A method for real-time monitoring of data of plasma cutting and welding integrated equipment

Abstract

The present invention relates to the field of data processing technology, and more specifically, to a method for real-time data monitoring of plasma cutting and welding integrated equipment, the method comprising: obtaining readings of multiple sensors electrically connected to the equipment to obtain a process parameter sequence; obtaining a neighborhood set of the parameter based on the neighbor index of any parameter in the process parameter sequence, the neighbor index being the value of the average difference between the parameter and the previous multiple parameters after rounding; correcting the neighbor index using the stability of the neighborhood set, the corrected neighbor index being negatively correlated with the stability, and obtaining the outlier coefficient of any parameter based on the corrected neighbor index using the ABOD algorithm, so as to monitor abnormal parameters according to the comparison result of the outlier coefficient and a preset threshold. The present invention can realize accurate monitoring of abnormal parameters of equipment.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and more specifically, to a method for real-time data monitoring of plasma cutting and welding integrated equipment. Background Art

[0002] Plasma cutting and welding equipment is an advanced manufacturing equipment that integrates cutting and welding functions and is widely used in many fields. Due to complex factors such as high temperature, high pressure and high speed operation during plasma cutting and welding, the equipment is prone to failure or deviation from the preset working state. For example, fluctuations in current and gas pressure may lead to unstable cutting quality, and abnormal temperature may cause material deformation or equipment damage. Therefore, it is of vital importance to monitor and warn of abnormal conditions in a timely manner during the production process.

[0003] In the related art, a method for temperature monitoring of an intelligent cutting head is disclosed in a patent application document with publication number CN117668734A, which includes: obtaining temperature data of the intelligent cutting head in a working state; obtaining a temperature fluctuation curve composed of the temperature data; dividing the temperature data into intervals according to the changing characteristics of the temperature fluctuation curve; obtaining the outlier degree of the temperature data according to the degree of temperature difference between each temperature data and all temperature data in its own interval, and then obtaining an adaptive k value for each temperature data; performing LOF abnormal data monitoring on all temperature data using the adaptive k value to obtain all abnormal temperature data; and performing temperature monitoring on the intelligent cutting head according to the abnormal temperature data.

[0004] However, the above scheme does not fully consider the influence of noise when detecting abnormal temperature, and the working environment of the cutting head may be disturbed by many factors, and the influence of noise is relatively large. This may lead to a decrease in the accuracy of outlier calculation, resulting in the inability to accurately identify abnormal data, affecting the accuracy of abnormal data monitoring of the equipment. Summary of the invention

[0005] In order to solve the problem of being unable to accurately monitor abnormal data of equipment, the present invention provides a real-time data monitoring method for plasma cutting and welding integrated equipment. The method comprises:

[0006] Obtaining readings of a plurality of sensors electrically connected to the device to obtain a sequence of process parameters;

[0007] Based on the neighbor index of any parameter in the process parameter sequence, a neighborhood set of the parameter is obtained, where the neighbor index is a rounded value of the average difference between the parameter and multiple previous parameters;

[0008] The neighbor index is corrected by using the stability of the neighborhood set. The corrected neighbor index is negatively correlated with the stability. Based on the corrected neighbor index, the outlier coefficient of any parameter is obtained by using the ABOD algorithm, so as to monitor the abnormal parameters according to the comparison result between the outlier coefficient and the preset threshold.

[0009] The method for obtaining stability is: calculate the average change rate of any dimension data in the neighborhood set, obtain the degree of fluctuation of the dimension data in the neighborhood set, and use the weight to obtain the weighted value of the degree of fluctuation. The weight is: , Neighborhood set Neidi The weight of the volatility of the dimensional data, Neighborhood set Neidi Dimensional data and The correlation of the data in each dimension, , Neighborhood Sets Neidi Dimensional data and The degree of fluctuation of the data in each dimension, is a normalization function; the weighted differences of the volatility of data of different dimensions in the neighborhood set are summed, and the normalized value of the reciprocal of the accumulated sum is used as stability.

