Fault analysis method and system based on production equipment data

Abstract

The invention relates to the technical field of equipment safety, in particular to a fault analysis method and system based on production equipment data, and the method comprises the following steps: obtaining a real-time equipment operation state data set of target production equipment, and obtaining a fault analysis result based on the real-time equipment operation state data set of the target production equipment; determining fault correlation degree information of the operation state of the equipment; the target fault degree of the target production equipment is obtained based on the real-time equipment operation state data of the target production equipment and the fault correlation degree information of the equipment operation state, so that the action of aligning the time of different equipment operation state data is avoided, fitting data is not required to be generated by time alignment, the data accuracy is further improved, and the production efficiency is improved. And the accuracy of equipment fault analysis is further improved.

Description

Technical Field

[0001] The present invention relates to the field of equipment safety technology, and in particular to a fault analysis method and system based on production equipment data. Background Art

[0002] Currently, most domestic manufacturing companies still rely on manual oversight of their production systems and equipment. This is especially true for equipment repairs, which rely on post-fault repairs. This not only fails to identify potential faults but also hinders timely repairs. This leads to low efficiency and significant costs, severely impacting the company's production profitability. Prognostic and Health Management (PHM) systems are currently effective tools for early warning and diagnosis of equipment failures. PHM systems can utilize a variety of deployment strategies, including centralized, decentralized, and hierarchical, using algorithms and models to predict and manage system health.

[0003] Patent CN112462734A in the prior art discloses a method and model for predicting and analyzing faults in industrial production equipment, including the following steps: classifying equipment and collecting operational data; collecting the failure rate, fault category, and fault cause within each cycle for general equipment; collecting real-time operational data for key equipment; and performing equipment fault prediction and analysis. This invention relates to the field of industrial production equipment technology. The method and model for predicting and analyzing faults in industrial production equipment collects data from the equipment end, then transmits the data to a fault analysis system, analyzes the cause of the fault through the fault analysis system, and then receives fault warning information through a fault warning unit and sends an alarm. It can be seen that while the above solution considers classifying equipment, it does not consider the time alignment of the operational status data of different equipment. Time alignment requires the generation of fitted data, which is often inaccurate and affects the accuracy of equipment fault analysis. Summary of the Invention

[0004] The present invention provides a method and system for fault analysis based on production equipment data, the method comprising the following steps:

[0005] S100, obtaining a real-time device operating status data set A=(A1, A2, A3) of a target production device, where A1 is a real-time temperature set of the target production device, A2 is a real-time voltage set of the target production device, and A3 is a real-time current set of the target production device.

[0006] S200, based on A, determine the fault correlation information F=(F1, F2, F3) of the equipment operating status, where F1 is the fault correlation between the real-time time corresponding to A1 and the real-time voltage corresponding to A2, F2 is the fault correlation between the real-time voltage corresponding to A2 and the real-time current corresponding to A3, and F3 is the fault correlation between the real-time current corresponding to A3 and the real-time temperature corresponding to A1.

[0007] S300 , based on A and F, obtaining a target failure degree U of the target production equipment for use in failure analysis of the target equipment.

[0008] The present invention also protects a fault analysis system based on production equipment data, the system comprising:

[0009] The first execution module is used to obtain a real-time device operation status data set A=(A1, A2, A3) of the target production equipment, where A1 is the real-time temperature set of the target production equipment, A2 is the real-time voltage set of the target production equipment, and A3 is the real-time current set of the target production equipment.

[0010] The second execution module is used to determine the fault correlation information F=(F1, F2, F3) of the equipment operation status based on A, where F1 is the fault correlation between the real-time time corresponding to A1 and the real-time voltage corresponding to A2, F2 is the fault correlation between the real-time voltage corresponding to A2 and the real-time current corresponding to A3, and F3 is the fault correlation between the real-time current corresponding to A3 and the real-time temperature corresponding to A1.

[0011] The third execution module is configured to obtain a target failure degree U of the target production equipment based on A and F, so as to be used for failure analysis of the target equipment.

