CN-122024424-A - SMT equipment monitoring method and system based on digital twinning

CN122024424ACN 122024424 ACN122024424 ACN 122024424ACN-122024424-A

Abstract

The invention relates to the technical field of equipment monitoring, and discloses a digital twinning-based SMT equipment monitoring method and a digital twinning-based SMT equipment monitoring system, wherein the method comprises the following steps: acquiring equipment operation data and time information thereof, carrying out time serialization on the equipment operation data based on the time information, analyzing the variation trend of an operation monitoring waveform, extracting the fluctuation amplitude and the abnormal duration of the operation monitoring waveform, determining the monitoring data according to the occurrence density and the abnormal duration of an abnormal waveform section in unit time, mapping the equipment operation data to a three-dimensional simulation model based on the monitoring data and the component identification corresponding to the equipment operation data, synchronously updating the three-dimensional simulation model at the time generated by a treatment instruction, determining alarm information based on the abnormal waveform section, and judging whether to adjust the alarm frequency of the alarm information based on the historical occurrence times of the abnormal waveform section. The invention ensures the reliability of monitoring the SMT equipment.

Inventors

Luo Bingqin
LIANG ZHENHUA
JIANG HUI

Assignees

深圳市永信达科技股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260327

Claims (10)

1. A digital twinning-based SMT device monitoring method, comprising: Acquiring equipment operation data and time information thereof, carrying out time serialization on the equipment operation data based on the time information, and determining an operation monitoring waveform; analyzing the variation trend of the operation monitoring waveform, extracting the fluctuation amplitude and the abnormal duration of the operation monitoring waveform, determining an abnormal waveform section based on the fluctuation amplitude and the abnormal duration, and determining monitoring data according to the occurrence density of the abnormal waveform section in unit time and the abnormal duration; Determining a monitoring intensity value corresponding to the abnormal waveform section based on the component identification corresponding to the monitoring data and the equipment operation data, determining a treatment instruction according to the monitoring intensity value, mapping the equipment operation data to a three-dimensional simulation model based on the time information, and synchronously updating the three-dimensional simulation model at the time generated by the treatment instruction; and determining alarm information based on the abnormal waveform section, and judging whether to adjust the alarm frequency of the alarm information based on the historical occurrence times of the abnormal waveform section.
2. A digital twinned based SMT device monitoring method according to claim 1, wherein when time-serializing device operation data based on said time information, determining an operation monitoring waveform comprises: And determining data points of the same operation parameter type in the equipment operation data, sorting the data points based on the time information, resampling the sorted data points based on a time step, supplementing middle sampling points when the time interval between adjacent sampling points is larger than the time step, arranging the resampled operation parameter values according to time sequence, and determining an operation monitoring waveform.
3. A digital twinning-based SMT device monitoring method according to claim 2, comprising, when analyzing a trend of variation of said operation monitoring waveform: And determining the parameter variation in unit time based on the numerical value difference of adjacent time points in the operation monitoring waveform, fitting the parameter variation in a time window, determining the variation slope in a corresponding time window, carrying out continuity check on the variation slope of the adjacent time window, and determining the variation trend of the operation monitoring waveform when the slope direction and the amplitude are consistent in the adjacent time window.
4. A digital twinning-based SMT device monitoring method according to claim 3, comprising, when extracting the fluctuation amplitude and anomaly duration of said operation monitoring waveform: Obtaining a maximum value and a minimum value of the operation monitoring waveform in the time window, determining a fluctuation amplitude according to a difference value between the maximum value and the minimum value, comparing the fluctuation amplitude with a fluctuation interval corresponding to a stable operation stage of the equipment, and determining an amplitude offset; And when the change slope of the operation monitoring waveform or the amplitude offset exceeds the corresponding threshold value, determining an abnormal starting time, detecting whether the change slope and the amplitude offset continuously exceed the corresponding threshold value in a sampling period, and determining an abnormal ending time when the change slope and the amplitude offset fall back into the corresponding threshold value, and determining the duration of the abnormality according to the time interval of the abnormal starting time and the abnormal ending time.
5. A digital twinning-based SMT device monitoring method according to claim 4, wherein when determining an abnormal waveform section based on said fluctuation amplitude and abnormal duration, comprising: when the fluctuation amplitude and the abnormal duration simultaneously meet the abnormal combination, determining that the corresponding time period is an abnormal waveform section, merging the abnormal waveform sections with continuous time, and when the time interval of the adjacent abnormal waveform sections is smaller than the merging interval, splicing the abnormal waveform sections, and determining the abnormal waveform sections.
6. A digital twinning-based SMT device monitoring method according to claim 5, wherein when determining monitoring data according to the occurrence density of said abnormal waveform section in a unit time and said abnormal duration, comprising: Counting the occurrence times of the abnormal waveform section, determining the occurrence density of the abnormal waveform section according to the occurrence times and a counting time range, comparing the abnormal duration with a duration interval, determining a duration grade, comparing the occurrence density with a density interval, determining a frequency grade, and determining the monitoring data according to the duration grade and the frequency grade.
7. A digital twinning-based SMT device monitoring method according to claim 6, wherein when determining a monitored intensity value corresponding to said abnormal waveform segment, comprising: Determining corresponding monitoring intensity values from preset monitoring intensity mapping rules according to the abnormal grades, correlating the monitoring intensity values with part identifiers in the equipment operation data, and determining the monitoring intensity values corresponding to different parts.
8. A digital twinned based SMT device monitoring method according to claim 7, wherein when said three-dimensional simulation model is updated synchronously at the time of generation of said treatment instructions, comprising: And mapping the running state of the corresponding time point in the equipment running data to the three-dimensional simulation model based on the time information, and updating the running state parameters of the corresponding parts in the three-dimensional simulation model according to all the monitoring intensity values.
9. A digital twin based SMT device monitoring method according to claim 8, wherein when judging whether to adjust the warning frequency of said warning information based on the historical number of occurrences of said abnormal waveform section, comprising: And updating the abnormal waveform section to the accumulated occurrence frequency in the historical operation process, judging that the alarm frequency of the alarm information is adjusted when the accumulated occurrence frequency is larger than or equal to the accumulated occurrence frequency threshold, and judging that the alarm frequency of the alarm information is not adjusted when the accumulated occurrence frequency is smaller than the accumulated occurrence frequency threshold.
10. A digital twinning-based SMT device monitoring system for applying a digital twinning-based SMT device monitoring method according to any one of claims 1-9, comprising: the acquisition unit is configured to acquire equipment operation data and time information thereof, and time-series the equipment operation data based on the time information to determine an operation monitoring waveform; an analysis unit configured to analyze a variation trend of the operation monitoring waveform, extract a fluctuation amplitude and an abnormal duration of the operation monitoring waveform, determine an abnormal waveform section based on the fluctuation amplitude and the abnormal duration, and determine monitoring data according to an occurrence density of the abnormal waveform section in a unit time and the abnormal duration; The processing unit is configured to determine a monitoring intensity value corresponding to the abnormal waveform section based on the monitoring data and the component identification corresponding to the equipment operation data, determine a treatment instruction according to the monitoring intensity value, map the equipment operation data to a three-dimensional simulation model based on the time information, and synchronously update the three-dimensional simulation model at the time generated by the treatment instruction; And a monitoring unit configured to determine alarm information based on the abnormal waveform section and determine whether to adjust an alarm frequency of the alarm information based on the historical number of occurrences of the abnormal waveform section.

Description

SMT equipment monitoring method and system based on digital twinning Technical Field The invention relates to the technical field of equipment monitoring, in particular to a digital twinning-based SMT equipment monitoring method and system. Background With the improvement of the automation and informatization degree of the SMT production line, higher requirements are put forward on the monitoring of equipment, the existing monitoring mode is remained on a data display or threshold alarm level, the dynamic change trend of the operation of the equipment is difficult to capture, a unified quantification mechanism is lacking in the duration time, the occurrence frequency and the accumulation relation among multiple anomalies of the abnormal state, the difference between occasional fluctuation and continuous degradation is difficult to distinguish due to the lack of grading and combined evaluation of the abnormal characteristics, and even if the existing monitoring mode introduces a digital twin technology to carry out visual display on the three-dimension of the equipment, the linkage relation between the existing monitoring mode and the abnormal evaluation result is lacking, the mapping and synchronous updating of the data and the abnormal information which are operated in real time cannot be carried out, so that the virtual simulation environment can reflect the operation of the equipment, but the change state of the operation risk of the equipment cannot be reflected. Therefore, there is a need to design a digital twinning-based SMT device monitoring method and system to solve the problems in the prior art. Disclosure of Invention In view of this, the invention provides a digital twinning-based SMT equipment monitoring method and system, which aims to solve the problems that dynamic change trend of equipment operation is difficult to capture, a unified quantification mechanism is lacking in duration time, occurrence frequency of abnormal states and accumulation relation among multiple anomalies, and linkage between digital twinning lack and an anomaly evaluation result causes that a virtual simulation environment can reflect equipment actions but cannot reflect change states of equipment operation risks. In one aspect, the invention provides a digital twinning-based SMT device monitoring method, comprising: Acquiring equipment operation data and time information thereof, carrying out time serialization on the equipment operation data based on the time information, and determining an operation monitoring waveform; analyzing the variation trend of the operation monitoring waveform, extracting the fluctuation amplitude and the abnormal duration of the operation monitoring waveform, determining an abnormal waveform section based on the fluctuation amplitude and the abnormal duration, and determining monitoring data according to the occurrence density of the abnormal waveform section in unit time and the abnormal duration; Determining a monitoring intensity value corresponding to the abnormal waveform section based on the component identification corresponding to the monitoring data and the equipment operation data, determining a treatment instruction according to the monitoring intensity value, mapping the equipment operation data to a three-dimensional simulation model based on the time information, and synchronously updating the three-dimensional simulation model at the time generated by the treatment instruction; and determining alarm information based on the abnormal waveform section, and judging whether to adjust the alarm frequency of the alarm information based on the historical occurrence times of the abnormal waveform section. Further, when the device operation data is time-serialized based on the time information, determining the operation monitoring waveform includes: And determining data points of the same operation parameter type in the equipment operation data, sorting the data points based on the time information, resampling the sorted data points based on a time step, supplementing middle sampling points when the time interval between adjacent sampling points is larger than the time step, arranging the resampled operation parameter values according to time sequence, and determining an operation monitoring waveform. Further, when analyzing the trend of the operation monitoring waveform, the method includes: And determining the parameter variation in unit time based on the numerical value difference of adjacent time points in the operation monitoring waveform, fitting the parameter variation in a time window, determining the variation slope in a corresponding time window, carrying out continuity check on the variation slope of the adjacent time window, and determining the variation trend of the operation monitoring waveform when the slope direction and the amplitude are consistent in the adjacent time window. Further, when extracting the fluctuation amplitude and the abnormal duration of the operat