[0010] The present invention utilizes the ABOD algorithm for anomaly detection based on angles to detect abnormal parameters of equipment, which can capture complex abnormal patterns in multidimensional data and is not easily affected by noise or scale differences of data of different dimensions. In the process of using the algorithm to identify abnormal parameters of equipment, a variety of data are combined to determine the neighborhood size when performing outlier measurement on each parameter, which can accurately measure the local characteristics of each parameter, further ensure the accuracy of the determined abnormal parameters, and realize accurate monitoring of abnormal parameters of equipment.

[0011] Preferably, the fluctuation degree satisfies the following relationship:

[0012] ;

[0013] In the formula, Neighborhood set Neidi The degree of fluctuation of the data in each dimension; , Neighborhood Sets Neidi The dimension Data and The value of the data; is the total number of data in the neighborhood set; for the summation symbol; is the absolute value symbol.

[0014] The present invention calculates the degree of fluctuation of each dimensional data in the neighborhood set and uses the fluctuation difference of data of different dimensions to measure the stability of the neighborhood set. It can capture the complex correlation and fluctuation law in multidimensional data, thereby ensuring the accuracy of the determined stability.

[0015] Preferably, the method for obtaining the stability of the neighborhood set includes:

[0016] The product of the fluctuation degree of any dimension data in the neighborhood set and the weight of the fluctuation degree of the dimension data is used as the weighted value of the fluctuation degree of the dimension data in the neighborhood set;

[0017] The stability of the neighborhood set satisfies the following relationship:

[0018] ;

[0019] In the formula, is the stability of the neighborhood set; , They are the first Dimensional data and The weighted value of the volatility of the data in each dimension; is the number of dimensions in the neighborhood set; for the summation symbol; The natural constant An exponential function with base .

[0020] The present invention can reduce the influence of the magnitude difference of data of different dimensions by calculating the weighted value of each dimensional data in the neighborhood set.

[0021] Preferably, when calculating the weight of the fluctuation degree of any dimensional data in the neighborhood set, the method for obtaining the correlation between different dimensional data includes:

[0022] The DTW distance between data of different dimensions in the neighborhood set is obtained to obtain the correlation between the corresponding two-dimensional data.

[0023] Preferably, the neighbor index is modified by using the stability of the neighborhood set, including:

[0024] Get the preset hyperparameters. If the stability of the neighborhood set is greater than or equal to the hyperparameters, the neighbor indexes before and after the correction are the same.

[0025] If the stability of the neighborhood set is less than the hyperparameter, the modified neighbor index satisfies the following relationship:

[0026] ;

[0027] In the formula, is the modified neighbor index; is the neighbor index before correction; is the stability of the neighborhood set; The symbol for rounding up.

[0028] The present invention corrects the neighbor index of the parameter so that the corrected neighbor index is more consistent with the distribution of actual data and task requirements.

[0029] Preferably, the method for obtaining the neighbor index includes:

[0030] Each parameter in the process parameter sequence is regarded as a vector, and the modulus length of each vector is calculated to obtain the value of the corresponding parameter. The data segment composed of multiple continuous parameters before any parameter is regarded as the local range of any parameter; the neighbor index of any parameter satisfies the following relationship:

[0031] ;

[0032] In the formula, is the first in the process parameter sequence Neighborhood index of the parameters; is the first in the process parameter sequence The value of a parameter; is the first The value of the data; is the total number of data within the local range; for the summation symbol; The symbol for rounding up.

[0033] The present invention sets the neighborhood index of the corresponding parameter in combination with the distribution density of the data around the parameter, which can better adapt to the difference in data distribution.

[0034] Preferably, monitoring abnormal parameters according to the comparison result of the outlier coefficient with the preset threshold value includes:

[0035] Obtaining a preset threshold value, when the outlier coefficient of any parameter in the process parameter sequence is greater than or equal to the preset threshold value, the parameter is determined to be an abnormal parameter;

[0036] When the outlier coefficient of any parameter in the process parameter sequence is less than a preset threshold, the parameter is determined to be a normal parameter.

[0037] The present invention has the following effects:

[0038] 1. The present invention takes into account the complex working environment and noise influence of the equipment, monitors the anomalies of multiple dimensional data of the equipment, and can effectively capture complex abnormal patterns. At the same time, it avoids the interference of noise and scale differences between different dimensional data on the results, thereby ensuring the accuracy of the determined abnormal parameters.

[0039] 2. When the present invention uses the ABOD algorithm to measure the outlier of each parameter, it can flexibly and adaptively adjust the neighborhood size according to the distribution characteristics of the current parameters, which can effectively avoid the misjudgment of outliers caused by the neighborhood being too large or too small, thereby accurately identifying abnormal parameters and realizing accurate monitoring of abnormal parameters of equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0041] Figure 1 It is a schematic flow chart of the steps of a method for real-time monitoring of data of a plasma cutting and welding integrated device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0043] The specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0044] Reference Figure 1 A method for real-time monitoring of data of a plasma cutting and welding integrated device includes steps S1 to S4, which are specifically as follows:

[0045] S1: Obtain readings of multiple sensors electrically connected to the equipment to obtain a process parameter sequence.

[0046] Optionally, the readings of the temperature sensor, pressure sensor and current sensor electrically connected to the plasma cutting and welding integrated device can be collected at a certain frequency, such as 1 Hz, so as to obtain three process parameters at each sampling moment and form a process parameter sequence. The present invention does not specifically limit the frequency of data collection.

[0047] In an exemplary embodiment of the present invention, each parameter in the process parameter sequence includes data of multiple dimensions, and each dimension corresponds to a type of sensor one by one.

[0048] Among them, each parameter in the process parameter sequence contains data in three dimensions: temperature, voltage and current, that is, each parameter is a triplet consisting of the temperature, pressure and current of the plasma cutting and welding integrated equipment obtained at the corresponding sampling moment.

[0049] S2: Based on the neighbor index of any parameter in the process parameter sequence, a neighborhood set of the parameter is obtained, where the neighbor index is a rounded value of the average difference between the parameter and multiple previous parameters.

[0050] Among them, the neighbor index refers to the data that can determine the number of neighbor points of the parameters in the process parameter sequence. The setting of the neighbor index of each parameter is related to the distribution discreteness of the surrounding data of the corresponding parameter. If the distribution of the surrounding data of any parameter is relatively dense, setting a smaller neighbor index is sufficient to reflect the local characteristics of the parameter; on the contrary, if the distribution of the surrounding data of any parameter is relatively discrete, it is necessary to set a larger neighbor index to ensure that there is sufficient data basis when analyzing the local characteristics of the parameter.

[0051] In an exemplary embodiment of the present invention, the neighbor index of any parameter in the process parameter sequence may be determined by the following steps:

[0052] Each parameter in the process parameter sequence is regarded as a vector, and the modulus length of each vector is calculated to obtain the value of the corresponding parameter. The data segment composed of multiple continuous parameters before any parameter is regarded as the local range of any parameter; the neighbor index of any parameter satisfies the following relationship:

[0053] ;

[0054] In the formula, is the first in the process parameter sequence Neighborhood index of the parameters; is the first in the process parameter sequence The value of a parameter; is the first The value of the data; is the total number of data within the local range; for the summation symbol; The symbol for rounding up.

[0055] in, Reflects the first The average difference between the parameter and the data in the local range, the larger the value, the If the distribution of the surrounding data of the first parameter is relatively discrete, a larger neighbor index needs to be set; on the contrary, the smaller the value, the smaller the neighbor index. The denser the distribution of data around a parameter, the smaller the neighbor index needs to be set.

[0056] Optionally, a data segment consisting of parameters collected within 1 minute before the sampling moment corresponding to any parameter may be used as the local range of the any parameter.

[0057] Optionally, since the parameters in the process parameter sequence are all triples including temperature data, pressure data, and current data, the triplet can be regarded as a one-dimensional vector containing 3 elements, so that the modulus length of each parameter corresponding to the vector can be used as the value of the corresponding parameter. It should be noted that the method for determining the modulus length of the vector is a prior art and is not described in detail in this embodiment.

[0058] In another embodiment, the average difference between any parameter and the parameters in the local range may be rounded down. This embodiment does not specifically limit the selected rounding method.

[0059] In another embodiment, the difference between any parameter and the previous parameters may also be determined by other methods, such as by directly performing a difference instead of performing a square root of the square difference.