[0012] Compared with the prior art, the present invention has at least the following beneficial effects:

[0013] A fault analysis method based on production equipment data of the present invention comprises the following steps: S100, obtaining a real-time equipment operating status data set A=(A1, A2, A3) of a target production equipment, where A1 is a real-time temperature set of the target production equipment, A2 is a real-time voltage set of the target production equipment, and A3 is a real-time current set of the target production equipment; S200, determining fault correlation information F=(F1, F2, F3) of the equipment operating status based on A, where F1 is a fault correlation between a real-time time corresponding to A1 and a real-time voltage corresponding to A2, F2 is a fault correlation between a real-time voltage corresponding to A2 and a real-time current corresponding to A3, and F3 is a fault correlation between a real-time current corresponding to A3 and a real-time temperature corresponding to A1; S300, obtaining a target fault degree U of the target production equipment based on A and F. It can be seen that the time alignment of operating status data of different equipment is avoided, and no time alignment is required to generate fitting data, thereby improving the accuracy of the data and further improving the accuracy of equipment fault analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0015] Figure 1 A flowchart of a fault analysis method based on production equipment data provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 making any creative efforts shall fall within the scope of protection of the present invention.

[0017] Example 1

[0018] like Figure 1 As shown, this embodiment provides a fault analysis method based on production equipment data, the method comprising the following steps:

[0019] S100, obtain the real-time equipment operation status data set A=(A1, A2, A3) of the target production equipment, A1 is the real-time temperature set of the target production equipment, A2 is the real-time voltage set of the target production equipment, and A3 is the real-time current set of the target production equipment. Further understanding: A1=(A 11,……,A 1x ,……,A 1p ), A 1x is the real-time temperature of the xth target production equipment, the value range of x is 1 to p, and p is the real-time temperature number of the target production equipment; A2= (A 21 ,……,A 2y ,……,A 2q ), A 2y is the real-time voltage of the yth target production equipment, the value range of y is 1 to q, q is the real-time voltage quantity of the target production equipment; A3= (A 31 ,……,A 3z ,……,A 3g ), A 3z is the real-time current of the zth target production equipment, the value range of z is 1 to g, and g is the real-time current quantity of the target production equipment.

[0020] S200, based on A, determine the fault correlation information F=(F1, F2, F3) of the equipment operating status, where F1 is the fault correlation between the real-time time corresponding to A1 and the real-time voltage corresponding to A2, F2 is the fault correlation between the real-time voltage corresponding to A2 and the real-time current corresponding to A3, and F3 is the fault correlation between the real-time current corresponding to A3 and the real-time temperature corresponding to A1.

[0021] Specifically, step S200 also includes the following steps:

[0022] S201, pre-process A to obtain the key equipment operation status data set A 0 , where A 0 =(A 0 1, A 0 2, A 0 3), A 0 1 is the critical temperature set of the target production equipment, A 0 2 is the key voltage set of the target production equipment, A 0 3 is the key current set of the target production equipment; it is further understood as: preprocessing the real-time equipment operation status data to obtain the key operation status data, wherein the preprocessing includes: data missing value processing, data deduplication processing, data outlier processing, etc. Those skilled in the art are aware of any data missing value processing, data deduplication processing, data outlier processing and other methods in the prior art, which will not be repeated here.

[0023] S202, A 01 is input into the first production equipment fault analysis model to obtain the first fault type set K1 corresponding to the target production equipment; the first production equipment fault analysis model is an LSTM model trained by the sample temperature data set, wherein the sample temperature data set includes a plurality of sample temperature data, each sample temperature data includes: the degree of change of the sample temperature, the mean value of the sample temperature, the change trend of the sample temperature, the peak value of the sample temperature, the valley value of the sample temperature and a plurality of sample production equipment fault types and the occurrence probability value of each production equipment fault type, for example, any sample temperature data B={B 0 1, B 0 2, B 0 3, B 0 4, B 0 5, B1, ..., B i ,……,B m}, B 0 1 is the degree of change in sample temperature, B 0 2 is the mean temperature of the sample, B 0 3 is the changing trend of sample temperature, B 0 4 is the peak temperature of the sample, B 0 5 is the valley value of the sample temperature, B i It refers to the i-th production equipment failure type and its occurrence probability value corresponding to the sample temperature data. The value range of i is 1 to m, and m is the number of production equipment failure types generated based on temperature. Those skilled in the art know any method for training LSTM models and will not repeat them here.

