CN113515402A - Fault information classification method and device for engineering equipment and engineering equipment - Google Patents
Fault information classification method and device for engineering equipment and engineering equipment Download PDFInfo
- Publication number
- CN113515402A CN113515402A CN202110638518.8A CN202110638518A CN113515402A CN 113515402 A CN113515402 A CN 113515402A CN 202110638518 A CN202110638518 A CN 202110638518A CN 113515402 A CN113515402 A CN 113515402A
- Authority
- CN
- China
- Prior art keywords
- fault
- test set
- data
- classification
- record data
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 74
- 238000012360 testing method Methods 0.000 claims abstract description 77
- 238000013145 classification model Methods 0.000 claims abstract description 46
- 238000012549 training Methods 0.000 claims abstract description 17
- 238000007781 pre-processing Methods 0.000 claims abstract description 8
- 238000013528 artificial neural network Methods 0.000 claims description 17
- 238000004422 calculation algorithm Methods 0.000 claims description 15
- 238000012545 processing Methods 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 6
- 238000007621 cluster analysis Methods 0.000 claims description 5
- 230000000306 recurrent effect Effects 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 241000393496 Electra Species 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 16
- 238000004458 analytical method Methods 0.000 description 15
- 238000010586 diagram Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 9
- 238000012423 maintenance Methods 0.000 description 9
- 238000004590 computer program Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 4
- 238000013507 mapping Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 238000003064 k means clustering Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005507 spraying Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/0703—Error or fault processing not based on redundancy, i.e. by taking additional measures to deal with the error or fault not making use of redundancy in operation, in hardware, or in data representation
- G06F11/079—Root cause analysis, i.e. error or fault diagnosis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24147—Distances to closest patterns, e.g. nearest neighbour classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
- G06N20/10—Machine learning using kernel methods, e.g. support vector machines [SVM]
-
- 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
The embodiment of the invention provides a method and a device for classifying fault information of a crane hydraulic system and engineering equipment. The fault information classification method comprises the following steps: acquiring fault record data; preprocessing the fault record data to extract a test set; carrying out data annotation on the fault record data in the test set so as to label fault phenomena and fault reasons corresponding to the fault phenomena, wherein one fault phenomenon corresponds to at least one fault reason; training the test set after the data marking to obtain a classification model of a fault phenomenon and a corresponding fault reason; and carrying out classification operation on the fault record data in the actual use process according to the classification model so as to determine the fault reason.
Description
Technical Field
The invention relates to the technical field of engineering equipment, in particular to a fault information classification method and device for engineering equipment and the engineering equipment.
Background
Usually, engineering equipment is equipped with a CRM system, and fault maintenance process data including "fault description", "fault phenomenon", "fault checking process, step", "fault occurrence reason, analysis", and the like are recorded in the CRM system. The field information contains the reasons of the fault phenomena, and research personnel can carry out design improvement in a targeted manner based on the information (fault reasons), so that the quality of parts is improved, and the number of product faults is reduced. However, the data recorded in these fields is unstructured data (text data), and cannot be directly subjected to batch statistical analysis, and developers need to analyze fault causes item by item, and count the types of causes of the fault and the proportion of each type. The mode has the disadvantages of large workload, large artificial subjective factors and low efficiency, and is difficult to systematically and comprehensively know the essential reasons of the fault phenomenon.
Disclosure of Invention
In order to solve the technical problem, the invention aims to provide a fault information classification method and device for engineering equipment and the engineering equipment.
In order to achieve the above object, in a first aspect of the present invention, there is provided a fault information classification method for engineering equipment, including: acquiring historical fault record data; preprocessing historical fault record data to extract a test set; carrying out data annotation on fault record data in the test set to classify target test sets of different fault types; carrying out model training on the target test set to obtain a classification model corresponding to the target test set; and carrying out classification operation on the actual fault record data according to the classification model so as to determine the fault reason.
In the embodiment of the present application, performing model training on the target test set to obtain a classification model corresponding to the target test set includes: performing feature coding on the target test set according to the data label; training the target test set subjected to feature coding through a neural network classification model to obtain a classification model; the neural network classification model is any one of a graph neural network, a recurrent neural network or a support vector machine.