[0060] Furthermore, after determining the neighbor index of any parameter, the neighbor index of any parameter can be recorded as , then the process of obtaining the neighborhood set of the parameter is: first, determine the distance between the remaining parameters and any parameter; then, select the neighboring set of the parameter in order from small to large. parameters with a smaller distance from any of the parameters; finally, the selected The data set consisting of the parameters and the arbitrary parameter is taken as the neighborhood set of the arbitrary parameter.

[0061] S3: The neighbor index is corrected using the stability of the neighborhood set. The corrected neighbor index is negatively correlated with stability.

[0062] Among them, stability refers to the data used to evaluate the accuracy of the neighbor index. For example, when the stability of the neighborhood set of any parameter is high, it means that the accuracy of the neighbor index of the parameter is high; on the contrary, when the stability of the neighborhood set of any parameter is low, it means that the accuracy of the neighbor index of the parameter is low, and the neighbor index of the parameter needs to be corrected to make the setting of the neighbor index more in line with the distribution of actual data and task requirements.

[0063] Specifically, the stability of the neighborhood set of any parameter can be determined by the following steps:

[0064] Step 1: Calculate the average change rate of any dimension data in the neighborhood set to obtain the degree of fluctuation of the dimension data in the neighborhood set;

[0065] Specifically, for the neighborhood set of any parameter in the process parameter sequence, the fluctuation degree of any dimension data in the neighborhood set satisfies the following relationship:

[0066] ;

[0067] In the formula, Neighborhood set Neidi The degree of fluctuation of the data in each dimension; , Neighborhood Sets Neidi The dimension Data and The value of the data; is the total number of data in the neighborhood set; for the summation symbol; is the absolute value symbol.

[0068] in, Reflects the neighborhood set Neidi In the dimension The rate of change of individual data; Reflects the neighborhood set Neidi The average change rate of the dimension data. The larger the value, the higher the neighborhood set. The fluctuation of data in each dimension is relatively large.

[0069] Step 2: Use weights to obtain the weighted value of the volatility;

[0070] It should be noted that, since there may be magnitude differences between data of different dimensions within a neighborhood set, the present invention can reduce the impact of magnitude differences of data of different dimensions by calculating the weight of the data of each dimension and adjusting the importance of the data of each dimension.

[0071] Specifically, for the fluctuation degree of any dimension of data in the neighborhood set, the weight of the fluctuation degree satisfies the following relationship:

[0072] ;

[0073] In the formula, Neighborhood set Neidi The weight of the volatility of the dimensional data, Neighborhood set Neidi Dimensional data and The correlation of the data in each dimension, , Neighborhood Sets Neidi Dimensional data and The degree of fluctuation of the data in each dimension, is the normalization function.

[0074] In an exemplary embodiment of the present invention, the correlation between data of different dimensions in a neighborhood set may be determined by the following steps:

[0075] The DTW distance between data of different dimensions in the neighborhood set is obtained to obtain the correlation between the corresponding two-dimensional data.

[0076] Optionally, other methods, such as mutual information (MI), can also be used to measure the correlation between data of different dimensions in the neighborhood set. It should be noted that the process of using the dynamic time warping (DTW) algorithm to obtain the DTW distance between data to measure the correlation between the corresponding two data is a prior art and will not be described in detail in this embodiment.

[0077] Step 3: Sum the weighted differences in the volatility of data of different dimensions in the neighborhood set, and use the normalized value of the reciprocal of the cumulative sum as stability.

[0078] In an exemplary embodiment of the present invention, the stability of a neighborhood set of any parameter may be determined by the following steps:

[0079] The product of the fluctuation degree of any dimension data in the neighborhood set and the weight of the fluctuation degree of the dimension data is used as the weighted value of the fluctuation degree of the dimension data in the neighborhood set;

[0080] The stability of the neighborhood set satisfies the following relationship:

[0081] ;

[0082] In the formula, is the stability of the neighborhood set; , They are the first Dimensional data and The weighted value of the volatility of the data in each dimension; is the number of dimensions in the neighborhood set; for the summation symbol; The natural constant An exponential function with base .

[0083] in, It reflects the difference in weighted values ​​of the fluctuation degree of data of different dimensions in the neighborhood set; It reflects the cumulative sum of the weighted differences in the degree of fluctuation of data of different dimensions in the neighborhood set. The smaller the value, the smaller the difference in the degree of fluctuation of the data of each dimension in the neighborhood set, which further indicates that the data in the neighborhood set is relatively stable, and the corresponding stability of the neighborhood set is relatively high.

[0084] Furthermore, after the stability of the neighborhood set of any parameter is determined, the neighborhood index of the parameter can be corrected using the stability of the neighborhood set.

[0085] In an exemplary embodiment of the present invention, the neighbor index of any parameter may be modified by the following steps:

[0086] Get the preset hyperparameters. If the stability of the neighborhood set is greater than or equal to the hyperparameters, the neighbor indexes before and after correction are the same. If the stability of the neighborhood set is less than the hyperparameters, the corrected neighbor index satisfies the following relationship:

[0087] ;

[0088] In the formula, is the modified neighbor index; is the neighbor index before correction; is the stability of the neighborhood set; The round-up symbol.

[0089] Optionally, the preset hyperparameter can be set to 0.6. When the stability of the neighborhood set of any parameter is greater than or equal to 0.6, it means that the stability of the neighborhood set of the parameter under the neighbor index of the parameter is high, which further indicates that the accuracy of the neighbor index of the parameter is high, and there is no need to correct the neighbor index of the parameter.

[0090] Optionally, if the stability of the neighborhood set of any parameter is less than 0.6, it means that the accuracy of the neighbor index of the parameter is low, and it is necessary to increase the neighbor index of the parameter to accurately measure the local characteristics of the parameter by increasing the data basis for analyzing the local characteristics of the parameter.

[0091] S4: Based on the corrected nearest neighbor index, the outlier coefficient of any parameter is obtained using the ABOD algorithm, so as to monitor the abnormal parameter according to the comparison result between the outlier coefficient and the preset threshold.

[0092] It should be noted that the working environment and operating conditions of the plasma cutting and welding integrated equipment are usually dynamically changing. For example, in different operation stages (such as cutting, welding, adjustment, etc.), the parameters of the equipment (such as current, air pressure, temperature) will have different change patterns, resulting in the distribution and structure of the data may change over time. The basic principle of the outlier metric algorithm (Angle-Based Outlier Detection, ABOD) is to identify outliers based on the angle relationship between the data. The neighborhood structure of the data points has a great influence on the detection results, especially when the data points have different distribution characteristics in different areas. The fixed neighborhood used by the traditional ABOD algorithm when identifying outliers cannot adapt to the dynamic change characteristics of the data of the plasma cutting and welding integrated equipment, resulting in certain errors in the identification of outliers. Therefore, the present invention improves the algorithm to identify abnormal parameters in the plasma cutting and welding integrated equipment by dynamically adjusting the neighborhood index of each parameter, so as to realize data monitoring of the plasma cutting and welding integrated equipment.

[0093] Specifically, the corrected neighbor index of each parameter calculated can be used as the neighborhood size when the ABOD algorithm determines the outlier coefficient of the corresponding parameter, so that the outlier coefficient of each parameter can be obtained. Among them, the process of obtaining the outlier coefficient of data using the ABOD algorithm based on the known neighborhood is a prior art, and the present invention will not be described in detail here.

[0094] In an exemplary embodiment of the present invention, the detection of abnormal parameters can be achieved by the following steps:

[0095] A preset threshold is obtained. When the outlier coefficient of any parameter in the process parameter sequence is greater than or equal to the preset threshold, the parameter is determined to be an abnormal parameter; when the outlier coefficient of any parameter in the process parameter sequence is less than the preset threshold, the parameter is determined to be a normal parameter.

[0096] Optionally, the preset threshold can be set to 1.5. When the outlier coefficient of any parameter is greater than or equal to 1.5, the parameter is judged to be an abnormal parameter, and an alarm is issued to remind relevant personnel to take corresponding measures in a timely manner; when the outlier coefficient of any parameter is less than 1.5, the parameter is judged to be a normal parameter, thereby realizing data monitoring of the plasma cutting and welding integrated equipment.

