Working condition analysis method and system of numerical control machining system

Abstract

The invention discloses a working condition analysis method and system for a numerical control machining system, and relates to the technical field of intelligent analysis, and the method comprises the following steps: constructing a working state parameter data set and a confidence interval set of working state parameters, marking the working state parameter data, calculating the probability corresponding to each fault when the fault occurs, and calculating the confidence interval set of the working state parameters; and fitting the probability set to obtain a final probability, calculating the probability of occurrence of a correlation value of the machine tool in the first time period, training a machine learning model used for feeding back a fault type, and outputting a machine tool state prediction result in combination with the final probability. Working state parameters are marked as correlation values and ircorrelation values, so that data classification is clearer, different fault type occurrence probabilities are fitted through multiple regression analysis to obtain a final probability, a probability result is more accurate, the fault type is predicted by training a machine learning model, and the fault classification accuracy is improved. Therefore, the analysis of the working condition is more advanced, and the resource waste caused by wrong production is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent analysis, and in particular to a working condition analysis method and system for a numerical control machining system. Background Art

[0002] In recent years, the technical level of the CNC system industry has been steadily improving. High precision, high efficiency and high reliability have become the important development direction of CNC systems. The intelligence level of CNC systems has been continuously improved, with stronger autonomous learning and adaptive capabilities, which can better meet complex and changing processing needs.

[0003] At present, a Chinese invention patent with publication number CN 103197609A discloses a dynamic feature modeling method for CNC machining. The method models dynamic features based on FunctionBlock, constructs feature unctionBlock and service FunctionBlock, and realizes the expression of intermediate states of machining features, processes and working parameters during machining, dynamic safety domain values ​​of parameters, and adjustment strategies exceeding the domain values. At the same time, it also realizes the acquisition and analysis of real-time information during machining. However, the relevant technology does not make a detailed classification of machine tool fault types, lacks comprehensiveness of analysis, does not predict machine tool faults, lacks the forward-looking nature of analysis, and is prone to erroneous production, resulting in waste of resources. Summary of the invention

[0004] The technical problem solved by the present invention is that the related technology does not classify the fault types of machine tools in detail, lacks comprehensive analysis, does not predict machine tool faults, lacks forward-looking analysis, and easily causes erroneous production, thereby resulting in waste of resources.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions: In the first aspect, a method for analyzing the working condition of a numerical control machining system comprises the following steps: Step S1, numbering the machine tools, obtaining the working state parameters of the machine tools based on the time series, and preprocessing the working state parameters to construct a working state parameter data set; Step S2, respectively obtaining a maximum threshold value of the working state parameter and a minimum threshold value of the working state parameter, and constructing a confidence interval set of the working state parameter; Step S3, comparing the preprocessed working state parameter with the confidence interval of the working state parameter, marking the preprocessed working state parameter data belonging to the confidence interval of the working state parameter as a relevant value, and marking the preprocessed working state parameter data not belonging to the confidence interval of the working state parameter as an irrelevant value; Step S4, obtaining irrelevant values, calling a fault type database, obtaining fault types corresponding to irrelevant values, counting the number of occurrences of each fault type in the first time period, calculating the total number of occurrences, calculating the probability of each fault corresponding to the fault when the fault occurs, dividing the first time period several times, obtaining a probability set corresponding to each fault when the fault occurs, fitting the probability set, and obtaining a final probability; Step S5, obtaining relevant values ​​and irrelevant values ​​in the first time period, sorting the relevant values ​​and the irrelevant values ​​in descending order according to the time series, and calculating the probability of the relevant values ​​appearing in the machine tool in the first time period; Step S6, training a machine learning model for feedback of fault types, wherein the machine learning model uses the fault type as output, and repeatedly tests and trains the machine learning model until the accuracy reaches a first threshold, and outputs a machine tool state prediction result in combination with the final probability.

[0006] As a preferred solution of the working condition analysis method of a numerical control machining system described in the present invention, the step S1 includes the following sub-steps: Step S11, numbering the machine tools, wherein the numbers are natural numbers; Step S12, acquiring the working state parameters of the machine tool based on the time series, and performing abnormal data elimination and normalization processing on the working state parameters, so that the working state parameters are greater than or equal to 0 and less than or equal to 1, and constructing a working state parameter data set, wherein the working state parameters include spindle speed, spindle temperature, feed current, feed speed, spindle load value and feed shaft load value.