In the embodiment of the present application, performing classification operation on actual fault record data according to a classification model to determine a fault cause includes: extracting a type of record set from actual fault record data according to the fault phenomenon; classifying operation is carried out on the record set through the corresponding classification model so as to obtain a feature code of a fault reason in the record set; counting the number of each type of feature codes; and multiplying the number of each type of feature codes by the corresponding weight to obtain a weight, and sequencing the weights to determine the fault reason.
In the embodiment of the present application, the data labeling of the fault record data in the test set to classify the target test set of different fault types includes: selecting the same fault type as a test set from the preprocessed test set; performing cluster analysis on the test set to divide the test set into a plurality of categories; and extracting a preset number of samples from the test set of each category to perform data annotation so as to obtain a target test set.
In the embodiment of the present application, the clustering analysis employs any one of a k-means algorithm and an Affinity prediction algorithm.
In an embodiment of the present application, preprocessing the fault log data to extract the test set includes: performing text feature extraction on the fault record data to obtain text data associated with fault processing; and smoothing and denoising the text data to extract a test set.
In the embodiment of the application, the text feature extraction is a word vector algorithm.
In this embodiment of the present application, the method for classifying fault information further includes: receiving feedback of a fault reason; and according to the feedback result, carrying out weighted correction on the weight of the feature code corresponding to the fault reason.
In a second aspect of the present invention, a fault information classification device for engineering equipment is provided, the fault information classification device includes a processor, and the processor is configured to execute the above fault information classification method.
In a third aspect of the invention, a piece of engineering equipment is provided, comprising the control device for the support leg assembly.
In a fourth aspect of the present invention, there is provided a machine-readable storage medium having stored thereon instructions for enabling a processor to execute the above-described fault information classification method for an engineering device when executed by the processor.
According to the technical scheme, the fault information classification method for the engineering equipment is provided, and the classification model related to the fault phenomenon is trained according to the past historical fault records, so that the fault reason is directly determined. Compared with the current working mode of mainly processing fault data recorded in the CRM by means of manual analysis, the automatic classification method for the fault reasons of the crane hydraulic system based on semantic analysis and data clustering is established, the essential reasons of fault phenomena can be quickly and accurately analyzed, research and development personnel are guided to specifically and mainly solve main problems, and therefore product quality is improved.
Additional features and advantages of embodiments of the invention will be set forth in the detailed description which follows.
Drawings
The accompanying drawings, which are included to provide a further understanding of the embodiments of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the embodiments of the invention without limiting the embodiments of the invention. In the drawings:
FIG. 1 illustrates a simplified connection topology of a fault information classification system provided by an embodiment of the present invention;
fig. 2 is a flowchart of a fault information classification method for engineering equipment according to an embodiment of the present invention;
fig. 3 is a detailed flowchart of step S102 in the fault information classification method according to the embodiment of the present invention;
fig. 4 is a detailed flowchart of step S103 in the fault information classification method according to the embodiment of the present invention;
fig. 5 is a specific flowchart of step S104 in the fault information classification method according to the embodiment of the present invention;
fig. 6 is a detailed flowchart of step S105 in the fault information classification provided in the embodiment of the present invention; and
fig. 7 is another flowchart in the fault information classification method according to the embodiment of the present invention.
Description of the reference numerals
200. A fault information classification system; 201. A fault classification module;
202. an upper computer; 203. operation platform
100. A CRM module;
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular device structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.
The following detailed description of embodiments of the invention refers to the accompanying drawings. It should be understood that the detailed description and specific examples, while indicating embodiments of the invention, are given by way of illustration and explanation only, not limitation.
For convenience of understanding, the first embodiment of the present invention first provides a fault information classification system. Referring to fig. 1, fig. 1 is a simple connection topology diagram of a fault information classification system according to an embodiment of the present invention;
the fault information classification system 200 includes:
the fault classification module 201: the CRM module 100 is used for receiving fault record data from the CRM module 100 and classifying according to the fault record data so as to determine a fault reason;
an upper computer 202: receiving the fault cause determined by the fault classification module 20, and generating a corresponding solution decision;
and the operation platform 203 displays the fault reason and the solution decision, and performs manual correction on the fault reason and the solution decision according to the situation judged by manual practice.
It can be understood that in the embodiment of the present invention, it is necessary to use the system to electrically connect the fault information classification system 200 to the CRM module 100, so as to read the corresponding fault record data in the CRM module. The CRM module is common customer relationship management in engineering equipment, and refers to a process that an enterprise coordinates the interaction between the enterprise and a customer on sales, marketing and service by using a corresponding information technology and an internet technology to improve core competitiveness, so that the management mode of the CRM module is improved, and innovative and personalized customer interaction and service are provided for the customer. Customer relationships refer to the collection of information that occurs, develops around the customer lifecycle. CRM in the embodiment of the invention mainly refers to customer relationship management in a customer service process. The customer service is mainly used for rapidly and timely obtaining information of problem customers and customer historical problem records and the like, and the main functions comprise customer feedback, solution, satisfaction investigation and the like. For example, the CRM records mainly refer to product fault maintenance records recorded in the CRM system and related to a hydraulic system in engineering equipment.
Further, the fault record data of the CRM module 100 are read through the fault classification module 201, the fault record data are automatically classified according to a classification model preset by the fault classification module 201 and uploaded to the upper computer 202, after the upper computer 202 receives the fault record data, a solution decision can be automatically or manually generated by background personnel according to experience, the solution decision is sent to the operating platform 203 to be displayed, the classification model is subjected to feedback correction, secondary training is carried out on the classification model, and the accuracy of the classification model in the fault classification module 201 is improved.
To sum up, the fault information classification system 200 provided in the embodiment of the present invention reads the fault record data of the CRM module 100 and generates a solution decision according to the fault record data under the condition that the original CRM module 100 can generate the fault record data, thereby reducing the difficulty of fault analysis and simplifying the tedious work of analyzing the fault cause one by one in the past.
Based on the feature that the fault record data of the past CRM module 100 are all presented in the form of text data, that is, the fault record data is field information, the invention can provide a fault information classification method based on semantic analysis and data clustering, so that the essential cause of the fault phenomenon can be known more accurately. In one general inventive concept, a fault information classification method for establishing a fault cause of a crane hydraulic system mainly comprises the following steps:
firstly, acquiring a fault data record of a CRM module: and the after-sale service department sends after-sale service personnel to a site to analyze the fault phenomenon and reason, maintains and solves the fault, and fills the relevant fault information into a CRM system of a company. The fault information to be filled in includes: host number, equipment model, fault code, fault system, component, part, fault description, fault mode, fault grade, fault phenomenon, fault inspection process, fault occurrence reason analysis and the like.
And then preprocessing is carried out on the hydraulic system fault records recorded by the CRM module, so that some training records can be selected and selected, and a test set required by the hydraulic system fault reason automatic classification system is artificially manufactured and established. And training a classification model between data such as 'fault phenomenon', 'fault inspection process' and 'fault occurrence reason analysis' and a classification label by using the test set.
By utilizing the model, in the actual use process of the engineering equipment, each fault record of the CRM module is automatically analyzed in sequence to obtain the reason type of each fault phenomenon, and then the record quantity of each fault type is counted, so that the main fault reason for generating each fault phenomenon can be known; after the main reasons are transmitted to the upper computer, the upper computer outputs a solution decision to the operating platform according to the preset mapping relation between the fault phenomenon and the solution decision according to the main reasons of the fault phenomenon; and maintenance personnel maintain and repair according to self experience and actual conditions according to the fault reasons and solution decisions reflected by the operation platform, and correct the classification model and the mapping relation model of the upper computer according to the real reflection.
Referring to fig. 2, fig. 2 is a flowchart of a fault information classification method for engineering equipment according to an embodiment of the present invention; therefore, the embodiment of the invention provides a fault information classification method, which comprises the following steps:
s101, acquiring historical fault record data;
step S102, preprocessing historical fault record data to extract a test set;
step S103, carrying out data annotation on the fault record data in the test set according to fault types so as to classify different target test sets;
s104, performing model training on the target test set to obtain a classification model corresponding to the target test set;
and step S105, carrying out classification operation on the actual fault record data according to the classification model so as to determine the fault reason.
Specifically, historical fault record data in the past is firstly pulled from an original CRM module, because the historical fault record data is huge and complex, the historical fault record data needs to be preprocessed to extract an available test set (a total test set), the test set is internally covered with data of multiple fault types, the test set is further subjected to data labeling, namely, the fault record data is labeled according to manual experience, and the test set is classified according to the fault types according to the labels to obtain target test sets of different fault types; after the target test set is obtained, the corresponding classification models are respectively trained according to the difference of the fault types, and the classification models are applied to the subsequent actual fault record data of the engineering equipment for classification operation, so that the fault reason is determined. It can be understood that, through steps S101 to S105, the corresponding classification model can be trained according to the historical fault record data in the original CRM module, so that in subsequent use, the fault reason reflected by the actual fault record data can be determined according to the classification model, time and labor are avoided for maintenance personnel to check the fault record data, the efficiency of equipment operation and maintenance is increased, and the operation and maintenance cost is reduced.
Referring to fig. 3, fig. 3 is a flowchart illustrating a step S102 of a fault information classification method according to an embodiment of the present invention; preprocessing the fault log data to extract a test set includes:
step S1021, text feature extraction is carried out on the fault record data to obtain text data relevant to fault processing;
and step S1022, smoothing and denoising the text data to extract a test set.
In particular, the test set is used to train the semantic classification model and test the accuracy of the model. The fault log data needs to be preprocessed to remove unreasonable data and less representative data. In the execution of steps S1021 to S1022, the test set is selected from the failure record data recorded by the CRM module of the engineering device in recent years, and since the failure record data includes multiple types of data, text data related to failure processing is extracted through text feature extraction, where the text feature extraction may adopt a word vector algorithm such as word2vec, and a segment vector algorithm such as paragraph2 vec.
In a specific embodiment, the fault record data is loaded first, and then the text feature extraction is performed, for example, by optionally using a countvectorer function (countvectorer function: a Chinese extraction instruction), wherein the countvectorer function takes the fault record data as a corpus, selects the fault record data according to the word frequency ordering in the corpus from high to low, and represents the text feature by counting the number of occurrences of words and using a sparse matrix of the number of occurrences of words. All the words are counted, the number of times of each word appears, the number of the words is the column of the obtained sparse matrix, and if the word to be extracted is the text characteristic of 'pipeline joint-oil leakage', the number of times is counted for 5000 times.
Referring to fig. 4, fig. 4 is a flowchart illustrating a step S103 of the fault information classification method according to the embodiment of the present invention; the target test set for carrying out data annotation on fault record data in the test set to classify different fault types comprises the following steps:
step S1031, selecting the same fault type as a test set in the preprocessed test set;
step S1032, performing cluster analysis on the test set to divide the test set into a plurality of categories;
step S1033, extracting a preset number of samples from the test set of each category, and performing data annotation to obtain a target test set.
The clustering analysis may adopt any one of a K-means algorithm (K-means algorithm: K-means clustering algorithm) and an Affinity Propagation algorithm (AP algorithm, also called neighbor Propagation algorithm or Affinity Propagation algorithm).
In a specific example, assuming that 5000 pieces of data are recorded in the fault record data of the CRM module, wherein the recorded data of one type, such as the phenomenon of 'pipe joint-oil leakage', are subjected to cluster analysis calculation, assuming that 3 types are obtained: c1, C2, C3, the number of records in each category being 1500, 2000 respectively. Next, 20 data were randomly selected in the C1, C2, and C3 sets, respectively, for a total of 60 data. Finally, according to CRM field information of the 60 records, such as "fault phenomenon", "fault checking process", and "fault occurrence cause analysis", the experts in the relevant direction analyze and judge the 60 CRM records, determine and label the specific cause category of each fault record, and it should be noted that all the data are structured data, and thus, a test set (total 60 samples) corresponding to the fault phenomenon, i.e., the target test set mentioned in step S1033, is completed. It will be appreciated that this process is referred to as data annotation. The advantage of this approach is that only a small amount of data needs to be labeled to cover all of the sample space.
Referring to fig. 5, fig. 5 is a flowchart illustrating a step S104 of the fault information classification method according to the embodiment of the present invention; the model training of the target test set to obtain a classification model corresponding to the target test set comprises the following steps:
s1041, performing feature coding on the target test set according to the data label;
and step S1042, training the target test set subjected to the feature coding through a neural network classification model to obtain a classification model.
The neural network classification model is any one of a graph neural network, a recurrent neural network or a support vector machine. Specifically, the method for training the classification model may adopt a general natural language processing method, and in a specific example, the method is applied to a test set corresponding to the "pipeline joint-oil leakage" fault phenomenon (assuming that 60 samples are provided in total). And carrying out feature coding on three columns of data, namely 'fault phenomenon', 'fault checking process' and 'fault occurrence cause analysis', by using a BERT pre-training model to obtain a feature-coded test set (still 60 samples). And then, training a neural network classification model by using the feature-coded test set. The neural network classification model may employ a GNN (graph neural network) or an RNN (recurrent neural network). Finally, a fault reason classification model corresponding to the pipeline joint-oil leakage is obtained as follows: and the composite model is composed of a characteristic coding model and a neural network classification model, namely the classification model.
Referring to fig. 6, fig. 6 is a flowchart illustrating a step S105 in the fault information classification according to an embodiment of the present invention; the step of carrying out classification operation on the actual fault record data according to the classification model so as to determine the fault reason comprises the following steps:
step S1051, extracting a kind of record set from the actual fault record data according to the fault phenomenon;
step S1052, performing classification operation on the record set through the corresponding classification model to obtain a feature code of a fault reason in the record set;
step S1053, counting the number of each type of feature codes;
and step S1054, multiplying the number of each type of feature codes by the corresponding weight to obtain a weight, and sequencing the weights to determine the fault reason.
The fault types comprise fault phenomena and fault reasons; one fault phenomenon corresponds to at least one fault reason; the feature encoding model includes any one of a BERT model, an ELECTRA model, a transform model. Specifically, continuing with the example of "pipe joint-oil leak" recorded in the CRM system, the main process of applying the trained fault cause classification model for automatic analysis is:
assuming that the set of the "pipeline joint-oil leakage" records in the CRM system has 5000 samples, each sample is calculated by applying a training model corresponding to the "pipeline joint-oil leakage" to obtain a category code of each record, thereby obtaining a category label of each record, then counting the number of records of each category label, and displaying a "pipeline joint-oil leakage" fault reason analysis result in the form of the following table, as shown in table 1 below.
TABLE 1
In some embodiments, the weight correspondences are all 1, and at this time, only the number of feature codes of each class needs to be sorted, so as to determine the main failure cause. As can be seen from the above table, the "line joint-oil leakage" failure phenomenon is mainly caused by two reasons, that is, "the tightening force at the time of makeup does not reach the standard" and "the line joint is damaged".
In some embodiments, the weights corresponding to each type of feature code are different, for example, by numbering the above, different types of codes respectively correspond to different weights, for example, a type code 3 corresponds to a weight a, a type code 4 corresponds to a weight b, a type code 1 corresponds to a weight c, and a type code 2 corresponds to a weight d, and then the ratio of the different types of codes is multiplied by the weight to obtain an average value, and the average value is normalized to obtain a weight, and finally the weight is sorted according to the size of the weight, so as to determine the main cause of failure.
Referring to fig. 7, fig. 7 is another flowchart of a fault information classification method according to an embodiment of the present invention; the fault information classification method further comprises the following steps:
step S106, receiving the feedback of the failure reason;
and step S107, carrying out weighted correction on the weight of the feature code corresponding to the fault reason according to the feedback result.
For different problems, according to the explanation of the above system embodiment, after the upper computer receives the failure reason, the upper computer generates a corresponding solution decision according to a preset mapping relationship, and if the failure phenomenon is mainly caused by two reasons, namely "tightening force does not reach the standard during assembly" and "pipeline joint damage", the solution decision to be responded is: 1. and 2, improving the quality of the assembly workshop, and improving the quality of the purchased pipeline joint by a purchasing department. At this time, the maintenance personnel takes the two pieces of information as reference, and carries out maintenance and feedback according to the actual situation.
According to the reflected solution decision and whether the fault reason is correct or not, maintenance personnel can feed back on the operation platform, when the fault reason determined by the classification module is correct each time, the weight of the corresponding feature code is correspondingly increased, the increased proportion can be preset in a percentage range, for example, 5%, otherwise, when the determined fault reason is wrong, the weight of the corresponding feature code is reduced, namely, the classification model is continuously optimized by performing weighting correction on the weight of the feature code corresponding to the fault reason according to the feedback result, so that the accuracy of classification pre-judgment is improved.
In summary, compared with the current working mode of mainly processing fault data recorded in a CRM by means of manual analysis, the method for automatically classifying the fault reasons of the crane hydraulic system based on semantic analysis and data clustering can quickly and accurately analyze the essential reasons of fault phenomena, guide research and development personnel to specifically and mainly solve main problems, and accordingly improve product quality.
It will also be appreciated by those skilled in the art that if the method or apparatus of the present invention is used, it is simply a matter of variation, or it is a combination of the aforementioned methods with additional functionality, or it is a matter of substitution for alternative types of materials, environments, or geometries of the various components, etc.; or the products formed by the components are integrally arranged; or a detachable design; it is within the scope of the present invention to replace the methods and apparatus of the present invention with any method/apparatus/device that combines the components to form a method/apparatus/device with specific functionality.
The device also comprises a memory, the fault information classification method for the engineering equipment can be stored in the memory as a program unit, and the processor executes the program unit stored in the memory to realize corresponding functions.
The processor comprises a kernel, and the kernel calls the corresponding program unit from the memory. The kernel can be set to be one or more than one, and the spraying arm of the gas water heater is controlled to clean the tableware according to the tableware image by adjusting the kernel parameters.
The memory may include volatile memory in a computer readable medium, Random Access Memory (RAM) and/or nonvolatile memory such as Read Only Memory (ROM) or flash memory (flash RAM), and the memory includes at least one memory chip.
An embodiment of the present invention provides a machine-readable storage medium on which a program is stored, the program implementing a fault information classification method for engineering equipment when being executed by a processor.
The embodiment of the invention provides a processor, which is used for running a program, wherein the fault information classification method for engineering equipment is executed when the program runs.
The embodiment of the invention also provides engineering equipment, which comprises the operation monitoring device for the engineering equipment described in the embodiment.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, apparatus, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In a typical configuration, a computing device includes one or more processors (CPUs), an input/output interface, an internet of things interface, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). The memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.
It should be noted that all the flow direction indicators (such as upper, lower, left, right, front and rear … …) in the embodiment of the present invention are only used to explain the relative geometrical relationship, movement, etc. of the components in a specific posture (as shown in the figure), and if the specific posture is changed, the flow direction indicator is changed accordingly.
In addition, the descriptions related to "first", "second", etc. in the present invention are for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In addition, the meaning of "and/or" appearing throughout is to include three juxtapositions, exemplified by "A and/or B" including either scheme A, or scheme B, or a scheme in which both A and B are satisfied. In addition, technical solutions between various embodiments may be combined with each other, but must be realized by a person skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination should not be considered to exist, and is not within the protection scope of the present invention.
The above are merely examples of the present application and are not intended to limit the present application. Various modifications and changes may occur to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.
Claims (12)
1. A fault information classification method for engineering equipment is characterized by comprising the following steps:
acquiring historical fault record data;
preprocessing the historical fault record data to extract a test set;
carrying out data annotation on the fault record data in the test set so as to classify target test sets of different fault types;
training the target test set to obtain a classification model corresponding to the target test set;
and carrying out classification operation on the actual fault record data according to the classification model so as to determine the fault reason.
2. The method of claim 1, wherein the training the target test set to obtain the classification model corresponding to the target test set comprises:
performing feature coding on the target test set according to data marking;
training the target test set subjected to feature coding through a neural network classification model to obtain a classification model;
wherein the neural network classification model comprises any one of a graph neural network, a recurrent neural network, and a support vector machine.
3. The fault information classification method according to claim 2, wherein the fault category includes a fault phenomenon and a fault cause; one fault phenomenon corresponds to at least one fault reason; the feature encoding model includes any one of a BERT model, an ELECTRA model, a transform model.
4. The method for classifying fault information according to claim 3, wherein the step of performing classification operation on the actual fault record data according to the classification model to determine the fault cause comprises:
extracting a type of record set from the actual fault record data according to the fault phenomenon;
classifying operation is carried out on the record set through a corresponding classification model so as to obtain a feature code of a fault reason in the record set;
counting the number of each type of feature codes;
and multiplying the number of each type of feature codes by the corresponding weight to obtain a weight, and sequencing the weights to determine the fault reason.
5. The method for classifying fault information according to claim 3, wherein the step of performing data labeling on the fault record data in the test set to classify target test sets of different fault categories comprises:
selecting the same fault type as a test set from the preprocessed test set;
performing cluster analysis on the test set to divide the test set into a plurality of categories;
and extracting a preset number of samples from the test set of each category to perform data annotation so as to obtain a target test set.
6. The method of classifying fault information according to claim 5, wherein the cluster analysis employs any one of a k-means algorithm and an Affinity prediction algorithm.
7. The method of classifying fault information according to claim 2, wherein the preprocessing the fault log data to extract a test set comprises:
performing text feature extraction on the fault record data to obtain text data associated with fault processing;
and smoothing and denoising the text data to extract a test set.
8. The method according to claim 2, wherein the text feature extraction is a word vector algorithm.
9. The fault information classification method according to any one of claims 4 to 7, characterized in that the fault information classification method further comprises:
receiving feedback of the failure cause;
and according to the feedback result, carrying out weighted correction on the weight of the feature code corresponding to the fault reason.
10. A fault information classification apparatus for engineering equipment, characterized in that the fault information classification apparatus comprises a processor configured to execute the fault information classification method for engineering equipment according to any one of claims 1 to 8.
11. An engineering equipment, characterized by comprising the fault information classification apparatus for engineering equipment according to claim 9.
12. A machine-readable storage medium having stored thereon instructions for enabling a processor to execute the fault information classification method for engineering equipment according to any one of claims 1 to 8 when the instructions are executed by the processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110638518.8A CN113515402A (en) | 2021-06-08 | 2021-06-08 | Fault information classification method and device for engineering equipment and engineering equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110638518.8A CN113515402A (en) | 2021-06-08 | 2021-06-08 | Fault information classification method and device for engineering equipment and engineering equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113515402A true CN113515402A (en) | 2021-10-19 |
Family
ID=78065748
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110638518.8A Pending CN113515402A (en) | 2021-06-08 | 2021-06-08 | Fault information classification method and device for engineering equipment and engineering equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113515402A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115049081A (en) * | 2022-06-18 | 2022-09-13 | 深圳市蓝晨科技股份有限公司 | Notebook computer detection management method, system and storage medium |
| CN116431378A (en) * | 2023-06-13 | 2023-07-14 | 成都实时技术股份有限公司 | Product embedded system fault data analysis method |
| CN116910478A (en) * | 2023-07-14 | 2023-10-20 | 西安科技大学 | Lithium ion battery accident tracing method and system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766879A (en) * | 2017-09-30 | 2018-03-06 | 中国南方电网有限责任公司 | The MLP electric network fault cause diagnosis methods of feature based information extraction |
| CN110781671A (en) * | 2019-10-29 | 2020-02-11 | 西安科技大学 | Knowledge mining method for intelligent IETM fault maintenance record text |
| CN111651601A (en) * | 2020-06-02 | 2020-09-11 | 全球能源互联网研究院有限公司 | Training method and classification method for fault classification model of power information system |
| CN112612890A (en) * | 2020-12-28 | 2021-04-06 | 交控科技股份有限公司 | Fault identification method and device for railway vehicle |
-
2021
- 2021-06-08 CN CN202110638518.8A patent/CN113515402A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107766879A (en) * | 2017-09-30 | 2018-03-06 | 中国南方电网有限责任公司 | The MLP electric network fault cause diagnosis methods of feature based information extraction |
| CN110781671A (en) * | 2019-10-29 | 2020-02-11 | 西安科技大学 | Knowledge mining method for intelligent IETM fault maintenance record text |
| CN111651601A (en) * | 2020-06-02 | 2020-09-11 | 全球能源互联网研究院有限公司 | Training method and classification method for fault classification model of power information system |
| CN112612890A (en) * | 2020-12-28 | 2021-04-06 | 交控科技股份有限公司 | Fault identification method and device for railway vehicle |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115049081A (en) * | 2022-06-18 | 2022-09-13 | 深圳市蓝晨科技股份有限公司 | Notebook computer detection management method, system and storage medium |
| CN116431378A (en) * | 2023-06-13 | 2023-07-14 | 成都实时技术股份有限公司 | Product embedded system fault data analysis method |
| CN116431378B (en) * | 2023-06-13 | 2023-08-18 | 成都实时技术股份有限公司 | Product embedded system fault data analysis method |
| CN116910478A (en) * | 2023-07-14 | 2023-10-20 | 西安科技大学 | Lithium ion battery accident tracing method and system |
| CN116910478B (en) * | 2023-07-14 | 2024-01-30 | 西安科技大学 | Lithium ion battery accident tracing method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113515402A (en) | Fault information classification method and device for engineering equipment and engineering equipment | |
| CN108764808A (en) | Data Analysis Services system and its on-time model dispositions method | |
| CN110502445B (en) | Software fault severity level judging method and device and model training method and device | |
| CN110633222A (en) | Method and device for determining regression test case | |
| CN113010389A (en) | Training method, fault prediction method, related device and equipment | |
| US20140046709A1 (en) | Methods and systems for evaluating technology assets | |
| US20140081680A1 (en) | Methods and systems for evaluating technology assets using data sets to generate evaluation outputs | |
| CN105868956A (en) | Data processing method and device | |
| CN109063433A (en) | Recognition methods, device and the readable storage medium storing program for executing of fictitious users | |
| CN111242430A (en) | Power equipment supplier evaluation method and device | |
| CN107066302A (en) | Defect inspection method, device and service terminal | |
| JP6975086B2 (en) | Quality evaluation method and quality evaluation equipment | |
| CN110765007A (en) | Crash information online analysis method for android application | |
| CN108399545B (en) | Method and device for detecting quality of electronic commerce platform | |
| CN112783762B (en) | Software quality assessment method, device and server | |
| CN113377640B (en) | Method, medium, device and computing equipment for explaining model under business scene | |
| JP5640796B2 (en) | Name identification support processing apparatus, method and program | |
| CN109165327A (en) | Interactive method, device and computer readable storage medium | |
| KR102499182B1 (en) | Loan regular auditing system using artificia intellicence | |
| CN113962216A (en) | Text processing method and device, electronic equipment and readable storage medium | |
| KR102217092B1 (en) | Method and apparatus for providing quality information of application | |
| CN111143220B (en) | Training system and method for software test | |
| Marson et al. | Quality function deployment (QFD) | |
| CN113238908A (en) | Server performance test data analysis method and related device | |
| CN111400375A (en) | Business opportunity mining method and device based on financial service data |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20211019 |