[0024] Preferably, B 0 1Meet the following conditions:

[0025] , where D t It is the real-time temperature of the tth sample in a certain collected sample real-time temperature data table. The value range of t is 1 to v, and v is a certain collected sample real-time temperature data table.

[0026] Preferably, B 0 2Meet the following conditions:

[0027] ;

[0028] Preferably, B 0 3. Meet the following conditions:

[0029] , where △TD is the time interval for collecting a sample real-time temperature data table; when t=v, D t+1 -D t =0.

[0030] Preferably, B 0 4 is the peak value of the real-time temperature of all samples in a certain collected sample real-time temperature data table. Those skilled in the art know any method for obtaining the peak value of temperature data, which will not be described here.

[0031] Preferably, B 0 5 is the valley value of the real-time temperature of all samples in a certain collected sample real-time temperature data table. Those skilled in the art know any method for obtaining the valley value of temperature data, which will not be described here.

[0032] Preferably, in step S202, A 0 1 is processed to obtain A 0 1 corresponds to the degree of change, A 0 1 corresponding to the mean, A 0 1 corresponding to the change trend, A 0 1 corresponding to the peak value, A 0 1 corresponds to the valley value, A 0 1 corresponds to the degree of change, A 0 1 corresponding to the mean, A 0 1 corresponding to the change trend, A 0 1 corresponding to the peak value, A 0 1 is input into the first production equipment fault analysis model to obtain the first fault type set K1 corresponding to the target production equipment, that is, K1={K 11 ,……,K 1a ,……,K 1b}, K 1a is the ath first fault type and its probability corresponding to the target production equipment. The value range of a is 1 to b, and b is the number of first fault types corresponding to the target production equipment.

[0033] S203, A 0 2 is input into the second production equipment fault analysis model to obtain the second fault type set K2 corresponding to the target production equipment; the second production equipment fault analysis model is an LSTM model trained by the sample voltage data set, wherein the sample voltage data set includes a plurality of sample voltage data, each sample voltage data includes: the degree of change of the sample voltage, the mean of the sample voltage, the change trend of the sample voltage, the peak value of the sample voltage, the valley value of the sample voltage and a plurality of sample production equipment fault types and the occurrence probability value of each production equipment fault type, for example, any sample voltage data C={C 0 1, C 0 2, C 0 3. C 0 4, C 0 5, C1, ..., C j ,……,Cn}, C 0 1 is the degree of change of the sample voltage, C 0 2 is the mean value of the sample voltage, C 0 3 is the variation trend of the sample voltage, C 0 4 is the peak value of the sample voltage, C 0 5 is the valley value of the sample voltage, C j It refers to the j-th production equipment failure type and its occurrence probability value corresponding to the sample voltage data. The value range of j is 1 to n, and n is the number of production equipment failure types generated based on voltage. Those skilled in the art know any method for training LSTM models and will not repeat them here.

[0034] Preferably, C 0 1Meet the following conditions:

[0035] , where Q g It is the g-th sample real-time voltage in a certain collected sample real-time voltage data table. The value range of g is 1 to s, and s is a certain collected sample real-time voltage data table.

[0036] Preferably, C 0 2Meet the following conditions:

[0037] ;

[0038] Preferably, C 0 3. Meet the following conditions:

[0039] , where △TH is the acquisition time interval of a sample real-time voltage data table; when g=s, Q g+1 -Q g =0.

[0040] Preferably, C 0 4 is the peak value of all sample real-time voltages in a certain collected sample real-time voltage data table. Those skilled in the art know any method for obtaining the peak value of voltage data, which will not be described here in detail.

[0041] Preferably, C 0 5 is the valley value of all sample real-time voltages in a certain collected sample real-time voltage data table. Those skilled in the art know any method for obtaining the valley value of voltage data, which will not be described in detail here.

