CN117574328B - Coupling slip analysis method and system based on torque data - Google Patents
Coupling slip analysis method and system based on torque data Download PDFInfo
- Publication number
- CN117574328B CN117574328B CN202410050838.5A CN202410050838A CN117574328B CN 117574328 B CN117574328 B CN 117574328B CN 202410050838 A CN202410050838 A CN 202410050838A CN 117574328 B CN117574328 B CN 117574328B
- Authority
- CN
- China
- Prior art keywords
- knowledge
- information
- regression analysis
- fragment
- segment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000004458 analytical method Methods 0.000 title claims abstract description 108
- 230000008878 coupling Effects 0.000 title claims abstract description 24
- 238000010168 coupling process Methods 0.000 title claims abstract description 24
- 238000005859 coupling reaction Methods 0.000 title claims abstract description 24
- 239000012634 fragment Substances 0.000 claims abstract description 166
- 238000000611 regression analysis Methods 0.000 claims abstract description 145
- 238000012545 processing Methods 0.000 claims abstract description 67
- 238000013473 artificial intelligence Methods 0.000 claims abstract description 44
- 230000006870 function Effects 0.000 claims description 50
- 238000000605 extraction Methods 0.000 claims description 35
- 238000000034 method Methods 0.000 claims description 28
- 230000004913 activation Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 6
- 238000004891 communication Methods 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 abstract description 18
- 238000012549 training Methods 0.000 description 11
- 238000011478 gradient descent method Methods 0.000 description 3
- 238000003491 array Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000002349 favourable effect Effects 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/27—Regression, e.g. linear or logistic regression
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
According to the torque data-based coupling slip analysis method and system, regression analysis is performed on the knowledge fragment set in the first aspect by using a regression analysis thread, so that a regression analysis result is obtained; node identification is carried out on at least part of knowledge fragment information of the first multi-aspect knowledge fragment set by utilizing the regression analysis result, and a second multi-aspect knowledge fragment set is obtained based on the identified knowledge fragment information; and analyzing the second multi-aspect knowledge fragment set by utilizing the preset artificial intelligence analysis thread to obtain analysis result information of the slipping data to be processed, wherein the analysis result information of the slipping data to be processed can be directly modeled, so that the processing speed of slipping data analysis is improved, and the spatial information and the node information can be combined and staggered through node identification, so that the analysis processing is performed on the basis, and the accuracy of slipping data analysis is improved.
Description
Technical Field
The application relates to the technical field of data analysis, in particular to a method and a system for analyzing coupling slip based on torque data.
Background
The coupling is also called a coupling. Mechanical parts for firmly connecting the driving shaft and the driven shaft in different mechanisms to rotate together and transmitting motion and torque. Sometimes also to couple the shaft with other parts (e.g., gears, pulleys, etc.). The two halves are usually combined, respectively connected by a key or a tight fit, fastened at the two shaft ends, and then connected in some way. The coupling can also compensate offset (including axial offset, radial offset, angular offset or comprehensive offset) between two shafts due to inaccurate manufacturing and installation, deformation or thermal expansion during operation and the like; impact and vibration are alleviated.
In the actual operation process, the coupling may be interfered by various factors in the working process, so that the coupling may slip, and the reason for the slip is a technical problem which is difficult to solve at present.
Disclosure of Invention
In order to improve the technical problems in the related art, the application provides a method and a system for analyzing the slip of a coupler based on torque data.
In a first aspect, a method for analyzing a slip of a coupling based on torque data is provided, the method comprising: obtaining slipping data to be processed; extracting knowledge segments from the slipping data to be processed by using an artificial intelligent analysis thread which is set in advance to obtain a first multi-aspect knowledge segment set, wherein the first multi-aspect knowledge segment set covers knowledge segment information on different nodes corresponding to the slipping data to be processed; carrying out regression analysis on the knowledge fragment set in the first aspect by using a regression analysis thread to obtain a regression analysis result; node identification is carried out on at least part of knowledge fragment information of the first multi-aspect knowledge fragment set by utilizing the regression analysis result, and a second multi-aspect knowledge fragment set is obtained based on the identified knowledge fragment information; and analyzing the second multi-aspect knowledge fragment set by using the preset artificial intelligence analysis thread to obtain analysis result information of the skid data to be processed.
Further, before the node identification is performed on at least part of knowledge piece information of the first multi-aspect knowledge piece set by using the regression analysis result, and a second multi-aspect knowledge piece set is obtained based on the identified knowledge piece information, the method further comprises: carrying out regression analysis on the knowledge fragment set of the first aspect by using a confidence coefficient regression analysis thread to obtain confidence coefficient information; the node identification is performed on at least part of knowledge segment information of the first multi-aspect knowledge segment set by using the regression analysis result, and a second multi-aspect knowledge segment set is obtained based on the identified knowledge segment information, including: node identification is carried out on at least part of knowledge segment information of the first multi-aspect knowledge segment set by utilizing the regression analysis result; performing function processing on the identified knowledge piece information by utilizing the confidence information; and obtaining a second multi-aspect knowledge fragment set based on the knowledge fragment information processed by the function.
Further, the directions of the first multi-aspect knowledge fragment set include a node direction and a direction set in advance; the node identification of at least a portion of knowledge segment information of the first set of aspects of knowledge segments using the regression analysis results includes: selecting at least one group of knowledge piece information from the first multi-aspect knowledge piece set according to a preset direction, wherein each group of knowledge piece information comprises knowledge piece information corresponding to different nodes in the same preset direction; and identifying the at least one set of knowledge piece information in the node direction by using the regression analysis result.
Further, the preset direction is an analysis direction; and/or the regression analysis result comprises a first number of identification values, and the at least one set of knowledge piece information comprises a first number of sets of first knowledge piece information; the identifying the at least one set of knowledge piece information in the node direction using the regression analysis result includes: and identifying the first knowledge fragment information of the a-th group in the node direction by using the identification value of the a-th group in the regression analysis result to obtain second knowledge fragment information of the a-th group, wherein a is a positive integer not greater than the first number.