[0007] As a preferred solution of the working condition analysis method of a numerical control machining system described in the present invention, step S2 includes the following sub-steps: Step S21, calling the machine tool database to obtain the maximum threshold value and the minimum threshold value of the working state parameter respectively; Step S22, taking the minimum threshold of the working state parameter as the lower limit of the interval and the maximum threshold of the working state parameter as the upper limit of the interval, construct a confidence interval set of the working state parameter, wherein the confidence interval set includes a spindle speed confidence interval, a spindle temperature confidence interval, a feed current confidence interval, a feed speed confidence interval, a spindle load value confidence interval, and a feed shaft load value confidence interval.

[0008] As a preferred solution of the working condition analysis method of a numerical control machining system described in the present invention, step S3 includes the following sub-steps: Step S31, respectively obtaining the preprocessed spindle speed, the preprocessed spindle temperature, the preprocessed feed current, the preprocessed feed speed, the preprocessed spindle load value and the preprocessed feed shaft load value; Step S32, comparing the preprocessed spindle speed with the spindle speed confidence interval, marking the preprocessed spindle speed that belongs to the spindle speed confidence interval as a first relevant value, and marking the preprocessed spindle speed that does not belong to the spindle speed confidence interval as a first irrelevant value; Step S33, loop step S32, respectively mark the second related value, third related value, fourth related value, fifth related value and sixth related value corresponding to the pretreated spindle temperature, pretreated feed current, pretreated feed speed, pretreated spindle load value and pretreated feed shaft load value, and respectively mark the second unrelated value, third unrelated value, fourth unrelated value, fifth unrelated value and sixth unrelated value corresponding to the pretreated spindle temperature, pretreated feed current, pretreated feed speed, pretreated spindle load value and pretreated feed shaft load value.

[0009] As a preferred solution of the working condition analysis method of a numerical control machining system described in the present invention, step S4 includes the following sub-steps: Step S41, calling the fault type database, inputting the first irrelevant value, and obtaining the corresponding fault type, wherein the fault type is spindle head wear; Step S42, looping step S41, respectively obtaining the fault types corresponding to the second irrelevant value, the third irrelevant value, the fourth irrelevant value, the fifth irrelevant value and the sixth irrelevant value, wherein the fault types are respectively bearing damage, feed shaft wear, servo system failure, spindle rotation heating and driver circuit damage; Step S43, counting the number of times each fault type occurs in the first time period, calculating the total number of times, respectively calculating the proportion of the number of times each fault type occurs in the total number of times, calculating the probability of each fault corresponding to the fault when the fault occurs, and dividing the first time period several times to obtain a probability set corresponding to each fault when the fault occurs, wherein the probability set includes the first probability to the Nth probability.

[0010] As a preferred solution of the working condition analysis method of a numerical control machining system described in the present invention, the step S43 includes the following sub-steps: Step S431, using the first threshold as the initial segmentation step length, and increasing the segmentation step length by the first value, to divide the first time period into a plurality of sub-time periods; Step S432, calculating the probability of each fault occurring during the first time period, recorded as a first probability; Step S433, repeat step S432, calculate the second probability to the Nth probability, and fit the first probability to the Nth probability into the final probability through a multiple regression algorithm.

[0011] As a preferred solution of the working condition analysis method of a numerical control machining system described in the present invention, step S5 includes the following sub-steps: Step S51, obtaining relevant values ​​and irrelevant values ​​within a first time period, and sorting the relevant values ​​and irrelevant values ​​within the first time period in descending order according to the time series; Step S52, calculating the probability of the relevant value appearing when the machine tool is running, the probability of the relevant value appearing in the machine tool within the first time period represents the probability that the machine tool is in a non-faulty running state, and the calculation expression of the probability of the relevant value appearing is:

[0012] in, is the probability of the relevant value appearing, is the number of relevant values ​​in the i-th working state parameter data in the preprocessed working state parameter data, is the number of irrelevant values ​​in the i-th working state parameter data in the preprocessed working state parameter data, .

[0013] As a preferred solution of the working condition analysis method of a numerical control machining system described in the present invention, step S6 includes the following sub-steps: Step S61, dividing the working state parameter data set into a training set and a test set according to a certain ratio; Step S62, taking the time point as input and the predicted fault type as output, training a machine learning model for feedback of the fault type, and repeatedly testing and training the machine learning model until the accuracy reaches a first threshold; Step S63: The fault type and the final probability corresponding to the fault type are used as the output result of the machine tool status.

[0014] In a second aspect, a working condition analysis system of a numerical control machining system includes an acquisition module, an analysis module, a prediction module and a terminal module; The acquisition module is used to acquire the working state parameters of the machine tool, pre-process the working state parameters, and transmit the pre-processed working state parameters to the analysis module and the prediction module; The analysis module is used to construct a confidence interval set of working state parameters, calculate the final probability and the probability of the machine tool having related values ​​in the first time period, the final probability being the probability of each fault type occurring, and transmit the final probability to the terminal module; The prediction module is used to predict the fault type of the machine tool and transmit the fault type of the machine tool to the terminal module; The terminal module is used to display the final probability and failure type.

[0015] In a third aspect, the present invention provides an electronic device comprising a memory, a processor and storage in the memory, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in any one of the methods described above are executed.

[0016] In a fourth aspect, the present invention provides a storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the steps in any one of the methods described above are executed.

[0017] The beneficial effects of the present invention are as follows: by constructing confidence intervals, the preprocessed working status parameters are marked as relevant values ​​and irrelevant values, so that the classification of data is clearer, the first time period is divided multiple times to obtain a probability set of different fault types, and the probabilities of occurrence of different fault types are fitted through multivariate regression analysis to obtain the final probability, so that the probability result is more accurate, and the fault type is predicted by training a machine learning model, so that the analysis of the working condition is more advanced, which is conducive to scientific production and reduces the waste of resources caused by erroneous production. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A basic flow chart of a method for analyzing a working condition of a numerical control machining system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, but not all of the embodiments.

[0020] Example 1, reference Figure 1 , as an embodiment of the present invention, provides a method for analyzing working conditions of a numerical control machining system, comprising the following steps: Step S1, numbering the machine tools, obtaining the working state parameters of the machine tools based on the time series, and preprocessing the working state parameters to construct a working state parameter data set; Step S2, respectively obtaining a maximum threshold value of the working state parameter and a minimum threshold value of the working state parameter, and constructing a confidence interval set of the working state parameter; Step S3, comparing the preprocessed working state parameter with the confidence interval of the working state parameter, marking the preprocessed working state parameter data belonging to the confidence interval of the working state parameter as a relevant value, and marking the preprocessed working state parameter data not belonging to the confidence interval of the working state parameter as an irrelevant value; Step S4, obtaining irrelevant values, calling a fault type database, obtaining fault types corresponding to irrelevant values, counting the number of occurrences of each fault type in the first time period, calculating the total number of occurrences, calculating the probability of each fault corresponding to the fault when the fault occurs, dividing the first time period several times, obtaining a probability set corresponding to each fault when the fault occurs, fitting the probability set, and obtaining a final probability; Step S5, obtaining relevant values ​​and irrelevant values ​​in the first time period, sorting the relevant values ​​and the irrelevant values ​​in descending order according to the time series, and calculating the probability of the relevant values ​​appearing in the machine tool in the first time period; Step S6, training a machine learning model for feedback of fault types, wherein the machine learning model uses the fault type as output, and repeatedly tests and trains the machine learning model until the accuracy reaches a first threshold, and outputs a machine tool state prediction result in combination with the final probability.

[0021] The present invention marks the preprocessed working state parameters as relevant values ​​and irrelevant values ​​by constructing confidence intervals, so that the data classification is clearer, and the first time period is divided multiple times to obtain a probability set of different fault types. The occurrence probabilities of different fault types are fitted through multivariate regression analysis to obtain the final probability, so that the probability result is more accurate. By training a machine learning model to predict the fault type, the analysis of the working condition is more advanced, which is conducive to scientific production and reduces the waste of resources caused by erroneous production.

[0022] The step S1 includes the following sub-steps: Step S11, numbering the machine tools, wherein the numbers are natural numbers; Step S12, acquiring the working state parameters of the machine tool based on the time series, and performing abnormal data elimination and normalization processing on the working state parameters, so that the working state parameters are greater than or equal to 0 and less than or equal to 1, and constructing a working state parameter data set, wherein the working state parameters include spindle speed, spindle temperature, feed current, feed speed, spindle load value and feed shaft load value.

[0023] In the specific implementation, abnormal data removal is based on the Z-score algorithm, which assumes that the conveyor belt speed data and the horizontal conveyor belt inclination angle data obey the Gaussian distribution, and the abnormal data are the data points distributed at the tail far away from the average value of the data.

[0024] The step S2 includes the following sub-steps: Step S21, calling the machine tool database to obtain the maximum threshold value and the minimum threshold value of the working state parameter respectively; Step S22, taking the minimum threshold of the working state parameter as the lower limit of the interval and the maximum threshold of the working state parameter as the upper limit of the interval, construct a confidence interval set of the working state parameter, wherein the confidence interval set includes a spindle speed confidence interval, a spindle temperature confidence interval, a feed current confidence interval, a feed speed confidence interval, a spindle load value confidence interval, and a feed shaft load value confidence interval.

[0025] The step S3 includes the following sub-steps: Step S31, respectively obtaining the preprocessed spindle speed, the preprocessed spindle temperature, the preprocessed feed current, the preprocessed feed speed, the preprocessed spindle load value and the preprocessed feed shaft load value; Step S32, comparing the preprocessed spindle speed with the spindle speed confidence interval, marking the preprocessed spindle speed that belongs to the spindle speed confidence interval as a first relevant value, and marking the preprocessed spindle speed that does not belong to the spindle speed confidence interval as a first irrelevant value; Step S33, loop step S32, respectively mark the second related value, third related value, fourth related value, fifth related value and sixth related value corresponding to the pretreated spindle temperature, pretreated feed current, pretreated feed speed, pretreated spindle load value and pretreated feed shaft load value, and respectively mark the second unrelated value, third unrelated value, fourth unrelated value, fifth unrelated value and sixth unrelated value corresponding to the pretreated spindle temperature, pretreated feed current, pretreated feed speed, pretreated spindle load value and pretreated feed shaft load value.

[0026] In a specific implementation, the preprocessed working status parameters are marked as relevant values ​​and irrelevant values ​​by constructing confidence intervals, so that the classification of data is clearer.

[0027] The step S4 includes the following sub-steps: Step S41, calling the fault type database, inputting the first irrelevant value, and obtaining the corresponding fault type, wherein the fault type is spindle head wear; Step S42, looping step S41, respectively obtaining the fault types corresponding to the second irrelevant value, the third irrelevant value, the fourth irrelevant value, the fifth irrelevant value and the sixth irrelevant value, wherein the fault types are respectively bearing damage, feed shaft wear, servo system failure, spindle rotation heating and driver circuit damage; Step S43, counting the number of times each fault type occurs in the first time period, calculating the total number of times, respectively calculating the proportion of the number of times each fault type occurs in the total number of times, calculating the probability of each fault corresponding to the fault when the fault occurs, and dividing the first time period several times to obtain a probability set corresponding to each fault when the fault occurs, wherein the probability set includes the first probability to the Nth probability.

[0028] The step S43 includes the following sub-steps: Step S431, using the first threshold as the initial segmentation step length, and increasing the segmentation step length by the first value, to divide the first time period into a plurality of sub-time periods; Step S432, calculating the probability of each fault occurring during the first time period, recorded as a first probability; Step S433, repeat step S432, calculate the second probability to the Nth probability, and fit the first probability to the Nth probability into the final probability through a multiple regression algorithm.

[0029] In a specific implementation, the first time period is divided multiple times to obtain probability sets of different fault types, and the occurrence probabilities of different fault types are fitted through multivariate regression analysis to obtain the final probability, so that the probability result is more accurate.

[0030] The step S5 includes the following sub-steps: Step S51, obtaining relevant values ​​and irrelevant values ​​within a first time period, and sorting the relevant values ​​and irrelevant values ​​within the first time period in descending order according to the time series; Step S52, calculating the probability of the relevant value appearing when the machine tool is running, the probability of the relevant value appearing in the machine tool within the first time period represents the probability that the machine tool is in a non-faulty running state, and the calculation expression of the probability of the relevant value appearing is:

[0031] in, is the probability of the relevant value appearing, is the number of relevant values ​​in the i-th working state parameter data in the preprocessed working state parameter data, is the number of irrelevant values ​​in the i-th working state parameter data in the preprocessed working state parameter data, .

[0032] In a specific implementation, the probability of the relevant value appearing in the machine tool within the first time period represents the probability that the machine tool operating state is not faulty. The present invention divides the machine tool operating conditions into two situations: faulty and non-faulty, and predicts the fault type and the corresponding probability of the fault type for the fault situation.

[0033] The step S6 comprises the following sub-steps: Step S61, dividing the working state parameter data set into a training set and a test set according to a certain ratio; Step S62, taking the time point as input and the predicted fault type as output, training a machine learning model for feedback of the fault type, and repeatedly testing and training the machine learning model until the accuracy reaches a first threshold; Step S63: The fault type and the final probability corresponding to the fault type are used as the output result of the machine tool status.

[0034] A working condition analysis system for a numerical control machining system, comprising a collection module, an analysis module, a prediction module and a terminal module; The acquisition module is used to acquire the working state parameters of the machine tool, pre-process the working state parameters, and transmit the pre-processed working state parameters to the analysis module and the prediction module; The analysis module is used to construct a confidence interval set of working state parameters, calculate the final probability and the probability of the machine tool having related values ​​in the first time period, the final probability being the probability of each fault type occurring, and transmit the final probability to the terminal module; The prediction module is used to predict the fault type of the machine tool and transmit the fault type of the machine tool to the terminal module; The terminal module is used to display the final probability and failure type.

[0035] The present invention classifies the fault types of machine tools in detail, improves the comprehensiveness of analysis, predicts machine tool faults, improves the proactiveness of analysis, reduces situations that cause erroneous production, and thus reduces waste of resources.

[0036] It should be appreciated that embodiments of the present invention may be implemented or enforced by a combination of computer hardware and software, or by computer instructions stored in a non-transitory computer-readable memory. The method may be implemented in a computer program using standard programming techniques - including a non-transitory computer-readable storage medium configured with a computer program, wherein the storage medium so configured causes the computer to operate in a specific and predefined manner - according to the methods and drawings described in the specific embodiments. Each program may be implemented in a high-level procedural or object-oriented programming language to communicate with a computer system. However, if desired, the program may be implemented in an assembly or machine language. In any case, the language may be a compiled or interpreted language. In addition, the program may be run on a programmed ASIC for this purpose.

[0037] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for analyzing working conditions of a numerical control machining system, characterized in that: The following steps are involved: Step S1, numbering the machine tools, obtaining the working state parameters of the machine tools based on the time series, and preprocessing the working state parameters to construct a working state parameter data set; Step S2, respectively obtaining a maximum threshold value of the working state parameter and a minimum threshold value of the working state parameter, and constructing a confidence interval set of the working state parameter; Step S3, comparing the preprocessed working state parameter with the confidence interval of the working state parameter, marking the preprocessed working state parameter data belonging to the confidence interval of the working state parameter as a relevant value, and marking the preprocessed working state parameter data not belonging to the confidence interval of the working state parameter as an irrelevant value; Step S4, obtaining irrelevant values, calling a fault type database, obtaining fault types corresponding to irrelevant values, counting the number of occurrences of each fault type in the first time period, calculating the total number of occurrences, calculating the probability of each fault corresponding to the fault when the fault occurs, dividing the first time period several times, obtaining a probability set corresponding to each fault when the fault occurs, fitting the probability set, and obtaining a final probability; Step S5, obtaining relevant values ​​and irrelevant values ​​in the first time period, sorting the relevant values ​​and the irrelevant values ​​in descending order according to the time series, and calculating the probability of the relevant values ​​appearing in the machine tool in the first time period; Step S6, training a machine learning model for feedback of fault types, wherein the machine learning model uses the fault type as output, and repeatedly tests and trains the machine learning model until the accuracy reaches a first threshold, and outputs a machine tool state prediction result in combination with the final probability.

2. The method for analyzing the working condition of a numerical control machining system according to claim 1, characterized in that: The step S1 includes the following sub-steps: Step S11, numbering the machine tools, wherein the numbers are natural numbers; Step S12, acquiring the working state parameters of the machine tool based on the time series, and performing abnormal data elimination and normalization processing on the working state parameters, so that the working state parameters are greater than or equal to 0 and less than or equal to 1, and constructing a working state parameter data set, wherein the working state parameters include spindle speed, spindle temperature, feed current, feed speed, spindle load value and feed shaft load value.

3. The method for analyzing the working condition of a numerical control machining system according to claim 1, characterized in that: The step S2 includes the following sub-steps: Step S21, calling the machine tool database to obtain the maximum threshold value and the minimum threshold value of the working state parameter respectively; Step S22, taking the minimum threshold of the working state parameter as the lower limit of the interval and the maximum threshold of the working state parameter as the upper limit of the interval, construct a confidence interval set of the working state parameter, wherein the confidence interval set includes a spindle speed confidence interval, a spindle temperature confidence interval, a feed current confidence interval, a feed speed confidence interval, a spindle load value confidence interval, and a feed shaft load value confidence interval.

4. The method for analyzing working conditions of a numerical control machining system according to claim 1, characterized in that: The step S3 includes the following sub-steps: Step S31, respectively obtaining the preprocessed spindle speed, the preprocessed spindle temperature, the preprocessed feed current, the preprocessed feed speed, the preprocessed spindle load value and the preprocessed feed shaft load value; Step S32, comparing the preprocessed spindle speed with the spindle speed confidence interval, marking the preprocessed spindle speed that belongs to the spindle speed confidence interval as a first relevant value, and marking the preprocessed spindle speed that does not belong to the spindle speed confidence interval as a first irrelevant value; Step S33, loop step S32, respectively mark the second related value, third related value, fourth related value, fifth related value and sixth related value corresponding to the pretreated spindle temperature, pretreated feed current, pretreated feed speed, pretreated spindle load value and pretreated feed shaft load value, and respectively mark the second unrelated value, third unrelated value, fourth unrelated value, fifth unrelated value and sixth unrelated value corresponding to the pretreated spindle temperature, pretreated feed current, pretreated feed speed, pretreated spindle load value and pretreated feed shaft load value.

5. The method for analyzing working conditions of a numerical control machining system according to claim 1, characterized in that: The step S4 includes the following sub-steps: Step S41, calling the fault type database, inputting the first irrelevant value, and obtaining the corresponding fault type, wherein the fault type is spindle head wear; Step S42, looping step S41, respectively obtaining the fault types corresponding to the second irrelevant value, the third irrelevant value, the fourth irrelevant value, the fifth irrelevant value and the sixth irrelevant value, wherein the fault types are respectively bearing damage, feed shaft wear, servo system failure, spindle rotation heating and driver circuit damage; Step S43, counting the number of times each fault type occurs in the first time period, calculating the total number of times, respectively calculating the proportion of the number of times each fault type occurs in the total number of times, calculating the probability of each fault corresponding to the fault when the fault occurs, and dividing the first time period several times to obtain a probability set corresponding to each fault when the fault occurs, wherein the probability set includes the first probability to the Nth probability.

6. A method for analyzing working conditions of a numerical control machining system according to claim 5, characterized in that: The step S43 includes the following sub-steps: Step S431, using the first threshold as the initial segmentation step length, and increasing the segmentation step length by the first value, to divide the first time period into a plurality of sub-time periods; Step S432, calculating the probability of each fault occurring during the first time period, recorded as a first probability; Step S433, repeat step S432, calculate the second probability to the Nth probability, and fit the first probability to the Nth probability into the final probability through a multiple regression algorithm.

7. The method for analyzing working conditions of a numerical control machining system according to claim 1, characterized in that: The step S5 comprises the following sub-steps: Step S51, obtaining relevant values ​​and irrelevant values ​​within a first time period, and sorting the relevant values ​​and irrelevant values ​​within the first time period in descending order according to the time series; Step S52, calculating the probability of the relevant value appearing when the machine tool is running, the probability of the relevant value appearing in the machine tool within the first time period represents the probability that the machine tool is in a non-faulty running state, and the calculation expression of the probability of the relevant value appearing is: ； in, is the probability of the relevant value appearing, is the number of relevant values ​​in the i-th working state parameter data in the preprocessed working state parameter data, is the number of irrelevant values ​​in the i-th working state parameter data in the preprocessed working state parameter data, ; The step S6 includes the following sub-steps: Step S61, dividing the working state parameter data set into a training set and a test set according to a certain ratio; Step S62, using the time point as an input and the predicted fault type as an output, training a machine learning model for feedback of the fault type, and repeatedly testing and training the machine learning model until the accuracy reaches a first threshold; Step S63: The fault type and the final probability corresponding to the fault type are used as the output result of the machine tool status.

8. A working condition analysis system for a numerical control machining system, characterized in that: It includes acquisition module, analysis module, prediction module and terminal module; The acquisition module is used to acquire the working state parameters of the machine tool, pre-process the working state parameters, and transmit the pre-processed working state parameters to the analysis module and the prediction module; The analysis module is used to construct a confidence interval set of working state parameters, calculate the final probability and the probability of the machine tool having related values ​​in the first time period, the final probability being the probability of each fault type occurring, and transmit the final probability to the terminal module; The prediction module is used to predict the fault type of the machine tool and transmit the fault type of the machine tool to the terminal module; The terminal module is used to display the final probability and failure type.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the method for analyzing the working condition of a numerical control machining system according to any one of claims 1 to 8 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for analyzing the working condition of a numerical control machining system according to any one of claims 1 to 8 is implemented.

Images