[0042] Preferably, in step S203, A 0 2 is processed to obtain A 0 2 The corresponding degree of change, A 0 2 corresponding to the mean, A 02 The corresponding change trend, A 0 2 corresponding peak value, A 0 2 corresponding to the valley value, A 0 2 The corresponding degree of change, A 0 2 corresponding to the mean, A 0 2 The corresponding change trend, A 0 2 corresponding peak value, A 0 2 is input into the second production equipment fault analysis model to obtain the second fault type set K2 corresponding to the target production equipment, that is, K2={K 21 ,……,K 2c ,……,K 2d}, K 2c is the cth second fault type and its probability corresponding to the target production equipment. The value range of c is 1 to d, and d is the number of second fault types corresponding to the target production equipment.

[0043] S204, A 0 3 is input into the third production equipment fault analysis model to obtain the third fault type set K3 corresponding to the target production equipment; the third production equipment fault analysis model is an LSTM model trained by the sample current data set, wherein the sample current data set includes a plurality of sample current data, each of which includes: the degree of change of the sample current, the mean value of the sample current, the change trend of the sample current, the peak value of the sample current, the valley value of the sample current and a plurality of sample production equipment fault types and the occurrence probability value of each production equipment fault type. For example, any sample voltage data L={L 0 1, L 0 2, L 0 3. L 0 4. L 0 5, L1, ..., L r ,……,L w}, L 0 1 is the degree of change of the sample current, L 0 2 is the average value of the sample current, L 0 3 is the changing trend of the sample current, L 0 4 is the peak value of the sample current, L 0 5 is the valley value of the sample current, L r It refers to the rth production equipment failure type and its occurrence probability value corresponding to the sample current data. The value range of r is 1 to w, and w is the number of production equipment failure types generated based on the current. Those skilled in the art know any method for training the LSTM model, which will not be repeated here.

[0044] Preferably, L 0 1Meet the following conditions:

[0045] , where P α It is the αth sample real-time current in a certain collected sample real-time current data table. The value range of α is 1 to β, and β is a certain collected sample real-time current data table.

[0046] Preferably, L 0 2Meet the following conditions:

[0047] ;

[0048] Preferably, L 0 3. Meet the following conditions:

[0049] , where △TU is the acquisition time interval of a sample real-time current data table; when α=β, P α+1 -P β =0.

[0050] Preferably, L 0 4 is the peak value of all sample real-time currents in a certain collected sample real-time voltage data table. Those skilled in the art know any method for obtaining the peak value of current data, which will not be described here.

[0051] Preferably, L 0 5 is the valley value of all sample real-time currents in a certain collected sample real-time voltage data table. Those skilled in the art know any method for obtaining the valley value of current data, which will not be described in detail here.

[0052] Preferably, in step S204, A 0 3 is processed to obtain A 0 3 corresponding to the degree of change, A 0 3 corresponding to the mean, A 0 3 Corresponding change trends, A 0 3 corresponding peak value, A 0 3 corresponding to the valley value, A 0 3 corresponding to the degree of change, A 0 3 corresponding to the mean, A 0 3 Corresponding change trends, A 0 3 corresponding peak value, A 0 3 is input into the third production equipment fault analysis model to obtain the third fault type set K3 corresponding to the target production equipment, that is, K3={K 31 ,……,K 3e ,……,K 3f}, K 3eis the e-th third fault type and its probability corresponding to the target production equipment. The value of e ranges from 1 to f, and f is the number of third fault types corresponding to the target production equipment.

[0053] S205: Based on K1, K2 and K3, obtain F.

[0054] Specifically, step S205 also includes the following steps:

[0055] S2051, match K1 and K2 to obtain the first identical fault type set △K1={△K 11 ,……,△K 1δ ,……△K 1η}, △K 1δ is the δth first identical fault type, where the value range of δ is 1 to η, and η is the number of first identical fault types. The first identical fault type refers to a fault type generated by temperature that is consistent with a fault type generated by voltage, and the difference between the probability of occurrence of the corresponding fault type generated by temperature and the probability of occurrence of the corresponding fault type generated by voltage is less than a first preset threshold. Those skilled in the art know how to set the threshold according to actual needs, and will not elaborate on this. It can be further understood as follows: K 1a The corresponding fault type and K 2c The corresponding fault types are consistent and K 1a The corresponding probability value and K 2c The difference between the corresponding occurrence probability values is less than a first preset threshold.

[0056] S2052, match K2 and K3 to obtain the second identical fault type set △K2={△K 21 ,……,△K 2φ ,……△K 2ᵞ}, △K 2φ is the second fault type of the φth, and the value range of φ is 1 to , is the number of the second identical fault types, wherein the second identical fault type means that the fault type generated by the voltage is consistent with the fault type generated by the current and the difference between the probability of occurrence of the corresponding fault type generated by the voltage and the probability of occurrence of the corresponding fault type generated by the current is less than the second preset threshold. Those skilled in the art know how to set the threshold according to actual needs and will not elaborate on it here; it can be further understood as: K 2c The corresponding fault type and K 3e The corresponding fault types are consistent and K 2c The corresponding probability value and K 3e The difference between the corresponding occurrence probability values is less than a second preset threshold.

[0057] S2053, match K1 and K3 to obtain the third identical fault type set △K3={△K 31 ,……,△K 3θ ,……△K 3ε}, △K 3θ is the θth third identical fault type, θ ranges from 1 to ε, ε is the number of third identical fault types, wherein the third identical fault type means that the fault type caused by temperature is consistent with the fault type caused by current and the difference between the probability of occurrence of the corresponding fault type caused by temperature and the probability of occurrence of the corresponding fault type caused by current is less than a third preset threshold. Those skilled in the art know how to set the threshold according to actual needs, which will not be repeated here; it can be further understood as: K 1a The corresponding fault type and K 3e The corresponding fault types are consistent and K 1a The corresponding probability value and K 3e The difference between the corresponding occurrence probability values is less than a third preset threshold.

[0058] S2054: Based on △K1, obtain F1, where F1 meets the following conditions:

[0059] , where △T1 is the time interval for collecting the real-time temperature of the target production equipment, and △T2 is the time interval for collecting the real-time voltage of the target production equipment. is △K 1δ The mean of the probability of occurrence of the corresponding temperature-generated fault type and the probability of occurrence of the corresponding voltage-generated fault type.

[0060] S2055: Based on △K2, obtain F2, where F2 meets the following conditions:

[0061] , where △T3 is the time interval for collecting the real-time current of the target production equipment. is △K 2φ The mean of the probability of occurrence of the corresponding voltage-generated fault type and the probability of occurrence of the corresponding current-generated fault type.

[0062] S2056, based on △K3, obtain F3, where F3 meets the following conditions:

[0063] ,in, is △K 3θ The mean of the probability of occurrence of the corresponding temperature-generated fault type and the probability of occurrence of the corresponding current-generated fault type.

[0064] S300 , based on A and F, obtaining a target failure degree U of the target production equipment for use in failure analysis of the target equipment.

[0065] Specifically, step S300 also includes the following steps:

[0066] S301, based on A, obtain the probability set of target fault types corresponding to A, △A, △A=(△A1,……,△A h ,……,△A k ), △A h is the probability of occurrence of the hth target fault type, where h ranges from 1 to k, and k is the number of target fault types. Further understanding: the target fault type is any fault type in the intersection of the fault type set corresponding to A1, the fault type set corresponding to A2, and the fault type set corresponding to A3. The probability of occurrence of the target fault type is the average of the probability of occurrence of the target fault type in the fault type set corresponding to A1, the fault type set corresponding to A2, and the fault type set corresponding to A3. The methods for obtaining the fault type set corresponding to A1, the fault type set corresponding to A2, and the fault type set corresponding to A3 refer to the methods for obtaining K1, K2, and K3, respectively, and are not repeated here.

[0067] S302: Obtain U based on ΔA and F, where U meets the following conditions:

[0068] .

[0069] As described above, this embodiment provides a fault analysis method based on production equipment data, the method comprising the following steps: S100, obtaining a real-time equipment operating status data set A=(A1, A2, A3) of a target production equipment, where A1 is the real-time temperature set of the target production equipment, A2 is the real-time voltage set of the target production equipment, and A3 is the real-time current set of the target production equipment; S200, determining fault correlation information F=(F1, F2, F3) of the equipment operating status based on A, where F1 is the fault correlation between the real-time time corresponding to A1 and the real-time voltage corresponding to A2, F2 is the fault correlation between the real-time voltage corresponding to A2 and the real-time current corresponding to A3, and F3 is the fault correlation between the real-time current corresponding to A3 and the real-time temperature corresponding to A1; S300, obtaining a target fault level U of the target production equipment based on A and F. It can be seen that the time alignment of the operating status data of different equipment is avoided, and no time alignment is required to generate fitted data, thereby improving the accuracy of the data and further improving the accuracy of the equipment fault analysis.

[0070] Example 2

[0071] This second embodiment provides a fault analysis system based on production equipment data. The system executes the fault analysis method based on production equipment data provided in the first embodiment. The system includes:

[0072] The first execution module is used to obtain a real-time device operation status data set A=(A1, A2, A3) of the target production equipment, where A1 is the real-time temperature set of the target production equipment, A2 is the real-time voltage set of the target production equipment, and A3 is the real-time current set of the target production equipment.

[0073] The second execution module is used to determine the fault correlation information F=(F1, F2, F3) of the equipment operation status based on A, where F1 is the fault correlation between the real-time time corresponding to A1 and the real-time voltage corresponding to A2, F2 is the fault correlation between the real-time voltage corresponding to A2 and the real-time current corresponding to A3, and F3 is the fault correlation between the real-time current corresponding to A3 and the real-time temperature corresponding to A1.

[0074] The third execution module is configured to obtain a target failure degree U of the target production equipment based on A and F, so as to be used for failure analysis of the target equipment.

[0075] Although some specific embodiments of the present invention have been described in detail by way of examples, it should be understood by those skilled in the art that the above examples are for illustration only and are not intended to limit the scope of the present invention. It should also be understood by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the present invention. The scope of the present invention is defined by the accompanying technical solutions.

Claims

1. A fault analysis method based on production equipment data, characterized in that: The method comprises the following steps: S100, obtaining a real-time equipment operating status data set A=(A1, A2, A3) of a target production equipment, where A1 is the real-time temperature set of the target production equipment, A2 is the real-time voltage set of the target production equipment, and A3 is the real-time current set of the target production equipment; S200, based on A, determining fault correlation information F = (F1, F2, F3) of the device operating status, where F1 is the fault correlation between the real-time time corresponding to A1 and the real-time voltage corresponding to A2, F2 is the fault correlation between the real-time voltage corresponding to A2 and the real-time current corresponding to A3, and F3 is the fault correlation between the real-time current corresponding to A3 and the real-time temperature corresponding to A1; S300: Based on A and F, a target failure degree U of the target production equipment is obtained for use in failure analysis of the target equipment. Step S300 also includes the following steps: S301, based on A, obtain the probability set of the target fault type corresponding to A, △A, △A=(△A1,……,△A h ,……,△A k ), △A h is the probability of occurrence of the hth target fault type, where h ranges from 1 to k, and k is the number of target fault types; S302: Obtain U based on ΔA and F, where U meets the following conditions: 。 2. A fault analysis method based on production equipment data according to claim 1, characterized in that: A1=(A 11 ,……,A 1x ,……,A 1p ), A 1x is the real-time temperature of the xth target production equipment, the value range of x is 1 to p, and p is the real-time temperature number of the target production equipment.

3. A fault analysis method based on production equipment data according to claim 1, characterized in that: A2=(A 21 ,……,A 2y ,……,A 2q ), A 2y is the real-time voltage of the y-th target production equipment, the value range of y is 1 to q, and q is the real-time voltage number of the target production equipment.

4. A fault analysis method based on production equipment data according to claim 1, characterized in that: A3=(A 31 ,……,A 3z ,……,A 3g ), A 3z is the real-time current of the zth target production equipment, the value range of z is 1 to g, and g is the real-time current quantity of the target production equipment.

5. The method for fault analysis based on production equipment data according to claim 1, characterized in that: The following steps are also included in step S200: S201, pre-process A to obtain the key equipment operation status data set A 0 , where A 0 =(A 0 1, A 0 2, A 0 3), A 0 1 is the critical temperature set of the target production equipment, A 0 2 is the key voltage set of the target production equipment, A 0 3 is the key current set of the target production equipment; S202, A 0 1 is input into the first production equipment fault analysis model to obtain the first fault type set K1 corresponding to the target production equipment; S203, A 0 2 is input into the second production equipment fault analysis model to obtain the second fault type set K2 corresponding to the target production equipment; S204, A 0 3 is input into the third production equipment fault analysis model to obtain the third fault type set K3 corresponding to the target production equipment; S205: Based on K1, K2 and K3, obtain F.

6. A method for fault analysis based on production equipment data according to claim 5, characterized in that: The following steps are also included in step S205: S2051, match K1 and K2 to obtain the first identical fault type set △K1={△K 11 ,……,△K 1δ ,……△K 1η }, △K 1δ is the δth first identical fault type, where the value of δ ranges from 1 to η, and η is the number of first identical fault types, wherein the first identical fault type refers to a fault type caused by temperature that is consistent with a fault type caused by voltage, and the difference between the probability of occurrence of the corresponding fault type caused by temperature and the probability of occurrence of the corresponding fault type caused by voltage is less than a first preset threshold; S2052, match K2 and K3 to obtain the second identical fault type set △K2={△K 21 ,……,△K 2φ ,……△K 2ᵞ }, △K 2φ is the second fault type of the φth fault, and the value range of φ is 1 to , is the number of the second identical fault types, wherein the second identical fault type refers to a fault type generated by voltage that is consistent with a fault type generated by current and a difference between the probability of occurrence of the corresponding fault type generated by voltage and the probability of occurrence of the corresponding fault type generated by current is less than a second preset threshold; S2053, match K1 and K3 to obtain the third identical fault type set △K3={△K 31 ,……,△K 3θ ,……△K 3ε }, △K 3θ is the θth third identical fault type, where θ ranges from 1 to ε, and ε is the number of third identical fault types, wherein the third identical fault type means that the fault type caused by temperature is consistent with the fault type caused by current, and the difference between the probability of occurrence of the corresponding fault type caused by temperature and the probability of occurrence of the corresponding fault type caused by current is less than a third preset threshold; S2054, based on △K1, obtain F1; S2055, based on △K2, obtain F2; S2056, based on △K3, obtain F3.

7. The method for fault analysis based on production equipment data according to claim 5, characterized in that: In step S202, the first generation equipment failure analysis model is an LSTM model trained through a sample temperature data set, wherein the sample temperature data set includes a plurality of sample temperature data, and each sample temperature data includes: the degree of change of the sample temperature, the mean of the sample temperature, the change trend of the sample temperature, the peak value of the sample temperature, the valley value of the sample temperature, and a plurality of sample production equipment failure types and the probability value of occurrence of each production equipment failure type.

8. The method for fault analysis based on production equipment data according to claim 5, characterized in that: In step S203, the second generation equipment failure analysis model is an LSTM model trained using a sample voltage data set, wherein the sample voltage data set includes a number of sample voltage data, and each sample voltage data includes: the degree of change of the sample voltage, the mean of the sample voltage, the change trend of the sample voltage, the peak value of the sample voltage, the valley value of the sample voltage, and a number of sample production equipment failure types and the probability value of occurrence of each production equipment failure type.

9. The method for fault analysis based on production equipment data according to claim 5, characterized in that: In step S204, the third generation equipment failure analysis model is an LSTM model trained using a sample current data set, wherein the sample current data set includes a plurality of sample current data, and each sample current data includes: the degree of change of the sample current, the mean of the sample current, the change trend of the sample current, the peak value of the sample current, the valley value of the sample current, and a plurality of sample production equipment failure types and the probability value of occurrence of each production equipment failure type.

10. A fault analysis system based on production equipment data, characterized in that: The system comprises: The first execution module is used to obtain a real-time device operating status data set A=(A1, A2, A3) of the target production equipment, where A1 is the real-time temperature set of the target production equipment, A2 is the real-time voltage set of the target production equipment, and A3 is the real-time current set of the target production equipment; The second execution module is used to determine, based on A, fault correlation information F=(F1, F2, F3) of the equipment operating status, where F1 is the fault correlation between the real-time time corresponding to A1 and the real-time voltage corresponding to A2, F2 is the fault correlation between the real-time voltage corresponding to A2 and the real-time current corresponding to A3, and F3 is the fault correlation between the real-time current corresponding to A3 and the real-time temperature corresponding to A1; The third execution module is configured to obtain a target failure degree U of the target production equipment based on A and F, so as to be used for failure analysis of the target equipment.

Images