Further, the identifying the a-th group of the first knowledge piece information in the node direction by using the a-th identification value in the regression analysis result, and obtaining the a-th group of the second knowledge piece information includes: obtaining a numerical range to which the a-th identification value belongs, wherein the difference between the upper limit value and the lower limit value of the numerical range is a preset numerical value; identifying the upper limit value node units of the first knowledge fragment information of the a group along the node direction to obtain the third knowledge fragment information of the a group, and identifying the lower limit value node units of the first knowledge fragment information of the a group along the node direction to obtain the fourth knowledge fragment information of the a group; the difference between the a-th identification value and the lower limit value is used as a confidence level to determine the third knowledge fragment information of the a-th group, a first function result of the a-th group is obtained, and the difference between the upper limit value and the a-th identification value is used as a confidence level to determine the fourth knowledge fragment information of the a-th group, and a second function result of the a-th group is obtained; and calculating a splicing result between the a-th group first function result and the a-th group second function result to determine the second knowledge piece information as the a-th group.
Further, the slip data to be processed includes a second number of slip factors, and the confidence information includes a second number of confidence values; the performing function processing on the identified knowledge piece information by using the confidence information comprises the following steps: carrying out function processing on the knowledge fragment value corresponding to the b node in the current knowledge fragment information by using the b confidence value in the confidence information to obtain the corresponding knowledge fragment information after the function processing; wherein b is a positive integer not greater than the second number.
Further, the obtaining a second multi-aspect knowledge fragment set based on the knowledge fragment information after the function processing includes: and using the knowledge segment information processed by the function and the unrecognized knowledge segment information in the first plurality of knowledge segments to form the second plurality of knowledge segments.
Further, the performing regression analysis on the first multi-aspect knowledge segment set by using a confidence coefficient regression analysis thread to obtain confidence coefficient information includes: simplifying the knowledge fragment set in the first aspect by using a first simplifying unit of the confidence coefficient regression analysis thread to obtain a first simplifying result; carrying out convolution processing on the first simplified result by using a first convolution layer of the confidence coefficient regression analysis thread to obtain a first knowledge segment extraction result; and carrying out nonlinear processing on the first knowledge segment extraction result by using a first activation layer of the confidence coefficient regression analysis thread to obtain the confidence coefficient information.
Further, the performing regression analysis on the knowledge segment set of the first aspect by using a regression analysis thread to obtain a regression analysis result includes: simplifying the knowledge fragment set in the first aspect by using a second simplifying unit of the regression analysis thread to obtain a second simplifying result; carrying out convolution processing on the second simplified result by utilizing a second convolution layer of the regression analysis thread to obtain a second knowledge segment extraction result; carrying out knowledge segment connection on the second knowledge segment extraction result by using a first full connection layer of the regression analysis thread to obtain a first knowledge segment connection result; performing nonlinear processing on the first knowledge segment connection result by using a second activation layer of the regression analysis thread to obtain a nonlinear processing result; carrying out knowledge segment connection on the nonlinear processing result by using a second full-connection layer of the regression analysis thread to obtain a second knowledge segment connection result; and carrying out nonlinear processing on the second knowledge segment connection result by using a third activation layer of the regression analysis thread to obtain the regression analysis result.
Further, the preset artificial intelligence analysis thread comprises at least one convolution layer; the knowledge segment extraction is performed on the slip data to be processed by using a preset artificial intelligence analysis thread to obtain a first multi-aspect knowledge segment set, which comprises the following steps: extracting knowledge segments from the slipping data to be processed by using a convolution layer of an artificial intelligent analysis thread which is set in advance to obtain a first multi-aspect knowledge segment set; if the number of convolution layers of the preset artificial intelligence analysis thread is more than 1, after the second multi-aspect knowledge fragment set is obtained, and before the analysis of the second multi-aspect knowledge fragment set by the preset artificial intelligence analysis thread is performed to obtain analysis result information of the skid data to be processed, the method further includes: performing knowledge segment extraction on the second multi-aspect knowledge segment set by using a convolution layer which does not execute knowledge segment extraction in the preset artificial intelligence analysis thread to obtain a new first multi-aspect knowledge segment set; executing the step of carrying out regression analysis on the new knowledge fragment set in the first aspect by using a regression analysis thread to obtain a regression analysis result and the subsequent step to obtain a new knowledge fragment set in the second aspect; repeatedly executing the convolution layers which do not execute knowledge fragment extraction to execute knowledge fragment extraction steps on the second multi-aspect knowledge fragment set until all convolution layers of the preset artificial intelligence analysis thread complete knowledge fragment extraction steps on the new second multi-aspect knowledge fragment set; the analyzing the second multi-aspect knowledge fragment set by using the preset artificial intelligence analysis thread to obtain analysis result information of the skid data to be processed, wherein the analysis result information comprises: and analyzing the knowledge fragment set in the second aspect by utilizing the fully-connected layer of the preset artificial intelligent analysis thread to obtain analysis result information of the skidding data to be processed.
Further, the slip data to be processed includes a plurality of slip factors, the knowledge segment extraction is performed on the slip data to be processed by using an artificial intelligence analysis thread set in advance, and obtaining a first multi-aspect knowledge segment set includes: the knowledge segments are extracted by utilizing the preset artificial intelligent analysis thread to obtain a knowledge segment set corresponding to each sliding factor; and splicing the knowledge fragment sets according to the nodes corresponding to the knowledge fragment sets in the skidding data to be processed to obtain the knowledge fragment sets in the first aspect.
In a second aspect, a torque data based coupling slip analysis system is provided comprising a processor and a memory in communication with each other, the processor being configured to read a computer program from the memory and execute the computer program to implement the method described above.
According to the torque data-based coupling slip analysis method and system, node information of slip data to be processed can be directly modeled, the processing speed of slip data analysis can be improved, and through node identification, spatial information and node information can be combined and staggered, so that analysis processing is performed on the basis, and the accuracy of slip data analysis can be improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present application and therefore should not be considered limiting the scope, and that other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for analyzing a slip of a coupling based on torque data according to an embodiment of the present application.
Detailed Description
In order to better understand the technical solutions described above, the following detailed description of the technical solutions of the present application is provided through the accompanying drawings and specific embodiments, and it should be understood that the specific features of the embodiments and embodiments of the present application are detailed descriptions of the technical solutions of the present application, and not limit the technical solutions of the present application, and the technical features of the embodiments and embodiments of the present application may be combined with each other without conflict.
Referring to fig. 1, a method for analyzing a slip of a coupling based on torque data is shown, and may include the following steps S11 to S15.
Step S11: and obtaining the slip data to be processed.
In this embodiment of the present application, the slip data to be processed may include a plurality of slip factors, for example, the slip data to be processed includes 4 slip factors, or the slip data to be processed includes 8 slip factors, or the slip data to be processed includes 16 slip factors, etc., which are not specifically limited herein. In one implementation scenario, the slip data to be processed may be monitored slip data captured by a monitoring camera, so as to perform behavior analysis on a target object in the monitored slip data. In another implementation scenario, the slip data to be processed may be slip data in a slip database to classify the slip data in the slip database.
Step S12: and extracting knowledge segments from the slipping data to be processed by using a preset artificial intelligent analysis thread to obtain a first multi-aspect knowledge segment set.
Illustratively, knowledge segments may be understood as features.
In a specific implementation scenario, in order to further reduce the thread parameters and reduce the processing load, thereby increasing the processing speed and the convergence speed during training, and avoiding over-fitting, the artificial intelligence analysis thread set in advance may be a two-dimensional neural artificial intelligence analysis thread.
In the embodiment of the application, the first multi-aspect knowledge segment set covers knowledge segment information on different nodes corresponding to the skid data to be processed. In one implementation scenario, in order to reduce the processing load of knowledge segment extraction on the skid data to be processed and improve the processing speed of skid data analysis, knowledge segment extraction can be respectively performed on a plurality of skid factors of the skid data to be processed through an artificial intelligence analysis thread set in advance to obtain a knowledge segment set corresponding to each skid factor, so that a plurality of knowledge segment sets are directly spliced according to nodes corresponding to the knowledge segment sets in the skid data to be processed to obtain a first multi-aspect knowledge segment set. For example, if the slip data to be processed includes 8 slip factors, knowledge segment extraction can be performed on the 4 slip factors by using an artificial intelligence analysis thread set in advance to obtain knowledge segment sets of each slip factor, so that the 4 knowledge segment sets are directly spliced according to nodes corresponding to the 4 knowledge segment sets in the slip data to be processed to obtain a first multi-aspect knowledge segment set.
Step S13: and carrying out regression analysis on the knowledge fragment set in the first aspect by using a regression analysis thread to obtain a regression analysis result.
The slipping data is often focused on continuous behavior of the target object, so that the time information and the space information can be integrated in order to better obtain the inherent node semantics of the slipping data. Therefore, in the embodiment of the application, the regression analysis thread is adopted for regression analysis to obtain the regression analysis result, so that node identification is performed based on the regression analysis result later, and therefore integration of time information and space is completed. The regression analysis thread can be specifically an artificial intelligence analysis thread which is set in advance, so that regression analysis can be performed on the first multi-aspect knowledge fragment set through the artificial intelligence analysis thread which is set in advance, and a regression analysis result is directly obtained.
Further, the simplified result can be subjected to convolution processing by using a convolution layer of the regression analysis thread to obtain a knowledge fragment extraction result, and the convolution layer of the regression analysis thread can specifically cover convolution kernels with the same number as the number of frames of the skid data to be processed.
Step S14: and carrying out node identification on at least part of knowledge fragment information of the first multi-aspect knowledge fragment set by using a regression analysis result, and obtaining a second multi-aspect knowledge fragment set based on the identified knowledge fragment information.
In one implementation scenario, at least part of the knowledge segment information may be segmented along a preset direction (e.g., an analysis direction) in order to identify information corresponding to different nodes in at least part of the knowledge segment information, thereby integrating the temporal information and the spatial information and improving accuracy of the skid data analysis.
In one implementation scenario, in order to reduce the amount of calculation for identification and increase the processing speed of the analysis of the skid data, at least one set of knowledge piece information may be selected from the first multi-aspect knowledge piece set according to a preset direction (such as an analysis direction), where each set of knowledge piece information includes knowledge piece information corresponding to different nodes in the same preset direction (such as the analysis direction), at this time, the second fully-connected layer of the regression analysis thread may cover neurons having the same number as the selected number of sets of knowledge piece information, so that the number of elements in the regression analysis result is the same as the selected number of sets of knowledge piece information, and further, each element in the regression analysis result may be used to identify at least one set of knowledge piece information in the node direction, for example, identify one node unit in the node direction, or identify two node units in the node direction, and so on.
After node identification is performed on at least part of knowledge segment information of the first multi-aspect knowledge segment set by using the regression analysis result, at least part of knowledge segment information after node identification can be spliced with part of knowledge segment information, which is not subjected to node identification, in the first multi-aspect knowledge segment set, so that a second multi-aspect knowledge segment set is obtained. In a specific implementation scenario, the knowledge segment information obtained after the node identification of at least part of the knowledge segment information with the analysis number which is not identified by the node may be spliced with part of the knowledge segment information with the analysis number which is not identified by the node, so as to obtain a second multi-aspect knowledge segment set.
Step S15: and analyzing the second multi-aspect knowledge fragment set by using a preset artificial intelligent analysis thread to obtain analysis result information of the skid data to be processed.
In an implementation scenario, in order to improve accuracy of the skid data analysis, the number of convolution layers of the artificial intelligence analysis thread set in advance may be more than 1, such as 2, 3, 4, etc., so before the analysis of the second multi-aspect knowledge fragment set, the convolution layer of the artificial intelligence analysis thread set in advance may be used to extract the knowledge fragment set of the second multi-aspect knowledge fragment, so as to obtain a new first multi-aspect knowledge fragment set, specifically, the new first multi-aspect knowledge fragment set may keep the dimension unchanged in the node direction, and the regression analysis is performed on the new first multi-aspect knowledge fragment set by using the regression analysis thread, so as to obtain a regression analysis result and a subsequent step, so as to obtain a new second multi-aspect knowledge fragment set, and the knowledge fragment extraction step is continuously repeated on the second multi-aspect knowledge fragment set by using the convolution layer of the artificial intelligence analysis thread set in advance until all the previously set of the convolution layers of the artificial intelligence analysis thread complete the extraction step on the new second multi-aspect knowledge fragment set, and finally, the skid data analysis result is obtained by using the set of the artificial intelligence analysis thread set. Taking the example that the preset artificial intelligence analysis thread comprises 3 convolution layers as the example, after the slipping data to be processed is subjected to knowledge segment extraction through the first convolution layer of the preset artificial intelligence analysis thread to obtain a first multi-aspect knowledge segment set, node identification is performed through the related steps to obtain a second multi-aspect knowledge segment set, before the analysis processing is performed by using the fully connected layer of the preset artificial intelligence analysis thread, the second multi-aspect knowledge segment set can be further input into the second convolution layer to perform knowledge segment extraction to obtain a new first multi-aspect knowledge segment set, node identification is performed on the new first multi-aspect knowledge segment set through the related steps to obtain a new second multi-aspect knowledge segment set, similarly, node identification is performed on the new second multi-aspect knowledge segment set through the third convolution layer to obtain a new first multi-aspect knowledge segment set, node identification is performed on the new first multi-aspect knowledge segment set through the related steps at this moment, and the latest analysis result of the fully connected layer of the preset artificial intelligence analysis thread can be performed on all the newly-aspect knowledge segment sets to obtain slipping data of the fully connected layer analysis thread. Of course, in other embodiments, to reduce the amount of computation, the node identification step may be added only after a portion of the convolutional layer.
According to the scheme, the knowledge fragment set in the first aspect is obtained by extracting the knowledge fragments of the slipping data to be processed, the knowledge fragment set in the first aspect covers knowledge fragment information on different nodes corresponding to the slipping data to be processed, regression analysis is carried out on the knowledge fragment set in the first aspect by using a regression analysis thread to obtain regression analysis results, at least part of knowledge fragment information in the knowledge fragment set in the first aspect is subjected to node identification by using the regression analysis results, the knowledge fragment set in the second aspect is obtained based on the identified knowledge fragment information, and further the node information of the slipping data to be processed can be directly modeled, so that the processing speed of the slipping data analysis is improved, and the spatial information and the node information can be jointly staggered by the node identification, so that analysis processing is carried out on the basis, and the accuracy of the slipping data analysis is improved.
In this embodiment of the present invention, the regression analysis result includes a first number of identification values, and at least a portion of the first multi-aspect knowledge fragment set may be further divided into a first number of groups of first knowledge fragment information along a preset direction (such as an analysis direction), and then the a-th identification value in the regression analysis result may be used to identify the a-th group of first knowledge fragment information in the node direction, so as to obtain a-th group of second knowledge fragment information, where a is a positive integer not greater than the first number. At least part of the first multi-aspect knowledge segment set includes 2 groups of first knowledge segment information, the 1 st identification value in the regression analysis result can be used for identifying the 1 st group of first knowledge segment information in the node direction to obtain the 1 st group of second knowledge segment information, and the 2 nd identification value in the regression analysis result can be used for identifying the 2 nd group of first knowledge segment information in the node direction to obtain the 2 nd group of second knowledge segment information, and when the first number is other values, the same can be said, and no example is given here.
Specifically, the method may include the steps of:
step S141: the value range to which the a-th identification value belongs is obtained, and the difference between the upper limit value and the lower limit value of the value range is a preset value.
Step S142: and identifying the upper limit value node units of the a-th group first knowledge fragment information along the node direction to obtain a-th group third knowledge fragment information, and identifying the lower limit value node units of the a-th group first knowledge fragment information along the node direction to obtain a-th group fourth knowledge fragment information.
Step S143: and determining the difference between the a-th identification value and the lower limit value as the confidence coefficient to perform function processing on the a-th group third knowledge fragment information to obtain a-th group first function result, and determining the difference between the upper limit value and the a-th identification value as the confidence coefficient to perform function processing on the a-th group fourth knowledge fragment information to obtain a-th group second function result.
Step S144: and calculating a splicing result between the first function result of the a group and the second function result of the a group to determine the second knowledge piece information of the a group.
In addition, in one implementation scenario, since the node recognition is performed on each group of the first knowledge piece information in a group unit, a symmetrical recognition strategy can be adopted during training, that is, only half of recognition values can be trained during training, and the other half of recognition values are obtained by performing conversion calculation on the recognition values, so that the processing load during training can be reduced.
Different from the foregoing embodiment, the a-th group of the third knowledge piece information is obtained by obtaining a numerical range to which the a-th identification value belongs, and the difference between the upper limit value and the lower limit value of the numerical range is a preset numerical value, identifying the upper limit value node units along the node direction by the a-th group of the first knowledge piece information, and identifying the lower limit value node units along the node direction by the a-th group of the first knowledge piece information, so as to obtain the a-th group of the fourth knowledge piece information; the difference between the a-th identification value and the lower limit value is determined as confidence level to perform function processing on the a-th first knowledge fragment information to obtain a-th first function result, and the difference between the upper limit value and the a-th identification value is determined as confidence level to perform function processing on the a-th fourth knowledge fragment information to obtain a-th second function result; and calculating a splicing result between the first function result of the a group and the second function result of the a group to determine the second knowledge fragment information of the a group, so that the first knowledge fragment information can be conveniently and rapidly identified and processed, and the processing speed of skid data analysis can be improved.
Specifically, the method may include the steps of:
step S51: and obtaining the slip data to be processed.
Reference may be made in particular to the relevant steps of the previous embodiments.
Step S52: and extracting knowledge segments from the slipping data to be processed by using a preset artificial intelligent analysis thread to obtain a first multi-aspect knowledge segment set.
In the embodiment of the application, the first multi-aspect knowledge segment set covers knowledge segment information on different nodes corresponding to the skid data to be processed. Reference may be made in particular to the relevant steps of the previous embodiments.
Step S53: and carrying out regression analysis on the knowledge fragment set in the first aspect by using a regression analysis thread to obtain a regression analysis result.
The first manifold knowledge segment set may be subjected to regression analysis by a regression analysis thread, and in particular, reference may be made to the relevant steps in the foregoing embodiments.
Step S54: and carrying out regression analysis on the first multi-aspect knowledge fragment set by using a confidence coefficient regression analysis thread to obtain confidence coefficient information.
During node identification, knowledge segments at the first end and the last end of the first knowledge segment information may be removed, so in order to measure the importance degree of each knowledge segment in the first knowledge segment information after node identification again, so as to better obtain long-range information, a attention mechanism may be adopted to perform re-function processing on each knowledge segment in the first knowledge segment information after node identification, so that confidence information needs to be obtained. The confidence level information may be obtained by performing a regression analysis on the first set of knowledge segments in the first aspect using a confidence level regression analysis thread.
In one implementation scenario, the confidence regression analysis thread may include a simplified unit, a convolution layer, and an activation layer connected in sequence. Therefore, the confidence regression analysis thread only covers 3 layers, and only the convolution layer covers the thread parameters, so that the thread structure can be simplified to a certain extent, the thread parameters are reduced, the thread capacity can be reduced, the convergence speed is further improved, the overfitting is avoided, the model obtained through training is as accurate as possible, and the accuracy of the slipping data analysis can be further improved.
Specifically, the first multi-aspect knowledge segment set may be simplified by a simplification unit of the confidence regression analysis thread to obtain a simplified result. In a specific implementation scenario, the simplification unit may be an average pooling layer, and reference may be made to the relevant steps in the foregoing embodiments. Further, the simplified result can be subjected to convolution processing by using a convolution layer of the regression analysis thread to obtain a knowledge fragment extraction result, 1 convolution kernel can be covered in the convolution layer of the regression analysis thread, and the knowledge fragment extraction result is subjected to nonlinear processing by using an activation layer of the confidence coefficient regression analysis thread to obtain confidence coefficient information.
Step S55: and carrying out node identification on at least part of knowledge segment information of the knowledge segment set in the first aspect by using the regression analysis result.
Step S56: and carrying out function processing on the information of the identified knowledge segments by using the confidence information.
In one implementation scenario, the slip data to be processed may specifically include a second number of slip factors, the confidence information may include a second number of confidence values, the second number may specifically be 4, 8, 16, etc., and is not specifically limited herein. Therefore, during function processing, the b-th confidence value in the confidence coefficient information can be used for performing function processing on the knowledge segment value corresponding to the b-th node in the current knowledge segment information to obtain the corresponding group of knowledge segment information after function processing, wherein b is a positive integer not greater than the second number.
Step S57: and obtaining a second multi-aspect knowledge fragment set based on the knowledge fragment information after the function processing.
After node recognition and function processing, a second multi-aspect knowledge fragment set corresponding to the first multi-aspect knowledge fragment set can be obtained. In one implementation scenario, the second set of multiple knowledge segments may be composed using the function-processed knowledge segment information and the unrecognized knowledge segment information in the first set of multiple knowledge segments. Specifically, the knowledge segment information after the function processing and the unrecognized knowledge segment information in the first aspect of knowledge segment set can be spliced to obtain a second aspect of knowledge segment set. The resulting second set of multiple knowledge segments has the same size as the first set of multiple knowledge segments. In addition, if the knowledge segment information in the knowledge segment set in the first aspect is all subjected to node identification processing, the knowledge segment information after the function processing may be directly combined to determine the knowledge segment set in the second aspect.
Step S58: and analyzing the second multi-aspect knowledge fragment set by using a preset artificial intelligent analysis thread to obtain analysis result information of the skid data to be processed.
Different from the foregoing embodiment, the confidence coefficient regression analysis thread is used to carry out regression analysis on the first multi-aspect knowledge fragment set to obtain confidence coefficient information, the regression analysis result is used to identify nodes of at least part of knowledge fragment information of the first multi-aspect knowledge fragment set, the confidence coefficient information is used to carry out function processing on the identified knowledge fragment information, and the second multi-aspect knowledge fragment set is obtained based on the knowledge fragment information after the function processing, so that knowledge fragment information of spatial and node joint interleaving can be directly obtained through the steps of identification and function processing, and the processing speed and accuracy of the skid data analysis are improved.
The model training method for skid data analysis can be specifically implemented by electronic equipment with processing functions such as a microcomputer, a server, a tablet computer and the like, or by program codes executed by a processor. Specifically, the method may include the steps of:
step S71: sample slip data was obtained.
In this embodiment of the present application, the sample slip data may include a number of slip factors, for example, may include 4 slip factors, or may also include 8 slip factors, or may also include 16 slip factors, which are not specifically limited herein.
Step S72: and extracting knowledge fragments from the sample slipping data by using a preset artificial intelligent analysis thread to obtain a knowledge fragment set with multiple aspects of the first sample.
Step S73: and carrying out regression analysis on the knowledge fragment set in the first sample aspects by using a regression analysis thread to obtain a regression analysis result.
The thread structure of the regression analysis thread may refer to the relevant steps in the foregoing embodiments, and will not be described herein. In an implementation scenario, the confidence coefficient regression analysis thread may also be used to perform regression analysis on the knowledge fragment set with multiple aspects of the first sample to obtain the confidence coefficient information, and the thread structure of the confidence coefficient regression analysis thread may refer to the relevant steps in the foregoing embodiment, which is not described herein again.
Step S74: and carrying out node identification on at least part of knowledge fragment information of the knowledge fragment set in the first sample aspect by using a regression analysis result, and obtaining the knowledge fragment set in the second sample aspect based on the identified knowledge fragment information.
The specific implementation step of performing node identification on at least part of knowledge segment information of the knowledge segment set in the first sample aspect by using the regression analysis result may refer to the relevant step in the foregoing embodiment, which is not described herein again. In an implementation scenario, the confidence information may be used to perform a function processing on the identified knowledge segment information, and based on the knowledge segment information after the function processing, a knowledge segment set with multiple aspects of the second sample may be obtained, and specific reference may be made to the relevant steps in the foregoing embodiment, which is not described herein again.
In one implementation scenario, the pre-set artificial intelligence analysis thread may include at least one convolution layer, and knowledge segment extraction may be performed on the sample slip data by using a convolution layer of the pre-set artificial intelligence analysis thread, to obtain a knowledge segment set of the first sample aspects. In a specific implementation scenario, the number of convolution layers of the preset artificial intelligence analysis thread may be more than 1, then knowledge fragment extraction may be performed on the knowledge fragment set in multiple aspects of the second sample by using the convolution layers that do not perform knowledge fragment extraction in the preset artificial intelligence analysis thread, to obtain a new knowledge fragment set in multiple aspects of the first sample, and performing regression analysis on the new knowledge fragment set in multiple aspects of the first sample by using the regression analysis thread to obtain a regression analysis result, and a subsequent step, so as to obtain a new knowledge fragment set in multiple aspects of the second sample, and further repeating the steps until all convolution layers of the preset artificial intelligence analysis thread complete the knowledge fragment extraction step on the new knowledge fragment set in multiple aspects of the second sample.
Step S75: and analyzing the knowledge fragment set in multiple aspects of the second sample by using an artificial intelligence analysis thread which is set in advance to obtain analysis result information of sample slip data.
Step S76: and calculating a loss value by using the preset labeling information and analysis result information.
Step S77: and adjusting parameters of an artificial intelligence analysis thread and a regression analysis thread which are set in advance based on the loss value.
In one implementation scenario, as in the previous step, the confidence coefficient regression analysis thread may be further used to perform regression analysis on the knowledge segment set in the first sample aspect to obtain confidence coefficient information, so that the identified knowledge segment information is subjected to function processing by using the confidence coefficient information, and the second sample multidimensional knowledge segment information is obtained based on the knowledge segment information after the function processing, so that parameters of the artificial intelligent analysis thread, the regression analysis thread and the confidence coefficient regression analysis thread set in advance may be further adjusted based on the loss value. Specifically, parameters of a convolution layer and a full connection layer in an artificial intelligence analysis thread which are set in advance can be adjusted, parameters of the convolution layer and the full connection layer in a regression analysis thread can be adjusted, and parameters of the convolution layer in a confidence coefficient regression analysis thread can be adjusted. In particular, gradient descent methods may be employed to adjust parameters, such as batch gradient descent methods, random gradient descent methods.
In one implementation scenario, after the parameters are adjusted, the above step S72 and the subsequent steps may be further performed again until the calculated loss value satisfies the training end condition set in advance. Specifically, the training end condition set in advance may include: the loss value is smaller than a previously set loss threshold value, and the loss value is not reduced, or the previously set training end condition may further include: the number of parameter adjustment reaches a preset number of times threshold, or preset training end conditions may further include: the thread performance is tested using the test slip data to meet a pre-set requirement (e.g., accuracy reaches a pre-set accuracy threshold).
According to the scheme, the knowledge fragment set in the first sample multiple aspects is obtained by extracting the knowledge fragments from the sample slipping data, the knowledge fragment set in the first sample multiple aspects covers knowledge fragment information on different nodes corresponding to the sample slipping data, regression analysis is carried out on the knowledge fragment set in the first sample multiple aspects by using a regression analysis thread to obtain a regression analysis result, node identification is carried out on at least part of knowledge fragment information in the knowledge fragment set in the first sample multiple aspects by using the regression analysis result, a knowledge fragment set in the second sample multiple aspects is obtained based on the identified knowledge fragment information, further, node information of the sample slipping data can be directly modeled, the speed of model training can be improved, and spatial information and node information can be combined and staggered by means of node identification, so that analysis processing is carried out on the basis, and the accuracy of slipping data analysis can be improved subsequently.
On the basis of the above, a coupling slip analysis device based on torque data is provided, the device comprising:
the data acquisition module is used for acquiring the slip data to be processed;
the knowledge segment extraction module is used for extracting knowledge segments from the slipping data to be processed by utilizing an artificial intelligence analysis thread which is set in advance to obtain a first multi-aspect knowledge segment set, wherein the first multi-aspect knowledge segment set covers knowledge segment information on different nodes corresponding to the slipping data to be processed;
the result regression analysis module is used for carrying out regression analysis on the knowledge fragment set in the first aspect by utilizing a regression analysis thread to obtain a regression analysis result;
the knowledge segment identification module is used for carrying out node identification on at least part of knowledge segment information of the first multi-aspect knowledge segment set by utilizing the regression analysis result, and obtaining a second multi-aspect knowledge segment set based on the identified knowledge segment information;
and the result analysis module is used for analyzing the knowledge fragment set in the second aspect by utilizing the preset artificial intelligence analysis thread to obtain analysis result information of the skid data to be processed.
On the above basis, a clutch slip analysis system based on torque data is shown, comprising a processor and a memory in communication with each other, the processor being adapted to read a computer program from the memory and execute it to implement the method described above.
On the basis of the above, there is also provided a computer readable storage medium on which a computer program stored which, when run, implements the above method.
In summary, based on the above scheme, the node information of the slip data to be processed can be directly modeled, which is favorable for improving the processing speed of the slip data analysis, and the spatial information and the node information can be jointly staggered through node identification, so that the analysis processing is performed on the basis, and the accuracy of the slip data analysis is favorable for improving.
It should be appreciated that the systems and modules thereof shown above may be implemented in a variety of ways. For example, in some embodiments, the system and its modules may be implemented in hardware, software, or a combination of software and hardware. Wherein the hardware portion may be implemented using dedicated logic; the software portions may then be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or special purpose design hardware. Those skilled in the art will appreciate that the methods and systems described above may be implemented using computer executable instructions and/or embodied in processor control code, such as provided on a carrier medium such as a magnetic disk, CD or DVD-ROM, a programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The system and its modules of the present application may be implemented not only with hardware circuitry, such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., but also with software, such as executed by various types of processors, and with a combination of the above hardware circuitry and software (e.g., firmware).
It should be noted that, the advantages that may be generated by different embodiments may be different, and in different embodiments, the advantages that may be generated may be any one or a combination of several of the above, or any other possible advantages that may be obtained.
Claims (10)
1. A method for analyzing slip of a coupling based on torque data, the method comprising:
obtaining slipping data to be processed;
extracting knowledge segments from the slipping data to be processed by using an artificial intelligent analysis thread which is set in advance to obtain a first multi-aspect knowledge segment set, wherein the first multi-aspect knowledge segment set covers knowledge segment information on different nodes corresponding to the slipping data to be processed;
carrying out regression analysis on the knowledge fragment set in the first aspect by using a regression analysis thread to obtain a regression analysis result;
node identification is carried out on at least part of knowledge fragment information of the first multi-aspect knowledge fragment set by utilizing the regression analysis result, and a second multi-aspect knowledge fragment set is obtained based on the identified knowledge fragment information;
and analyzing the second multi-aspect knowledge fragment set by using the preset artificial intelligence analysis thread to obtain analysis result information of the skid data to be processed.
2. The torque data based coupling slip analysis method of claim 1, further comprising, prior to said node identifying at least a portion of knowledge piece information of said first set of knowledge pieces using said regression analysis results and deriving a second set of knowledge pieces based on said identified knowledge piece information: carrying out regression analysis on the knowledge fragment set of the first aspect by using a confidence coefficient regression analysis thread to obtain confidence coefficient information;
the node identification is performed on at least part of knowledge segment information of the first multi-aspect knowledge segment set by using the regression analysis result, and a second multi-aspect knowledge segment set is obtained based on the identified knowledge segment information, including:
node identification is carried out on at least part of knowledge segment information of the first multi-aspect knowledge segment set by utilizing the regression analysis result;
performing function processing on the identified knowledge piece information by utilizing the confidence information;
and obtaining a second multi-aspect knowledge fragment set based on the knowledge fragment information processed by the function.
3. The torque data based coupling slip analysis method of claim 2, wherein the directions of the first set of aspects of knowledge segments include a node direction and a pre-set direction; the node identification of at least a portion of knowledge segment information of the first set of aspects of knowledge segments using the regression analysis results includes:
Selecting at least one group of knowledge piece information from the first multi-aspect knowledge piece set according to a preset direction, wherein each group of knowledge piece information comprises knowledge piece information corresponding to different nodes in the same preset direction;
and identifying the at least one set of knowledge piece information in the node direction by using the regression analysis result.
4. The method for analyzing the slip of the coupling based on the torque data according to claim 3, wherein the preset direction is an analysis direction; and/or the regression analysis result comprises a first number of identification values, and the at least one set of knowledge piece information comprises a first number of sets of first knowledge piece information; the identifying the at least one set of knowledge piece information in the node direction using the regression analysis result includes: and identifying the first knowledge fragment information of the a-th group in the node direction by using the identification value of the a-th group in the regression analysis result to obtain second knowledge fragment information of the a-th group, wherein a is a positive integer not greater than the first number.
5. The method of claim 4, wherein the identifying the a-th set of the first knowledge piece information in the node direction by using the a-th identification value in the regression analysis result, and obtaining the a-th set of the second knowledge piece information comprises:
Obtaining a numerical range to which the a-th identification value belongs, wherein the difference between the upper limit value and the lower limit value of the numerical range is a preset numerical value;
identifying the upper limit value node units of the first knowledge fragment information of the a group along the node direction to obtain the third knowledge fragment information of the a group, and identifying the lower limit value node units of the first knowledge fragment information of the a group along the node direction to obtain the fourth knowledge fragment information of the a group;
the difference between the a-th identification value and the lower limit value is used as a confidence level to determine the third knowledge fragment information of the a-th group, a first function result of the a-th group is obtained, and the difference between the upper limit value and the a-th identification value is used as a confidence level to determine the fourth knowledge fragment information of the a-th group, and a second function result of the a-th group is obtained;
and calculating a splicing result between the a-th group first function result and the a-th group second function result to determine the second knowledge piece information as the a-th group.
6. A method of analyzing a coupling slip based on torque data according to claim 3, wherein the slip data to be processed comprises a second number of slip factors, and the confidence information comprises a second number of confidence values; the performing function processing on the identified knowledge piece information by using the confidence information comprises the following steps: carrying out function processing on the knowledge fragment value corresponding to the b node in the current knowledge fragment information by using the b confidence value in the confidence information to obtain the corresponding knowledge fragment information after the function processing; wherein b is a positive integer not greater than the second number.
7. The method for analyzing the slip of the coupling based on the torque data according to claim 2, wherein the obtaining the second multi-aspect knowledge segment set based on the knowledge segment information after the function processing includes: and using the knowledge segment information processed by the function and the unrecognized knowledge segment information in the first plurality of knowledge segments to form the second plurality of knowledge segments.
8. The torque data based coupling slip analysis method of claim 2, wherein performing regression analysis on the first set of knowledge segments in multiple aspects using a confidence regression analysis thread to obtain confidence information comprises:
simplifying the knowledge fragment set in the first aspect by using a first simplifying unit of the confidence coefficient regression analysis thread to obtain a first simplifying result;
carrying out convolution processing on the first simplified result by using a first convolution layer of the confidence coefficient regression analysis thread to obtain a first knowledge segment extraction result;
and carrying out nonlinear processing on the first knowledge segment extraction result by using a first activation layer of the confidence coefficient regression analysis thread to obtain the confidence coefficient information.
9. The method for analyzing the slip of the coupling based on the torque data according to claim 1, wherein the regression analysis of the knowledge segment set of the first aspect by using a regression analysis thread to obtain a regression analysis result comprises:
simplifying the knowledge fragment set in the first aspect by using a second simplifying unit of the regression analysis thread to obtain a second simplifying result;
carrying out convolution processing on the second simplified result by utilizing a second convolution layer of the regression analysis thread to obtain a second knowledge segment extraction result; carrying out knowledge segment connection on the second knowledge segment extraction result by using a first full connection layer of the regression analysis thread to obtain a first knowledge segment connection result;
performing nonlinear processing on the first knowledge segment connection result by using a second activation layer of the regression analysis thread to obtain a nonlinear processing result;
carrying out knowledge segment connection on the nonlinear processing result by using a second full-connection layer of the regression analysis thread to obtain a second knowledge segment connection result;
nonlinear processing is carried out on the second knowledge segment connection result by utilizing a third activation layer of the regression analysis thread to obtain the regression analysis result;
Wherein the preset artificial intelligence analysis thread comprises at least one convolution layer; the knowledge segment extraction is performed on the slip data to be processed by using a preset artificial intelligence analysis thread to obtain a first multi-aspect knowledge segment set, which comprises the following steps: extracting knowledge segments from the slipping data to be processed by using a convolution layer of an artificial intelligent analysis thread which is set in advance to obtain a first multi-aspect knowledge segment set;
if the number of convolution layers of the preset artificial intelligence analysis thread is more than 1, after the second multi-aspect knowledge fragment set is obtained, and before the analysis of the second multi-aspect knowledge fragment set by the preset artificial intelligence analysis thread is performed to obtain analysis result information of the skid data to be processed, the method further includes: performing knowledge segment extraction on the second multi-aspect knowledge segment set by using a convolution layer which does not execute knowledge segment extraction in the preset artificial intelligence analysis thread to obtain a new first multi-aspect knowledge segment set; executing the step of carrying out regression analysis on the new knowledge fragment set in the first aspect by using a regression analysis thread to obtain a regression analysis result and the subsequent step to obtain a new knowledge fragment set in the second aspect;
Repeatedly executing the convolution layers which do not execute the knowledge fragment extraction to extract the knowledge fragments of the second multi-aspect knowledge fragment set until all convolution layers of the artificial intelligence analysis thread which are set in advance complete the knowledge fragment extraction step of the new second multi-aspect knowledge fragment set; the analyzing the second multi-aspect knowledge fragment set by using the preset artificial intelligence analysis thread to obtain analysis result information of the skid data to be processed, wherein the analysis result information comprises: analyzing the knowledge fragment set in the second aspect by utilizing the fully-connected layer of the preset artificial intelligent analysis thread to obtain analysis result information of the skidding data to be processed;
the method comprises the steps of obtaining to-be-processed slip data, wherein the to-be-processed slip data comprises a plurality of slip factors, extracting knowledge segments from the to-be-processed slip data by using an artificial intelligence analysis thread which is set in advance, and obtaining a first multi-aspect knowledge segment set comprises: the knowledge segments are extracted by utilizing the preset artificial intelligent analysis thread to obtain a knowledge segment set corresponding to each sliding factor; and splicing the knowledge fragment sets according to the nodes corresponding to the knowledge fragment sets in the skidding data to be processed to obtain the knowledge fragment sets in the first aspect.
10. A coupling slip analysis system based on torque data, comprising a processor and a memory in communication with each other, the processor being adapted to read a computer program from the memory and execute the computer program to implement the method of any one of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410050838.5A CN117574328B (en) | 2024-01-15 | 2024-01-15 | Coupling slip analysis method and system based on torque data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410050838.5A CN117574328B (en) | 2024-01-15 | 2024-01-15 | Coupling slip analysis method and system based on torque data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117574328A CN117574328A (en) | 2024-02-20 |
| CN117574328B true CN117574328B (en) | 2024-03-29 |
Family
ID=89890364
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410050838.5A Active CN117574328B (en) | 2024-01-15 | 2024-01-15 | Coupling slip analysis method and system based on torque data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117574328B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019094721A2 (en) * | 2017-11-09 | 2019-05-16 | Strong Force Iot Portfolio 2016, Llc | Methods and systems for the industrial internet of things |
| CN209085657U (en) * | 2017-08-02 | 2019-07-09 | 强力物联网投资组合2016有限公司 | For data gathering system related or industrial environment with chemical production technology |
| CN112469649A (en) * | 2018-06-26 | 2021-03-09 | 意大利技术联盟有限责任公司 | Method for predicting the presence of product defects during an intermediate processing step of a thin product wound into a roll |
| CN113626594A (en) * | 2021-07-16 | 2021-11-09 | 上海齐网网络科技有限公司 | Operation and maintenance knowledge base establishing method and system based on multiple intelligent agents |
| CN114424167A (en) * | 2019-05-06 | 2022-04-29 | 强力物联网投资组合2016有限公司 | Platform for promoting intelligent development of industrial Internet of things system |
| WO2023065545A1 (en) * | 2021-10-19 | 2023-04-27 | 平安科技（深圳）有限公司 | Risk prediction method and apparatus, and device and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11389171B2 (en) * | 2006-11-21 | 2022-07-19 | David S. Goldsmith | Integrated system for the infixion and retrieval of implants |
| US9645575B2 (en) * | 2013-11-27 | 2017-05-09 | Adept Ai Systems Inc. | Method and apparatus for artificially intelligent model-based control of dynamic processes using probabilistic agents |
-
2024
- 2024-01-15 CN CN202410050838.5A patent/CN117574328B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN209085657U (en) * | 2017-08-02 | 2019-07-09 | 强力物联网投资组合2016有限公司 | For data gathering system related or industrial environment with chemical production technology |
| WO2019094721A2 (en) * | 2017-11-09 | 2019-05-16 | Strong Force Iot Portfolio 2016, Llc | Methods and systems for the industrial internet of things |
| CN112469649A (en) * | 2018-06-26 | 2021-03-09 | 意大利技术联盟有限责任公司 | Method for predicting the presence of product defects during an intermediate processing step of a thin product wound into a roll |
| CN114424167A (en) * | 2019-05-06 | 2022-04-29 | 强力物联网投资组合2016有限公司 | Platform for promoting intelligent development of industrial Internet of things system |
| CN113626594A (en) * | 2021-07-16 | 2021-11-09 | 上海齐网网络科技有限公司 | Operation and maintenance knowledge base establishing method and system based on multiple intelligent agents |
| WO2023065545A1 (en) * | 2021-10-19 | 2023-04-27 | 平安科技（深圳）有限公司 | Risk prediction method and apparatus, and device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| "风电联轴器极限扭矩下膜片组振动特性分析";李钦奉等;《风电联轴器极限扭矩下膜片组振动特性分析》;20200930;第49卷(第9期);19-22 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117574328A (en) | 2024-02-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2024021246A1 (en) | Cross-device incremental bearing fault diagnosis method based on continuous learning | |
| CN110285969A (en) | The rolling bearing fault of polynomial kernel implanting feature distribution adaptation migrates diagnostic method | |
| CN113111820B (en) | Rotary part fault diagnosis method and device based on improved CNN and relation module | |
| CN113673346B (en) | Motor vibration data processing and state identification method based on multiscale SE-Resnet | |
| Chadha et al. | Time series based fault detection in industrial processes using convolutional neural networks | |
| Ainapure et al. | Deep learning-based cross-machine health identification method for vacuum pumps with domain adaptation | |
| CN112488179A (en) | Rotary machine fault diagnosis method based on GRU | |
| CN111191838A (en) | Industrial equipment state control method and device integrating artificial intelligence algorithm | |
| Zhang et al. | A new method for diagnosing motor bearing faults based on Gramian Angular Field Image Coding and improved CNN-ELM | |
| CN116028876A (en) | Rolling bearing fault diagnosis method based on transfer learning | |
| CN115470717A (en) | Method, device, equipment and storage medium for predicting remaining life of robot | |
| CN117574328B (en) | Coupling slip analysis method and system based on torque data | |
| CN115290326A (en) | Rolling bearing fault intelligent diagnosis method | |
| CN115791174B (en) | Rolling bearing abnormality diagnosis method, system, electronic equipment and storage medium | |
| CN113541985A (en) | Internet of things fault diagnosis method, training method of model and related device | |
| CN116929815A (en) | Equipment working state monitoring system and method based on Internet of things | |
| CN115859077A (en) | Multi-feature fusion motor small sample fault diagnosis method under variable working conditions | |
| CN114841196A (en) | Mechanical equipment intelligent fault detection method and system based on supervised learning | |
| CN116150186B (en) | Communication Internet of things management method and system based on big data | |
| Yu et al. | Fault diagnosis of wearable temperature sensors based on multi-scale feature extraction | |
| Saha et al. | Enhancing Bearing Fault Diagnosis Using Transfer Learning and Random Forest Classification: A Comparative Study on Variable Working Conditions | |
| Ahsan et al. | Advanced Fault Diagnosis in Rotating Machines Using 2D Grayscale Images with Improved Deep Convolutional Neural Networks | |
| CN115329906B (en) | Fault diagnosis method and device based on graph regularization network | |
| CN116543538B (en) | Internet of things fire-fighting electrical early warning method and early warning system | |
| CN117155706B (en) | Network abnormal behavior detection method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |