index weight, +.>For vibration influencing index weight, +.>Is the index weight of the press-fitting efficiency.
Specifically, in the process of calculating the quality index Q of the plug press-fitting, the weight value of each index is set according to the actual demand and the historical experience data of the press-fitting process. These weight values reflect the relative importance of the different indices in evaluating the overall press-fit quality. For example, if press-fit stability is considered to be the most critical quality criterion during a certain press-fit, the weight of the press-fit stability index S will be set higher. The press-fit stability index S reflects the stability of the pressure during press-fitting, which is an important parameter for evaluating the press-fit quality. A stable press-fitting process means that the contact between the plug and the pipe is uniform and stable, thus ensuring a good sealing effect. Its weight valueAccording to actual demand and experience data to ensureThe extent of influence of S in the overall index Q matches its actual importance. The vibration influence index V describes vibration characteristics during press fitting. Excessive vibration may cause damage to the plug or misalignment of the position, affecting the quality of the press fit. Its weight value +.>The degree of influence of vibration on the overall press-fit quality is determined. The press-fit efficiency index E then reflects the efficiency of the press-fit process, including the location of the maximum pressure point and the duration of the vibration. This helps to evaluate the time efficiency of the press-fitting process. Its weight value +.>The importance of efficiency in the overall evaluation is shown. The integrated plug press-fit quality index Q provides a comprehensive and quantitative evaluation standard through the weighted fusion of the three core parameters, and reflects the press-fit quality trend in real time.
The method for evaluating the press-fitting quality of the plug in real time and sending out a warning prompt comprises the following steps of:
comparing the calculated quality index Q of the press fitting of the plug with a preset quality threshold; if Q is higher than a preset threshold, judging the press-fit quality to be qualified; if the Q is lower than a preset threshold, judging that the press-fit quality is unqualified.
Specifically, the qualification represents that the current press-fitting process meets the quality requirement, and no extra measures are needed; otherwise, the disqualification is possibly caused by insufficient or overlarge press-fitting force, if the press-fitting force is insufficient, the sealing between the plug and the pipeline is possibly not tight, leakage is easily generated, the safety and the performance of the equipment are affected, and if the press-fitting force is overlarge, the deformation and even the damage of the plug or the pipeline are possibly caused, and the safety and the performance of the equipment are also affected; the disqualification can be caused by inaccurate press-fitting positions, and if the press-fitting positions of the plugs deviate from preset positions, poor sealing effect or difficult installation of other connecting parts can be caused; the reason for the disqualification may also be excessive vibration, which may affect the press-fit quality and may even cause damage to the equipment if abnormal vibration occurs during the press-fit, which may be caused by equipment failure or improper operation.
The method for evaluating the press-fitting quality of the plug in real time and sending out a warning prompt further comprises the following steps:
and continuously monitoring the change of the press-fit quality index Q, and if the Q has a descending trend and is close to a preset threshold value, sending out a warning prompt.
In particular, the quality control of the press-fitting process is ensured to be in a desired stable state by continuously monitoring the change in the press-fitting quality index Q. In this process, the monitoring system will constantly acquire and analyze real-time data of the press-fit quality index Q. The monitoring system can identify the variation trend of the Q value, and evaluate whether the Q value has a descending trend by calculating the moving average of the Q value.
The warning prompt includes:
depending on the implementation evaluation, the operator is alerted by an audible warning that prompt intervention or examination is required.
Specifically, when the monitoring system detects that the plug press quality index Q has a decreasing trend and approaches to or even falls below a preset threshold, the system automatically triggers a warning mechanism. An audible alert is one of the most efficient notification means that can quickly and directly alert the operator. The audible warning is typically composed of a preset alarm tone or voice prompt. The alert tone typically has a harsher tone to ensure that the operator's attention is drawn in a noisy production environment.
By means of the audible warning, the operator can quickly learn about the problems present and intervene in time, performing the necessary checks or interventions. For example, they may stop the current press-fitting operation, check the working state of the press-fitting device, adjust the device parameters, or check whether the raw materials and working environment conditions are satisfactory. The timeliness and intuitiveness of the sound warning greatly improve the efficiency of problem discovery and response, thereby being beneficial to reducing the generation of unqualified products and ensuring the press-fit quality and the production stability.
The warning cue further comprises:
according to the real-time evaluation, various index anomalies are displayed in bright colors and icon changes on a monitoring interface and possible causes of problems are provided in text form.
Specifically, when the system detects that the press-fit quality index Q is abnormal or close to a preset threshold value, the monitoring interface attracts the attention of an operator through obvious color and icon changes. For example, if the press-fit quality index Q decreases, the associated icon on the monitoring interface may become red, accompanied by the display of a warning symbol. To assist operators and maintenance personnel in quickly identifying and locating problems, the monitoring interface not only displays changes in anomaly indices, but also provides possible causes of the problems in text form. These text prompts may include, but are not limited to: too high pressure, too low pressure, vibration anomalies, positional deviations, and the like.
The above embodiments are merely illustrative of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention, and various modifications and improvements made by those skilled in the art to the technical solution of the present invention should fall within the scope of protection defined by the claims of the present invention without departing from the spirit of the design of the present invention.
Claims (10)
1. The quality detection method for the press fitting of the plug is characterized by comprising the following steps of:
collecting position data, pressure data and vibration data in the plug press-fitting process in real time, and drawing a pressure position curve and a vibration curve in the press-fitting time period;
extracting and analyzing the pressure position curve to obtain plug pressure point characteristic parameters, wherein the pressure point characteristic parameters comprise a pressure stabilizing interval, a pressure fluctuation range and a maximum pressure point position;
extracting and analyzing the vibration curve to obtain plug vibration characteristic parameters, wherein the vibration characteristic parameters comprise plug vibration frequency, plug amplitude and vibration duration;
carrying out fusion analysis on the pressure point characteristic parameters and the vibration characteristic parameters based on a random forest model, evaluating the press-fitting quality of the plug in real time and sending out warning prompt;
the fusion analysis specifically comprises the following steps:
taking the pressure stabilizing interval and plug amplitude as input characteristics, and calculating to obtain a press-fit stability index S;
calculating a vibration index V in the press-fitting process by taking a pressure fluctuation range and a plug vibration frequency as input characteristics;
calculating a press-fitting efficiency index E by taking the position of the maximum pressure point and the vibration duration as input characteristics;
and carrying out weighted average on the press-fit stability index S, the vibration influence index V and the press-fit efficiency index E according to preset weight values, and calculating to obtain a plug press-fit quality index Q.
2. The quality detection method for press fitting of a plug according to claim 1, wherein the collecting position data, time data, pressure data and vibration data in the process of press fitting of the plug in real time specifically comprises the following steps:
the position change of the plug is monitored in real time through a multi-dimensional camera, and position data are obtained;
monitoring pressure change in the pressure packaging process in real time through a pressure sensor to obtain pressure data;
vibration information in the press fitting process is monitored and recorded in real time through a vibration sensor, and vibration data are obtained.
3. The quality inspection method of plug press-fitting according to claim 1, wherein the drawing of the pressure position curve and the vibration curve in the press-fitting time period specifically includes the steps of:
the acquired position data are sorted according to the time sequence, and time sequence position data are generated;
the acquired pressure data are sorted according to the time sequence, and time sequence pressure data are generated;
aligning the time axis of the time-series position data with the time-series pressure data, and drawing a pressure position curve based on the time sequence;
and arranging the acquired vibration data according to a time sequence, and drawing a vibration curve of the time sequence.
4. The quality inspection method of a plug press according to claim 3, wherein the extracting and analyzing the pressure position curve specifically comprises the following steps:
determining key points in the pressure position curve through an extreme point detection algorithm, wherein the key points comprise a pressure starting point, a pressure stabilizing point, a maximum pressure point and corresponding position information;
calculating a pressure stabilizing section in the plug press-fitting process according to the determined key points;
calculating a pressure fluctuation range, wherein the fluctuation range is determined by the difference between the maximum value and the minimum value of the pressure;
and determining the position of the maximum pressure point according to the position information of the key point.
5. The quality inspection method of a plug press according to claim 1, wherein the extracting and analyzing the vibration curve specifically comprises the steps of:
performing frequency domain analysis on the vibration data through Fourier transformation to determine plug vibration frequency in the plug press-fitting process;
a peak detection algorithm is applied in a time domain to identify the peak value and the valley value of the vibration curve, and plug amplitude is further calculated;
the duration of the vibration is determined by comparing the mean value of the vibration amplitude within the window with a preset vibration amplitude threshold.
6. The quality detection method for press-fitting of a plug according to claim 1, wherein the step of calculating the quality index Q of press-fitting of a plug by weighted-averaging the press-fitting stability index S, the vibration influence index V, and the press-fitting efficiency index E according to a preset weight value comprises the following steps:
setting weight values of all indexes according to actual requirements and experience data in the press-fitting process;
the obtained press-fit stability index S, vibration influence index V and press-fit efficiency index E are weighted according to corresponding weight values, and a specific calculation formula of the plug press-fit quality index Q is as follows:
，
wherein,for press-fit stability index weight, +.>For vibration influencing index weight, +.>Is the index weight of the press-fitting efficiency.
7. The quality inspection method of the press-fitting of the plug according to claim 1, wherein the real-time evaluation of the quality of the press-fitting of the plug and the issuing of the warning prompt include the steps of:
comparing the calculated quality index Q of the press fitting of the plug with a preset quality threshold; if Q is higher than a preset threshold, judging the press-fit quality to be qualified; if the Q is lower than a preset threshold, judging that the press-fit quality is unqualified.
8. The quality inspection method of the press-fitting of the plug according to claim 7, wherein the real-time evaluation of the quality of the press-fitting of the plug and the warning of the quality of the press-fitting of the plug are further provided with the steps of:
and continuously monitoring the change of the press-fit quality index Q, and if the Q has a descending trend and is close to a preset threshold value, sending out a warning prompt.
9. The quality inspection method of a plug press according to claim 8, wherein the warning prompt includes:
depending on the implementation evaluation, the operator is alerted by an audible warning that prompt intervention or examination is required.
10. The quality inspection method of a plug press of claim 9, wherein the warning prompt further comprises:
according to the real-time evaluation, various index anomalies are displayed in bright colors and icon changes on a monitoring interface and the cause of the problem is provided in a text form.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311421194.8A CN117150274B (en) | 2023-10-31 | 2023-10-31 | Quality detection method for press fitting of plug |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311421194.8A CN117150274B (en) | 2023-10-31 | 2023-10-31 | Quality detection method for press fitting of plug |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117150274A CN117150274A (en) | 2023-12-01 |
| CN117150274B true CN117150274B (en) | 2024-01-26 |
Family
ID=88901161
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311421194.8A Active CN117150274B (en) | 2023-10-31 | 2023-10-31 | Quality detection method for press fitting of plug |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117150274B (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114167838A (en) * | 2021-12-03 | 2022-03-11 | 西安电子科技大学 | Multi-scale health assessment and fault prediction method for servo system |
| CN115615732A (en) * | 2022-11-11 | 2023-01-17 | 湖北中烟工业有限责任公司 | Quality detector abnormal state monitoring method and system |
| CN116861369A (en) * | 2023-09-04 | 2023-10-10 | 山东捷瑞数字科技股份有限公司 | Air compressor health assessment method and system for industrial internet heterogeneous data fusion |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI449883B (en) * | 2011-02-10 | 2014-08-21 | Univ Nat Taiwan Science Tech | Method for analyzing structure safety |
-
2023
- 2023-10-31 CN CN202311421194.8A patent/CN117150274B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114167838A (en) * | 2021-12-03 | 2022-03-11 | 西安电子科技大学 | Multi-scale health assessment and fault prediction method for servo system |
| CN115615732A (en) * | 2022-11-11 | 2023-01-17 | 湖北中烟工业有限责任公司 | Quality detector abnormal state monitoring method and system |
| CN116861369A (en) * | 2023-09-04 | 2023-10-10 | 山东捷瑞数字科技股份有限公司 | Air compressor health assessment method and system for industrial internet heterogeneous data fusion |
Non-Patent Citations (2)
| Title |
|---|
| 基于功率曲线分析与神经网络的风电机组故障预警方法;乔福宇;马良玉;马永光;;中国测试(第08期);全文 * |
| 基于经验模态分解与深度森林的液压泵健康评估;李志远;黄亦翔;刘成良;李彦明;贡亮;;机械与电子(第05期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117150274A (en) | 2023-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116772944B (en) | Intelligent monitoring system and method for gas distribution station | |
| CN116320832B (en) | Monitoring equipment fault monitoring method and device | |
| CN111354366B (en) | Abnormal sound detection method and abnormal sound detection device | |
| US7872584B2 (en) | Analyzing smoke or other emissions with pattern recognition | |
| CN113982850B (en) | Fan comprehensive health analysis method and system integrating high-low frequency signals | |
| CN114754824A (en) | Monitoring and early warning method and system for wire drawing machine | |
| CN108168659B (en) | Intelligent diagnosis method for vortex shedding flowmeter | |
| CN116028887B (en) | Analysis method of continuous industrial production data | |
| CN114419837A (en) | Safety production early warning system and method thereof | |
| CN115220396A (en) | Intelligent monitoring method and system for numerical control machine tool | |
| CN117791876A (en) | Substation equipment operation state monitoring method and abnormal control system | |
| CN113177646A (en) | Power distribution equipment online monitoring method and system based on self-adaptive edge proxy | |
| CN115356970A (en) | Dynamic loop monitoring system and method based on real-time data and operation and maintenance data | |
| CN117150274B (en) | Quality detection method for press fitting of plug | |
| CN117373478A (en) | Voiceprint recognition system for transformer | |
| CN117523799A (en) | Intelligent monitoring method and system for potential safety hazards | |
| CN116271930A (en) | Remote control system for alcohol refining process | |
| CN114962393B (en) | Intelligent operation management and control system and method for redried tobacco leaf packaging hydraulic system | |
| CN115221924A (en) | Time-series multi-mode-based industrial equipment anomaly detection intelligent recognition algorithm framework | |
| CN106650154B (en) | Nuclear power conventional island equipment reliability monitoring method based on thermal performance index | |
| CN113325824A (en) | Regulating valve abnormity identification method and system based on threshold monitoring | |
| CN112814890A (en) | Method for detecting pump machine fault based on voiceprint and vibration | |
| TWI827176B (en) | System for detecting state of cut-off wheels and method of detecting state of cut-off wheels | |
| CN118197497B (en) | Intelligent automatic control system of punching machine | |
| CN117074845B (en) | Electronic component quality detection platform |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |