A real-time detection method for abnormal pressure at laser cutting outlet

Abstract

The present invention discloses a real-time detection method for abnormal outlet pressure of laser cutting. Firstly, a sliding calculation method is adopted to traverse the inlet pressure data of a protective gas source, and a verification segment set is preliminarily screened according to the fluctuation of the inlet pressure of the gas source; then, the verification segment is extracted and all time points are traversed to perform abnormal outlet pressure detection of the laser cutting equipment, and it is judged whether the current outlet pressure is abnormal based on the fluctuation of the inlet pressure; all verification segments are traversed and all abnormal outlet pressure segments are marked; the abnormal outlet pressure detection method designed by the present invention makes abnormal judgment on the outlet pressure based on different possible situations of the inlet pressure, such as stability, upward fluctuation, downward fluctuation, etc., and finally obtains the abnormal verification segment, and the judgment result is accurate.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent welding and cutting, and in particular relates to a real-time detection method for abnormal pressure at a laser cutting outlet. Background Art

[0002] In the process of industrial laser cutting, in order to ensure product quality and work efficiency, a stable shielding gas flow is required to protect the cutting operation. In the prior art, there is a control device for shielding gas flow, which can control the shielding gas outlet pressure of the laser cutting equipment to the target value. However, when the inlet pressure of the existing shielding gas outlet pressure control system fluctuates, the outlet pressure will also be abnormal. This abnormal fluctuation is difficult to detect, but it is of great significance for the improvement of the outlet pressure control logic. Therefore, it is necessary to design a method for detecting abnormal outlet pressure to detect abnormal fluctuations in outlet pressure. Summary of the invention

[0003] Purpose of the invention: In view of the problems existing in the above-mentioned background technology, the present invention provides a real-time detection method for abnormal outlet pressure of laser cutting. First, several sections to be checked are divided based on the inlet pressure, and then all elements in the check section are traversed to judge the abnormal outlet pressure based on the changing relationship between the inlet pressure and the outlet pressure.

[0004] Technical solution: A real-time detection method for abnormal laser cutting outlet pressure, comprising the following steps:

[0005] Step S1, obtaining laser cutting timing data, and generating a verification segment set based on sliding calculation; the laser cutting timing data includes the shielding gas inlet pressure and outlet pressure; calculating the median of the inlet pressure of all sliding windows, and generating a median set, judging and generating a number of verification segments based on the median set, and storing them in the verification segment set;

[0006] Step S2, extract the verification segment and traverse all time points, perform outlet pressure anomaly detection, and determine whether the current outlet pressure is abnormal based on the inlet pressure fluctuation; traverse all verification segments and mark all outlet pressure abnormal segments.

[0007] Furthermore, the specific method of generating the check segment set based on sliding calculation in step S1 includes:

[0008] Step S1.1, real-time acquisition of a laser cutting time series data set, wherein the time series data set includes the inlet pressure and outlet pressure corresponding to each time series index index; setting a sliding window and sliding the time series data set in sequence with a window length ws, calculating the median of the inlet pressure in each sliding window, generating a median combination, and presetting an offset offset;

[0009] Step S1.2, set pointers left and right, where left points to the first element in the median set, and right points to the second element in the median set; calculate the difference between the elements pointed to by left and right, and when the absolute value of the difference is less than the preset threshold th1, it means that the inlet pressure fluctuates slightly, at this time right is shifted right by one position, and the difference is calculated repeatedly until the right pointer exceeds the inlet pressure median set;

[0010] Step S1.3, when the absolute value of the difference is greater than or equal to the preset threshold value th1, it means that the median of the inlet pressure fluctuates significantly. At this time, the corresponding time series data set is intercepted with left as the starting point and right+offset as the end point to generate a verification segment; after the generation is completed, move left to the right position, move right one position to the right, and repeat the calculation of the difference in step S1.2; finally, obtain several verification segments and store them in the verification segment set.

[0011] Furthermore, the method for detecting outlet pressure abnormality in real time in step S2 specifically includes:

[0012] Step S2.1, extract a check segment and check it in sequence, calculate the inlet pressure difference and outlet pressure difference between adjacent indexes in sequence, and store them in a difference set; when the difference set is less than the offset, shift the check position right by one position and re-acquire the difference set, otherwise execute step S2.2;

[0013] Step S2.2, calculate the mean of all inlet pressure differences in the difference set, recorded as the inlet pressure difference mean; calculate the mean of the outlet pressure differences, recorded as the outlet pressure difference mean; calculate the difference between the outlet pressure and the inlet pressure at the offset position before the current calibration position, recorded as the follow-up pressure difference;

[0014] Step S2.2.1, when the absolute value of the mean inlet pressure difference is less than or equal to the preset threshold value th2, it means that the inlet pressure is relatively stable, and the outlet pressure is judged at this time; when the absolute value of the follow-up pressure difference is greater than the preset threshold value th3, or the absolute value of the mean outlet pressure difference is greater than the preset threshold value th4, it means that the outlet pressure is abnormal;

[0015] Step S2.2.2, when the absolute value of the mean inlet pressure difference is greater than the preset threshold value th2, further judge, if the mean inlet pressure difference is greater than 0, it means that the inlet pressure rises, at this time calculate the following pressure difference and the mean outlet pressure difference and judge as follows: when the absolute value of the following pressure difference is less than or equal to the preset threshold value th5 and the absolute value of the mean outlet pressure difference is greater than the preset threshold value th6, it means that the outlet pressure is abnormal, and the number of bad points is increased by 1;

[0016] Step S2.2.3, when the mean value of the inlet pressure difference is less than 0, it means that the inlet pressure decreases, and the following pressure difference is further determined; when the absolute value of the following pressure difference is less than or equal to the preset threshold value th7, it means that the outlet pressure fluctuates with the inlet pressure. At this time, if the absolute value of the difference between the mean value of the inlet pressure difference and the mean value of the outlet pressure difference is greater than the preset threshold value th8, it means that the outlet pressure is abnormal, and the number of bad points is increased by 1;

[0017] Step S2.2.4: When the absolute value of the follow pressure difference is greater than the preset threshold value th7, further judgment is made. If the mean value of the outlet pressure difference is greater than the preset threshold value th9, it means that the outlet airflow is unstable and the outlet pressure is abnormal, and the number of bad points is increased by 1.

[0018] Furthermore, after completing a round of verification in step S2, the first element in the difference set is removed, and the verification position is shifted right by one position, and the pressure anomaly detection in steps S2.1 and S2.2 is repeated until the verification position exceeds the verification segment length; the number of all bad points is counted and compared with the preset tolerance, and when the number of bad points is greater than the tolerance, the corresponding verification segment is marked as abnormal.

[0019] Furthermore, all the check segments in the check segment set are checked repeatedly, and all abnormal check segments are obtained and marked in the time series data set.

[0020] The method for real-time detection of abnormal laser cutting outlet pressure adopted by the present invention has the following beneficial effects:

[0021] Based on the control of the shielding gas outlet pressure by the outlet pressure control system, an accurate real-time detection method for pressure anomaly is proposed for the outlet pressure anomaly caused by the inlet pressure fluctuation. By collecting laser cutting time series data and generating several verification segments based on the inlet pressure, an accurate method for judging outlet pressure anomaly is given for different verification segments according to the different change trends of the inlet pressure. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a flow chart of the real-time detection method for abnormal laser cutting outlet pressure provided by the present invention. DETAILED DESCRIPTION

[0023] In view of the stable outlet pressure required during the laser cutting operation, there are related control systems in the prior art that can control the outlet pressure at a stable set threshold. However, due to the change in the inlet pressure, the outlet pressure often cannot be maintained at a steady value for a long time, and it is also impossible to effectively monitor whether the outlet pressure is abnormal. In view of the above problems, the present invention provides a real-time detection method for abnormal outlet pressure of laser cutting, which first obtains a set of sections to be inspected through the inlet pressure, and then determines whether the outlet pressure is abnormal based on the follow-up relationship between the inlet pressure and the outlet pressure.

[0024] The present invention will be further explained below in conjunction with the accompanying drawings.

[0025] like Figure 1 As shown, the present invention provides a method for real-time detection of abnormal pressure at a laser cutting outlet, comprising the following steps:

[0026] Step S1, obtaining laser cutting timing data, and generating a verification segment set based on sliding calculation.

[0027] Step S1.1, first collect the time series data set in the laser cutting operation in real time through the external sensor system, and the time series data set includes the inlet pressure and outlet pressure corresponding to each time series index index. The inlet pressure is the shielding gas pressure at the input laser cutting equipment, and the outlet pressure is the shielding gas output end pressure of the laser cutting equipment. Set a sliding window and slide the time series data set in sequence with the window length ws. Each slide calculates the median of the inlet pressure in the window to generate an inlet pressure median set. Preset offset offset.

[0028] Step S1.2, initialize pointers left and right, where left points to the first element in the median set and right points to the second element in the median set, calculate the difference between the elements pointed to by left and right, and when the absolute value of the difference is less than the preset threshold th1, it means that the inlet pressure fluctuation is not large. At this time, right is shifted right by one position, and the difference is calculated repeatedly until the right pointer exceeds the inlet pressure median set.

[0029] Step S1.3, when the absolute value of the difference is greater than or equal to the preset threshold value th1, it means that the median of the inlet pressure fluctuates significantly. At this time, a check segment is generated with left as the starting point and right+offset as the end point. The offset represents the boundary of the original inlet pressure data corresponding to the median pointed to by the pointer. The original time series data is corresponded to [left-offset, right+offset], and the check segment can be truncated. In order to avoid duplication, the corresponding original time series data is intercepted with [left, right+offset] in this embodiment. After the generation is completed, move left to the right position, right one position to the right, and repeat the calculation of the difference in step S1.2. Finally, several check segments are obtained and stored in the check segment set.

[0030] Step S2: After obtaining the verification segment set, the outlet pressure abnormality is judged based on the inlet pressure and outlet pressure values. Specifically,

[0031] Step S2.1, extract a check segment and check it in sequence, calculate the inlet pressure difference and outlet pressure difference between adjacent indexes in sequence, and store them in a difference set; when the difference set is less than the offset offset, shift the check position right by one position, otherwise execute step S2.2.

[0032] Step S2.2, calculate the mean of all inlet pressure differences in the difference set, recorded as the inlet pressure difference mean; calculate the mean of the outlet pressure differences, recorded as the outlet pressure difference mean; and simultaneously calculate the difference between the outlet pressure and the inlet pressure at the offset position before the current calibration position, i.e., the following pressure difference.

[0033] When dividing the verification segment in step S1, in order to avoid duplication, [left, right+offset] is taken as the index of the corresponding verification segment. However, when actually judging the pressure abnormality, the time series data corresponding to the offset before the current verification position should be used as the verification point.

[0034] Step S2.2.1, when the absolute value of the mean inlet pressure difference is less than or equal to the preset threshold value th2, it means that the inlet pressure is relatively stable at this time, and correspondingly, the outlet pressure should also remain stable. At this time, it is necessary to detect the outlet pressure. When the absolute value of the following pressure difference is greater than the preset threshold value th3, or the absolute value of the mean outlet pressure difference is greater than the preset threshold value th4, it means that the outlet pressure does not follow the inlet pressure in time, or the outlet pressure is unstable, which are abnormal conditions, and the number of bad points is increased by 1.

[0035] Step S2.2.2: When the absolute value of the inlet pressure difference mean is greater than the preset threshold value th2, it means that the inlet pressure is unstable. If the inlet pressure difference mean is greater than 0, it means that the inlet pressure is rising. When the inlet pressure fluctuates upward, the outlet pressure should be stabilized at a fixed value according to the setting of the external control system. If the outlet pressure also rises at this time, it means that the outlet pressure is abnormal. The specific judgment is as follows:

[0036] When the absolute value of the follow-up pressure difference is less than or equal to the preset threshold value th5 and the absolute value of the outlet pressure difference mean is greater than the preset threshold value th6, it indicates that an abnormality exists and the number of bad points increases by 1.

[0037] Step S2.2.3, when the mean value of the inlet pressure difference is less than 0, it means that the inlet pressure continues to decrease. At this time, it is necessary to further judge the following pressure difference. When the absolute value of the following pressure difference is less than or equal to the preset threshold value th7, it means that the outlet pressure fluctuates following the inlet pressure. At this time, if the absolute value of the difference between the mean value of the inlet pressure difference and the mean value of the outlet pressure difference is greater than the preset threshold value th8, it means that the outlet pressure fluctuates following the inlet pressure, and the outlet pressure of the entire verification section fluctuates following the inlet pressure, so there is an abnormality in the outlet pressure, and the number of bad points increases by 1.

[0038] Step S2.2.4: When the absolute value of the following pressure difference is greater than the preset threshold value th7, it means that the outlet pressure does not follow the inlet pressure in time. At this time, it is necessary to further determine whether the outlet pressure is stable. If the mean value of the outlet pressure difference is greater than the preset threshold value th9, it means that the outlet pressure fluctuates and there is an abnormality. The number of bad points increases by 1. Otherwise, the outlet pressure is stable and no abnormality occurs.

[0039] Step S2.3: After completing a round of verification, remove the first element in the difference set, and shift the verification position rightward by one position, and repeat the pressure anomaly detection in steps S2.1 and S2.2 until the verification position exceeds the verification segment length. At this time, count all the bad points and compare them with the preset tolerance. When the bad point number is greater than the tolerance, mark the corresponding verification segment as abnormal.

[0040] Step S2.4: Repeat steps S2.1 to S2.3 for all check segments in the check segment set, obtain abnormal check segments, and mark them in the time series data set.

[0041] The real-time detection method for abnormal outlet pressure of laser cutting provided by the present invention can judge the abnormality of outlet pressure by analyzing the time series data of inlet pressure and outlet pressure. During the laser cutting process, the outlet pressure is controlled by an external device. When the inlet pressure fluctuates, the outlet pressure will have various abnormalities. The present invention designs a two-step real-time detection method for abnormal outlet pressure. First, the inlet pressure fluctuation is detected based on the median of the inlet pressure, and the remaining steady-state segments are eliminated. Then, the verification is performed in sequence based on the screened verification segments. Based on the different situations that the inlet pressure may be stable, upward fluctuation, downward fluctuation, etc., the outlet pressure is judged abnormally, and finally the abnormal verification segment is obtained, and the abnormal verification segment is marked in the time series data table.

[0042] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for real-time detection of abnormal pressure at a laser cutting outlet, characterized in that: The following steps are involved: Step S1, obtaining laser cutting timing data, and generating a verification segment set based on sliding calculation; the laser cutting timing data includes the shielding gas inlet pressure and outlet pressure; calculating the median of the inlet pressure of all sliding windows, and generating a median set, judging and generating a number of verification segments based on the median set, and storing them in the verification segment set; Step S2, extract the verification segment and traverse all time points to perform outlet pressure anomaly detection, and determine whether the current outlet pressure is abnormal based on the inlet pressure fluctuation; Traverse all the calibration sections and mark all sections with abnormal outlet pressure; The outlet pressure abnormality detection method specifically comprises: Step S2.1, extract a check segment and check it in sequence, calculate the inlet pressure difference and outlet pressure difference between adjacent indexes in sequence, and store them in a difference set; when the difference set is less than the offset, shift the check position right by one position and re-acquire the difference set, otherwise execute step S2.2; Step S2.2, calculate the mean of all inlet pressure differences in the difference set, recorded as the inlet pressure difference mean; calculate the mean of the outlet pressure differences, recorded as the outlet pressure difference mean; calculate the difference between the outlet pressure and the inlet pressure at the offset position before the current calibration position, recorded as the follow-up pressure difference; Step S2.2.1, when the absolute value of the mean inlet pressure difference is less than or equal to the preset threshold value th2, it means that the inlet pressure is relatively stable, and the outlet pressure is judged at this time; when the absolute value of the follow-up pressure difference is greater than the preset threshold value th3, or the absolute value of the mean outlet pressure difference is greater than the preset threshold value th4, it means that the outlet pressure is abnormal; Step S2.2.2, when the absolute value of the mean inlet pressure difference is greater than the preset threshold value th2, further judge, if the mean inlet pressure difference is greater than 0, it means that the inlet pressure rises, at this time calculate the following pressure difference and the mean outlet pressure difference and judge as follows: when the absolute value of the following pressure difference is less than or equal to the preset threshold value th5 and the absolute value of the mean outlet pressure difference is greater than the preset threshold value th6, it means that the outlet pressure is abnormal, and the number of bad points is increased by 1; Step S2.2.3, when the mean value of the inlet pressure difference is less than 0, it means that the inlet pressure decreases, and the following pressure difference is further determined; when the absolute value of the following pressure difference is less than or equal to the preset threshold value th7, it means that the outlet pressure fluctuates with the inlet pressure. At this time, if the absolute value of the difference between the mean value of the inlet pressure difference and the mean value of the outlet pressure difference is greater than the preset threshold value th8, it means that the outlet pressure is abnormal, and the number of bad points is increased by 1; Step S2.2.4: When the absolute value of the follow pressure difference is greater than the preset threshold value th7, further judgment is made. If the mean value of the outlet pressure difference is greater than the preset threshold value th9, it means that the outlet airflow is unstable and the outlet pressure is abnormal, and the number of bad points is increased by 1.

2. A method for real-time detection of abnormal laser cutting outlet pressure according to claim 1, characterized in that: The specific method for generating a check segment set based on sliding calculation in step S1 includes: Step S1.1, real-time acquisition of a laser cutting time series data set, wherein the time series data set includes the inlet pressure and outlet pressure corresponding to each time series index index; setting a sliding window and sliding the time series data set in sequence with a window length ws, calculating the median of the inlet pressure in each sliding window, generating a median combination, and presetting an offset offset; Step S1.2, set pointers left and right, where left points to the first element in the median set, and right points to the second element in the median set; calculate the difference between the elements pointed to by left and right, and when the absolute value of the difference is less than the preset threshold th1, it means that the inlet pressure fluctuates slightly, at this time right is shifted right by one position, and the difference is calculated repeatedly until the right pointer exceeds the inlet pressure median set; Step S1.3, when the absolute value of the difference is greater than or equal to the preset threshold value th1, it means that the median of the inlet pressure fluctuates significantly. At this time, the corresponding time series data set is intercepted with left as the starting point and right+offset as the end point to generate a verification segment; after the generation is completed, move left to the right position, move right one position to the right, and repeat the calculation of the difference in step S1.2; finally, obtain several verification segments and store them in the verification segment set.

3. The method for real-time detection of abnormal laser cutting outlet pressure according to claim 1, characterized in that: After completing a round of verification in step S2, remove the first element in the difference set, and shift the inspection position right by one position, and repeat the pressure anomaly detection in steps S2.1 and S2.2 until the verification position exceeds the verification segment length; count all bad points and compare them with the preset tolerance. When the bad point number is greater than the tolerance, mark the corresponding verification segment as abnormal.

4. A method for real-time detection of abnormal laser cutting outlet pressure according to claim 3, characterized in that: Repeat the check for all check segments in the check segment set, obtain all abnormal check segments, and mark them in the time series data set.

Images