[0097] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0098] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for real-time data monitoring of plasma cutting and welding integrated equipment, characterized in that: include: Obtaining readings of a plurality of sensors electrically connected to the device to obtain a sequence of process parameters; Based on a neighbor index of any parameter in the process parameter sequence, a neighborhood set of the parameter is obtained, wherein the neighbor index is a rounded value of an average difference between the parameter and a plurality of previous parameters; The neighbor index is corrected by using the stability of the neighborhood set, wherein the corrected neighbor index is negatively correlated with the stability, and based on the corrected neighbor index, an outlier coefficient of any parameter is obtained by using the ABOD algorithm, so as to monitor abnormal parameters according to a comparison result between the outlier coefficient and a preset threshold value; The method for obtaining the stability is: calculating the average change rate of any dimension data in the neighborhood set to obtain the degree of fluctuation of the dimension data in the neighborhood set, which satisfies the following relationship: ; Neighborhood set Neidi The degree of fluctuation of the data in each dimension; , Neighborhood Sets Neidi The dimension Data and The value of the data; is the total number of data in the neighborhood set; for the summation symbol; is the absolute value symbol; The weighted value of the volatility is obtained using the weights, wherein: , Neighborhood set Neidi The weight of the volatility of the dimensional data, Neighborhood set Neidi Dimensional data and The correlation of the data in each dimension, Neighborhood set Neidi The degree of fluctuation of the data in each dimension, is a normalization function; the difference in weighted values ​​of the fluctuation degree of data of different dimensions in the neighborhood set is summed, and the normalized value of the reciprocal of the accumulated sum is used as the stability.

2. The method for real-time data monitoring of plasma cutting and welding integrated equipment according to claim 1, characterized in that: The method for obtaining the stability of the neighborhood set includes: The product of the fluctuation degree of any dimension data in the neighborhood set and the weight of the fluctuation degree of the dimension data is used as the weighted value of the fluctuation degree of the dimension data in the neighborhood set; The stability of the neighborhood set satisfies the following relationship: ； In the formula, is the stability of the neighborhood set; , They are the first Dimensional data and The weighted value of the volatility of the data in each dimension; is the number of dimensions in the neighborhood set; for the summation symbol; The natural constant An exponential function with base .

3. The method for real-time data monitoring of plasma cutting and welding integrated equipment according to claim 2, characterized in that: When calculating the weight of the fluctuation degree of any dimensional data in the neighborhood set, the method for obtaining the correlation between different dimensional data includes: The DTW distances between the data of different dimensions in the neighborhood set are obtained to obtain the correlation between the corresponding two-dimensional data.

4. The method for real-time data monitoring of plasma cutting and welding integrated equipment according to claim 2, characterized in that: The method of modifying the neighbor index by utilizing the stability of the neighborhood set includes: Obtaining a preset hyperparameter, if the stability of the neighborhood set is greater than or equal to the hyperparameter, the neighbor indexes before and after correction are the same; If the stability of the neighborhood set is less than the hyperparameter, the modified neighbor index satisfies the following relationship: ； In the formula, is the modified neighbor index; is the neighbor index before correction; is the stability of the neighborhood set; The symbol for rounding up.

5. The method for real-time data monitoring of plasma cutting and welding integrated equipment according to claim 1, characterized in that: The method for obtaining the neighbor index includes: Each parameter in the process parameter sequence is taken as a vector, and the modulus length of each vector is calculated to obtain the value of the corresponding parameter. The data segment consisting of multiple continuous parameters before any parameter is taken as the local range of any parameter; the neighbor index of any parameter satisfies the following relationship: ； In the formula, is the first in the process parameter sequence Neighborhood index of the parameters; is the first in the process parameter sequence The value of a parameter; is the first The value of the data; is the total number of data within the local range; for the summation symbol; The symbol for rounding up.

6. The method for real-time data monitoring of plasma cutting and welding integrated equipment according to claim 1, characterized in that: The monitoring of abnormal parameters according to the comparison result between the outlier coefficient and the preset threshold value includes: Obtaining a preset threshold value, and when the outlier coefficient of any parameter in the process parameter sequence is greater than or equal to the preset threshold value, determining the parameter as an abnormal parameter; When the outlier coefficient of any parameter in the process parameter sequence is less than the preset threshold, the parameter is determined to be a normal parameter.

7. The method for real-time data monitoring of plasma cutting and welding integrated equipment according to claim 1, characterized in that: Each parameter in the process parameter sequence includes data of multiple dimensions, and each dimension corresponds to a type of sensor one by one.

Images